Flag of the European Union EU Clinical Trials Register Help

Clinical trials

The European Union Clinical Trials Register   allows you to search for protocol and results information on:
  • interventional clinical trials that were approved in the European Union (EU)/European Economic Area (EEA) under the Clinical Trials Directive 2001/20/EC
  • clinical trials conducted outside the EU/EEA that are linked to European paediatric-medicine development

    • EU/EEA interventional clinical trials approved under or transitioned to the Clinical Trial Regulation 536/2014 are publicly accessible through the
    Clinical Trials Information System (CTIS).

    The EU Clinical Trials Register currently displays   44334   clinical trials with a EudraCT protocol, of which   7366   are clinical trials conducted with subjects less than 18 years old.   The register also displays information on   18700   older paediatric trials (in scope of Article 45 of the Paediatric Regulation (EC) No 1901/2006).

    Phase 1 trials conducted solely on adults and that are not part of an agreed paediatric investigation plan (PIP) are not publicly available (see Frequently Asked Questions ).  
     
    Examples: Cancer AND drug name. Pneumonia AND sponsor name.
    How to search [pdf]
    Search Tips: Under advanced search you can use filters for Country, Age Group, Gender, Trial Phase, Trial Status, Date Range, Rare Diseases and Orphan Designation. For these items you should use the filters and not add them to your search terms in the text field.
    Advanced Search: Search tools
     

    < Back to search results

    Download PDF

    Clinical Trial Results:
    A Phase 3 Randomized, Double-Blind, Active-Controlled, Multicenter Study of the Long-Term Safety and Efficacy of Subcutaneous Administration of Tanezumab in Subjects With Osteoarthritis of the hip or Knee

    Summary
    EudraCT number
    		 2012-003721-22
    Trial protocol
    		 BG   SK   LT   HR  
    Global end of trial date
    27 Feb 2019

    Results information
    Results version number
    		 v1(current)
    This version publication date
    12 Mar 2020
    First version publication date
    12 Mar 2020
    Other versions

    Trial information

    		 Close Top of page
    Trial identification
    Sponsor protocol code
    A4091058
    Additional study identifiers
    ISRCTN number
    		 -
    US NCT number
    		 NCT02528188
    WHO universal trial number (UTN)
    		 -
    Sponsors
    Sponsor organisation name
    Pfizer Inc.
    Sponsor organisation address
    235 E 42nd Street, New York, United States, NY 10017
    Public contact
    Pfizer ClinicalTrials.gov Call Center, Pfizer Inc., 001 18007181021, ClinicalTrials.gov_Inquiries@pfizer.com
    Scientific contact
    Pfizer ClinicalTrials.gov Call Center, Pfizer Inc., 001 18007181021, ClinicalTrials.gov_Inquiries@pfizer.com
    Paediatric regulatory details
    Is trial part of an agreed paediatric investigation plan (PIP)
    No
    Does article 45 of REGULATION (EC) No 1901/2006 apply to this trial?
    No
    Does article 46 of REGULATION (EC) No 1901/2006 apply to this trial?
    No
    Results analysis stage
    Analysis stage
    Final
    Date of interim/final analysis
    27 Feb 2019
    Is this the analysis of the primary completion data?
    No
    Global end of trial reached?
    Yes
    Global end of trial date
    27 Feb 2019
    Was the trial ended prematurely?
    No
    General information about the trial
    Main objective of the trial
    To characterize the long-term risk of joint safety events in subjects with osteoarthritis of the knee or hip who received tanezumab 2.5 milligram (mg) or tanezumab 5 mg subcutaneously (SC) versus nonsteroidal anti-inflammatory drugs (NSAIDs) treatment (naproxen 500 mg twice in a day [BID], celecoxib 100 mg BID, or diclofenac extended release [ER] 75 mg BID) over the course of 56-weeks of treatment using a composite endpoint (includes adjudication outcomes of rapidly progressive osteoarthritis type-1 or type-2, subchondral insufficiency fracture (SPONK), primary osteonecrosis, or pathological fracture) and to demonstrate superior efficacy of tanezumab 2.5 mg and tanezumab 5 mg SC versus NSAID treatment (naproxen 500 mg BID, celecoxib 100 mg BID, or diclofenac ER 75 mg BID) at Week 16.
    Protection of trial subjects
    The study was in compliance with the ethical principles derived from the Declaration of Helsinki and in compliance with all International Council for Harmonisation (ICH) Good Clinical Practice (GCP) Guidelines. All the local regulatory requirements pertinent to safety of trial subjects were followed.
    Background therapy
    		 -
    Evidence for comparator
    		 -
    Actual start date of recruitment
    21 Jul 2015
    Long term follow-up planned
    No
    Independent data monitoring committee (IDMC) involvement?
    Yes
    Population of trial subjects
    Number of subjects enrolled per country
    Country: Number of subjects enrolled
    Australia: 49
    Country: Number of subjects enrolled
    Brazil: 67
    Country: Number of subjects enrolled
    Bulgaria: 28
    Country: Number of subjects enrolled
    Colombia: 3
    Country: Number of subjects enrolled
    Croatia: 1
    Country: Number of subjects enrolled
    Japan: 200
    Country: Number of subjects enrolled
    Korea, Republic of: 34
    Country: Number of subjects enrolled
    Lithuania: 13
    Country: Number of subjects enrolled
    New Zealand: 50
    Country: Number of subjects enrolled
    Peru: 36
    Country: Number of subjects enrolled
    Philippines: 4
    Country: Number of subjects enrolled
    Russian Federation: 28
    Country: Number of subjects enrolled
    Serbia: 45
    Country: Number of subjects enrolled
    Slovakia: 4
    Country: Number of subjects enrolled
    Taiwan: 6
    Country: Number of subjects enrolled
    Ukraine: 78
    Country: Number of subjects enrolled
    United States: 2350
    Worldwide total number of subjects
    2996
    EEA total number of subjects
    46
    Number of subjects enrolled per age group
    In utero
    0
    Preterm newborn - gestational age < 37 wk
    0
    Newborns (0-27 days)
    0
    Infants and toddlers (28 days-23 months)
    0
    Children (2-11 years)
    0
    Adolescents (12-17 years)
    0
    Adults (18-64 years)
    1986
    From 65 to 84 years
    997
    85 years and over
    13

    Subject disposition

    		 Close Top of page
    Recruitment
    Recruitment details
    		 -

    Pre-assignment
    Screening details
    		 A total of 3021 subjects were enrolled into the study. 2996 subjects received study treatment.

    Period 1
    Period 1 title
    Overall Study (overall period)
    Is this the baseline period?
    		 Yes
    Allocation method
    Randomised - controlled
    Blinding used
    		 Double blind
    Roles blinded
    		 Investigator, Subject

    Arms
    Are arms mutually exclusive
    Yes

    Arm title
    		 Tanezumab 2.5 mg
    Arm description
    		 Tanezumab (RN624 or PF-04383119) 2.5 mg injection administered SC once every 8 weeks, from Baseline (Day 1) up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, BID, from Baseline up to Week 56.
    Arm type
    		 Experimental

    Investigational medicinal product name
    Tanezumab
    Investigational medicinal product code
    RN624 or PF-04383119
    Other name
    Pharmaceutical forms
    Injection
    Routes of administration
    Subcutaneous use
    Dosage and administration details
    Subjects received 2.5 mg injection SC once every 8 weeks

    Arm title
    		 Tanezumab 5 mg
    Arm description
    		 Tanezumab (RN624 or PF-04383119) 5 mg injection administered SC once every 8 weeks, from Baseline up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, twice daily, from Baseline up to week 56.
    Arm type
    		 Experimental

    Investigational medicinal product name
    Tanezumab
    Investigational medicinal product code
    RN624 or PF-04383119
    Other name
    Pharmaceutical forms
    Injection
    Routes of administration
    Subcutaneous use
    Dosage and administration details
    Subjects received 5 mg injection SC once every 8 weeks

    Arm title
    		 NSAID
    Arm description
    		 Non-steroidal anti-inflammatory drug (NSAID) tablets (naproxen 500 mg, celecoxib 100 mg, or diclofenac ER 75 mg), administered orally, twice daily, from Baseline up to week 56 and placebo matched to tanezumab (RN624 or PF-04383119) injection administered SC once every 8 weeks, from Baseline up to Week 48.
    Arm type
    		 Experimental

    Investigational medicinal product name
    Naproxen 500 mg, Celecoxib 100 mg, or diclofenac ER 75 mg
    Investigational medicinal product code
    Other name
    Pharmaceutical forms
    Tablet
    Routes of administration
    Oral use
    Dosage and administration details
    Subjects received Naproxen 500 mg, celecoxib 100 mg, or diclofenac ER 75 mg tablet orally, twice daily.

    Number of subjects in period 1
    		Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Started
    1002
    998
    996
    Completed
    741
    729
    757
    Not completed
    261
    269
    239
         Adverse event, serious fatal
    4
    4
    -
         Consent withdrawn by subject
    97
    104
    100
         Adverse event, non-fatal
    23
    22
    8
         Unspecified
    89
    91
    74
         Lost to follow-up
    25
    21
    31
         Protocol deviation
    4
    6
    4
         Lack of efficacy
    19
    21
    22

    Baseline characteristics

    		 Close Top of page
    Baseline characteristics reporting groups
    Reporting group title
    Tanezumab 2.5 mg
    Reporting group description
    		 Tanezumab (RN624 or PF-04383119) 2.5 mg injection administered SC once every 8 weeks, from Baseline (Day 1) up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, BID, from Baseline up to Week 56.

    Reporting group title
    Tanezumab 5 mg
    Reporting group description
    		 Tanezumab (RN624 or PF-04383119) 5 mg injection administered SC once every 8 weeks, from Baseline up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, twice daily, from Baseline up to week 56.

    Reporting group title
    NSAID
    Reporting group description
    		 Non-steroidal anti-inflammatory drug (NSAID) tablets (naproxen 500 mg, celecoxib 100 mg, or diclofenac ER 75 mg), administered orally, twice daily, from Baseline up to week 56 and placebo matched to tanezumab (RN624 or PF-04383119) injection administered SC once every 8 weeks, from Baseline up to Week 48.

    Reporting group values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID Total
    Number of subjects
    		 1002 998 996 2996
    Age categorical
    Units: Subjects
        In utero
    		 0 0 0 0
        Preterm newborn infants (gestational age < 37 wks)
    		 0 0 0 0
        Newborns (0-27 days)
    		 0 0 0 0
        Infants and toddlers (28 days-23 months)
    		 0 0 0 0
        Children (2-11 years)
    		 0 0 0 0
        Adolescents (12-17 years)
    		 0 0 0 0
        Adults (18-64 years)
    		 673 637 676 1986
        From 65-84 years
    		 325 356 316 997
        85 years and over
    		 4 5 4 13
    Age Continuous
    Units: years
        arithmetic mean (standard deviation)
    		 60.30 ( 9.17 ) 61.15 ( 9.57 ) 60.25 ( 9.46 ) -
    Sex: Female, Male
    Units: Subjects
        Female
    		 637 654 662 1953
        Male
    		 365 344 334 1043
    Race (NIH/OMB)
    Units: Subjects
        American Indian or Alaska Native
    		 0 0 0 0
        Asian
    		 110 95 99 304
        Native Hawaiian or Other Pacific Islander
    		 0 0 0 0
        Black or African American
    		 166 162 186 514
        White
    		 705 712 680 2097
        More than one race
    		 0 0 0 0
        Unknown or Not Reported
    		 21 29 31 81
    Ethnicity (NIH/OMB)
    Units: Subjects
        Hispanic or Latino
    		 181 179 192 552
        Not Hispanic or Latino
    		 821 819 804 2444
        Unknown or Not Reported
    		 0 0 0 0

    End points

    		 Close Top of page
    End points reporting groups
    Reporting group title
    Tanezumab 2.5 mg
    Reporting group description
    		 Tanezumab (RN624 or PF-04383119) 2.5 mg injection administered SC once every 8 weeks, from Baseline (Day 1) up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, BID, from Baseline up to Week 56.

    Reporting group title
    Tanezumab 5 mg
    Reporting group description
    		 Tanezumab (RN624 or PF-04383119) 5 mg injection administered SC once every 8 weeks, from Baseline up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, twice daily, from Baseline up to week 56.

    Reporting group title
    NSAID
    Reporting group description
    		 Non-steroidal anti-inflammatory drug (NSAID) tablets (naproxen 500 mg, celecoxib 100 mg, or diclofenac ER 75 mg), administered orally, twice daily, from Baseline up to week 56 and placebo matched to tanezumab (RN624 or PF-04383119) injection administered SC once every 8 weeks, from Baseline up to Week 48.

    Primary: Percentage of Subjects With Adjudicated Primary Composite Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Percentage of Subjects With Adjudicated Primary Composite Joint Safety Outcome
    End point description
    Any subject with incidence of an adjudicated outcome of primary osteonecrosis, rapidly progressive osteoarthritis (OA) type 1 or type 2, subchondral insufficiency fracture, or pathological fracture. Rapidly progressive OA type 1 events were those that the Adjudication Committee considered to have significant loss of joint space width (JSW) (greater than or equal to [>=] 2 millimeters [mm]) within approximately 1 year without gross structural failure. Rapidly progressive OA type 2 events were those considered to have abnormal loss/destruction of bone including limited or total collapse of at least one subchondral surface (e.g., medial femoral condyle) that is not normally present in conventional end-stage OA. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Primary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: percentage of subjects
        number (confidence interval 95%)
    3.9 (2.8 to 5.3)
    7.1 (5.6 to 8.9)
    1.5 (0.8 to 2.5)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0123
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    2.39
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.58
         upper limit
    		 4.68
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    5.61
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 3.55
         upper limit
    		 8.14

    Primary: Observation Time-Adjusted Event Rate of Subjects With Adjudicated Primary Composite Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Observation Time-Adjusted Event Rate of Subjects With Adjudicated Primary Composite Joint Safety Outcome
    End point description
    Observation time was defined as the start day of first SC study medication until either the (i) date of completion of or withdrawal from study, if a subject did not have the event, or (ii) date of the event (earliest event within each subject in the case of multiple events). Primary joint safety outcome included subjects with adjudicated outcome of primary osteonecrosis, rapidly progressive OA type 1 or type 2, subchondral insufficiency fracture, or pathological fracture. Event rate was calculated as the number of events per 1000 subject-years at risk. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Primary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: events per 1000 subject-years
        number (confidence interval 95%)
    38.3 (28.0 to 52.5)
    71.5 (56.7 to 90.2)
    14.8 (8.9 to 24.6)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0012
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    23.5
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 9.3
         upper limit
    		 37.7
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    56.7
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 38.4
         upper limit
    		 74.9

    Primary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Week 16

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Week 16
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. The WOMAC pain subscale is a 5-item questionnaire used to assess the amount of pain experienced due to OA of index joint (knee or hip) during past 48 hours. It was calculated as the mean of scores from 5 individual questions, which may not be a whole (integer) number, scored on a numerical rating scale (NRS). Scores for each question and WOMAC Pain subscale score on NRS ranged from 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. Intent to treat (ITT) population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Primary
    End point timeframe
    Baseline, Week 16
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
        least squares mean (standard error)
    -3.22 ( 0.11 )
    -3.33 ( 0.11 )
    -3.07 ( 0.11 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. analysis of covariance (ANCOVA) model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [1]
    P-value
    		 = 0.1597
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.36
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Notes
    [1] - A graphical testing procedure was applied to maintain Type I error. Tanezumab 5 mg versus NSAID was tested first and if all primary endpoints were found significant, then the testing was continued for Tanezumab 2.5 mg versus NSAID and the key secondary endpoint (>=50% reduction from baseline in WOMAC Pain at Week 16). Primary endpoints were tested sequentially within a dose in order of WOMAC pain, WOMAC physical function, and Patient’s Global Assessment (PGA).
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [2]
    P-value
    		 = 0.0148
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.26
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.46
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Notes
    [2] - A graphical testing procedure was applied to maintain Type I error. Tanezumab 5 mg versus NSAID was tested first and if all primary endpoints were found significant, then the testing was continued for Tanezumab 2.5 mg versus NSAID and the key secondary endpoint (>=50% reduction from baseline in WOMAC Pain at Week 16). Primary endpoints were tested sequentially within a dose in order of WOMAC pain, WOMAC physical function, and PGA.

    Primary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Week 16

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Week 16
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. Physical function refers to subject's ability to move around and perform usual activities of daily living. The WOMAC physical function subscale is a 17-item questionnaire used to assess the degree of difficulty experienced due to OA in index joint (knee or hip) during past 48 hours. It was calculated as mean of the scores from 17 individual questions, which may not be a whole (integer) number, scored on a NRS. Scores for each question and WOMAC physical function subscale score on NRS ranged from 0 (no difficulty) to 10 (extreme difficulty), where higher scores indicated extreme difficulty/worse physical function. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Primary
    End point timeframe
    Baseline, Week 16
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
        least squares mean (standard error)
    -3.27 ( 0.11 )
    -3.39 ( 0.11 )
    -3.08 ( 0.11 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [3]
    P-value
    		 = 0.0691
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.4
         upper limit
    		 0.02
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Notes
    [3] - A graphical testing procedure was applied to maintain Type I error. Tanezumab 5 mg versus NSAID was tested first and if all primary endpoints were found significant, then the testing was continued for Tanezumab 2.5 mg versus NSAID and the key secondary endpoint (>=50% reduction from baseline in WOMAC Pain at Week 16). Primary endpoints were tested sequentially within a dose in order of WOMAC pain, WOMAC physical function, and PGA.
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [4]
    P-value
    		 = 0.003
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.31
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.52
         upper limit
    		 -0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Notes
    [4] - A graphical testing procedure was applied to maintain Type I error. Tanezumab 5 mg versus NSAID was tested first and if all primary endpoints were found significant, then the testing was continued for Tanezumab 2.5 mg versus NSAID and the key secondary endpoint (>=50% reduction from baseline in WOMAC Pain at Week 16). Primary endpoints were tested sequentially within a dose in order of WOMAC pain, WOMAC physical function, and PGA.

    Primary: Change from Baseline in Patient’s Global Assessment (PGA) of Osteoarthritis at Week 16

    		 Close Top of page
    End point title
    		 Change from Baseline in Patient’s Global Assessment (PGA) of Osteoarthritis at Week 16
    End point description
    PGA of OA was assessed by asking a question from subjects: “Considering all the ways your OA in your knee or hip (index joint) affects you, how are you doing today? "Subjects responded on a scale ranging from 1-5, using Interactive Response Technology (IRT), where 1=very good (no symptom and no limitation of normal activities), 2= good (mild symptoms and no limitation of normal activities), 3= fair (moderate symptoms and limitation of some normal activities), 4= poor (severe symptoms and inability to carry out most normal activities), and 5= very poor (very severe symptoms and inability to carry out all normal activities). Higher scores indicated worsening of condition. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Primary
    End point timeframe
    Baseline, Week 16
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
        least squares mean (standard error)
    -0.96 ( 0.04 )
    -0.97 ( 0.04 )
    -0.94 ( 0.04 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [5]
    P-value
    		 = 0.6332
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.09
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.04
    Notes
    [5] - A graphical testing procedure was applied to maintain Type I error. Tanezumab 5 mg versus NSAID was tested first and if all primary endpoints were found significant, then the testing was continued for Tanezumab 2.5 mg versus NSAID and the key secondary endpoint (>=50% reduction from baseline in WOMAC Pain at Week 16). Primary endpoints were tested sequentially within a dose in order of WOMAC pain, WOMAC physical function, and PGA.
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [6]
    P-value
    		 = 0.3431
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.11
         upper limit
    		 0.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.04
    Notes
    [6] - A graphical testing procedure was applied to maintain Type I error. Tanezumab 5 mg versus NSAID was tested first and if all primary endpoints were found significant, then the testing was continued for Tanezumab 2.5 mg versus NSAID and the key secondary endpoint (>=50% reduction from baseline in WOMAC Pain at Week 16). Primary endpoints were tested sequentially within a dose in order of WOMAC pain, WOMAC physical function, and PGA.

    Secondary: Percentage of Subjects With Adjudicated Secondary Composite Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Percentage of Subjects With Adjudicated Secondary Composite Joint Safety Outcome
    End point description
    Any subject with incidence of an adjudicated outcome of primary osteonecrosis, rapidly progressive OA type 2, subchondral insufficiency fracture, or pathological fracture. Rapidly progressive OA type 2 events were those considered to have abnormal loss/destruction of bone including limited or total collapse of at least one subchondral surface (e.g., medial femoral condyle) that is not normally present in conventional end-stage OA. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: percentage of subjects
        number (confidence interval 95%)
    1.0 (0.5 to 1.8)
    2.2 (1.4 to 3.3)
    0.5 (0.2 to 1.2)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4082
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    0.5
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.75
         upper limit
    		 2.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0238
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    1.7
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.31
         upper limit
    		 3.63

    Secondary: Observation Time-Adjusted Event Rate of Subjects With Adjudicated Secondary Composite Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Observation Time-Adjusted Event Rate of Subjects With Adjudicated Secondary Composite Joint Safety Outcome
    End point description
    Observation time was defined as the start day of first SC study medication until either the (i) date of completion of or withdrawal from study, if a subject did not have the event, or (ii) date of the event (earliest event within each subject in the case of multiple events). Secondary joint safety outcome included primary osteonecrosis, rapidly progressive OA (type-2), subchondral insufficiency fracture, or pathological fracture. Event rate was calculated as the number of events per 1000 subject-years at risk. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: events per 1000 subject-years
        number (confidence interval 95%)
    9.7 (5.2 to 18.1)
    21.8 (14.4 to 33.1)
    4.9 (2.1 to 11.8)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2035
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    4.8
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -2.6
         upper limit
    		 12.2
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.001
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    16.9
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 6.8
         upper limit
    		 27

    Secondary: Percentage of Subjects With Individual Adjudicated Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Percentage of Subjects With Individual Adjudicated Joint Safety Outcome
    End point description
    Any subject with incidence of an adjudicated outcome of rapidly progressive OA (type-1 only), rapidly progressive OA (type-2 only), rapidly progressive OA (type-1 or type-2 combined), subchondral insufficiency fracture, primary osteonecrosis, and pathological fracture. Rapidly progressive OA type 1 events were those that the Adjudication Committee considered to have significant loss of JSW >=2 mm within approximately 1 year without gross structural failure. Rapidly progressive OA type 2 events were those considered to have abnormal loss/destruction of bone including limited or total collapse of at least one subchondral surface (e.g., medial femoral condyle) that is not normally present in conventional end-stage OA. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: percentage of subjects
    number (confidence interval 95%)
        Rapidly Progressive OA Type 1 or 2
    3.2 (2.2 to 4.5)
    6.3 (4.9 to 8.0)
    1.2 (0.6 to 2.1)
        Rapidly Progressive OA type 1
    2.9 (1.9 to 4.1)
    4.9 (3.7 to 6.4)
    1.1 (0.6 to 2.0)
        Rapidly Progressive OA type 2
    0.3 (0.1 to 0.9)
    1.4 (0.8 to 2.3)
    0.1 (0.0 to 0.6)
        Primary Osteonecrosis
    0.1 (0.0 to 0.6)
    0.1 (0.0 to 0.6)
    0 (0.0 to 0.4)
        Pathological Fracture
    0 (0.0 to 0.4)
    0 (0.0 to 0.4)
    0 (0.0 to 0.4)
        Subchondral Insufficiency Fracture
    0.6 (0.2 to 1.3)
    0.7 (0.3 to 1.4)
    0.4 (0.1 to 1.0)
    Statistical analysis title
    		 Tanezumab 2.5 mg vs NSAID
    Statistical analysis description
    Rapidly progressive OA Type 1 or 2
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0248
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk Difference
    Point estimate
    1.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.31
         upper limit
    		 4.17
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1 or 2
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference
    Point estimate
    5.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 3.16
         upper limit
    		 7.54
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0366
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference
    Point estimate
    1.79
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.16
         upper limit
    		 3.92
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0001
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference
    Point estimate
    3.81
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.99
         upper limit
    		 6.12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 2
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6168
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference
    Point estimate
    0.2
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.76
         upper limit
    		 1.71
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 2
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0388
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference
    Point estimate
    1.3
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.17
         upper limit
    		 2.97
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Primary osteonecrosis
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7245
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference
    Point estimate
    0.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.74
         upper limit
    		 1.51
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Primary osteonecrosis
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7182
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    0.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.74
         upper limit
    		 1.52
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Subchondral insufficiency fracture
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6824
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    0.2
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.96
         upper limit
    		 1.9
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Subchondral insufficiency fracture
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5632
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    0.3
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.86
         upper limit
    		 2.03

    Secondary: Observation Time-Adjusted Event Rate of Subjects With Individual Adjudicated Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Observation Time-Adjusted Event Rate of Subjects With Individual Adjudicated Joint Safety Outcome
    End point description
    Observation time was defined as the start day of first SC study medication until either the (i) date of completion of or withdrawal from study, if a subject did not have the event, or (ii) date of the event (earliest event within each subject in the case of multiple events). Individual joint safety outcome included rapidly progressive OA (type-1 only), rapidly progressive OA (type-2 only), rapidly progressive OA (type-1 or type-2 combined), subchondral insufficiency fracture, primary osteonecrosis, and pathological fracture. Event rate was calculated as the number of events per 1000 subject-years at risk. 99999 =95% CI was not estimable since no subjects had events. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: events per 1000 subject-years
    number (confidence interval 95%)
        Rapidly Progressive OA Type 1 or 2
    31.4 (22.2 to 44.4)
    63.3 (49.5 to 81.1)
    11.9 (6.7 to 20.9)
        Rapidly Progressive OA Type 1
    28.4 (19.8 to 40.9)
    49.1 (37.1 to 65.0)
    10.9 (6.0 to 19.6)
        Rapidly Progressive OA Type 2
    2.9 (0.9 to 9.1)
    13.9 (8.2 to 23.4)
    1.0 (0.1 to 7.0)
        Primary Osteonecrosis
    1.0 (0.1 to 6.9)
    1.0 (0.1 to 7.0)
    0 (-99999 to 99999)
        Pathological Fracture
    0 (-99999 to 99999)
    0 (-99999 to 99999)
    0 (-99999 to 99999)
        Subchondral Insufficiency Fracture
    5.8 (2.6 to 13.0)
    6.9 (3.3 to 14.5)
    3.9 (1.5 to 10.5)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1 or 2
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0027
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    19.56
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 6.78
         upper limit
    		 32.35
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1 or 2
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    51.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 34.47
         upper limit
    		 68.5
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0047
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    17.58
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 5.39
         upper limit
    		 29.76
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 1
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    38.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 23.05
         upper limit
    		 53.4
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 2
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3214
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    1.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -1.89
         upper limit
    		 5.76
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Rapidly Progressive OA Type 2
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0008
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    12.88
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 5.36
         upper limit
    		 20.39
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Subchondral Insufficiency Fracture
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5394
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    1.9
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -4.17
         upper limit
    		 7.96
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Subchondral Insufficiency Fracture
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3636
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    2.98
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -3.44
         upper limit
    		 9.39

    Secondary: Percentage of Subjects With Total Joint Replacement or Adjudicated Primary Composite Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Percentage of Subjects With Total Joint Replacement or Adjudicated Primary Composite Joint Safety Outcome
    End point description
    Percentage of subjects with total joint replacement (hip or knee) or adjudicated primary composite joint safety outcomes were reported. Adjudicated primary composite joint safety outcomes included primary osteonecrosis, rapidly progressive OA type 1 or type 2, subchondral insufficiency fracture, or pathological fracture. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: percentage of subject
        number (confidence interval 95%)
    8.6 (6.9 to 10.5)
    13.1 (11.1 to 15.4)
    3.7 (2.6 to 5.1)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    The event of adjudicated primary osteonecrosis in the tanezumab 2.5 mg treatment group is not included in this analysis. Conclusions for this analysis do not change as the comparison to NSAID is already statistically significant in favor of NSAID.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0002
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    4.87
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 2.43
         upper limit
    		 7.74
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Exact methods for risk difference
    Parameter type
    		 Risk difference (RD)
    Point estimate
    9.41
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 6.73
         upper limit
    		 12.52

    Secondary: Observation Time-Adjusted Event Rate of Subjects With Total Joint Replacement or Adjudicated Primary Composite Joint Safety Outcome

    		 Close Top of page
    End point title
    		 Observation Time-Adjusted Event Rate of Subjects With Total Joint Replacement or Adjudicated Primary Composite Joint Safety Outcome
    End point description
    Observation time was defined as the start day of first SC study medication until either the (i) date of completion of or withdrawal from study, if a subject did not have the event, or (ii) date of the event (earliest event within each subject in the case of multiple events). Adjudicated primary composite joint safety outcomes included primary osteonecrosis, rapidly progressive OA type 1 or type 2, subchondral insufficiency fracture, or pathological fracture. Event rate was calculated as the number of events per 1000 subject-years at risk. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: events per 1000 subject-years
        number (confidence interval 95%)
    84.9 (68.7 to 104.9)
    132.5 (111.7 to 157.3)
    36.7 (26.6 to 50.6)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    The event of adjudicated primary osteonecrosis in the tanezumab 2.5 mg treatment group is not included in this analysis. Conclusions for this analysis do not change as the comparison to NSAID is already statistically significant in favor of NSAID.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    48.25
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 26.76
         upper limit
    		 69.74
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Poisson model for rate difference
    Parameter type
    		 Rate Difference
    Point estimate
    95.83
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 70.25
         upper limit
    		 121.42

    Secondary: Change From Baseline in Medial or Lateral Joint Space Width of the Index Knee (Kellgren-Lawrence Grade [KLG] 2 or 3) at Weeks 56 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Medial or Lateral Joint Space Width of the Index Knee (Kellgren-Lawrence Grade [KLG] 2 or 3) at Weeks 56 and 80
    End point description
    Change from baseline in JSW was defined as change in JSW compared to baseline in subjects with KLG 2 or 3 over the course of the study. It was measured radiographically in the medial and lateral tibiofemoral of knee in subjects with OA. KLG system was a method of classifying the severity of knee OA using five grades i.e. 0 [no radiographic features of OA], 1 [doubtful joint space narrowing (JSN) and possible osteophytic lipping], 2 [definite osteophytes and possible JSN on anteroposterior weight-bearing radiograph], 3 [multiple osteophytes, definite JSN, sclerosis, possible bony deformity], 4 [large osteophytes, marked JSN, severe sclerosis and definite bony deformity]. Higher grade indicating worse knee function. The number of subjects with progression of OA in the index knee are summarized separately by the compartment of OA at baseline (medial or lateral). Safety population was analyzed. ‘n’ =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    651
    668
    695
    Units: millimeter (mm)
    least squares mean (standard error)
        Change in Medial JSW at Week 56 (n= 486,506,523)
    -0.25 ( 0.03 )
    -0.34 ( 0.03 )
    -0.19 ( 0.03 )
        Change in Medial JSW at Week 80 (n =402,413,432)
    -0.33 ( 0.04 )
    -0.37 ( 0.04 )
    -0.25 ( 0.03 )
        Change in Lateral JSW at Week 56 (n =110,94,114)
    -0.26 ( 0.07 )
    -0.32 ( 0.07 )
    -0.27 ( 0.07 )
        Change in Lateral JSW at Week 80 (n= 88,75,98)
    -0.46 ( 0.08 )
    -0.32 ( 0.09 )
    -0.37 ( 0.08 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change in medial JSW at Week 56: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0979
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.15
         upper limit
    		 0.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.04
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change in medial JSW at Week 56: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.24
         upper limit
    		 -0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.04
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change in medial JSW at Week 80: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1162
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.17
         upper limit
    		 0.02
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change in medial JSW at Week 80: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0128
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.22
         upper limit
    		 -0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change in lateral JSW at Week 56: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8885
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.17
         upper limit
    		 0.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change in lateral JSW at Week 56: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6345
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.24
         upper limit
    		 0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change in lateral JSW at Week 80: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4406
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.09
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.32
         upper limit
    		 0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.12
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change in lateral JSW at Week 80: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7109
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.19
         upper limit
    		 0.28
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.12

    Secondary: Change From Baseline in Joint Space Width of the Index hip (Kellgren-Lawrence Grade 2 or 3) at Weeks 56 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Joint Space Width of the Index hip (Kellgren-Lawrence Grade 2 or 3) at Weeks 56 and 80
    End point description
    Change from baseline in JSW was defined as narrowing in JSW compared to baseline in subjects with KLG 2 or 3 over the course of the study. It was measured radiographically in the index hip in subjects with OA. KLG system was a method of classifying the severity of hip OA using five grades i.e.0 [no radiographic features of OA], 1 [doubtful JSN) and possible osteophytic lipping], 2 [definite osteophytes and possible JSN on anteroposterior weight-bearing radiograph], 3 [multiple osteophytes, definite JSN, sclerosis, possible bony deformity], 4 [large osteophytes, marked JSN, severe sclerosis and definite bony deformity]. Higher grade indicating worse hip function. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Number of subjects analysed’ =subjects who were evaluable for this endpoint. ‘n’=subjects who were evaluable at specified time.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    123
    132
    120
    Units: mm
    least squares mean (standard error)
        Change at Week 56 (n =110, 112, 107)
    -0.35 ( 0.06 )
    -0.40 ( 0.06 )
    -0.21 ( 0.06 )
        Change at Week 80 (n =89, 89, 86)
    -0.46 ( 0.07 )
    -0.35 ( 0.07 )
    -0.28 ( 0.07 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    243
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.102
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.3
         upper limit
    		 0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    252
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.023
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.35
         upper limit
    		 -0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 80: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    243
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0645
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 0.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 80: ANCOVA model included treatment, baseline JSW as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    252
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5005
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.26
         upper limit
    		 0.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1

    Secondary: Number of Subjects With Progression of Osteoarthritis in the Index Knee (Kellgren-Lawrence Grade 2 or 3) According to Bland and Altman Method at Weeks 56 and 80

    		 Close Top of page
    End point title
    		 Number of Subjects With Progression of Osteoarthritis in the Index Knee (Kellgren-Lawrence Grade 2 or 3) According to Bland and Altman Method at Weeks 56 and 80
    End point description
    Progression of OA according to Bland-Altman as defined by a decrease JSW >=1.96 times within-subject standard deviation of change in JSW. The number of subjects with progression of OA in the index knee are summarized separately by the compartment of OA at baseline (medial or lateral). Kellgren-Lawrence grade system was a method of classifying the severity of knee OA using five grades i.e. 0 [no radiographic features of OA], 1 [doubtful joint space narrowing (JSN) and possible osteophytic lipping], 2 [definite osteophytes and possible JSN on anteroposterior weight-bearing radiograph], 3 [multiple osteophytes, definite JSN, sclerosis, possible bony deformity], 4 [large osteophytes, marked JSN, severe sclerosis and definite bony deformity]. Higher grade indicating worse knee function. Safety population was analyzed. ‘Number of subjects analysed’ =subjects who were evaluable for this endpoint. ‘n’=subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    651
    668
    695
    Units: subjects
        Decreased medial JSW at Week 56 (n=486,506,523)
    33
    43
    20
        Decreased medial JSW at Week 80 (n=402,413,432)
    29
    38
    16
        Decreased lateral JSW at Week 56 (n=110,94,114)
    5
    8
    9
        Decreased lateral JSW at Week 80 (n=88,75,98)
    9
    4
    7
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Decrease in medial JSW at Week 56: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0358
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.85
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.04
         upper limit
    		 3.29
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Decrease in medial JSW at Week 56: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0021
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    2.37
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.37
         upper limit
    		 4.12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Decrease in medial JSW at Week 80: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0301
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    2.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.07
         upper limit
    		 3.77
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Decrease in medial JSW at Week 80: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0016
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    2.65
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.45
         upper limit
    		 4.85
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Decrease in lateral JSW at Week 56: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3002
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.55
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.18
         upper limit
    		 1.7
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Decrease in lateral JSW at Week 56: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8997
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.39
         upper limit
    		 2.89
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Decrease in lateral JSW at Week 80: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1346
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4559
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.53
         upper limit
    		 4.18
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Decrease in lateral JSW at Week 80: Logistic regression model included treatment, and baseline JSW as covariate.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1363
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5996
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.71
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.2
         upper limit
    		 2.54

    Secondary: Number of Subjects With Progression of Osteoarthritis in the Index Hip (Kellgren-Lawrence Grade 2 or 3) According to Bland and Altman Method at Weeks 56 and 80

    		 Close Top of page
    End point title
    		 Number of Subjects With Progression of Osteoarthritis in the Index Hip (Kellgren-Lawrence Grade 2 or 3) According to Bland and Altman Method at Weeks 56 and 80
    End point description
    Progression of OA according to Bland-Altman methodology as defined by a decrease in JSW >=1.96 times within-subject standard deviation of the change in JSW in the index hip. The number of subjects with progression of OA in the index hip per Bland-Altman methodology are reported. Kellgren-Lawrence grade system was a method of classifying the severity of hip OA using five grades i.e. 0 (no radiographic features of OA), 1 (doubtful JSN and possible osteophytic lipping), 2 (definite osteophytes and possible JSN on anteroposterior weight-bearing radiograph), 3 (multiple osteophytes, definite JSN, sclerosis, possible bony deformity), 4 (large osteophytes, marked JSN, severe sclerosis and definite bony deformity). Higher grade indicating worse hip function. Safety population was analyzed. ‘Number of subjects analysed’ =subjects who were evaluable for this endpoint. ‘n’=subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    123
    132
    120
    Units: subjects
        Week 56 (n = 110, 112, 107)
    10
    10
    3
        Week 80 (n = 89, 89, 86)
    9
    9
    3
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Decrease at week 56: Logistic regression model included treatment, and baseline JSW as covariate
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    243
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0714
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    3.37
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 12.65
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Decrease at week 56: Logistic regression model included treatment, and baseline JSW as covariate
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    252
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0681
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    3.42
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.91
         upper limit
    		 12.84
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Decrease at week 80: Logistic regression model included treatment, and baseline JSW as covariate
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    243
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0967
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    3.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 11.95
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Decrease at week 80: Logistic regression model included treatment, and baseline JSW as covariate
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    252
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0976
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    3.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 11.9

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Weeks 2, 4, 8, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Weeks 2, 4, 8, 24, 32, 40, 48 and 56
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. The WOMAC pain subscale is a 5-item questionnaire used to assess the amount of pain experienced due to OA of index joint (knee or hip) during past 48 hours. It was calculated as the mean of scores from 5 individual questions scored on a NRS, which may not be a whole (integer) number. Scores for each question and WOMAC Pain subscale score on NRS ranged from 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on scale
    least squares mean (standard error)
        Change at Week 2
    -1.65 ( 0.08 )
    -1.49 ( 0.08 )
    -1.55 ( 0.08 )
        Change at Week 4
    -2.25 ( 0.09 )
    -2.29 ( 0.09 )
    -1.98 ( 0.09 )
        Change at Week 8
    -2.41 ( 0.10 )
    -2.65 ( 0.10 )
    -2.27 ( 0.10 )
        Change at Week 24
    -2.73 ( 0.13 )
    -2.86 ( 0.13 )
    -2.67 ( 0.13 )
        Change at Week 32
    -2.64 ( 0.13 )
    -2.68 ( 0.13 )
    -2.57 ( 0.13 )
        Change at Week 40
    -2.56 ( 0.13 )
    -2.57 ( 0.13 )
    -2.52 ( 0.13 )
        Change at Week 48
    -2.54 ( 0.13 )
    -2.48 ( 0.13 )
    -2.47 ( 0.13 )
        Change at Week 56
    -2.44 ( 0.13 )
    -2.37 ( 0.13 )
    -2.42 ( 0.14 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2212
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.27
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4557
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.23
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0029
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.27
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.45
         upper limit
    		 -0.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0005
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.32
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.5
         upper limit
    		 -0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1273
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.38
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.57
         upper limit
    		 -0.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6349
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.19
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1339
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.44
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6237
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4224
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 0.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7526
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.22
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7328
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.22
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5888
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.19
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9345
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.28
         upper limit
    		 0.26
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8782
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.29
         upper limit
    		 0.25
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline WOMAC pain subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7076
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.22
         upper limit
    		 0.32
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Week 64
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. The WOMAC pain subscale is a 5-item questionnaire used to assess the amount of pain experienced due to OA of index joint (knee or hip) during past 48 hours. It was calculated as the mean of scores from 5 individual questions scored on a NRS, which may not be a whole (integer) number. Scores for each question and WOMAC Pain subscale score on NRS ranged from 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    7.01 ( 1.12 )
    7.02 ( 1.12 )
    6.96 ( 1.08 )
        Change at Week 64 (n =437, 419, 445)
    -3.47 ( 2.45 )
    -3.12 ( 2.40 )
    -3.85 ( 2.07 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Weeks 2, 4, 8, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Weeks 2, 4, 8, 24, 32, 40, 48 and 56
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. Physical function refers to subject’s ability to move around and perform usual activities of daily living. The WOMAC physical function subscale is a 17-item questionnaire used to assess the degree of difficulty experienced due to OA in index joint (knee or hip) during past 48 hours. It was calculated as mean of the scores from 17 individual questions, which may not be a whole (integer) number, scored on a NRS. Scores for each question and WOMAC physical function subscale score on NRS ranged from 0 (no difficulty) to 10 (extreme difficulty), where higher scores indicated extreme difficulty/worse physical function. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 2
    -1.76 ( 0.08 )
    -1.64 ( 0.08 )
    -1.55 ( 0.08 )
        Change at Week 4
    -2.29 ( 0.09 )
    -2.31 ( 0.09 )
    -1.96 ( 0.09 )
        Change at Week 8
    -2.46 ( 0.10 )
    -2.69 ( 0.10 )
    -2.27 ( 0.10 )
        Change at Week 24
    -2.78 ( 0.13 )
    -2.88 ( 0.13 )
    -2.66 ( 0.13 )
        Change at Week 32
    -2.66 ( 0.13 )
    -2.67 ( 0.13 )
    -2.55 ( 0.13 )
        Change at Week 40
    -2.56 ( 0.13 )
    -2.57 ( 0.13 )
    -2.50 ( 0.13 )
        Change at Week 48
    -2.56 ( 0.14 )
    -2.49 ( 0.13 )
    -2.45 ( 0.13 )
        Change at Week 56
    -2.45 ( 0.14 )
    -2.36 ( 0.13 )
    -2.41 ( 0.14 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.015
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 -0.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3286
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.25
         upper limit
    		 0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0004
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.32
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.5
         upper limit
    		 -0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.35
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.53
         upper limit
    		 -0.17
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0517
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 0
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.42
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.61
         upper limit
    		 -0.23
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3621
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 0.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0832
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.47
         upper limit
    		 0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4072
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.38
         upper limit
    		 0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3404
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.13
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.39
         upper limit
    		 0.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6344
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.34
         upper limit
    		 0.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5756
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.35
         upper limit
    		 0.19
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4394
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.38
         upper limit
    		 0.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7747
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.23
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7305
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.32
         upper limit
    		 0.22
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint and highest KLG and NSAID) as fixed effects, baseline WOMAC physical function subscale and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.733
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.22
         upper limit
    		 0.32
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Week 64
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. Physical function refers to subject’s ability to move around and perform usual activities of daily living. The WOMAC physical function subscale is a 17-item questionnaire used to assess the degree of difficulty experienced due to OA in index joint (knee or hip) during past 48 hours. It was calculated as mean of the scores from 17 individual questions, which may not be a whole (integer) number, scored on NRS. Scores for each question and WOMAC physical function subscale score on NRS ranged from 0 (no difficulty) to 10 (extreme difficulty), higher scores indicated extreme difficulty/worse physical function. ITT population was analyzed. “n” =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000,995,994)
    7.09 ( 1.07 )
    7.08 ( 1.11 )
    6.99 ( 1.09 )
        Change at Week 64 (n =437, 419, 445)
    -3.42 ( 2.40 )
    -3.12 ( 2.41 )
    -3.81 ( 2.12 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Patient’s Global Assessment (PGA) of Osteoarthritis at Weeks 2, 4, 8, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Patient’s Global Assessment (PGA) of Osteoarthritis at Weeks 2, 4, 8, 24, 32, 40, 48 and 56
    End point description
    PGA of OA was assessed by asking a question from subjects: “Considering all the ways your OA in your knee or hip (index joint) affects you, how are you doing today?" Subjects responded on a scale ranging from 1-5, using IRT, where 1=very good (no symptom and no limitation of normal activities), 2= good (mild symptoms and no limitation of normal activities), 3= fair (moderate symptoms and limitation of some normal activities), 4= poor (severe symptoms and inability to carry out most normal activities), and 5= very poor (very severe symptoms and inability to carry out all normal activities). Higher scores indicated worsening of condition. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 2
    -0.67 ( 0.03 )
    -0.67 ( 0.03 )
    -0.63 ( 0.03 )
        Change at Week 4
    -0.81 ( 0.03 )
    -0.84 ( 0.03 )
    -0.69 ( 0.03 )
        Change at Week 8
    -0.77 ( 0.03 )
    -0.85 ( 0.03 )
    -0.76 ( 0.03 )
        Change at Week 24
    -0.74 ( 0.05 )
    -0.79 ( 0.05 )
    -0.74 ( 0.05 )
        Change at Week 32
    -0.72 ( 0.05 )
    -0.71 ( 0.05 )
    -0.72 ( 0.05 )
        Change at Week 40
    -0.70 ( 0.05 )
    -0.69 ( 0.05 )
    -0.69 ( 0.05 )
        Change at Week 48
    -0.70 ( 0.05 )
    -0.66 ( 0.05 )
    -0.67 ( 0.05 )
        Change at Week 56
    -0.65 ( 0.05 )
    -0.60 ( 0.05 )
    -0.66 ( 0.05 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2159
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.02
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.03
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2049
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.02
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.03
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0002
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.19
         upper limit
    		 -0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.03
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.21
         upper limit
    		 -0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.03
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7799
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.08
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.03
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0061
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.09
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.16
         upper limit
    		 -0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.03
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9718
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.1
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 24: MMultiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3292
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.14
         upper limit
    		 0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.983
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.1
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9137
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.09
         upper limit
    		 0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8784
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.11
         upper limit
    		 0.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9995
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.1
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6648
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.13
         upper limit
    		 0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.728
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.09
         upper limit
    		 0.12
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8856
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.09
         upper limit
    		 0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Change at Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline PGA and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2814
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.05
         upper limit
    		 0.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.05

    Secondary: Change From Baseline in Patient’s Global Assessment (PGA) of Osteoarthritis at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Patient’s Global Assessment (PGA) of Osteoarthritis at Week 64
    End point description
    PGA of OA was assessed by asking a question from subjects: “Considering all the ways your OA in your knee or hip (index joint) affects you, how are you doing today?" Subjects responded on a scale ranging from 1-5, using IRT, where 1=very good (no symptom and no limitation of normal activities), 2= good (mild symptoms and no limitation of normal activities), 3= fair (moderate symptoms and limitation of some normal activities), 4= poor (severe symptoms and inability to carry out most normal activities), and 5= very poor (very severe symptoms and inability to carry out all normal activities). Higher scores indicated worsening of condition. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    3.49 ( 0.61 )
    3.46 ( 0.60 )
    3.44 ( 0.59 )
        Change at Week 64 (n =437, 419, 445)
    -0.79 ( 0.96 )
    -0.64 ( 0.98 )
    -0.95 ( 0.96 )
    No statistical analyses for this end point

    Secondary: Percentage of Subjects Meeting Outcome Measures in Arthritis Clinical Trials-Osteoarthritis Research Society International (OMERACT-OARSI) Responder Index at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64

    		 Close Top of page
    End point title
    		 Percentage of Subjects Meeting Outcome Measures in Arthritis Clinical Trials-Osteoarthritis Research Society International (OMERACT-OARSI) Responder Index at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point description
    OMERACT-OARSI responders=if the change (improvement) from baseline to week of interest was >= 50% and >=2 units in either WOMAC pain subscale or physical function subscale score; if change (improvement) from baseline to week of interest was >=20% and >=1 unit in at least 2 of the following: 1) WOMAC pain subscale score, 2) WOMAC physical function subscale score, 3) PGA of OA. WOMAC pain subscale assess amount of pain experienced (score: 0 [no pain] to 10 [extreme pain], higher score = more pain), WOMAC physical function subscale assess degree of difficulty experienced (score: 0 [no difficulty] to 10 [extreme difficulty], higher score = worse physical function) and PGA of OA (score: 1 [very good] to 5 [very poor], higher score =worse condition). Missing data was imputed using mixed baseline/last observation carried forward (BOCF/LOCF). ITT population was analyzed. Overall number of subjects analyzed=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points.
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: percentage of subjects
    number (not applicable)
        Week 2 (n =1001, 996, 996)
    46.7
    43.7
    44.8
        Week 4 (n =1001, 996, 996)
    62.6
    62.7
    56.4
        Week 8 (n =1001, 996, 996)
    67.5
    70.3
    64.4
        Week 16 (n =1001, 996, 996)
    78.2
    78.3
    75.1
        Week 24 (n =1001, 996, 995)
    62.4
    64.8
    61.3
        Week 32 (n =1001, 996, 995)
    59.2
    59.9
    58.6
        Week 40 (n =1001, 996, 995)
    58.4
    58.7
    58.2
        Week 48 (n =1001, 996, 995)
    57.4
    56.2
    57.3
        Week 56 (n =1001, 996, 995)
    56.5
    54.5
    56.0
        Week 64 (n =437, 420, 446)
    79.2
    75.2
    86.5
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4691
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5451
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.95
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0059
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.29
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.54
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0057
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.29
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.55
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1584
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.95
         upper limit
    		 1.38
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.006
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.3
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.58
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1117
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.96
         upper limit
    		 1.46
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1004
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.97
         upper limit
    		 1.47
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6258
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.25
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1154
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.96
         upper limit
    		 1.39
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8018
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5697
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9557
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.84
         upper limit
    		 1.2
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8553
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9901
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.84
         upper limit
    		 1.2
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.587
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.95
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8302
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4823
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.12

    Secondary: Percentage of Subjects Achieving Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Reduction >=30 Percent (%), >=50%, >=70% and >=90% Response at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64

    		 Close Top of page
    End point title
    		 Percentage of Subjects Achieving Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Reduction >=30 Percent (%), >=50%, >=70% and >=90% Response at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point description
    Percentage of subjects with reduction in WOMAC pain intensity of at least 30%,50%,70% and 90% at Weeks 2, 4, 8,16,24,32,40,48,56 and 64 compared to baseline were classified as responders to WOMAC pain subscale and are reported here. WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. The WOMAC pain subscale is a 5-item questionnaire used to assess the amount of pain experienced due to OA of index joint (knee or hip) during past 48 hours. It was calculated as the mean of scores from 5 individual questions scored on a NRS. Scores for each question and WOMAC Pain subscale score on NRS ranged from 0 (no pain) to 10 (extreme pain), higher scores indicated higher pain. Missing data was imputed using mixed BOCF/LOCF. ITT population was analysed. ‘n’=subjects evaluable at specified time points.
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: percentage of subjects
    number (not applicable)
        Week 2: At least 30% reduction (n =1002,998,996)
    34.8
    30.5
    32.4
        Week 2: At least 50% reduction (n =1002,998,996)
    17.8
    16.5
    14.7
        Week 2: At least 70% reduction (n =1002,998,996)
    7.7
    7.1
    6.2
        Week 2: At least 90% reduction (n =1002,998,996)
    2.4
    2.5
    1.8
        Week 4: At least 30% reduction (n =1002,998,996)
    50.2
    49.5
    44.4
        Week 4: At least 50% reduction (n =1002,998,996)
    30.4
    30.5
    24.9
        Week 4: At least 70% reduction (n =1002,998,996)
    14.5
    16.4
    11.9
        Week 4: At least 90% reduction (n =1002,998,996)
    4.3
    4.9
    3.1
        Week 8: At least 30% reduction (n =1002,998,996)
    55.9
    59.0
    54.1
        Week 8: At least 50% reduction (n =1002,998,996)
    36.8
    39.3
    32.6
        Week 8: At least 70% reduction (n =1002,998,996)
    19.3
    22.4
    15.9
        Week 8: At least 90% reduction (n =1002,998,996)
    4.7
    6.6
    4.2
        Week 16: At least 30% reduction (n =1002,998,996)
    71.8
    72.9
    68.9
        Week 16: At least 50% reduction (n =1002,998,996)
    54.9
    56.5
    51.5
        Week 16: At least 70% reduction (n =1002,998,996)
    28.9
    35.0
    28.8
        Week 16: At least 90% reduction (n =1002,998,996)
    10.3
    12.7
    8.5
        Week 24: At least 30% reduction (n =1002,998,996)
    59.4
    61.1
    59.4
        Week 24: At least 50% reduction (n =1002,998,996)
    49.3
    49.4
    47.5
        Week 24: At least 70% reduction (n =1002,998,996)
    30.8
    33.8
    29.0
        Week 24: At least 90% reduction (n =1002,998,996)
    10.3
    13.3
    11.5
        Week 32: At least 30% reduction (n =1002,998,996)
    56.8
    55.7
    56.3
        Week 32: At least 50% reduction (n =1002,998,996)
    47.4
    45.8
    46.3
        Week 32: At least 70% reduction (n =1002,998,996)
    31.2
    31.5
    27.4
        Week 32: At least 90% reduction (n =1002,998,996)
    10.3
    12.9
    10.0
        Week 40: At least 30% reduction (n =1002,998,996)
    55.7
    54.6
    54.8
        Week 40: At least 50% reduction (n =1002,998,996)
    47.2
    45.2
    46.0
        Week 40: At least 70% reduction (n =1002,998,996)
    30.0
    30.4
    29.3
        Week 40: At least 90% reduction (n =1002,998,996)
    10.8
    12.0
    10.4
        Week 48: At least 30% reduction (n =1002,998,996)
    54.6
    52.9
    54.2
        Week 48: At least 50% reduction (n =1002,998,996)
    46.2
    43.2
    44.4
        Week 48: At least 70% reduction (n =1002,998,996)
    29.6
    29.4
    28.5
        Week 48: At least 90% reduction (n =1002,998,996)
    10.3
    11.4
    10.6
        Week 56: At least 30% reduction (n =1002,998,996)
    53.1
    51.2
    52.7
        Week 56: At least 50% reduction (n =1002,998,996)
    44.3
    41.5
    43.5
        Week 56: At least 70% reduction (n =1002,998,996)
    28.2
    27.0
    27.5
        Week 56: At least 90% reduction (n =1002,998,996)
    10.1
    10.5
    10.1
        Week 64: At least 30% reduction (n =437,419,445)
    73.0
    69.0
    81.3
        Week 64: At least 50% reduction (n =437,419,445)
    55.4
    47.3
    60.2
        Week 64: At least 70% reduction (n =437,419,445)
    31.1
    24.3
    34.2
        Week 64: At least 90% reduction (n =437,419,445)
    9.6
    7.9
    12.6
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2938
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.92
         upper limit
    		 1.33
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=30%: reduction OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3146
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.91
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.75
         upper limit
    		 1.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    P-value
    		 = 0.0748
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.98
         upper limit
    		 1.58
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.45
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.255
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.23
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.74
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.478
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.62
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4006
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.3
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.7
         upper limit
    		 2.42
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3089
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.38
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.74
         upper limit
    		 2.54
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0114
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.26
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.05
         upper limit
    		 1.5
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0239
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.23
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.03
         upper limit
    		 1.47
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0079
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.31
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.07
         upper limit
    		 1.6
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0068
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.32
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.6
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1037
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.96
         upper limit
    		 1.62
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0046
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.45
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.12
         upper limit
    		 1.88
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1971
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.36
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 2.19
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.048
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.59
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1
         upper limit
    		 2.52
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4744
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0336
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.02
         upper limit
    		 1.45
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0559
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.2
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1
         upper limit
    		 1.44
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0021
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.34
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.11
         upper limit
    		 1.61
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0535
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.26
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1
         upper limit
    		 1.59
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0003
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.53
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.22
         upper limit
    		 1.92
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6421
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.72
         upper limit
    		 1.7
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0207
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.6
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.07
         upper limit
    		 2.38
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1635
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.95
         upper limit
    		 1.39
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0529
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1
         upper limit
    		 1.47
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [7]
    P-value
    		 = 0.1322
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.96
         upper limit
    		 1.37
    Notes
    [7] - The two key secondary comparisons for 'Subjects with >=50% reduction from baseline in WOMAC Pain at Week 16' (tanezumab 2.5 mg treatment group versus NSAID and tanezumab 5 mg treatment group versus NSAID) could not be considered significant since preceding tests in the graphical testing procedure were not significant.
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority [8]
    P-value
    		 = 0.0262
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.02
         upper limit
    		 1.46
    Notes
    [8] - The two key secondary comparisons for 'Subjects with >=50% reduction from baseline in WOMAC Pain at Week 16' (tanezumab 2.5 mg treatment group versus NSAID and tanezumab 5 mg treatment group versus NSAID) could not be considered significant since preceding tests in the graphical testing procedure were not significant.
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9805
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0033
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.33
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.1
         upper limit
    		 1.61
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.159
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.92
         upper limit
    		 1.69
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0024
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.57
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.17
         upper limit
    		 2.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9932
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.2
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4374
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.29
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4406
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4078
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.29
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4071
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.31
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0248
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.03
         upper limit
    		 1.5
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3641
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.88
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.66
         upper limit
    		 1.16
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2497
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.17
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.53
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8682
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.21
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7317
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.97
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 1.16
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6493
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.24
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7973
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.98
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.82
         upper limit
    		 1.17
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0757
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.98
         upper limit
    		 1.45
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0578
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.99
         upper limit
    		 1.47
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.901
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.76
         upper limit
    		 1.37
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0548
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.32
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.99
         upper limit
    		 1.74
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7331
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.23
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8632
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.98
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.82
         upper limit
    		 1.18
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6262
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.25
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6817
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 1.15
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.799
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.24
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6786
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8069
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.78
         upper limit
    		 1.38
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2951
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.54
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9093
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.21
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5032
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4382
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5638
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.95
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6436
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.27
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.706
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7729
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.72
         upper limit
    		 1.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6405
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 1.42
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9046
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.21
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=30% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4491
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.93
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.78
         upper limit
    		 1.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7429
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.23
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=50% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3467
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.92
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.77
         upper limit
    		 1.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7624
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=70% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7686
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.97
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.18
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9384
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.74
         upper limit
    		 1.33
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=90% reduction: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8495
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.77
         upper limit
    		 1.38

    Secondary: Percentage of Subjects With Cumulative Percent Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Weeks 16, 24 and 56

    		 Close Top of page
    End point title
    		 Percentage of Subjects With Cumulative Percent Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale at Weeks 16, 24 and 56
    End point description
    WOMAC: Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. The WOMAC pain subscale is a 5-item questionnaire used to assess the amount of pain experienced due to OA of index joint during past 48 hours. It was calculated as the mean of scores from 5 individual questions scored on a NRS. Scores for each question and WOMAC Pain subscale score on NRS ranged from 0 (no pain) to 10 (extreme pain), higher scores indicated higher pain. Percentage of subjects with cumulative reduction (as %) (>0%; >= 10, 20,30,40,50,60,70,80 and 90%; = 100%) in WOMAC pain subscale from Baseline to Weeks 16,24,56 were reported, subjects (%) are reported more than once in categories specified. Missing data was imputed using mixed BOCF/LOCF. ITT population was analysed. ‘Number of subjects analysed’=subjects evaluable for this endpoint.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16, 24 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    995
    994
    Units: percentage of subjects
    number (not applicable)
        Week 16: >0%
    89.5
    87.6
    87.1
        Week 16: >=10%
    85.0
    82.8
    82.8
        Week 16: >=20%
    78.1
    78.3
    75.8
        Week 16: >=30%
    71.8
    72.9
    68.9
        Week 16: >=40%
    63.7
    63.5
    59.9
        Week 16: >=50%
    54.9
    56.5
    51.5
        Week 16: >=60%
    40.9
    44.8
    38.8
        Week 16: >=70%
    28.9
    35.0
    28.8
        Week 16: >=80%
    19.4
    23.9
    18.8
        Week 16: >=90%
    10.3
    12.7
    8.5
        Week 16: =100%
    4.4
    3.9
    3.3
        Week 24: >=0%
    66.7
    68.2
    64.8
        Week 24: >=10%
    64.9
    66.4
    63.4
        Week 24: >=20%
    62.2
    65.2
    62.1
        Week 24: >=30%
    59.4
    61.1
    59.4
        Week 24: >=40%
    55.2
    55.7
    54.7
        Week 24: >=50%
    49.3
    49.4
    47.5
        Week 24: >=60%
    40.7
    41.2
    38.1
        Week 24: >=70%
    30.8
    33.8
    29.0
        Week 24: >=80%
    20.6
    24.0
    20.2
        Week 24: >=90%
    10.3
    13.3
    11.5
        Week 24: =100%
    3.9
    4.5
    3.4
        Week 56: >=0%
    60.8
    59.1
    59.7
        Week 56: >=10%
    59.1
    57.0
    58.1
        Week 56: >=20%
    55.9
    54.8
    56.3
        Week 56: >=30%
    53.1
    51.2
    52.7
        Week 56: >=40%
    48.6
    46.8
    48.6
        Week 56: >=50%
    44.3
    41.5
    43.5
        Week 56: >=60%
    37.0
    33.8
    36.3
        Week 56: >=70%
    28.2
    27.0
    27.5
        Week 56: >=80%
    18.9
    19.1
    18.6
        Week 56: >=90%
    10.1
    10.5
    10.1
        Week 56: =100%
    4.5
    5.3
    4.1
    No statistical analyses for this end point

    Secondary: Percentage of Subjects Achieving Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale Reduction of >=30%, >=50%, >=70% and >=90% Response at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64

    		 Close Top of page
    End point title
    		 Percentage of Subjects Achieving Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale Reduction of >=30%, >=50%, >=70% and >=90% Response at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point description
    Percentage of subjects with reduction in WOMAC physical function compared to baseline were classified as responders. WOMAC: Self-administered,disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain,stiffness and physical function in subjects with OA.Physical function:Subject’s ability to move around and perform usual activities of daily living. WOMAC physical function subscale17-item questionnaire used to assess the degree of difficulty experienced due to OA in knee/hip during past 48 hours,calculated as mean of the scores from 17 individual questions scored on NRS. Scores for each question and WOMAC physical subscale on NRS ranged from 0 (no difficulty) to 10 (extreme difficulty), higher scores indicated extreme difficulty/worse physical function.Missing data was imputed using mixed BOCF/LOCF. ITT population was analysed. ‘Number of subjects analysed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points.
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    995
    994
    Units: percentage of subjects
    number (not applicable)
        Week 2: At least 30% reduction (n=1000,995,994)
    35.8
    32.1
    31.7
        Week 2: At least 50% reduction (n=1000,995,994)
    20.0
    17.0
    15.4
        Week 2: At least 70% reduction (n=1000,995,994)
    8.3
    8.2
    5.8
        Week 2: At least 90% reduction (n=1000,995,994)
    2.1
    3.2
    1.7
        Week 4: At least 30% reduction (n=1000,995,994)
    49.0
    49.1
    43.2
        Week 4: At least 50% reduction (n=1000,995,994)
    31.1
    31.3
    23.1
        Week 4: At least 70% reduction (n=1000,995,994)
    15.5
    15.8
    11.2
        Week 4: At least 90% reduction (n=1000,995,994)
    4.6
    5.4
    2.6
        Week 8: At least 30% reduction (n=1000,995,994)
    56.0
    59.5
    55.0
        Week 8: At least 50% reduction (n=1000,995,994)
    36.6
    40.0
    31.4
        Week 8: At least 70% reduction (n=1000,995,994)
    18.7
    21.3
    14.1
        Week 8: At least 90% reduction (n=1000,995,994)
    5.8
    7.1
    4.4
        Week 16: At least 30% reduction (n=1000,995,994)
    71.6
    71.8
    68.1
        Week 16: At least 50% reduction (n=1000,995,994)
    53.1
    55.8
    50.1
        Week 16: At least 70% reduction (n=1000,995,994)
    29.9
    34.3
    27.9
        Week 16: At least 90% reduction (n=1000,995,994)
    10.7
    13.4
    9.7
        Week 24: At least 30% reduction (n=1000,995,994)
    59.5
    61.3
    59.0
        Week 24: At least 50% reduction (n=1000,995,994)
    49.9
    48.2
    46.8
        Week 24: At least 70% reduction (n=1000,995,994)
    30.4
    32.7
    27.8
        Week 24: At least 90% reduction (n=1000,995,994)
    11.0
    13.0
    9.8
        Week 32: At least 30% reduction (n=1000,995,994)
    56.7
    56.6
    55.9
        Week 32: At least 50% reduction (n=1000,995,994)
    47.2
    45.7
    44.7
        Week 32: At least 70% reduction (n=1000,995,994)
    29.7
    30.2
    26.8
        Week 32: At least 90% reduction (n=1000,995,994)
    11.0
    13.1
    9.4
        Week 40: At least 30% reduction (n=1000,995,994)
    55.5
    55.5
    54.9
        Week 40: At least 50% reduction (n=1000,995,994)
    45.5
    45.0
    45.0
        Week 40: At least 70% reduction (n=1000,995,994)
    29.5
    29.1
    27.6
        Week 40: At least 90% reduction (n=1000,995,994)
    10.3
    13.2
    9.5
        Week 48: At least 30% reduction (n=1000,995,994)
    54.5
    53.3
    54.6
        Week 48: At least 50% reduction (n=1000,995,994)
    45.3
    43.5
    43.4
        Week 48: At least 70% reduction (n=1000,995,994)
    29.1
    27.9
    26.1
        Week 48: At least 90% reduction (n=1000,995,994)
    10.2
    12.0
    9.4
        Week 56: At least 30% reduction (n=1000,995,994)
    52.0
    51.1
    52.9
        Week 56: At least 50% reduction (n=1000,995,994)
    44.1
    41.3
    42.5
        Week 56: At least 70% reduction (n=1000,995,994)
    26.9
    26.4
    26.0
        Week 56: At least 90% reduction (n=1000,995,994)
    9.3
    10.5
    9.0
        Week 64: At least 30% reduction (n =437,419,445)
    71.4
    68.0
    78.2
        Week 64: At least 50% reduction (n =437,419,445)
    52.9
    44.6
    58.9
        Week 64: At least 70% reduction (n =437,419,445)
    31.4
    22.9
    33.9
        Week 64: At least 90% reduction (n =437,419,445)
    9.4
    7.9
    13.3
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0651
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.99
         upper limit
    		 1.44
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9032
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.84
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.01
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.36
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.72
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3728
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.42
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0434
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.44
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.01
         upper limit
    		 2.04
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0425
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.44
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.01
         upper limit
    		 2.04
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5442
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.64
         upper limit
    		 2.34
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0349
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.9
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.05
         upper limit
    		 3.45
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0108
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.26
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.05
         upper limit
    		 1.51
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.008
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.27
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.07
         upper limit
    		 1.52
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.5
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.22
         upper limit
    		 1.83
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.51
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.24
         upper limit
    		 1.85
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.006
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.45
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.11
         upper limit
    		 1.88
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0031
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.49
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.14
         upper limit
    		 1.93
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0235
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.76
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 2.88
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0021
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    2.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.31
         upper limit
    		 3.42
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7613
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.23
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0548
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1
         upper limit
    		 1.43
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0196
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.25
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.04
         upper limit
    		 1.51
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.45
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.2
         upper limit
    		 1.75
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0077
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.39
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.09
         upper limit
    		 1.77
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.65
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.3
         upper limit
    		 2.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1942
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.31
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.96
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0122
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.64
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.11
         upper limit
    		 2.43
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0977
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.97
         upper limit
    		 1.43
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0806
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.98
         upper limit
    		 1.44
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2097
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.94
         upper limit
    		 1.34
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0135
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.25
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.05
         upper limit
    		 1.49
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3571
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.33
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0025
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.34
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.11
         upper limit
    		 1.63
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4658
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.49
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0108
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.44
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.09
         upper limit
    		 1.9
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8393
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2977
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.92
         upper limit
    		 1.32
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1944
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.94
         upper limit
    		 1.34
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5714
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2472
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.92
         upper limit
    		 1.36
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0235
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.25
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.03
         upper limit
    		 1.51
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4074
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.13
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.51
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0296
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.37
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.03
         upper limit
    		 1.81
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7607
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.23
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7979
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2964
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.92
         upper limit
    		 1.31
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.695
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.24
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1985
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.93
         upper limit
    		 1.38
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1239
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.17
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.96
         upper limit
    		 1.42
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2762
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.58
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0121
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.44
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.91
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8443
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8472
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.898
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.21
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9385
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.19
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4261
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.32
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.525
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.3
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6273
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.45
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0132
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.43
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.9
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9095
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.18
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5098
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4488
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.28
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9757
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.19
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1843
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.94
         upper limit
    		 1.39
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4356
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.32
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6342
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.45
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.081
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.29
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.97
         upper limit
    		 1.73
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6527
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.15
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=30% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3857
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.92
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.77
         upper limit
    		 1.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.528
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.27
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=50% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5355
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7732
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.84
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=70% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9397
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.82
         upper limit
    		 1.23
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9044
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.75
         upper limit
    		 1.39
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56, >=90% reduction: Odds ratio and 95% CI estimated from logistic regression model. Logistic regression model included baseline WOMAC physical function subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3193
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.57

    Secondary: Percentage of Subjects With Cumulative Percent Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Weeks 16, 24 and 56

    		 Close Top of page
    End point title
    		 Percentage of Subjects With Cumulative Percent Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Physical Function Subscale at Weeks 16, 24 and 56
    End point description
    Percentage of subjects with cumulative reduction (in percent) in WOMAC physical function subscale from baseline to Weeks 16, 24 and 56 were reported. WOMAC:Self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. Physical function: subject’s ability to move around and perform usual activities of daily living. WOMAC physical function subscale:17-item questionnaire to assess the degree of difficulty experienced due to OA in index joint (knee or hip) during past 48 hours, calculated as mean of the scores from 17 individual questions scored on a NRS. Scores for each question and WOMAC Pain subscale on NRS ranged from 0 (no difficulty) to 10 (extreme difficulty), higher scores indicated extreme difficulty/worse physical function. Missing data was imputed using mixed BOCF/LOCF. ITT population was analysed. ‘Number of subjects analysed’=subjects evaluable for this endpoint.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16, 24 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    995
    994
    Units: percentage of subjects
    number (not applicable)
        Week 16: >0%
    90.0
    88.8
    87.4
        Week 16: >=10%
    85.0
    83.9
    81.4
        Week 16: >=20%
    78.4
    77.3
    73.7
        Week 16: >=30%
    71.6
    71.8
    68.1
        Week 16: >=40%
    63.7
    64.1
    61.0
        Week 16: >=50%
    53.1
    55.8
    50.1
        Week 16: >=60%
    41.4
    44.7
    39.8
        Week 16: >=70%
    29.9
    34.3
    27.9
        Week 16: >=80%
    20.8
    24.4
    17.9
        Week 16: >=90%
    10.7
    13.4
    9.7
        Week 16: =100%
    2.9
    3.3
    2.0
        Week 24: >=0%
    66.7
    68.6
    65.0
        Week 24: >=10%
    65.0
    66.6
    63.3
        Week 24: >=20%
    62.6
    64.2
    60.8
        qWeek 24: >=30%
    59.5
    61.3
    59.0
        Week 24: >=40%
    54.9
    56.0
    53.9
        Week 24: >=50%
    49.9
    48.2
    46.8
        Week 24: >=60%
    41.3
    42.0
    37.7
        Week 24: >=70%
    30.4
    32.7
    27.8
        Week 24: >=80%
    19.9
    22.6
    18.9
        Week 24: >=90%
    11.0
    13.0
    9.8
        Week 24: =100%
    3.0
    3.2
    2.7
        Week 56: >=0%
    61.1
    59.5
    60.1
        Week 56: >=10%
    59.3
    57.3
    57.8
        Week 56: >=20%
    56.1
    54.3
    55.4
        Week 56: >=30%
    52.0
    51.1
    52.9
        Week 56: >=40%
    48.5
    46.3
    48.9
        Week 56: >=50%
    44.1
    41.3
    42.5
        Week 56: >=60%
    36.7
    34.6
    34.8
        Week 56: >=70%
    26.9
    26.4
    26.0
        Week 56: >=80%
    17.1
    17.1
    17.4
        Week 56: >=90%
    9.3
    10.5
    9.0
        Week 56: =100%
    2.9
    3.6
    3.3
    No statistical analyses for this end point

    Secondary: Percentage of Subjects Achieving Improvement of >=2 Points in Patient's Global Assessment (PGA) of Osteoarthritis at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64

    		 Close Top of page
    End point title
    		 Percentage of Subjects Achieving Improvement of >=2 Points in Patient's Global Assessment (PGA) of Osteoarthritis at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point description
    PGA of OA was assessed by asking a question from subjects: “Considering all the ways your osteoarthritis in your knee or hip affects you, how are you doing today?" Subjects responded on a scale ranging from 1-5, where, 1=very good (no symptom and no limitation of normal activities), 2= good (mild symptoms and no limitation of normal activities), 3= fair (moderate symptoms and limitation of some normal activities), 4= poor (severe symptoms and inability to carry out most normal activities), and 5 = very poor (very severe symptoms and inability to carry out all normal activities). Higher scores indicated worse condition. Percentage of subjects with improvement of at least 2 points from baseline in PGA of OA were reported. Missing data was imputed using mixed BOCF/LOCF. ITT population was analysed. ‘Overall number of subjects analyzed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    995
    994
    Units: percentage of subjects
    number (not applicable)
        Week 2 (n= 1000, 995, 994)
    14.6
    15.6
    11.6
        Week 4 (n= 1000, 995, 994)
    21.4
    22.4
    15.9
        Week 8 (n= 1000, 995, 994)
    21.9
    23.7
    19.0
        Week 16 (n= 1000, 995, 994)
    29.1
    30.3
    28.2
        Week 24 (n= 1000, 995, 994)
    23.4
    24.8
    23.7
        Week 32 (n= 1000, 995, 994)
    23.7
    22.3
    23.6
        Week 40 (n= 1000, 995, 994)
    21.7
    21.7
    21.0
        Week 48 (n= 1000, 995, 994)
    22.0
    21.7
    21.1
        Week 56 (n= 1000, 995, 994)
    21.0
    19.7
    20.8
        Week 64 (n= 437, 419, 445))
    21.1
    17.4
    25.8
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1772
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.91
         upper limit
    		 1.62
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0154
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.42
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.07
         upper limit
    		 1.89
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0105
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.4
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.08
         upper limit
    		 1.81
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0002
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.63
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.26
         upper limit
    		 2.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4007
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.43
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0116
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.37
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.07
         upper limit
    		 1.75
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6279
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.95
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.76
         upper limit
    		 1.18
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4674
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.09
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.35
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3135
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.89
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.71
         upper limit
    		 1.12
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7581
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.3
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4504
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.92
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.73
         upper limit
    		 1.15
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2629
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.88
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.7
         upper limit
    		 1.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6763
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.95
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.75
         upper limit
    		 1.2
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9456
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.27
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.752
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.76
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9981
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.26
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5284
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.93
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.74
         upper limit
    		 1.17
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline PGA of OA scale, baseline diary average pain, and classification variables index joint, highest KLG, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.322
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.89
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.7
         upper limit
    		 1.12

    Secondary: Change From Baseline in Average Pain Score in the Index Joint at Weeks 1, 2, 3, 4, 6, 8, 10, 12, 16, 20, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Pain Score in the Index Joint at Weeks 1, 2, 3, 4, 6, 8, 10, 12, 16, 20, 24, 32, 40, 48 and 56
    End point description
    Subjects assessed their average pain in the index hip/knee in the past 24 hours using NRS, with a scale ranging from 0 (no pain) to 10 (worst possible pain). Higher scores indicated higher pain. Data for Weeks 20 through 56 represents averages of the values reported during the 4-week interval up to and including the given week. Change from baseline was calculated using the difference between each post-baseline weekly mean and the baseline mean score. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 1, 2, 3, 4, 6, 8, 10, 12, 16, 20, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 1
    -0.47 ( 0.05 )
    -0.56 ( 0.05 )
    -0.56 ( 0.05 )
        Change at Week 2
    -1.02 ( 0.07 )
    -0.97 ( 0.07 )
    -0.91 ( 0.07 )
        Change at Week 3
    -1.40 ( 0.08 )
    -1.30 ( 0.08 )
    -1.23 ( 0.08 )
        Change at Week 4
    -1.62 ( 0.09 )
    -1.65 ( 0.09 )
    -1.32 ( 0.09 )
        Change at Week 6
    -1.85 ( 0.09 )
    -1.97 ( 0.09 )
    -1.49 ( 0.09 )
        Change at Week 8
    -1.83 ( 0.10 )
    -2.04 ( 0.10 )
    -1.59 ( 0.10 )
        Change at Week 10
    -2.35 ( 0.10 )
    -2.46 ( 0.10 )
    -1.98 ( 0.10 )
        Change at Week 12
    -2.48 ( 0.10 )
    -2.55 ( 0.10 )
    -2.10 ( 0.10 )
        Change at Week 16
    -2.41 ( 0.10 )
    -2.52 ( 0.10 )
    -2.17 ( 0.11 )
        Change at Week 20
    -2.56 ( 0.11 )
    -2.60 ( 0.11 )
    -2.27 ( 0.11 )
        Change at Week 24
    -2.35 ( 0.13 )
    -2.41 ( 0.12 )
    -2.11 ( 0.12 )
        Change at Week 32
    -2.27 ( 0.13 )
    -2.26 ( 0.13 )
    -2.06 ( 0.13 )
        Change at Week 40
    -2.25 ( 0.13 )
    -2.20 ( 0.13 )
    -2.07 ( 0.13 )
        Change at Week 48
    -2.20 ( 0.13 )
    -2.10 ( 0.13 )
    -2.03 ( 0.13 )
        Change at Week 56
    -2.17 ( 0.13 )
    -2.03 ( 0.13 )
    -2.04 ( 0.13 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 1: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.112
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.09
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.02
         upper limit
    		 0.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.06
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 1: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9686
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.11
         upper limit
    		 0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.06
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1589
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.25
         upper limit
    		 0.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.472
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.2
         upper limit
    		 0.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 3: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.032
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.17
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 -0.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 3: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3336
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.24
         upper limit
    		 0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0005
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.3
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.47
         upper limit
    		 -0.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.33
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.5
         upper limit
    		 -0.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 6: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.37
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.55
         upper limit
    		 -0.18
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 6: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.66
         upper limit
    		 -0.3
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0115
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.42
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.45
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.63
         upper limit
    		 -0.26
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 10: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0002
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.36
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.56
         upper limit
    		 -0.17
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 10: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.67
         upper limit
    		 -0.28
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 12: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0002
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.37
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.57
         upper limit
    		 -0.18
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 12: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.45
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.64
         upper limit
    		 -0.25
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0238
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.44
         upper limit
    		 -0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0011
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.34
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.55
         upper limit
    		 -0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 20: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0064
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.3
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.51
         upper limit
    		 -0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 20: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0025
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.33
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.54
         upper limit
    		 -0.12
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0589
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.23
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.48
         upper limit
    		 0.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.12
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.018
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.29
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.53
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0944
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.47
         upper limit
    		 0.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1198
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.2
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.45
         upper limit
    		 0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1837
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.17
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.43
         upper limit
    		 0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.317
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.13
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.39
         upper limit
    		 0.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1873
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.17
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.43
         upper limit
    		 0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5719
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.18
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3571
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.39
         upper limit
    		 0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest KLG and NSAID) as fixed effects, baseline pain in the index joint as covariate, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9072
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.25
         upper limit
    		 0.28
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14

    Secondary: Change From Baseline in Average Pain Score in the Index Joint at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Pain Score in the Index Joint at Week 64
    End point description
    Subjects assessed their average pain in the index hip/knee in the past 24 hours using NRS, with a scale ranging from 0 (no pain) to 10 (worst possible pain). Higher scores indicated higher pain. Data represents averages of the values reported during the 8-week interval (4-week interval for weeks 20 to 56) up to and including the given week. Change from baseline was calculated using the difference between each post-baseline weekly mean and the baseline mean score. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n= 995, 992, 990 )
    6.76 ( 1.59 )
    6.77 ( 1.58 )
    6.76 ( 1.54 )
        Change at Week 64 (n =436, 416, 434)
    -3.01 ( 2.60 )
    -2.81 ( 2.71 )
    -3.24 ( 2.55 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Stiffness Subscale at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Stiffness Subscale at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. Stiffness was defined as a sensation of decreased ease of movement in the index joint (knee or hip). The WOMAC stiffness subscale is a 2-item questionnaire used to assess the amount of stiffness experienced due to OA in the index joint (knee or hip) during the past 48 hours. It was calculated as the mean of scores from 2 individual questions scored on NRS. Scores for each question and WOMAC stiffness subscale score on NRS ranged from 0 (no stiffness) to 10 (extreme stiffness), where higher scores indicated higher stiffness. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 2
    -1.79 ( 0.09 )
    -1.70 ( 0.09 )
    -1.48 ( 0.09 )
        Change at Week 4
    -2.32 ( 0.10 )
    -2.43 ( 0.10 )
    -1.95 ( 0.10 )
        Change at Week 8
    -2.46 ( 0.10 )
    -2.79 ( 0.10 )
    -2.16 ( 0.10 )
        Change at Week 16
    -3.32 ( 0.11 )
    -3.54 ( 0.11 )
    -3.10 ( 0.11 )
        Change at Week 24
    -2.77 ( 0.13 )
    -2.95 ( 0.13 )
    -2.63 ( 0.13 )
        Change at Week 32
    -2.68 ( 0.13 )
    -2.74 ( 0.13 )
    -2.52 ( 0.13 )
        Change at Week 40
    -2.58 ( 0.14 )
    -2.64 ( 0.14 )
    -2.46 ( 0.14 )
        Change at Week 48
    -2.60 ( 0.14 )
    -2.54 ( 0.13 )
    -2.44 ( 0.14 )
        Change at Week 56
    -2.46 ( 0.14 )
    -2.46 ( 0.14 )
    -2.42 ( 0.14 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0004
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.31
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.49
         upper limit
    		 -0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0134
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.4
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.38
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.56
         upper limit
    		 -0.19
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.67
         upper limit
    		 -0.29
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0031
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.29
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.49
         upper limit
    		 -0.1
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.63
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.82
         upper limit
    		 -0.43
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0425
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.43
         upper limit
    		 -0.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.44
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.65
         upper limit
    		 -0.23
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.298
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.14
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.4
         upper limit
    		 0.12
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0179
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.32
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.58
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2328
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.42
         upper limit
    		 0.1
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1019
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.22
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.49
         upper limit
    		 0.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4002
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.38
         upper limit
    		 0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2149
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.17
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.45
         upper limit
    		 0.1
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2442
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.43
         upper limit
    		 0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4609
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 0.17
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8129
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.24
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC stiffness subscales and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7883
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.32
         upper limit
    		 0.24
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Stiffness Subscale at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Stiffness Subscale at Week 64
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA. Stiffness was defined as a sensation of decreased ease of movement in the index joint (knee or hip). The WOMAC stiffness subscale is a 2-item questionnaire used to assess the amount of stiffness experienced due to OA in the index joint (knee or hip) during the past 48 hours. It was calculated as the mean of scores from 2 individual questions scored on NRS. Scores for each question and WOMAC stiffness subscale score on NRS ranged from 0 (no stiffness) to 10 (extreme stiffness), where higher scores indicated higher stiffness. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    7.15 ( 1.42 )
    7.20 ( 1.40 )
    7.09 ( 1.42 )
        Change at Week 64 (n =437, 419, 445)
    -3.31 ( 2.72 )
    -3.04 ( 2.64 )
    -3.66 ( 2.36 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Average Score at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Average Score at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA of index joint (knee or hip). WOMAC pain subscale assess amount of pain experienced (score: 0 [no pain] to 10 [extreme pain], higher score = more pain), WOMAC physical function subscale assess degree of difficulty experienced (score: 0 [no difficulty] to 10 [extreme difficulty], higher score = worse physical function) and WOMAC stiffness subscale assess the amount of stiffness experienced (score: 0 [no stiffness] to 10 [extreme stiffness], higher score = higher stiffness). WOMAC average score was the mean of WOMAC pain, physical function and stiffness subscale scores and ranges from 0 to 10, where higher scores indicated worse response. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 2
    -1.73 ( 0.08 )
    -1.61 ( 0.08 )
    -1.52 ( 0.08 )
        Change at Week 4
    -2.28 ( 0.09 )
    -2.34 ( 0.09 )
    -1.95 ( 0.09 )
        Change at Week 8
    -2.44 ( 0.10 )
    -2.71 ( 0.10 )
    -2.23 ( 0.10 )
        Change at Week 16
    -3.26 ( 0.11 )
    -3.41 ( 0.11 )
    -3.07 ( 0.11 )
        Change at Week 24
    -2.74 ( 0.13 )
    -2.88 ( 0.13 )
    -2.64 ( 0.13 )
        Change at Week 32
    -2.65 ( 0.13 )
    -2.69 ( 0.13 )
    -2.54 ( 0.13 )
        Change at Week 40
    -2.57 ( 0.13 )
    -2.58 ( 0.13 )
    -2.49 ( 0.13 )
        Change at Week 48
    -2.56 ( 0.13 )
    -2.48 ( 0.13 )
    -2.44 ( 0.13 )
        Change at Week 56
    -2.45 ( 0.13 )
    -2.38 ( 0.13 )
    -2.40 ( 0.13 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0119
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3073
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.24
         upper limit
    		 0.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0002
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.33
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.5
         upper limit
    		 -0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.39
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.56
         upper limit
    		 -0.22
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0251
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.39
         upper limit
    		 -0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.66
         upper limit
    		 -0.3
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0625
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.39
         upper limit
    		 0.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.34
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.54
         upper limit
    		 -0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4005
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.36
         upper limit
    		 0.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.053
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.25
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.5
         upper limit
    		 0
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4074
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.37
         upper limit
    		 0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2629
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.41
         upper limit
    		 0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5582
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.34
         upper limit
    		 0.18
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4907
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.09
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.35
         upper limit
    		 0.17
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4043
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.38
         upper limit
    		 0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7663
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.23
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7415
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 0.22
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC average scores and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8688
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.24
         upper limit
    		 0.29
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Average Score at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Average Score at Week 64
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA of index joint (knee or hip). WOMAC pain subscale assess amount of pain experienced (score: 0 [no pain] to 10 [extreme pain], higher score = more pain), WOMAC physical function subscale assess degree of difficulty experienced (score: 0 [no difficulty] to 10 [extreme difficulty], higher score = worse physical function) and WOMAC stiffness subscale assess the amount of stiffness experienced (score: 0 [no stiffness] to 10 [extreme stiffness], higher score = higher stiffness). WOMAC average score was the mean of WOMAC pain, physical function and stiffness subscale scores and ranges from 0 to 10, where higher scores indicated worse response. ITT population was analysed. ‘n’=subjects evaluable at specified time points for each arm,
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    7.09 ( 1.08 )
    7.10 ( 1.10 )
    7.01 ( 1.08 )
        Change at Week 64 (n =437, 419, 445)
    -3.40 ( 2.40 )
    -3.09 ( 2.37 )
    -3.77 ( 2.06 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Walking on a Flat Surface at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Walking on a Flat Surface at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA in index joint (knee or hip). Subjects answered a question: "How much pain have you had when walking on a flat surface?”. Subjects responded about the amount of pain they experienced when walking on a flat surface by using a NRS of 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 2
    -1.54 ( 0.08 )
    -1.39 ( 0.08 )
    -1.46 ( 0.08 )
        Change at Week 4
    -2.14 ( 0.10 )
    -2.15 ( 0.10 )
    -1.91 ( 0.10 )
        Change at Week 8
    -2.26 ( 0.10 )
    -2.47 ( 0.10 )
    -2.22 ( 0.10 )
        Change at Week 16
    -3.01 ( 0.11 )
    -3.13 ( 0.11 )
    -2.95 ( 0.11 )
        Change at Week 24
    -2.64 ( 0.13 )
    -2.76 ( 0.13 )
    -2.60 ( 0.13 )
        Change at Week 32
    -2.54 ( 0.13 )
    -2.54 ( 0.13 )
    -2.52 ( 0.14 )
        Change at Week 40
    -2.48 ( 0.14 )
    -2.42 ( 0.14 )
    -2.48 ( 0.14 )
        Change at Week 48
    -2.45 ( 0.14 )
    -2.34 ( 0.14 )
    -2.42 ( 0.14 )
        Change at Week 56
    -2.37 ( 0.14 )
    -2.21 ( 0.14 )
    -2.39 ( 0.14 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3622
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.25
         upper limit
    		 0.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3894
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.09
         upper limit
    		 0.24
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0137
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.23
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.42
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0106
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.24
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.43
         upper limit
    		 -0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7179
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.23
         upper limit
    		 0.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.012
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.25
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.44
         upper limit
    		 -0.05
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5372
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.28
         upper limit
    		 0.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0899
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.4
         upper limit
    		 0.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7646
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.3
         upper limit
    		 0.22
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2547
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.41
         upper limit
    		 0.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8806
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.29
         upper limit
    		 0.25
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8416
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.3
         upper limit
    		 0.24
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9981
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.28
         upper limit
    		 0.28
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6485
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.21
         upper limit
    		 0.34
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8317
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.3
         upper limit
    		 0.24
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5819
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.19
         upper limit
    		 0.35
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8538
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.26
         upper limit
    		 0.31
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when walking on a flat surface and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1981
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.1
         upper limit
    		 0.47
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.15

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Walking on a Flat Surface at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Walking on a Flat Surface at Week 64
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA in index joint (knee or hip). Subjects answered a question: "How much pain have you had when walking on a flat surface?”. Subjects responded about the amount of pain they experienced when walking on a flat surface by using a NRS of 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. ITT population was analysed. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    6.86 ( 1.33 )
    6.90 ( 1.34 )
    6.86 ( 1.30 )
        Change at Week 64 (n =437, 419, 445)
    -3.20 ( 2.78 )
    -2.69 ( 2.58 )
    -3.67 ( 2.32 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Going Up or Downstairs at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Going Up or Downstairs at Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA in index joint (knee or hip). Subjects answered a question: "How much pain have you had when going up or down the stairs?” Subjects responded about the amount of pain they experienced when going up or down stairs by using a NRS of 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    least squares mean (standard error)
        Change at Week 2
    -1.81 ( 0.08 )
    -1.66 ( 0.08 )
    -1.66 ( 0.09 )
        Change at Week 4
    -2.34 ( 0.10 )
    -2.43 ( 0.10 )
    -2.08 ( 0.10 )
        Change at Week 8
    -2.48 ( 0.10 )
    -2.81 ( 0.10 )
    -2.40 ( 0.10 )
        Change at Week 16
    -3.34 ( 0.11 )
    -3.50 ( 0.11 )
    -3.18 ( 0.12 )
        Change at Week 24
    -2.89 ( 0.13 )
    -3.03 ( 0.13 )
    -2.83 ( 0.14 )
        Change at Week 32
    -2.76 ( 0.14 )
    -2.84 ( 0.14 )
    -2.74 ( 0.14 )
        Change at Week 40
    -2.69 ( 0.14 )
    -2.74 ( 0.14 )
    -2.70 ( 0.14 )
        Change at Week 48
    -2.70 ( 0.14 )
    -2.63 ( 0.14 )
    -2.67 ( 0.14 )
        Change at Week 56
    -2.55 ( 0.14 )
    -2.47 ( 0.14 )
    -2.55 ( 0.14 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0846
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.02
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9749
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.18
         upper limit
    		 0.17
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0076
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.26
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.45
         upper limit
    		 -0.07
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0003
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.35
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.54
         upper limit
    		 -0.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4402
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.27
         upper limit
    		 0.12
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.41
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.61
         upper limit
    		 -0.21
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1543
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.16
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.38
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0053
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.31
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.53
         upper limit
    		 -0.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6445
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1439
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.2
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.47
         upper limit
    		 0.07
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8402
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.3
         upper limit
    		 0.25
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4439
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.11
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.38
         upper limit
    		 0.17
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9376
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.27
         upper limit
    		 0.29
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7614
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.33
         upper limit
    		 0.24
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8081
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.32
         upper limit
    		 0.25
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.15
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8078
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.25
         upper limit
    		 0.33
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.15
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9705
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.28
         upper limit
    		 0.29
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.15
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Multiple imputation method was applied for missing data, with imputation dependent on reason for missing data. ANCOVA model for imputed datasets included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WOMAC pain when going up or down stairs and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5785
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.21
         upper limit
    		 0.38
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.15

    Secondary: Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Going Up or Downstairs at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Pain Subscale Item: Pain When Going Up or Downstairs at Week 64
    End point description
    WOMAC: self-administered, disease-specific questionnaire which assesses clinically important, subject-relevant symptoms for pain, stiffness and physical function in subjects with OA in index joint (knee or hip). Subjects answered a question: "How much pain have you had when going up or down the stairs?” Subjects responded about the amount of pain they experienced when going up or down stairs by using a NRS of 0 (no pain) to 10 (extreme pain), where higher scores indicated higher pain. ITT population was analysed. Here, “n”=subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    7.89 ( 1.24 )
    7.88 ( 1.29 )
    7.83 ( 1.19 )
        Change at Week 64 (n =437, 419, 445)
    -3.28 ( 2.67 )
    -2.97 ( 2.69 )
    -3.70 ( 2.50 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Work Productivity and Activity Impairment Questionnaire for Osteoarthritis (WPAI:OA) Scores at Weeks 16, 24 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Work Productivity and Activity Impairment Questionnaire for Osteoarthritis (WPAI:OA) Scores at Weeks 16, 24 and 56
    End point description
    WPAI is 6-question subject rated questionnaire to determine the impact of OA on absenteeism, presenteeism, work productivity, and daily activity impairment for a period of 7 days prior to a visit. It yields 4 sub-scores: work time missed (absenteeism), impairment while working (presenteeism), overall work impairment (work productivity) and activity impairment (daily activity impairment). These sub-scores are expressed as an impairment percentage (range from 0 to 100), with higher numbers indicating greater impairment and less productivity. ITT population was analyzed. Here, Overall number of subjects analyzed’=subjects evaluable for this endpoint. Here, “n” =subjects evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 16, 24 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    932
    941
    930
    Units: units on a scale
    least squares mean (standard error)
        CAW 16:%Work Time Missed(n= 350,339,338)
    -2.33 ( 0.62 )
    -3.35 ( 0.64 )
    -2.92 ( 0.63 )
        CAW 16:%Impairment While Working(n=343,336,336)
    -28.07 ( 1.58 )
    -26.94 ( 1.61 )
    -26.59 ( 1.60 )
        CAW 16:%Overall Work Impairment(n=343,336,336)
    -28.67 ( 1.62 )
    -27.51 ( 1.65 )
    -27.04 ( 1.63 )
        CAW 16:%Activity Impairment(n=932,941,930)
    -30.59 ( 1.04 )
    -31.36 ( 1.04 )
    -29.38 ( 1.05 )
        CAW 24:%Work Time Missed(n=300,286,282)
    -2.70 ( 0.80 )
    -2.19 ( 0.81 )
    -2.73 ( 0.81 )
        CAW 24:%Impairment While Working(n=297,282,279)
    -25.34 ( 1.73 )
    -26.66 ( 1.74 )
    -25.15 ( 1.74 )
        CAW 24:%Overall Work Impairment(n=297,282,279)
    -26.05 ( 1.78 )
    -27.33 ( 1.80 )
    -25.90 ( 1.80 )
        CAW 24:%Activity Impairment(n=822,821,820)
    -29.88 ( 1.13 )
    -30.53 ( 1.13 )
    -29.76 ( 1.14 )
        CAW 56:%Work Time Missed(n=163,178,152)
    -0.12 ( 1.56 )
    -1.84 ( 1.47 )
    -0.81 ( 1.53 )
        CAW 56:%Impairment While Working(n=162,174,152)
    -31.49 ( 2.22 )
    -29.92 ( 2.12 )
    -34.59 ( 2.17 )
        CAW 56:%Overall Work Impairment(n=162,174,152)
    -31.21 ( 2.39 )
    -29.29 ( 2.28 )
    -34.26 ( 2.34 )
        CAW 56:%Activity Impairment|(n=480,486,477)
    -34.47 ( 1.42 )
    -32.91 ( 1.39 )
    -36.17 ( 1.41 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Work Time Missed: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4303
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.58
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.87
         upper limit
    		 2.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.74
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Work Time Missed: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5656
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.43
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -1.9
         upper limit
    		 1.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.75
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Work Time Missed: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.976
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -1.78
         upper limit
    		 1.83
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.92
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Work Time Missed: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5678
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.53
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -1.3
         upper limit
    		 2.36
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.93
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Work Time Missed: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6974
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.68
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -2.77
         upper limit
    		 4.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.76
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Work Time Missed: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1261
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    2.64
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.75
         upper limit
    		 6.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.72
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Impairment While Working: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.406
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -1.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -4.96
         upper limit
    		 2.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.78
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Impairment While Working: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.848
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.34
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -3.84
         upper limit
    		 3.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.78
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Impairment While Working: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.923
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.19
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -4
         upper limit
    		 3.63
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.94
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Impairment While Working: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4421
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -1.51
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -5.36
         upper limit
    		 2.34
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.96
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Impairment While Working: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2049
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    3.1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -1.7
         upper limit
    		 7.91
    Variability estimate
    Standard error of the mean
    Dispersion value
    2.44
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Impairment While Working: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0527
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    4.68
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.05
         upper limit
    		 9.41
    Variability estimate
    Standard error of the mean
    Dispersion value
    2.4
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Overall Work Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3699
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -1.63
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -5.21
         upper limit
    		 1.94
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.82
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Overall Work Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7945
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -4.06
         upper limit
    		 3.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.83
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Overall Work Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.941
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -4.13
         upper limit
    		 3.83
    Variability estimate
    Standard error of the mean
    Dispersion value
    2.03
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Overall Work Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.486
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -1.43
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -5.44
         upper limit
    		 2.59
    Variability estimate
    Standard error of the mean
    Dispersion value
    2.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Overall Work Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2448
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    3.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -2.1
         upper limit
    		 8.2
    Variability estimate
    Standard error of the mean
    Dispersion value
    2.62
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Overall Work Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0551
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    4.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.11
         upper limit
    		 10.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    2.58
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Activity Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.247
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -1.21
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -3.26
         upper limit
    		 0.84
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Percent Activity Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.057
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -1.98
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -4.01
         upper limit
    		 0.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.04
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Activity Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.917
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.12
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -2.36
         upper limit
    		 2.12
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.14
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Percent Activity Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4991
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -0.77
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -3
         upper limit
    		 1.46
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Activity Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1862
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2377
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    1.7
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -1.12
         upper limit
    		 4.52
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.44
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Percent Activity Impairment: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline WPAI score and baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1871
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0238
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    3.26
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.43
         upper limit
    		 6.08
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.44

    Secondary: Change From Baseline in Work Productivity and Activity Impairment Questionnaire for Osteoarthritis (WPAI:OA) Scores at Week 64

    		 Close Top of page
    End point title
    		 Change From Baseline in Work Productivity and Activity Impairment Questionnaire for Osteoarthritis (WPAI:OA) Scores at Week 64
    End point description
    WPAI is 6-question subject rated questionnaire to determine the impact of OA on absenteeism, presenteeism, work productivity, and daily activity impairment for a period of 7 days prior to a visit. It yields 4 sub-scores: work time missed (absenteeism), impairment while working (presenteeism), overall work impairment (work productivity) and activity impairment (daily activity impairment). These sub-scores are expressed as an impairment percentage (range from 0 to 100), with higher numbers indicating greater impairment and less productivity. CAW = Change at Week, (%) = Percent. ITT population was analyzed. Here, “n” =subjects evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline: % Work Time Missed (n=438,417,420)
    6.1 ( 15.81 )
    6.0 ( 15.56 )
    5.2 ( 14.54 )
        Baseline:%Impairment While Working (n=434,413,417)
    60.5 ( 20.39 )
    58.3 ( 20.78 )
    59.3 ( 18.97 )
        Baseline:%Overall Work Impairment (n=434,413,417)
    62.1 ( 21.02 )
    60.0 ( 21.37 )
    60.6 ( 19.78 )
        Baseline: % Activity Impairment (n=1000,995,994)
    68.3 ( 14.93 )
    67.9 ( 15.83 )
    66.7 ( 15.35 )
        CAW 64: % Work Time Missed(n=149,137,142)
    -1.8 ( 19.35 )
    4.1 ( 21.88 )
    -2.1 ( 16.65 )
        CAW 64:% Impairment While Working(n=146,135,141)
    -24.2 ( 27.69 )
    -20.7 ( 29.13 )
    -26.5 ( 26.24 )
        CAW 64:% Overall Work Impairment (n =146,135,141)
    -24.5 ( 28.77 )
    -19.2 ( 30.54 )
    -27.0 ( 27.22 )
        CAW 64: % Activity Impairment(n =450,427,453)
    -28.7 ( 26.68 )
    -24.1 ( 27.80 )
    -32.1 ( 24.51 )
    No statistical analyses for this end point

    Secondary: Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Mobility Domain

    		 Close Top of page
    End point title
    		 Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Mobility Domain
    End point description
    Number of subjects with mobility domain responses of EQ-5D-5L were provided. EQ-5D-5L is a standardized subjects completed questionnaire that measures health-related quality of life and translates that score into an index value or utility score. EQ-5D-5L consists of two components: a health state profile and an optional visual analogue scale (VAS). EQ-5D health state profile is comprised of 5 dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: 1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, and 5=extreme problems. Higher scores indicated greater levels of problems across the five dimensions. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 24, 40, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline:No problem in walking(n=1000,995,994)
    26
    20
    23
        Week 8:No problem in walking(n=956,966,956)
    223
    241
    216
        Week 16:No problem in walking(n=913,920,915)
    299
    319
    292
        Week 24:No problem in walking(n=817,816,813)
    259
    261
    260
        Week 40:No problem in walking(n=561,553,535)
    217
    211
    218
        Week 56:No problem in walking(n=458,458,459)
    157
    147
    170
        Week 64:No problem in walking(n=450,428,454)
    98
    66
    107
        Baseline:Slight problem in walking(n=1000,995,994)
    203
    192
    194
        Week 8:Slight problem in walking(n=956,966,956)
    374
    411
    392
        Week 16:Slight problem in walking(n=913,920,915)
    388
    371
    412
        Week 24:Slight problem in walking(n=817,816,813)
    308
    310
    337
        Week 40:Slight problem in walking(n=561,553,535)
    215
    209
    217
        Week 56:Slight problem in walking(n=458,458,459)
    205
    166
    189
        Week 64:Slight problem in walking(n=450,428,454)
    156
    156
    205
        Baseline:Moderate problem in walk(n=1000,995,994)
    567
    579
    588
        Week 8:Moderate problem in walk(n=956,966,956)
    318
    266
    301
        Week 16:Moderate problem in walk(n=913,920,915)
    199
    196
    185
        Week 24:Moderate problem in walk(n=817,816,813)
    216
    200
    186
        Week 40:Moderate problem in walk(n=561,553,535)
    110
    106
    91
        Week 56:Moderate problem in walk(n=458,458,459)
    77
    120
    91
        Week 64:Moderate problem in walk(n=450,428,454)
    150
    151
    121
        Baseline:Severe problem in walk(n=1000,995,994)
    204
    202
    189
        Week 8:Severe problem in walk(n=956,966,956)
    41
    48
    44
        Week 16:Severe problem in walk(n=913,920,915)
    27
    34
    26
        Week 24:Severe problem in walk(n=817,816,813)
    34
    43
    29
        Week 40:Severe problem in walk(n=561,553,535)
    19
    27
    9
        Week 56:Severe problem in walk(n=458,458,459)
    19
    24
    9
        Week 64:Severe problem in walking(n=450,428,454)
    45
    54
    21
        Baseline:Unable to walk(n=1000,995,994)
    0
    2
    0
        Week 8:Unable to walk(n=956,966,956)
    0
    0
    3
        Week 16:Unable to walk(n=913,920,915)
    0
    0
    0
        Week 24:Unable to walk(n=817,816,813)
    0
    2
    1
        Week 40:Unable to walk(n=561,553,535)
    0
    0
    0
        Week 56:Unable to walk(n=458,458,459)
    0
    1
    0
        Week 64:Unable to walk(n=450,428,454)
    1
    1
    0
    No statistical analyses for this end point

    Secondary: Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Self-Care Domain

    		 Close Top of page
    End point title
    		 Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Self-Care Domain
    End point description
    Number of subjects with self- care domain responses of EQ-5D-5L were provided. EQ-5D-5L is a standardized subjects completed questionnaire that measures health-related quality of life and translates that score into an index value or utility score. EQ-5D-5L consists of two components: a health state profile and an optional VAS. EQ-5D health state profile is comprised of 5 dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: 1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, and 5=extreme problems. Higher scores indicated greater levels of problems across the five dimensions. Washing or dressing =W/D. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 24, 40, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline:No problems in W/D(n=1000,995,994)
    251
    242
    270
        Week 8:No problems in W/D(n=956,966,956)
    551
    569
    542
        Week 16:No problems in W/D(n=913,920,915)
    610
    597
    583
        Week 24:No problems in W/D(n=817,816,813)
    504
    504
    527
        Week 40:No problems in W/D(n=561,553,535)
    377
    359
    371
        Week 56:No problems in W/D(n=458,458,459)
    305
    294
    291
        Week 64:No problems in W/D(n=450,428,454)
    233
    192
    264
        Baseline:Slight problems in W/D(n=1000,995,994)
    315
    295
    319
        Week 8:Slight problems in W/D(n=956,966,956)
    270
    261
    276
        Week 16:Slight problems in W/D(n=913,920,915)
    216
    231
    246
        Week 24:Slight problems in W/D(n=817,816,813)
    214
    200
    192
        Week 40:Slight problems in W/D(n=561,553,535)
    140
    136
    125
        Week 56:Slight problems in W/D(n=458,458,459)
    107
    115
    122
        Week 64:Slight problems in W/D(n=450,428,454)
    142
    136
    131
        Baseline:Moderate problems in W/D(n=1000,995,994)
    361
    389
    350
        Week 8:Moderate problems in W/D(n=956,966,956)
    126
    128
    134
        Week 16:Moderate problems in W/D(n=913,920,915)
    81
    87
    77
        Week 24:Moderate problems in W/D(n=817,816,813)
    91
    102
    86
        Week 40:Moderate problems in W/D(n=561,553,535)
    42
    54
    38
        Week 56:Moderate problems in W/D(n=458,458,459)
    42
    47
    40
        Week 64:Moderate problems in W/D(n=450,428,454)
    66
    89
    57
        Baseline:Severe problems in W/D(n=1000,995,994)
    73
    69
    55
        Week 8:Severe problems inW/D(n=956,966,956)
    8
    8
    3
        Week 16:Severe problems in W/D(n=913,920,915)
    6
    5
    9
        Week 24:Severe problems in W/D(n=817,816,813)
    7
    9
    8
        Week 40:Severe problems in W/D(n=561,553,535)
    1
    4
    1
        Week 56:Severe problems in W/D(n=458,458,459)
    3
    2
    5
        Week 64:Severe problems in W/D(n=450,428,454)
    8
    11
    2
        Baseline:Unable to wash or dress(n=1000,995,994)
    0
    0
    0
        Week 8:Unable to wash or dress(n=956,966,956)
    1
    0
    1
        Week 16:Unable to wash or dress(n=913,920,915)
    0
    0
    0
        Week 24:Unable to wash or dress(n=817,816,813)
    1
    1
    0
        Week 40;Unable to wash or dress(n=561,553,535)
    1
    0
    0
        Week 56:Unable to wash or dress(n=458,458,459)
    1
    0
    1
        Week 64:Unable to wash or dress(n=450,428,454)
    1
    0
    0
    No statistical analyses for this end point

    Secondary: Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Usual Activities Domain

    		 Close Top of page
    End point title
    		 Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Usual Activities Domain
    End point description
    Number of subjects with usual activities domain responses of EQ-5D-5L were provided. EQ-5D-5L is a standardized subjects completed questionnaire that measures health-related quality of life and translates that score into an index value or utility score. EQ-5D-5L consists of two components: a health state profile and an optional VAS. EQ-5D health state profile is comprised of 5 dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: 1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, and 5=extreme problems. Higher scores indicated greater levels of problems across the five dimensions. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 24, 40, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline:No problems (n=1000,995,994)
    22
    24
    38
        Week 8:No problems (n=956,966,956)
    229
    266
    221
        Week 16:No problems (n=913,920,915)
    302
    333
    310
        Week 24:No problems(n=817,816,813)
    262
    290
    273
        Week 40:No problems(n=561,553,535)
    225
    221
    218
        Week 56:No problems(n=458,458,459)
    155
    170
    182
        Week 64:No problems(n=450,428,454)
    101
    69
    129
        Baseline:Slight problems (n=1000,995,994)
    229
    218
    225
        Week 8:Slight problems (n=956,966,956)
    402
    411
    426
        Week 16:Slight problems (n=913,920,915)
    402
    382
    408
        Week 24:Slight problems (n=817,816,813)
    353
    315
    344
        Week 40:Slight problems (n=561,553,535)
    239
    213
    233
        Week 56:Slight problems (n=458,458,459)
    211
    179
    199
        Week 64:Slight problems (n=450,428,454)
    173
    163
    197
        Baseline:Moderate problems (n=1000,995,994)
    538
    551
    561
        Week 8:Moderate problems (n=956,966,956)
    292
    256
    274
        Week 16:Moderate problems (n=913,920,915)
    184
    182
    172
        Week 24:Moderate problems (n=817,816,813)
    174
    182
    166
        Week 40:Moderate problems (n=561,553,535)
    85
    97
    74
        Week 56:Moderate problems (n=458,458,459)
    79
    86
    69
        Week 64:Moderate problems (n=450,428,454)
    138
    155
    115
        Baseline:Severe problems (n=1000,995,994)
    208
    201
    169
        Week 8:Severe problems (n=956,966,956)
    33
    31
    35
        Week 16:Severe problems (n=913,920,915)
    24
    21
    24
        Week 24:Severe problems (n=817,816,813)
    27
    27
    29
        Week 40:Severe problems (n=561,553,535)
    12
    20
    10
        Week 56:Severe problems (n=458,458,459)
    13
    22
    9
        Week 64:Severe problems (n=450,428,454)
    37
    37
    12
        Baseline:Unable to do activities(n=1000,995,994)
    3
    1
    1
        Week 8:Unable to do activities (n=956,966,956)
    0
    2
    0
        Week 16:Unable to do activities(n=913,920,915)
    1
    2
    1
        Week 24:Unable to do activities(n=817,816,813)
    1
    2
    1
        Week 40:Unable to do activities(n=561,553,535)
    0
    2
    0
        Week 56:Unable to do activities(n=458,458,459)
    0
    1
    0
        Week 64:Unable to do activities(n=450,428,454)
    1
    4
    1
    No statistical analyses for this end point

    Secondary: Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Pain/Discomfort Domain

    		 Close Top of page
    End point title
    		 Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Pain/Discomfort Domain
    End point description
    Number of subjects with pain/discomfort domain responses of EQ-5D-5L were provided. EQ-5D-5L is a standardized subject completed questionnaire that measures health-related quality of life and translates that score into an index value or utility score. EQ-5D-5L consists of two components: a health state profile and an optional VAS. EQ-5D health state profile is comprised of 5 dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: 1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, and 5=extreme problems. Higher scores indicated greater levels of problems across the five dimensions. Pain or discomfort = P/D. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 24, 40, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline:No P/D(n=1000,995,994)
    6
    4
    5
        Week 8:No P/D(n=956,966,956)
    82
    102
    83
        Week 16:No P/D(n=913,920,915)
    128
    163
    131
        Week 24:No P/D(n=817,816,813)
    117
    148
    130
        Week 40:No P/D(n=561,553,535)
    97
    122
    110
        Week 56:No P/D(n=458,458,459)
    76
    90
    85
        Week 64:No P/D(n=450,428,454)
    45
    35
    62
        Baseline:Slight P/D(n=1000,995,994)
    81
    75
    86
        Week 8:Slight P/D(n=956,966,956)
    433
    465
    434
        Week 16:Slight P/D(n=913,920,915)
    508
    482
    515
        Week 24:Slight P/D(n=817,816,813)
    413
    384
    413
        Week 40:Slight P/D(n=561,553,535)
    298
    264
    308
        Week 56:Slight P/D(n=458,458,459)
    248
    211
    259
        Week 64:Slight P/D(n=450,428,454)
    169
    115
    191
        Baseline:Moderate P/D(n=1000,995,994)
    548
    574
    588
        Week 8:Moderate P/D(n=956,966,956)
    369
    327
    365
        Week 16:Moderate P/D(n=913,920,915)
    235
    225
    217
        Week 24:Moderate P/D(n=817,816,813)
    213
    218
    215
        Week 40:Moderate P/D(n=561,553,535)
    139
    130
    104
        Week 56:Moderate P/D(n=458,458,459)
    111
    128
    103
        Week 64:Moderate P/D(n=450,428,454)
    165
    191
    171
        Baseline:Severe P/D(n=1000,995,994)
    334
    314
    295
        Week 8:Severe P/D(n=956,966,956)
    68
    68
    71
        Week 16:Severe P/D(n=913,920,915)
    39
    44
    46
        Week 24:Severe P/D(n=817,816,813)
    70
    62
    51
        Week 40:Severe P/D(n=561,553,535)
    25
    30
    13
        Week 56:Severe P/D(n=458,458,459)
    23
    26
    9
        Week 64:Severe P/D(n=450,428,454)
    66
    76
    29
        Baseline:Extreme P/D(n=1000,995,994)
    31
    28
    20
        Week 8:Extreme P/D(n=956,966,956)
    4
    4
    3
        Week 16:Extreme P/D(n=913,920,915)
    3
    6
    6
        Week 24:Extreme P/D(n=817,816,813)
    4
    4
    4
        Week 40:Extreme P/D(n=561,553,535)
    2
    7
    0
        Week 56:Extreme P/D(n=458,458,459)
    0
    3
    3
        Week 64:Extreme P/D(n=450,428,454)
    5
    11
    1
    No statistical analyses for this end point

    Secondary: Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Anxiety/ Depression Domain

    		 Close Top of page
    End point title
    		 Number of Subjects With Responses to European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L): Anxiety/ Depression Domain
    End point description
    Number of subjects with anxiety/depression domain responses of EQ-5D-5L were provided. EQ-5D-5L is a standardized subject completed questionnaire that measures health-related quality of life and translates that score into an index value or utility score. EQ-5D-5L consists of two components: a health state profile and an optional VAS. EQ-5D health state profile is comprised of 5 dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: 1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, and 5=extreme problems. Higher scores indicated greater levels of problems across the five dimensions. Anxious/depressed = A/D. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 24, 40, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline:Not A/D(n=1000,995,994)
    560
    570
    585
        Week 8:Not A/D(n=956,966,956)
    693
    703
    664
        Week 16:Not A/D(n=913,920,915)
    680
    701
    701
        Week 24:Not A/D(n=817,816,813)
    611
    606
    599
        Week 40:Not A/D(n=561,553,535)
    442
    429
    400
        Week 56:Not A/D(n=458,458,459)
    351
    330
    338
        Week 64:Not A/D(n=450,428,454)
    308
    275
    315
        Baseline:Slight A/D(n=1000,995,994)
    252
    235
    236
        Week 8:Slight A/D(n=956,966,956)
    189
    180
    206
        Week 16:Slight A/D(n=913,920,915)
    170
    147
    151
        Week 24:Slight A/D(n=817,816,813)
    147
    131
    144
        Week 40:Slight A/D(n=561,553,535)
    92
    82
    107
        Week 56:Slight A/D(n=458,458,459)
    88
    90
    86
        Week 64:Slight A/D(n=450,428,454)
    104
    96
    100
        Baseline:Moderate A/D(n=1000,995,994)
    155
    151
    144
        Week 8:Moderate A/D(n=956,966,956)
    64
    71
    75
        Week 16:Moderate A/D(n=913,920,915)
    53
    62
    53
        Week 24:Moderate A/D(n=817,816,813)
    52
    66
    58
        Week 40:Moderate A/D(n=561,553,535)
    24
    35
    21
        Week 56:Moderate A/D(n=458,458,459)
    18
    33
    34
        Week 64:Moderate A/D(n=450,428,454)
    29
    46
    34
        Baseline:Severe A/D(n=1000,995,994)
    28
    37
    26
        Week 8:Severe A/D(n=956,966,956)
    9
    7
    9
        Week 16:Severe A/D(n=913,920,915)
    8
    6
    8
        Week 24:Severe A/D(n=817,816,813)
    7
    10
    11
        Week 40:Severe A/D(n=561,553,535)
    3
    6
    7
        Week 56:Severe A/D(n=458,458,459)
    0
    3
    1
        Week 64:Severe A/D(n=450,428,454)
    8
    9
    5
        Baseline:Extreme A/D(n=1000,995,994)
    5
    2
    3
        Week 8:Extreme A/D(n=956,966,956)
    1
    5
    2
        Week 16:Extreme A/D(n=913,920,915)
    2
    4
    2
        Week 24:Extreme A/D(n=817,816,813)
    0
    3
    1
        Week 40:Extreme A/D(n=561,553,535)
    0
    1
    0
        Week 56:Extreme A/D(n=458,458,459)
    1
    2
    0
        Week 64:Extreme A/D(n=450,428,454)
    1
    2
    0
    No statistical analyses for this end point

    Secondary: European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L) Overall Health Utility Score/Index Value

    		 Close Top of page
    End point title
    		 European Quality of Life-5 Dimensions-5 Levels (EQ-5D-5L) Overall Health Utility Score/Index Value
    End point description
    EQ-5D-5L: Standardized subject completed questionnaire that measures health-related quality of life and translates that score into an index value or utility score. It consists of two components: a health state profile and an optional VAS. EQ-5D health state profile comprises of 5 dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has 5 levels: 1=no problems, 2=slight problems, 3=moderate problems, 4=severe problems, 5=extreme problems. Responses from the 5 domains were used to calculate single utility index (overall health utility score) where values are <= 1. Overall health utility score for a subject with no problems in all 5 items is 1 for all countries (except for Zimbabwe where it is 0.9), and is reduced where a subject reports greater levels of problems across the 5 dimensions. ITT population. Here, 'n'=subjects who were evaluable at specified time point.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 24, 40, 56 and 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n = 992, 985, 980)
    0.61 ( 0.14 )
    0.61 ( 0.14 )
    0.62 ( 0.13 )
        Week 8 (n = 948, 957, 942)
    0.74 ( 0.12 )
    0.75 ( 0.13 )
    0.74 ( 0.13 )
        Week 16 (n = 905, 912, 901)
    0.77 ( 0.12 )
    0.78 ( 0.13 )
    0.77 ( 0.13 )
        Week 24 (n = 809, 808, 799)
    0.76 ( 0.14 )
    0.76 ( 0.15 )
    0.77 ( 0.14 )
        Week 40 (n = 554, 545, 523)
    0.79 ( 0.13 )
    0.78 ( 0.15 )
    0.80 ( 0.12 )
        Week 56 (n = 453, 452, 449)
    0.78 ( 0.12 )
    0.77 ( 0.15 )
    0.79 ( 0.12 )
        Week 64 (n = 444, 421, 445)
    0.72 ( 0.15 )
    0.69 ( 0.16 )
    0.75 ( 0.13 )
    No statistical analyses for this end point

    Secondary: Treatment Satisfaction Questionnaire Medicine Version II (TSQM v.II) Satisfaction With Effectiveness, Side Effects, Convenience, and Overall Satisfaction Responses

    		 Close Top of page
    End point title
    		 Treatment Satisfaction Questionnaire Medicine Version II (TSQM v.II) Satisfaction With Effectiveness, Side Effects, Convenience, and Overall Satisfaction Responses
    End point description
    TSQM v.II is a self-administered 11-item validated scale that quantified subject's level of satisfaction with study medication (scored on a 7-point Likert scale [1= extremely dissatisfied, 2=very dissatisfied, 3=dissatisfied, 4=somewhat satisfied, 5=satisfied, 6=very satisfied, 7=extremely satisfied]) and dissatisfaction with side effects (3 questions scored on 5 point Likert scale [1= extremely dissatisfied, 2=very dissatisfied, 3=somewhat dissatisfied, 4=slightly dissatisfied, 5=not at all dissatisfied] and 1 question on 2 point scale [0 =No, 1=Yes]). Subjects were asked to assess their level of satisfaction. The 11 questions of the TSQM were used to calculate the 4 endpoints of effectiveness, side Effects, convenience and global satisfaction, each scored on a 0-100 scale with 100 being the best level of satisfaction. ITT population. Overall number of subjects analyzed’=subjects evaluable for this endpoint. n=subjects who were evaluable at specified timepoint.
    End point type
    Secondary
    End point timeframe
    Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    939
    954
    936
    Units: units on a scale
    least squares mean (standard error)
        Week 16: Effectiveness (n=939, 954, 936)
    64.26 ( 1.03 )
    66.27 ( 1.02 )
    61.61 ( 1.03 )
        Week 16: Side Effects (n=86, 94, 84)
    68.61 ( 3.20 )
    73.32 ( 3.09 )
    71.03 ( 3.15 )
        Week 16: Convenience (n=939, 954, 936)
    75.50 ( 0.80 )
    75.78 ( 0.79 )
    73.70 ( 0.80 )
        Week 16: Global Satisfaction (n=939, 954, 936)
    70.32 ( 0.99 )
    70.69 ( 0.98 )
    67.13 ( 0.99 )
        Week 56: Effectiveness (n=498, 487, 489)
    69.79 ( 1.38 )
    67.91 ( 1.36 )
    67.64 ( 1.38 )
        Week 56: Side Effects (n=30, 38, 24)
    78.62 ( 5.28 )
    62.00 ( 4.88 )
    71.34 ( 5.14 )
        Week 56: Convenience (n=498, 487, 489)
    78.03 ( 1.12 )
    77.67 ( 1.10 )
    76.18 ( 1.11 )
        Week 56: Global Satisfaction (n=498, 487, 489)
    75.31 ( 1.27 )
    73.37 ( 1.25 )
    73.37 ( 1.26 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Effectiveness; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0142
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    2.66
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.53
         upper limit
    		 4.78
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Effectiveness; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 < 0.0001
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    4.67
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 2.56
         upper limit
    		 6.78
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Effectiveness; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1371
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    2.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.69
         upper limit
    		 4.99
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.45
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Effectiveness; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8524
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.27
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -2.6
         upper limit
    		 3.14
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.46
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Side Effects; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5253
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -2.42
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -9.93
         upper limit
    		 5.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    3.8
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Side Effects; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5381
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    2.29
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -5.04
         upper limit
    		 9.62
    Variability estimate
    Standard error of the mean
    Dispersion value
    3.71
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Side Effects; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2694
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    7.27
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -5.99
         upper limit
    		 20.54
    Variability estimate
    Standard error of the mean
    Dispersion value
    6.44
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Side Effects; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1349
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    -9.34
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -21.8
         upper limit
    		 3.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    6.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Convenience; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0264
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    1.8
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.21
         upper limit
    		 3.38
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.81
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Convenience; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0098
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    2.07
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.5
         upper limit
    		 3.65
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.8
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Convenience; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0937
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    1.85
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.31
         upper limit
    		 4.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.1
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Convenience; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1838
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    1.48
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.7
         upper limit
    		 3.67
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Global Satisfaction; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0025
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    3.18
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.12
         upper limit
    		 5.25
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.05
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Global Satisfaction; Week 16: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0007
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    3.55
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 1.51
         upper limit
    		 5.6
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.04
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    TSQM Global Satisfaction; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1875
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1373
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    1.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -0.62
         upper limit
    		 4.51
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.31
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    TSQM Global Satisfaction; Week 56: ANCOVA model included treatment, randomization stratification variables (index joint, highest Kellgren-Lawrence grade and NSAID) as fixed effects, baseline diary average pain as covariates, and study site as a random effect.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1890
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.996
    Method
    		 ANCOVA
    Parameter type
    		 LS Mean Difference
    Point estimate
    0.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 -2.59
         upper limit
    		 2.6
    Variability estimate
    Standard error of the mean
    Dispersion value
    1.32

    Secondary: Patient Reported Treatment Impact Assessment-Modified (mPRTI) Score at Weeks 16 and 56: Subject Global Preference Assessment- What is The Current or Most Recent Treatment You Were Receiving for Osteoarthritis Pain Before Enrolling?

    		 Close Top of page
    End point title
    		 Patient Reported Treatment Impact Assessment-Modified (mPRTI) Score at Weeks 16 and 56: Subject Global Preference Assessment- What is The Current or Most Recent Treatment You Were Receiving for Osteoarthritis Pain Before Enrolling?
    End point description
    The mPRTI is a self-administered questionnaire containing subject's global preference assessment (to assess previous treatment and preference to continue using the investigational product) and subject's willingness to use drug again assessment. To assess current or most recent treatment, subjects responded for, 1=injectable prescription medicines, 2=prescription medicines taken by mouth, 3=surgery, 4=prescription medicines and surgery and 5=no treatment. Number of subjects who responded for the specified question were reported. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Week 16:Injectable pres med(n=939,954,936)
    99
    98
    82
        Week 56:Injectable pres med(n=498,487,489)
    44
    47
    40
        Week 16:Pres med taken by mouth(n=939,954,936)
    611
    633
    647
        Week 56:Pres med taken by mouth(n=498,487,489)
    307
    296
    324
        Week 16:Surgery(n=939,954,936)
    7
    7
    9
        Week 56:Surgery(n=498,487,489)
    8
    4
    2
        Week 16:Pres med and surgery(n=939,954,936)
    33
    28
    27
        Week 56:Pres medicines and surgery(n=498,487,489)
    20
    18
    20
        Week 16:No treatment(n=939,954,936)
    189
    188
    171
        Week 56:No treatment(n=498,487,489)
    119
    122
    103
    No statistical analyses for this end point

    Secondary: Patient Reported Treatment Impact Assessment-Modified (mPRTI) Score at Weeks 16 and 56: Subject Global Preference Assessment- Overall, do You Prefer the Drug That you Received in This Study to Previous Treatment?

    		 Close Top of page
    End point title
    		 Patient Reported Treatment Impact Assessment-Modified (mPRTI) Score at Weeks 16 and 56: Subject Global Preference Assessment- Overall, do You Prefer the Drug That you Received in This Study to Previous Treatment?
    End point description
    The mPRTI is a self-administered questionnaire containing suject reported treatment impact assessment (to assess subject satisfaction), subject global preference assessment (to assess previous treatment and preference to continue using the investigational product) and subject willingness to use drug again assessment. To assess preference to continue using the investigational product, subjects responded using interactive response technology (IRT) on a 5 point likert scale from 1-5, where, 1= yes, I definitely prefer the drug that I am receiving now, 2= I have a slight preference for the drug that I am receiving now, 3= I have no preference either way, 4= I have a slight preference for my previous treatment, 5= No, I definitely prefer my previous treatment. Higher scores indicate lesser preference to use the investigational product. ITT population was analyzed. n= subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: Subjects
        Week 16:Definitely prefer studyDrug(n=939,954,936)
    577
    597
    531
        Week 56:Definitely prefer studyDrug(n=498,487,489)
    342
    323
    302
        Week 16:Slightly prefer study Drug(n=939,954,936)
    141
    169
    158
        Week 56:Slightly prefer study Drug(n=498,487,489)
    70
    75
    89
        Week 16:No preference either way(n=939,954,936)
    149
    114
    164
        Week 56:No preference either way(n=498,487,489)
    61
    65
    71
        Week 16:Slightly prefer my PT (n=939,954,936)
    28
    34
    36
        Week 56:Slightly prefer my PT(n=498,487,489)
    16
    16
    13
        Week 16:No,definitely prefer my PT(n=939,954,936)
    44
    40
    47
        Week 56:No,definitely prefer my PT(n=498,487,489)
    9
    8
    14
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0823
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0049
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0718
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1947
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval

    Secondary: Patient Reported Treatment Impact Assessment-Modified (mPRTI) Score at Weeks 16 and 56: Subject Willingness to Use Drug Again Assessment- Willing to use the Same Drug That you Have Received in This Study for Your Osteoarthritis Pain?

    		 Close Top of page
    End point title
    		 Patient Reported Treatment Impact Assessment-Modified (mPRTI) Score at Weeks 16 and 56: Subject Willingness to Use Drug Again Assessment- Willing to use the Same Drug That you Have Received in This Study for Your Osteoarthritis Pain?
    End point description
    The mPRTI is a self-administered questionnaire containing subject reported treatment impact assessment (to assess subject satisfaction), subject global preference assessment (to assess previous treatment and preference to continue using the investigational product) and subject willingness to use drug again assessment. To assess subject willingness to use drug again, subjects responded using IRT on a 5 point likert scale from 1-5, where, 1= yes, I would definitely want to use the same drug again, 2= I might want to use the same drug again, 3= I am not sure, 4= I might not want to use the same drug again, 5= no, I definitely would not want to use the same drug again. Higher scores indicate lesser willingness to use the investigational product. ITT population included all randomized subjects who received at least one dose of SC study medication (either Tanezumab or matching placebo). Here, “n” = subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Week 16:Definitely want to use SDA(n=939,954,936)
    627
    641
    560
        Week 56:Definitely want to use SDA(n=498,487,489)
    352
    341
    310
        Week 16:Might want to use SDA(n=939,954,936)
    138
    154
    169
        Week 56:Might want to use SDA(n=498,487,489)
    78
    75
    97
        Week 16:I am not sure(n=939,954,936)
    108
    96
    134
        Week 56:I am not sure(n=498,487,489)
    54
    46
    58
        Week 16:Might not want to use SDA(n=939,954,936)
    19
    21
    23
        Week 56:Might not want to use SDA(n=498,487,489)
    4
    11
    12
        Week 16: Wouldn't want to use SDA(n=939,954,936)
    47
    42
    50
        Week 56:Wouldn't want to use SDA(n=498,487,489)
    10
    14
    12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0229
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0029
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0266
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1835
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval

    Secondary: Number of Subjects Who Withdrew Due to Lack of Efficacy

    		 Close Top of page
    End point title
    		 Number of Subjects Who Withdrew Due to Lack of Efficacy
    End point description
    Number of subjects who withdrew from treatment due to lack of efficacy have been reported here. ITT population included all randomized subejcts who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
    60
    63
    91
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline diary average pain, baseline WOMAC pain score, classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0076
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.63
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.45
         upper limit
    		 0.88
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    OR and 95% CI estimated from logistic regression model. Logistic regression model included baseline diary average pain, baseline WOMAC pain score, classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0187
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.67
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.48
         upper limit
    		 0.94

    Secondary: Time to Discontinuation Due to Lack of Efficacy

    		 Close Top of page
    End point title
    		 Time to Discontinuation Due to Lack of Efficacy
    End point description
    Time to discontinuation due to lack of efficacy was defined as the time interval from the date of first study drug administration up to the date of discontinuation of subject from treatment due to lack of efficacy. 99999 =Due to the Kaplan-Meier estimate not reaching the level for discontinuation due to lack of efficacy, median and upper limit could not be calculated. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, ‘Overall number of subjects analyzed’ signifies subjects who discontinued from the study due to lack of efficacy.
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    60
    63
    91
    Units: days
        median (full range (min-max))
    99999 (7 to 99999)
    99999 (14 to 99999)
    99999 (14 to 99999)
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Missing data for the selected percentile(s) was due to the Kaplan-Meier estimate not reaching the level for discontinuation due to lack of efficacy.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    151
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0074
    Method
    		 Logrank
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Missing data for the selected percentile(s) was due to the Kaplan-Meier estimate not reaching the level for discontinuation due to lack of efficacy.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    154
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0162
    Method
    		 Logrank
    Confidence interval

    Secondary: Number of Subjects who Took Rescue Medication During Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Number of Subjects who Took Rescue Medication During Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point description
    In case of inadequate pain relief, acetaminophen/paracetamol up to 3000 mg per day and up to 3 days in a week between baseline and Week 16, and 3000 mg per day and up to 7 days per week between Week 16 and 64 could be taken as rescue medication. Number of subjects with any use of rescue medication during the particular study week were summarized. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘n’=subjects who were evaluable at specified time point for each arm respectively.
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Week 2 (n= 998, 994, 993)
    567
    548
    527
        Week 4 (n= 1002, 995, 995)
    481
    437
    469
        Week 8 (n= 1002, 998, 996)
    433
    377
    418
        Week 16 (n= 1002, 998, 996)
    353
    330
    352
        Week 24 (n= 1002, 998, 996)
    372
    358
    384
        Week 32 (n= 1002, 998, 996)
    391
    380
    390
        Week 40 (n= 1002, 998, 996)
    391
    388
    388
        Week 48 (n= 1002, 998, 996)
    391
    393
    389
        Week 56 (n= 1002, 998, 996)
    391
    408
    397
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1136
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.15
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.97
         upper limit
    		 1.38
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.391
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.08
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.9
         upper limit
    		 1.29
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7691
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.03
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.86
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.143
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.88
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.73
         upper limit
    		 1.05
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6454
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.04
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.87
         upper limit
    		 1.25
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0561
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.84
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.7
         upper limit
    		 1
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9056
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.82
         upper limit
    		 1.19
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2919
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.9
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.75
         upper limit
    		 1.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4976
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.78
         upper limit
    		 1.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2425
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.9
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.75
         upper limit
    		 1.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9242
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.19
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6621
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.15
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9977
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.2
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9762
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.84
         upper limit
    		 1.2
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9718
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.19
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8219
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.22
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6936
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.8
         upper limit
    		 1.16
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Odds ratio and 95%CI estimated from logistic regression model. Logistic regression model included baseline WOMAC pain subscale, baseline diary average pain, and classification variables index joint, highest Kellgren-Lawrence grade, NSAID and treatment.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5769
    Method
    		 Regression, Logistic
    Parameter type
    		 Odds ratio (OR)
    Point estimate
    1.05
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.88
         upper limit
    		 1.26

    Secondary: Number of Subjects Who Took Rescue Medication During Week 64

    		 Close Top of page
    End point title
    		 Number of Subjects Who Took Rescue Medication During Week 64
    End point description
    In case of inadequate pain relief, after Week 16, acetaminophen/paracetamol up to 3000 mg per day up to 7 days in a week could be taken as rescue medication and use was reported weekly via diary. Number of subjects with any use of rescue medication during Week 64 were summarized. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Number of subjects analysed’ =subjects who were evaluable for this endpoint.
    End point type
    Secondary
    End point timeframe
    Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    437
    417
    437
    Units: subject
    251
    268
    215
    No statistical analyses for this end point

    Secondary: Number of Days of Rescue Medication Used During Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56

    		 Close Top of page
    End point title
    		 Number of Days of Rescue Medication Used During Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point description
    In case of inadequate pain relief during the treatment period, acetaminophen/paracetamol up to 3000 mg per day and up to 3 days in a week between baseline and Week 16, and 3000 mg per day and up to 7 days per week between Week 16 and 64 could be taken as rescue medication. Number of days the subjects used the rescue medication during the particular study weeks were summarized. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8, 16, 24, 32, 40, 48 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: days
    least squares mean (standard error)
        Week 2
    2.31 ( 0.13 )
    2.29 ( 0.13 )
    2.26 ( 0.13 )
        Week 4
    1.80 ( 0.12 )
    1.70 ( 0.11 )
    1.86 ( 0.12 )
        Week 8
    1.65 ( 0.12 )
    1.42 ( 0.11 )
    1.65 ( 0.12 )
        Week 16
    1.29 ( 0.11 )
    1.25 ( 0.10 )
    1.39 ( 0.11 )
        Week 24
    1.56 ( 0.12 )
    1.56 ( 0.13 )
    1.65 ( 0.13 )
        Week 32
    1.67 ( 0.13 )
    1.66 ( 0.13 )
    1.78 ( 0.13 )
        Week 40
    1.70 ( 0.13 )
    1.71 ( 0.13 )
    1.76 ( 0.13 )
        Week 48
    1.68 ( 0.13 )
    1.76 ( 0.14 )
    1.74 ( 0.13 )
        Week 56
    1.73 ( 0.13 )
    1.85 ( 0.14 )
    1.74 ( 0.13 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7746
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    1.02
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.07
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8441
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.07
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.658
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.97
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2279
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.91
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.78
         upper limit
    		 1.06
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.07
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9986
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    1
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.84
         upper limit
    		 1.18
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0771
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.86
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.72
         upper limit
    		 1.02
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.07
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.4485
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.93
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.77
         upper limit
    		 1.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.2817
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.9
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.74
         upper limit
    		 1.09
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5426
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5385
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.13
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 32: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5041
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.94
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.79
         upper limit
    		 1.12
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 32: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.441
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.93
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.78
         upper limit
    		 1.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.08
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 40: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7426
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.97
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 1.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 40: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7469
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.97
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 1.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 48: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6784
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.96
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.81
         upper limit
    		 1.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 48: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.8822
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    1.01
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.85
         upper limit
    		 1.21
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9119
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.99
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.83
         upper limit
    		 1.18
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.5148
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    1.06
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.89
         upper limit
    		 1.26
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.09

    Secondary: Number of Days of Rescue Medication Used During Week 64

    		 Close Top of page
    End point title
    		 Number of Days of Rescue Medication Used During Week 64
    End point description
    In case of inadequate pain relief, after week 16, acetaminophen/paracetamol up to 3000 mg per day up to 7 days in a week could be taken as rescue medication and use was reported weekly via diary. Number of days the subjects used the rescue medication during Week 64 were summarized. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here ‘Overall number of subjects analyzed’ = subjects who took rescue medication.
    End point type
    Secondary
    End point timeframe
    Week 64
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    437
    417
    437
    Units: days
        arithmetic mean (standard deviation)
    2.0 ( 2.38 )
    2.3 ( 2.46 )
    1.7 ( 2.26 )
    No statistical analyses for this end point

    Secondary: Amount of Rescue Medication Used During Weeks 2, 4, 8 and 16

    		 Close Top of page
    End point title
    		 Amount of Rescue Medication Used During Weeks 2, 4, 8 and 16
    End point description
    In case of inadequate pain relief, acetaminophen/paracetamol up to 3000 mg per day up to 3 days in a week could be taken as rescue medication. The total dosage of acetaminophen in milligrams used during the specified week were summarized. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Weeks 2, 4, 8 and 16
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: milligrams
    least squares mean (standard error)
        Week 2
    2880.3 ( 353.05 )
    2898.7 ( 358.72 )
    3310.5 ( 410.42 )
        Week 4
    2107.8 ( 302.24 )
    1946.5 ( 277.29 )
    2814.1 ( 401.29 )
        Week 8
    1995.6 ( 322.53 )
    1628.8 ( 258.90 )
    2839.7 ( 446.00 )
        Week 16
    1696.4 ( 307.80 )
    1581.6 ( 284.12 )
    2320.0 ( 413.14 )
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 2: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3348
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.87
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.66
         upper limit
    		 1.15
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 2: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.3595
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.88
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.66
         upper limit
    		 1.16
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0854
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.75
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.54
         upper limit
    		 1.04
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0281
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.69
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.5
         upper limit
    		 0.96
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.12
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0595
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.7
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.49
         upper limit
    		 1.01
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.13
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.003
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.57
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.4
         upper limit
    		 0.83
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.11
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1998
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1389
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.73
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.48
         upper limit
    		 1.11
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.15
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Analysis was performed using negative binomial model with model terms of baseline WOMAC Pain subscale score, baseline diary average pain score, index joint, Kellgren Lawrence grade, NSAID and treatment group.
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.0709
    Method
    		 Negative binomial model
    Parameter type
    		 LS Mean Ratio
    Point estimate
    0.68
    Confidence interval
         level
    		 95%
         sides
    2-sided
         lower limit
    		 0.45
         upper limit
    		 1.03
    Variability estimate
    Standard error of the mean
    Dispersion value
    0.14

    Secondary: Health Care Resource Utilization (HCRU): Number of Visits of Services Directly Related to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Visits of Services Directly Related to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Visits of services directly related to osteoarthritis evaluated were: visits to primary care physician, neurologist, rheumatologist, physician assistant or nurse practitioner, pain specialist, orthopedist, physical therapist, chiropractor, alternative medicine or therapy, podiatrist, nutritionist/dietitian, radiologist, home healthcare services and other practitioner. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). Here, “n” signifies those subjects who were evaluable at specified time point for this endpoint for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: visits
    median (full range (min-max))
        Baseline: Primary Care Physician (n = 442,433,436)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 101.0)
    1.0 (1.0 to 150.0)
        Baseline: Neurologist (n = 10, 15, 12)
    1.0 (1.0 to 120.0)
    1.0 (1.0 to 90.0)
    1.0 (1.0 to 10.0)
        Baseline: Rheumatologist (n = 83, 108, 114)
    1.0 (1.0 to 15.0)
    2.0 (1.0 to 111.0)
    2.0 (1.0 to 6.0)
        Baseline:Physician Assistant or Nurse(n=35,38,30)
    1.0 (1.0 to 6.0)
    1.0 (1.0 to 122.0)
    1.0 (1.0 to 120.0)
        Baseline: Pain Specialist (n = 42, 51, 41)
    1.0 (1.0 to 100.0)
    2.0 (1.0 to 111.0)
    1.0 (1.0 to 23.0)
        Baseline: Orthopedist (n = 171, 189, 160)
    2.0 (1.0 to 18.0)
    1.0 (1.0 to 190.0)
    2.0 (1.0 to 100.0)
        Baseline: Physical Therapist (n = 57, 62, 51)
    4.0 (1.0 to 100.0)
    3.0 (1.0 to 111.0)
    3.0 (1.0 to 32.0)
        Baseline: Chiropractor (n = 48, 33, 43)
    3.0 (1.0 to 190.0)
    3.0 (1.0 to 30.0)
    3.0 (1.0 to 24.0)
        Baseline:Alternative Medicine/Therapy(n=33,50,52)
    2.0 (1.0 to 24.0)
    2.0 (1.0 to 111.0)
    2.0 (1.0 to 80.0)
        Baseline: Podiatrist (n = 10, 20, 13)
    1.0 (1.0 to 2.0)
    1.0 (1.0 to 91.0)
    1.0 (1.0 to 2.0)
        Baseline: Nutritionist/Dietitian (n = 8, 11, 14)
    1.0 (1.0 to 3.0)
    1.0 (1.0 to 3.0)
    1.0 (1.0 to 4.0)
        Baseline: Radiologist (n = 139, 152, 138)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 111.0)
    1.0 (1.0 to 5.0)
        Baseline: Home Healthcare Services (n = 7, 5, 7)
    2.0 (1.0 to 24.0)
    1.0 (1.0 to 111.0)
    3.0 (1.0 to 91.0)
        Baseline: Other Practitioner (n = 82, 98, 91)
    2.0 (1.0 to 190.0)
    2.0 (1.0 to 111.0)
    2.0 (1.0 to 111.0)
        Week 64: Primary Care Physician (n=231, 264, 223)
    1.0 (1.0 to 190.0)
    1.0 (1.0 to 101.0)
    1.0 (1.0 to 120.0)
        Week 64: Neurologist (n = 17, 20, 13)
    1.0 (1.0 to 110.0)
    1.0 (1.0 to 4.0)
    1.0 (1.0 to 83.0)
        Week 64: Rheumatologist (n = 27, 32, 35)
    1.0 (1.0 to 4.0)
    1.0 (1.0 to 4.0)
    1.0 (1.0 to 3.0)
        Week 64:Physician Assistant or Nurse(n =30,36,27)
    1.0 (1.0 to 10.0)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 101.0)
        Week 64: Pain Specialist (n = 37, 43, 29)
    1.0 (1.0 to 190.0)
    1.0 (1.0 to 144.0)
    1.0 (1.0 to 180.0)
        Week 64: Orthopedist (n = 146, 163, 121)
    1.0 (1.0 to 190.0)
    1.0 (1.0 to 30.0)
    1.0 (1.0 to 100.0)
        Week 64: Physical Therapist (n = 69, 50, 37)
    4.0 (1.0 to 180.0)
    4.5 (1.0 to 111.0)
    3.0 (1.0 to 100.0)
        Week 64: Chiropractor (n = 32, 25, 22)
    3.0 (1.0 to 21.0)
    2.0 (1.0 to 30.0)
    3.0 (1.0 to 111.0)
        Week 64:Alternative Medicine/Therapy(n =33,45,46)
    1.0 (1.0 to 11.0)
    2.0 (1.0 to 10.0)
    2.0 (1.0 to 190.0)
        Week 64: Podiatrist (n = 11, 18, 12)
    1.0 (1.0 to 5.0)
    1.0 (1.0 to 3.0)
    1.0 (1.0 to 2.0)
        Week 64: Nutritionist/Dietitian (n = 9, 8, 6)
    2.0 (1.0 to 8.0)
    1.0 (1.0 to 3.0)
    1.0 (1.0 to 3.0)
        Week 64: Radiologist (n = 63, 79, 58)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 30.0)
    1.0 (1.0 to 100.0)
        Week 64: Home Healthcare Services (n = 5, 9, 6)
    4.0 (1.0 to 9.0)
    4.0 (1.0 to 16.0)
    5.0 (2.0 to 90.0)
        Week 64: Other Practitioner (n = 74, 106, 80)
    1.0 (1.0 to 190.0)
    1.0 (1.0 to 190.0)
    1.0 (1.0 to 111.0)
        Week 80: Primary Care Physician (n =134,118,110)
    1.0 (1.0 to 102.0)
    1.0 (1.0 to 111.0)
    1.0 (1.0 to 101.0)
        Week 80: Neurologist (n = 7, 6, 3)
    1.0 (1.0 to 1.0)
    1.0 (1.0 to 6.0)
    1.0 (1.0 to 2.0)
        Week 80: Rheumatologist (n = 16, 33, 15)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 111.0)
    1.0 (1.0 to 101.0)
        Week 80:Physician Assistant or Nurse(n =25,10,9)
    1.0 (1.0 to 6.0)
    1.0 (1.0 to 4.0)
    1.0 (1.0 to 90.0)
        Week 80: Pain Specialist (n = 16, 19, 15)
    1.5 (1.0 to 4.0)
    2.0 (1.0 to 111.0)
    1.0 (1.0 to 190.0)
        Week 80: Orthopedist (n = 86, 79, 44)
    1.5 (1.0 to 101.0)
    1.0 (1.0 to 101.0)
    1.0 (1.0 to 100.0)
        Week 80: Physical Therapist (n = 28, 32, 19)
    8.0 (1.0 to 30.0)
    5.5 (1.0 to 21.0)
    3.0 (1.0 to 21.0)
        Week 80: Chiropractor (n = 20, 18, 11)
    3.0 (1.0 to 111.0)
    4.5 (1.0 to 61.0)
    3.0 (1.0 to 120.0)
        Week 80:Alternative Medicine/Therapy (n=12,15,18)
    3.5 (1.0 to 90.0)
    1.0 (1.0 to 90.0)
    2.0 (1.0 to 90.0)
        Week 80: Podiatrist (n = 12, 6, 5)
    1.0 (1.0 to 90.0)
    1.0 (1.0 to 111.0)
    3.0 (1.0 to 101.0)
        Week 80: Nutritionist/Dietitian (n = 5, 2, 4)
    1.0 (1.0 to 100.0)
    1.5 (1.0 to 2.0)
    1.0 (1.0 to 2.0)
        Week 80: Radiologist (n = 39, 36, 36)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 4.0)
    1.0 (1.0 to 101.0)
        Week 80: Home Healthcare Services (n = 4, 3, 2)
    2.5 (1.0 to 140.0)
    4.0 (1.0 to 18.0)
    1.0 (1.0 to 1.0)
        Week 80: Other Practitioner (n = 40, 36, 24)
    1.0 (1.0 to 190.0)
    1.0 (1.0 to 100.0)
    1.0 (1.0 to 100.0)
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Number of Subjects who Visited the Emergency Room Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Subjects who Visited the Emergency Room Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was number of subjects who visited the emergency room due to osteoarthritis. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘n’=subjects who were evaluable at specified time point for this endpoint for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline (n = 1001, 997, 995)
    15
    23
    11
        Week 64 (n = 773, 782, 799)
    10
    15
    5
        Week 80 (n = 432, 396, 424)
    4
    5
    2
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Number of Visits to the Emergency Room Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Visits to the Emergency Room Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was number of visits to the emergency room due to OA. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Overall number of subjects analysed’ =subjects who were evaluable for this endpoint. 'n’=subjects who were evaluable at specified time point for this endpoint for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    15
    23
    11
    Units: visits
    median (full range (min-max))
        Baseline (n = 15, 23, 11)
    1.0 (1.0 to 3.0)
    1.0 (1.0 to 222.0)
    1.0 (1.0 to 6.0)
        Week 64 (n = 10, 15, 5)
    1.0 (1.0 to 2.0)
    1.0 (1.0 to 3.0)
    1.0 (1.0 to 2.0)
        Week 80 (n = 4, 5, 2)
    1.0 (1.0 to 2.0)
    3.0 (1.0 to 11.0)
    1.0 (1.0 to 1.0)
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Number of Subjects Hospitalized Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Subjects Hospitalized Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was number of subjects who were hospitalized due to OA. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). 'n’=subjects who were evaluable at specified time point for this endpoint for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline (n = 1001, 997, 995)
    11
    6
    1
        Week 64 (n = 773, 782, 799)
    5
    11
    6
        Week 80 (n = 432, 396, 424)
    8
    12
    0
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Number of Nights Stayed in the Hospital Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Nights Stayed in the Hospital Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was number of nights stayed in the hospital due to OA. Here, 99999 =median and full range could not be estimated as no subjects were analyzed. ITT population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Overall number of subjects analysed’ =subjects who were evaluable for this endpoint. 'n’=subjects who were evaluable at specified time point for this endpoint for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    11
    12
    6
    Units: Nights
    median (full range (min-max))
        Baseline (n = 11, 6, 1)
    12.0 (1.0 to 15.0)
    9.0 (1.0 to 14.0)
    11.0 (11.0 to 11.0)
        Week 64 (n = 5, 11, 6)
    2.0 (1.0 to 20.0)
    2.0 (1.0 to 40.0)
    2.0 (1.0 to 4.0)
        Week 80 (n = 8, 12, 0)
    2.0 (1.0 to 26.0)
    2.0 (1.0 to 21.0)
    99999 (99999 to 99999)
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Number of Subjects who Used Any Aids/Devices for Doing Things Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Subjects who Used Any Aids/Devices for Doing Things Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was number of subjects who used any aids/devices for doing things. Aids such as walking aid, wheelchair, device or utensil for dress/bathe/eat and any other aids/devices. Response for each aid/device usage was in terms of Never (Ne), Rarely (R), Sometimes (S), Often (O) and Always (A). Device/Utensil =D/U. The ITT population included all randomized subjects who received at least one dose of subcutaneous study medication (either Tanezumab or placebo). 'n’=subjects who were evaluable at specified time point for this endpoint for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Baseline:Walking Aid Use-Ne (n=1001,997,995)
    852
    838
    851
        Baseline:Wheelchair Use- Ne(n=1001,997,995)
    992
    989
    988
        Baseline:D/U toDressBatheEat-Ne(n=1001,997,995)
    970
    976
    977
        Baseline:Other Aids/Devices-Ne(n=1001,997,995)
    932
    935
    921
        Week 64:Walking Aid Use-Ne(n=773,782,799)
    662
    662
    714
        Week 64: Wheelchair Use Ne(n=773,782,799)
    765
    776
    794
        Week 64: D/U to Dress Bathe Eat-Ne(n=773,782,799)
    760
    768
    792
        Week 64: Other Aids Or Devices Ne(n=773,782,799)
    733
    720
    771
        Week 80: Walking Aid Use Ne(n=432,396,424)
    373
    322
    386
        Week 80: Wheelchair Use Ne(n=432,396,424)
    430
    389
    421
        Week 80: D/U to Dress Bathe Eat Ne(n=432,396,424)
    425
    383
    422
        Week 80: Other Aids or Devices Ne(n=432,396,424)
    416
    375
    410
        Baseline: Walking Aid Use R(n=1001,997,995)
    27
    18
    24
        Baseline: Wheelchair Use R(n=1001,997,995)
    2
    2
    1
        Baseline: D/U to Dress Bathe Eat R(n=1001,997,995)
    2
    1
    0
        Baseline: Other Aids Or Devices R(n=1001,997,995)
    8
    7
    9
        Week 64: Walking Aid Use R(n=773,782,799)
    21
    9
    12
        Week 64: Wheelchair Use R(n=773,782,799)
    2
    1
    2
        Week 64: D/U to Dress Bathe Eat R(n=773,782,799)
    2
    1
    1
        Week 64: Other Aids Or Devices R(n=773,782,799)
    6
    9
    11
        Week 80: Walking Aid Use R(n=432,396,424)
    12
    10
    6
        Week 80: Wheelchair Use R(n=432,396,424)
    0
    2
    2
        Week 80: D/U to Dress Bathe Eat R(n=432,396,424)
    0
    0
    1
        Week 80: Other Aids Or Devices R(n=432,396,424)
    4
    3
    4
        Baseline: Walking Aid Use S(n=1001,997,995)
    71
    69
    75
        Baseline: Wheelchair Use S(n=1001,997,995)
    4
    6
    5
        Baseline: D/U to Dress Bathe Eat S(n=1001,997,995)
    7
    6
    6
        Baseline: Other Aids Or Devices S(n=1001,997,995)
    27
    22
    41
        Week 64: Walking Aid Use S(n=773,782,799)
    48
    47
    37
        Week 64: Wheelchair Use S(n=773,782,799)
    4
    2
    3
        Week 64: D/U to Dress Bathe Eat S(n=773,782,799)
    7
    7
    2
        Week 64: Other Aids Or Devices S(n=773,782,799)
    19
    26
    8
        Week 80: Walking Aid Use S(n=432,396,424)
    25
    25
    17
        Week 80: Wheelchair Use S(n=432,396,424)
    1
    4
    1
        Week 80: D/U to Dress Bathe Eat S(n=432,396,424)
    2
    6
    1
        Week 80: Other Aids Or Devices S(n=432,396,424)
    5
    11
    4
        Baseline: Walking Aid Use O(n=1001,997,995)
    32
    43
    26
        Baseline: Wheelchair Use O(n=1001,997,995)
    3
    0
    1
        Baseline: D/U to Dress Bathe Eat O(n=1001,997,995)
    16
    9
    7
        Baseline: Other Aids Or Devices O(n=1001,997,995)
    27
    22
    14
        Week 64: Walking Aid Use O(n=773,782,799)
    20
    30
    17
        Week 64: Wheelchair Use O(n=773,782,799)
    2
    1
    0
        Week 64: D/U to Dress Bathe Eat O(n=773,782,799)
    3
    5
    2
        Week 64: Other Aids Or Devices O(n=773,782,799)
    12
    20
    6
        Week 80: Walking Aid Use O(n=432,396,424)
    14
    17
    7
        Week 80: Wheelchair Use O(n=432,396,424)
    0
    0
    0
        Week 80: D/U to Dress Bathe Eat O(n=432,396,424)
    3
    5
    0
        Week 80: Other Aids Or Devices O(n=432,396,424)
    5
    4
    2
        Baseline: Walking Aid Use A(n=1001,997,995)
    19
    29
    19
        Baseline: Wheelchair Use A(n=1001,997,995)
    0
    0
    0
        Baseline: D/U to Dress Bathe Eat A(n=1001,997,995)
    6
    5
    5
        Baseline: Other Aids Or Devices A(n=1001,997,995)
    7
    11
    10
        Week 64: Walking Aid Use A(n=773,782,799)
    22
    34
    19
        Week 64: Wheelchair Use A(n=773,782,799)
    0
    2
    0
        Week 64: D/U to Dress Bathe Eat A(n=773,782,799)
    1
    1
    2
        Week 64: Other Aids Or Devices A(n=773,782,799)
    3
    7
    3
        Week 80: Walking Aid Use A(n=432,396,424)
    8
    22
    8
        Week 80: Wheelchair Use A(n=432,396,424)
    1
    1
    0
        Week 80: D/U to Dress Bathe Eat A(n=432,396,424)
    2
    2
    0
        Week 80: Other Aids Or Devices A(n=432,396,424)
    2
    3
    4
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Number of Subjects who Quit Job Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Number of Subjects who Quit Job Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was number of subjects who quit job due to OA. ITT population was analyzed. Here ‘Overall number of subjects analyzed’=subjects evaluable for this endpoint. Here, “n” =subjects evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    995
    Units: subject
        Baseline (n =1001, 997, 995)
    47
    55
    65
        Week 64 (n =773, 782, 799)
    28
    35
    26
        Week 80 (n =432, 396, 424)
    12
    18
    6
    No statistical analyses for this end point

    Secondary: Health Care Resource Utilization (HCRU): Duration Since Quitting Job Due to Osteoarthritis

    		 Close Top of page
    End point title
    		 Health Care Resource Utilization (HCRU): Duration Since Quitting Job Due to Osteoarthritis
    End point description
    Osteoarthritis HCRU assessed healthcare usage during the last 3 months (for Baseline and Week 80) and past 8 weeks (for Week 64). Domain evaluated was duration since quitting job due to OA. ITT population was analyzed. One additional subject apart from the ones who had responded for quitting job responded to duration since quitting job. Here ‘Overall number of subjects analyzed’=subjects evaluable for this endpoint. Here, “n” =subjects evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    47
    55
    66
    Units: years
    median (full range (min-max))
        Baseline (n =47, 55, 66)
    2.0 (0.1 to 15.4)
    1.8 (0.3 to 20.0)
    2.4 (0.1 to 80.0)
        Week 64 (n =33, 38, 32)
    2.4 (0.2 to 18.6)
    1.8 (0.1 to 31.0)
    4.0 (0.1 to 90.0)
        Week 80 (n =12, 18, 6)
    2.0 (0.3 to 30.4)
    2.0 (0.1 to 50.0)
    1.8 (0.2 to 20.4)
    No statistical analyses for this end point

    Secondary: Number of Subjects With Categorical Change From Baseline in Lower Extremity Activity Scale (LEAS) at Weeks 4, 8, 16, 24, 56 and 80

    		 Close Top of page
    End point title
    		 Number of Subjects With Categorical Change From Baseline in Lower Extremity Activity Scale (LEAS) at Weeks 4, 8, 16, 24, 56 and 80
    End point description
    LEAS is a self-administered scale to assess activity level in subjects with total knee arthroplasty. LEAS scale reflected 4 levels of lower-extremity activity: housebound, more ordinary walking about the house, walking about community,walking about the community as well as substantial work or exercise.It consisted of 12 questions resulting in 18-level scale that allowed subject to select a single description that most represented his/her self-perceived activity level.Final score was number of descriptor selected by subjects as being most representative of his/her activity level. Minimum possible score was 1 (entirely bedbound) and maximum possible score was 18 (currently competitive athlete). Higher score indicated increased activity. Categorical changes from baseline were reported in terms of improvement (Change >0), No change and worsening (Change <0). ITT. Overall number of subjects analyzed=subjects evaluable for this endpoint. 'n' =subjects evaluable at specified time point.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 4, 8, 16, 24, 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    995
    994
    Units: subjects
        Change at Week 4, Improvement (n =1000,995,994)
    423
    421
    411
        Change at Week 8, Improvement (n =1000,995,994)
    454
    443
    445
        Change at Week 16, Improvement (n =1000,995,994)
    488
    470
    477
        Change at Week 24, Improvement (n =1000,995,994)
    478
    458
    467
        Change at Week 56, Improvement (n =1000,995,994)
    486
    429
    461
        Change at Week 80, Improvement (n =438,408,432)
    220
    196
    227
        Change at Week 4, No Change (n =1000,995,994)
    370
    394
    369
        Change at Week 8, No Change n =1000,995,994)
    325
    362
    348
        Change at Week 16, No Change (n =1000,995,994)
    288
    312
    312
        Change at Week 24, No Change (n =1000,995,994)
    277
    302
    291
        Change at Week 56, No Change (n =1000,995,994)
    270
    314
    300
        Change at Week 80, No Change (n =438,408,432)
    105
    97
    125
        Change at Week 4, Worsening (n =1000,995,994)
    207
    180
    214
        Change at Week 8, Worsening (n =1000,995,994)
    221
    190
    201
        Change at Week 16, Worsening (n =1000,995,994)
    224
    213
    205
        Change at Week 24, Worsening (n =1000,995,994)
    245
    235
    236
        Change at Week 56, Worsening (n =1000,995,994)
    244
    252
    233
        Change at Week 80, Worsening (n =438,408,432)
    113
    115
    80
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 4: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6037
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 4: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1928
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 8: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7204
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 8: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7969
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 16: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7857
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 16: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.6627
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 24: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.9867
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 24: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.819
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 2.5 mg Vs NSAID
    Statistical analysis description
    Week 56: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 2.5 mg v NSAID
    Number of subjects included in analysis
    1994
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.7284
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval
    Statistical analysis title
    		 Tanezumab 5 mg Vs NSAID
    Statistical analysis description
    Week 56: Cochran-Mantel-Haenszel test for row mean scores differ, stratified by the combinations of the three stratification factors (index joint, highest Kellgren-Lawrence grade and NSAID)
    Comparison groups
    Tanezumab 5 mg v NSAID
    Number of subjects included in analysis
    1989
    Analysis specification
    Pre-specified
    Analysis type
    		 superiority
    P-value
    		 = 0.1545
    Method
    		 Cochran-Mantel-Haenszel
    Confidence interval

    Secondary: Change From Baseline in Average Daily Minutes of Physical Activity at Weeks 16 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Daily Minutes of Physical Activity at Weeks 16 and 56
    End point description
    Subject activity level was assessed using actigraphy. Subjects continuously wore the accelerometer (apart for water activities) in the morning until going to bed at night for 7 or 14 consecutive days while going about their usual daily activities. Subjects maintained a log (electronic or written) to record when the accelerometer was put on in the morning and removed at night (or if removed for any other purpose). Accelerometry analysis set = All subjects treated with tanezumab or matching placebo SC who had any baseline or post-baseline accelerometry data. ‘Overall number of subjects analysed’ = subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    36
    46
    44
    Units: minutes
    median (full range (min-max))
        Baseline (n = 36, 46, 44)
    97.0 (22.6 to 164.9)
    107.1 (50.5 to 195.5)
    99.2 (13.9 to 176.9)
        Change at Week 16 (n = 20, 29, 24)
    3.9 (-39.5 to 60.1)
    2.9 (-49.4 to 45.1)
    -4.2 (-45.5 to 67.8)
        Change at Week 56 (n = 7, 8, 10)
    -8.9 (-63.5 to 16.9)
    -10.1 (-25.4 to 30.2)
    3.9 (-17.7 to 15.3)
    No statistical analyses for this end point

    Secondary: Change From Baseline in Average Daily Physical Activity Counts at Weeks 16 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Daily Physical Activity Counts at Weeks 16 and 56
    End point description
    An average daily physical activity count was measured using actigraphy. Subjects continuously wore the accelerometer (apart for water activities) in the morning until going to bed at night for 7 or 14 consecutive days while going about their usual daily activities. Subjects maintained a log (electronic or written) to record when the accelerometer was put on in the morning and removed at night (or if removed for any other purpose). Accelerometry analysis set = All subjects treated with tanezumab or matching placebo SC who had any baseline or post-baseline accelerometry data. ‘Overall number of subjects analysed’ = subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    36
    46
    44
    Units: physical activity counts
    median (full range (min-max))
        Baseline (n = 36, 46, 44)
    75244 (12385 to 129724)
    95911 (25816 to 428586)
    74414 (8136.6 to 253672)
        Change at Week 16 (n = 20, 29, 24)
    -470.0 (-32693 to 225384)
    -2261 (-114000 to 125895)
    1202.9 (-70004 to 300461)
        Change at Week 56 (n = 7, 8, 10)
    -14552 (-42956 to 24607)
    -8313 (-60556 to 53888)
    4414.3 (-15896 to 56451)
    No statistical analyses for this end point

    Secondary: Change From Baseline in Average Daily Minutes of Moderate to Vigorous Physical Activity at Weeks 16 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Daily Minutes of Moderate to Vigorous Physical Activity at Weeks 16 and 56
    End point description
    An average daily physical activity count was measured using actigraphy which was then sorted into three intensity thresholds: light (100 – less than {<1500} counts moderate (1,500 – <6500 counts), and vigorous (>=6500 counts).subjects continuously wore the accelerometer (apart for water activities) in the morning until going to bed at night for 7 or 14 consecutive days while going about their usual daily activities. Subjects maintained a log (electronic or written) to record when the accelerometer was put on in the morning and removed at night (or if removed for any other purpose). Accelerometry analysis set = All subjects treated with tanezumab or matching placebo SC who had any baseline or post-baseline accelerometry data. ‘Number of subjects analysed’ = subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    36
    46
    44
    Units: minutes
    median (full range (min-max))
        Baseline (n = 36, 46, 44)
    41.2 (2.2 to 80.4)
    53.1 (7.3 to 156.0)
    41.9 (0.7 to 117.0)
        Change at Week 16 (n = 20, 29, 24)
    0.7 (-25.3 to 62.7)
    -1.6 (-72.2 to 78.8)
    -0.1 (-39.2 to 68.7)
        Change at Week 56 (n = 7, 8, 10)
    -3.8 (-27.5 to 23.5)
    2.7 (-27.5 to 38.5)
    7.4 (-13.4 to 40.7)
    No statistical analyses for this end point

    Secondary: Change From Baseline in Average Daily Minutes of Bouted (Sustained) Moderate to Vigorous Physical Activity at Weeks 16 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Daily Minutes of Bouted (Sustained) Moderate to Vigorous Physical Activity at Weeks 16 and 56
    End point description
    An average daily physical activity count was measured using actigraphy which was then sorted into three intensity thresholds: light (100 – <1,500 counts moderate (1,500 – <6,500 counts), and vigorous (>=6,500 counts).subjects continuously wore the accelerometer (apart for water activities) in the morning until going to bed at night for 7 or 14 consecutive days while going about their usual daily activities. Subjects maintained a log (electronic or written) to record when the accelerometer was put on in the morning and removed at night (or if removed for any other purpose). A “bout” of moderate to vigorous activity was defined as 10 or more consecutive minutes above the moderate physical activity level threshold, with allowance for interruptions of 1 or 2 minutes below the threshold. Accelerometry analysis set was analysed. ‘Overall number of subjects analysed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    36
    46
    44
    Units: minutes
    median (full range (min-max))
        Baseline (n =36,46,44)
    0.0 (0.0 to 16.5)
    0.0 (0.0 to 111.5)
    0.0 (0.0 to 42.4)
        Change at Week 16 (n =20,29,24)
    0.0 (-13.1 to 25.9)
    0.0 (-73.3 to 13.0)
    0.0 (-12.6 to 64.8)
        Change at Week 56 (n =7,8,10)
    0.0 (-5.8 to 0.0)
    -1.4 (-11.5 to 8.2)
    0.0 (-2.0 to 6.5)
    No statistical analyses for this end point

    Secondary: Change From Baseline in Average Daily Step Count at Weeks 16 and 56

    		 Close Top of page
    End point title
    		 Change From Baseline in Average Daily Step Count at Weeks 16 and 56
    End point description
    Average daily step count was measured using actigraphy. Subjects continuously wore the accelerometer (apart for water activities) in the morning until going to bed at night for 7 or 14 consecutive days while going about their usual daily activities. Subjects maintained a log (electronic or written) to record when the accelerometer was put on in the morning and removed at night (or if removed for any other purpose). Accelerometry analysis set = All subjects treated with tanezumab or matching placebo SC who had any baseline or post-baseline accelerometry data. ‘Overall number of subjects analysed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 16 and 56
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    36
    46
    44
    Units: step count
    median (full range (min-max))
        Baseline (n = 36, 46, 44)
    4851.0 (827.6 to 9206.5)
    5834.8 (2041.0 to 17895)
    4779.0 (1015.6 to 11759)
        Change at Week 16 (n = 20, 29, 24)
    350.9 (-4970 to 5017.9)
    87.8 (-9415 to 5849.9)
    -705.7 (-3286 to 6270.4)
        Change at Week 56 (n = 7, 8, 10)
    -1938 (-4055 to 138.3)
    -543.2 (-1856 to 3609.9)
    242.6 (-4411 to 3312.2)
    No statistical analyses for this end point

    Secondary: Number of Subjects With Treatment-Emergent Adverse Events (AEs) and Serious Adverse Events (SAEs)

    		 Close Top of page
    End point title
    		 Number of Subjects With Treatment-Emergent Adverse Events (AEs) and Serious Adverse Events (SAEs)
    End point description
    An AE was any untoward medical occurrence in a subject who received study drug without regard to possibility of causal relationship. SAE was an AE resulting in any of the following outcomes or deemed significant for any other reason: death; initial or prolonged inpatient hospitalization; life-threatening experience (immediate risk of dying); persistent or significant disability/incapacity; congenital anomaly. Treatment-emergent were events between first dose of study drug and up to week 80 that were absent before treatment or that worsened relative to pre-treatment state. AEs included both serious and non-serious AEs. Clinically significant physical examination abnormalities were reported as AEs. Safety population included all randomised subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        AEs
    681
    744
    666
        SAEs
    78
    110
    66
    No statistical analyses for this end point

    Secondary: Number of Subjects With Treatment-Related Treatment-Emergent Adverse Events (AEs) and Serious Adverse Events (SAEs)

    		 Close Top of page
    End point title
    		 Number of Subjects With Treatment-Related Treatment-Emergent Adverse Events (AEs) and Serious Adverse Events (SAEs)
    End point description
    Treatment-related AE was any untoward medical occurrence attributed to study drug in a subject who received study drug. SAE was an AE resulting in any of the following outcomes or deemed significant for any other reason: death; initial or prolonged inpatient hospitalization; life-threatening experience (immediate risk of dying); persistent or significant disability/incapacity; congenital anomaly. Treatment-emergent were events between first dose of study drug and up to week 80 that were absent before treatment or that worsened relative to pre-treatment state. Relatedness to study drug was assessed by the investigator. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo).
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: subjects
        Treatment Related AEs
    190
    250
    179
        Treatment Related SAEs
    7
    20
    7
    No statistical analyses for this end point

    Secondary: Number of Subjects With Laboratory Test Abnormalities With Regard to Normal Baseline

    		 Close Top of page
    End point title
    		 Number of Subjects With Laboratory Test Abnormalities With Regard to Normal Baseline
    End point description
    Primary Abnormality criteria: HGB, hematocrit, RBC count <0.8* LLN; Ery. mean corpuscular volume/Hg/ HGB conc., RBCs distribution width <0.9*LLN, >1.1*ULN; platelets <0.5*LLN,>1.75*ULN; WBC count<0.6*LLN, >1.5*ULN; Lymphocytes,Leukocytes,Neutrophils <0.8*LLN, >1.2*ULN; Basophils,Eosinophils,Monocytes>1.2*ULN; Prothrombin time/Intl. normalized ratio>1.1*ULN; total bilirubin>1.5*ULN; aspartate aminotransferase,ALT,gamma GT,LDH,alkaline phosphatase >3.0*ULN; total protein; albumin<0.8*LLN, >1.2*ULN; blood urea nitrogen,creatinine,Cholesterol,triglycerides >1.3*ULN; Urate>1.2*ULN; sodium<0.95*LLN,>1.05*ULN; potassium,chloride,calcium,magnesium,bicarbonate <0.9*LLN, >1.1*ULN; phosphate<0.8*LLN, >1.2*ULN; glucose<0.6*LLN, >1.5*ULN; HGB A1C >1.3*ULN; creatine kinase>2.0*ULN, specific gravity<1.003, >1.030; pH<4.5, >8; Urine Glucose, protein,HGB,bilirubin >=1; Ketones>=1;Urine erythrocytes,Leukocytes>=20. Safety population. Number of subjects analysed=subjects evaluable for this endpoint.
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    877
    897
    884
    Units: Subjects
    109
    102
    121
    No statistical analyses for this end point

    Secondary: Number of Subjects With Laboratory Test Abnormalities With Regard to Abnormal Baseline

    		 Close Top of page
    End point title
    		 Number of Subjects With Laboratory Test Abnormalities With Regard to Abnormal Baseline
    End point description
    Primary Abnormality criteria: hemoglobin; hematocrit; RBC count < 0.8*LLN; Ery. mean corpuscular volume/ hemoglobin/ HGB concentration, erythrocytes distribution width <0.9*LLN, >1.1*ULN; platelets <0.5*LLN,>1.75*upper limit of normal (ULN); white blood cell count<0.6*LLN, >1.5*ULN; Lymphocytes, Leukocytes, Neutrophils <0.8*LLN, >1.2*ULN; Basophils, Eosinophils, Monocytes >1.2*ULN; total bilirubin>1.5*ULN; aspartate aminotransferase, alanine aminotransferase, gamma GT,LDH, alkaline phosphatase >3.0*ULN; total protein; albumin<0.8*LLN, >1.2*ULN; blood urea nitrogen, creatinine, Cholesterol, triglycerides >1.3*ULN; Urate >1.2*ULN; sodium <0.95*LLN,>1.05*ULN; potassium, chloride, calcium, magnesium, bicarbonate <0.9*LLN, >1.1*ULN; phosphate <0.8*LLN, >1.2*ULN; glucose <0.6*LLN, >1.5*ULN; Hemoglobin A1C >1.3*ULN; creatine kinase >2.0*ULN; Nitrite >=1. Safety population was analysed. ‘Number of subjects analysed’=subjects evaluable for this endpoint.
    End point type
    Secondary
    End point timeframe
    Baseline up to Week 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    706
    731
    691
    Units: subjects
    78
    61
    84
    No statistical analyses for this end point

    Secondary: Change From Baseline in Blood Pressure (BP) at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Blood Pressure (BP) at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point description
    Measurement of BP included sitting systolic blood pressure (SBP) and diastolic blood pressure (DBP). Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: millimeters of mercury (mmHg)
    arithmetic mean (standard deviation)
        SBP: Baseline (n = 1002, 998, 996)
    128.9 ( 12.91 )
    129.3 ( 13.43 )
    128.8 ( 13.26 )
        SBP: Change at Week 2 (n = 973, 971, 966)
    -2.7 ( 11.69 )
    -4.2 ( 11.92 )
    -1.2 ( 11.24 )
        SBP: Change at Week 4 (n = 967, 955, 949)
    -4.0 ( 11.44 )
    -4.9 ( 12.68 )
    -1.8 ( 11.29 )
        SBP: Change at Week 8 (n = 930, 926, 918)
    -2.9 ( 12.16 )
    -3.8 ( 12.60 )
    -1.8 ( 11.68 )
        SBP: Change at Week 16 (n = 635, 637, 601)
    -3.0 ( 11.66 )
    -3.7 ( 12.95 )
    -1.3 ( 12.37 )
        SBP: Change at Week 24 (n = 545, 542, 528)
    -3.0 ( 12.35 )
    -3.1 ( 13.49 )
    -1.7 ( 11.83 )
        SBP: Change at Week 32 (n = 518, 512, 505)
    -2.8 ( 11.95 )
    -3.3 ( 13.70 )
    -1.7 ( 13.49 )
        SBP: Change at Week 40 (n = 492, 474, 482)
    -2.5 ( 11.35 )
    -3.8 ( 14.41 )
    -2.3 ( 13.42 )
        SBP: Change at Week 48 (n = 471, 449, 459)
    -2.7 ( 12.21 )
    -3.0 ( 13.29 )
    -2.2 ( 12.97 )
        SBP: Change at Week 56 (n = 446, 421, 445)
    -3.1 ( 11.76 )
    -3.4 ( 13.78 )
    -2.2 ( 12.64 )
        SBP: Change at Week 64 (n = 430, 407, 435)
    -2.1 ( 13.40 )
    -2.1 ( 13.56 )
    -2.8 ( 13.33 )
        SBP: Change at Week 80 (n = 423, 390, 415)
    -1.0 ( 13.24 )
    -1.3 ( 13.71 )
    -2.3 ( 13.31 )
        DBP: Baseline (n = 1002, 998, 996)
    79.3 ( 8.56 )
    79.1 ( 8.68 )
    79.3 ( 8.57 )
        DBP: Change at Week 2 (n = 973, 971, 966)
    -1.3 ( 8.29 )
    -2.1 ( 7.96 )
    -1.1 ( 7.84 )
        DBP: Change at Week 4 (n = 967, 955, 949)
    -2.2 ( 8.06 )
    -2.5 ( 8.10 )
    -1.4 ( 8.19 )
        DBP: Change at Week 8 (n = 930, 926, 918)
    -1.1 ( 7.84 )
    -1.7 ( 8.24 )
    -1.1 ( 8.23 )
        DBP: Change at Week 16 (n = 635, 637, 601)
    -1.3 ( 8.14 )
    -1.8 ( 8.53 )
    -1.1 ( 8.30 )
        DBP: Change at Week 24 (n = 545, 542, 528)
    -1.3 ( 8.39 )
    -1.7 ( 8.91 )
    -1.4 ( 8.26 )
        DBP: Change at Week 32 (n = 518, 512, 505)
    -1.3 ( 8.28 )
    -1.4 ( 9.14 )
    -1.2 ( 8.91 )
        DBP: Change at Week 40 (n = 492, 474, 482)
    -1.2 ( 8.07 )
    -2.0 ( 8.90 )
    -1.1 ( 8.69 )
        DBP: Change at Week 48 (n = 471, 449, 459)
    -0.9 ( 8.82 )
    -1.8 ( 8.89 )
    -1.5 ( 8.65 )
        DBP: Change at Week 56 (n = 446, 421, 445)
    -1.8 ( 8.61 )
    -1.9 ( 9.11 )
    -1.2 ( 8.69 )
        DBP: Change at Week 64 (n = 430, 407, 435)
    -0.8 ( 8.63 )
    -0.8 ( 9.23 )
    -1.7 ( 9.03 )
        DBP: Change at Week 80 (n = 423, 390, 415)
    -0.6 ( 8.73 )
    -0.6 ( 9.66 )
    -1.2 ( 9.35 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Heart Rate at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Heart Rate at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point description
    Heart rate was measured at sitting position. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    998
    996
    Units: beats per minute
    arithmetic mean (standard deviation)
        Baseline (n =1002, 998, 996)
    70.8 ( 9.09 )
    70.5 ( 9.66 )
    70.6 ( 9.36 )
        Change at Week 2 (n =973, 971, 966)
    1.8 ( 8.65 )
    2.0 ( 8.50 )
    1.1 ( 8.31 )
        Change at Week 4 (n =967, 955, 949)
    1.6 ( 9.00 )
    2.0 ( 9.03 )
    1.2 ( 8.75 )
        Change at Week 8 (n =930, 926, 917)
    0.7 ( 9.06 )
    0.8 ( 8.54 )
    0.1 ( 8.78 )
        Change at Week 16 (n =635, 637, 601)
    0.5 ( 9.37 )
    0.5 ( 9.14 )
    0.8 ( 8.86 )
        Change at Week 24 (n =545, 542, 528)
    0.4 ( 9.46 )
    0.7 ( 9.36 )
    0.8 ( 9.24 )
        Change at Week 32 (n =518, 512, 505)
    1.2 ( 9.67 )
    1.6 ( 9.34 )
    1.7 ( 9.70 )
        Change at Week 40 (n =492, 474, 482)
    1.2 ( 9.89 )
    1.6 ( 9.40 )
    1.4 ( 8.71 )
        Change at Week 48 (n =471, 448, 458)
    0.6 ( 9.86 )
    1.0 ( 9.34 )
    1.3 ( 9.44 )
        Change at Week 56 (n =446, 421, 445)
    0.2 ( 9.51 )
    0.1 ( 10.15 )
    -0.0 ( 9.28 )
        Change at Week 64 (n =430, 407, 435)
    1.5 ( 9.81 )
    1.5 ( 9.55 )
    0.5 ( 9.84 )
        Change at Week 80 (n =423, 390, 415)
    0.9 ( 10.24 )
    0.6 ( 9.97 )
    0.9 ( 9.61 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Electrocardiogram (ECG) Parameters at Weeks 56 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Electrocardiogram (ECG) Parameters at Weeks 56 and 80
    End point description
    A 12-lead ECG was recorded after subjects had rested for at least 5 minutes in the supine position in a quiet environment. All standard intervals (PR, QRS, QT, QTcF, QTcB, RR intervals) were collected. ECG abnormalities included: 1) QT interval, QT interval corrected using Bazett's formula (QTcB) and QT interval corrected using Fridericia's formula (QTcF): increase from baseline greater than (>) 30 millisecond (ms) or 60 ms; absolute value > 450 ms, >480 ms and > 500 ms; 2) heart rate (HR): absolute value <=50 bpm and decrease from baseline >=20 bpm; absolute value >=120 beats per minute (bpm) and increase from baseline >=20 bpm; 3)PR interval: absolute value >=220 ms and increase from baseline >=20 ms; 4) QRS interval: absolute value >= 120 ms. Safety population included all subjects treated with tanezumab or placebo SC. ‘Overall number of subjects analyzed’=subjects evaluable for this endpoint. n=subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    995
    995
    Units: milliseconds
    arithmetic mean (standard deviation)
        RR Interval: Baseline (n = 1002,995,995)
    940.5 ( 136.21 )
    940.1 ( 144.48 )
    936.1 ( 142.31 )
        RR Interval: Change at Week 56 (n = 440,413,441)
    -26.3 ( 123.23 )
    -22.6 ( 128.58 )
    -14.9 ( 128.73 )
        RR Interval: Change at Week 80 (n = 419,381,410)
    -33.6 ( 119.23 )
    -32.4 ( 126.08 )
    -34.3 ( 133.83 )
        PR Interval: Baseline (n = 996,990,991)
    165.0 ( 27.43 )
    165.9 ( 25.45 )
    163.9 ( 23.98 )
        PR Interval: Change at Week 56 (n =435,410,434)
    1.7 ( 12.98 )
    0.6 ( 13.33 )
    1.7 ( 14.40 )
        PR Interval: Change at Week 80 (n =412,378,401)
    0.3 ( 13.39 )
    -0.8 ( 14.70 )
    0.6 ( 13.80 )
        QRS Interval: Baseline (n = 1002,995,995)
    94.9 ( 13.12 )
    94.6 ( 13.65 )
    94.3 ( 13.16 )
        QRS Interval: Change at Week 56 (n = 440,413,441)
    0.2 ( 8.22 )
    0.4 ( 8.30 )
    -0.4 ( 7.39 )
        QRS Interval: Change at Week 80 (n = 419,381,410)
    -0.2 ( 8.12 )
    1.0 ( 8.64 )
    -0.1 ( 7.81 )
        QT Interval: Baseline (n =999,994,994)
    405.0 ( 28.86 )
    403.8 ( 30.08 )
    404.3 ( 29.33 )
        QT Interval: Change at Week 56 (n =437,413,440)
    -3.5 ( 24.07 )
    -4.5 ( 25.06 )
    -2.9 ( 26.58 )
        QT Interval: Change at Week 80 (n =417,381,408)
    -6.2 ( 22.56 )
    -6.8 ( 24.46 )
    -6.0 ( 25.34 )
        QTCB Interval: Baseline (n =999,994,994)
    419.3 ( 21.85 )
    418.5 ( 21.96 )
    419.7 ( 21.60 )
        QTCB Interval: Change at Week 56 (n =437,413,440)
    2.3 ( 18.82 )
    0.5 ( 17.83 )
    0.2 ( 19.66 )
        QTCB Interval: Change at Week 80(n =417,381,408)
    1.5 ( 19.37 )
    0.2 ( 17.98 )
    1.7 ( 19.06 )
        QTCF Interval: Baseline (n =999,994,994)
    414.2 ( 19.96 )
    413.3 ( 20.09 )
    414.3 ( 19.49 )
        QTCF Interval: Change at Week 56 (n =437,413,440)
    0.3 ( 16.24 )
    -1.2 ( 15.55 )
    -0.8 ( 17.50 )
        QTCF Interval: Change at Week 80 (n =417,381,408)
    -1.2 ( 16.16 )
    -2.1 ( 15.55 )
    -1.0 ( 16.21 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Heart Rate (as Assessed by ECG) at Weeks 56 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Heart Rate (as Assessed by ECG) at Weeks 56 and 80
    End point description
    Heart rate was measured at sitting position. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Overall number of subjects analyzed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1002
    995
    995
    Units: beats per minute
    arithmetic mean (standard deviation)
        Baseline (n = 1002, 995, 995)
    65.2 ( 9.70 )
    65.4 ( 10.30 )
    65.6 ( 10.36 )
        Change at Week 56 (n = 440, 413, 441)
    2.0 ( 9.10 )
    1.7 ( 9.72 )
    1.0 ( 9.95 )
        Change at Week 80 (n = 419, 381, 410)
    2.7 ( 9.00 )
    2.3 ( 9.34 )
    2.5 ( 10.02 )
    No statistical analyses for this end point

    Secondary: Number of Subjects With Confirmed Orthostatic Hypotension

    		 Close Top of page
    End point title
    		 Number of Subjects With Confirmed Orthostatic Hypotension
    End point description
    Orthostatic hypotension was defined as postural change (supine to standing) that met the following criteria: For systolic BP <=150 mmHg (mean supine): Reduction in systolic BP>=20 mmHg or reduction in diastolic BP>=10 mmHg at the 1 and/or 3 minute standing BP measurements. For systolic BP >150 mmHg (mean supine): Reduction in systolic BP>=30 mmHg or reduction in diastolic BP>=15 mmHg at the 1 and/or 3 minute standing BP measurements. If the 1 minute or 3 minute standing BP in a sequence met the orthostatic hypotension criteria, then that sequence was considered positive. If 2 of 2 or 2 of 3 sequences were positive, then orthostatic hypotension was considered confirmed. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Overall number of subjects analysed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1001
    996
    996
    Units: subjects
        Baseline (n = 1001, 996, 996)
    0
    3
    1
        Week 2 (n = 956, 957, 954)
    2
    4
    2
        Week 4 (n = 949, 938, 937)
    1
    1
    2
        Week 8 (n =917, 918, 914)
    0
    2
    1
        Week 16 (n =677, 697, 654)
    1
    1
    0
        Week 24 (n =544, 538, 525)
    0
    1
    1
        Week 32 (n = 511, 503, 497)
    2
    2
    0
        Week 40 (n =488, 472, 480)
    1
    1
    1
        Week 48 (n =467, 447, 459)
    2
    2
    0
        Week 56 (n =441, 419, 442)
    1
    1
    1
        Week 64 (n =429, 402, 425)
    0
    1
    3
        Week 80 (n =420, 386, 414)
    0
    1
    0
    No statistical analyses for this end point

    Secondary: Change From Baseline in Survey of Autonomic Symptom (SAS) Scores at Weeks 24, 56 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Survey of Autonomic Symptom (SAS) Scores at Weeks 24, 56 and 80
    End point description
    The SAS is a 12 item (11 for females) questionnaire, from which the total number of symptoms (0-12 for males and 0-11 for females) is calculated. Each positive symptom is rated from 1 (not at all) to 5 (a lot). The total impact score was the sum of all symptom rating scores, with 0 assigned where the subject did not have the particular symptom. The range for the total impact score is 0-60 for males and 0-55 for females, higher scores indicating higher impact. Safety population included all randomized subjects who received at least one dose of SC study medication (either tanezumab or matching placebo). ‘Overall number of subjects analysed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively. Here, number is abbreviated as '#', symptom impact score abbreviated as 'SIS', Week abbreviated as 'W'
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 24, 56 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    997
    996
    Units: units on a score
    arithmetic mean (standard deviation)
        # of symptoms reported: Baseline(n=1000,997,996)
    0.47 ( 0.76 )
    0.53 ( 0.78 )
    0.49 ( 0.76 )
        # symptoms reported:Change at W24(n=546,543,532)
    0.21 ( 1.13 )
    0.18 ( 1.22 )
    0.11 ( 1.20 )
        # symptoms reported:Change at W56(n=448,421,443)
    0.28 ( 1.19 )
    0.33 ( 1.34 )
    0.22 ( 1.21 )
        # symptoms reported:Change at W80(n=423,391,414)
    0.89 ( 1.39 )
    0.94 ( 1.43 )
    0.74 ( 1.22 )
        Total SIS: Baseline(n=1000,997,996)
    1.10 ( 1.84 )
    1.23 ( 1.89 )
    1.13 ( 1.82 )
        Total SIS: Change at W24(n=546,543,532)
    0.66 ( 2.90 )
    0.52 ( 3.19 )
    0.33 ( 2.99 )
        Total SIS: Change at W56(n=448,421,443)
    0.97 ( 3.29 )
    1.21 ( 4.01 )
    0.82 ( 3.40 )
        Total SIS: Change at W80(n=423,391,414)
    1.33 ( 3.85 )
    1.31 ( 4.36 )
    0.89 ( 3.65 )
    No statistical analyses for this end point

    Secondary: Change From Baseline in Neuropathy Impairment Score (NIS) at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80

    		 Close Top of page
    End point title
    		 Change From Baseline in Neuropathy Impairment Score (NIS) at Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point description
    NIS is a standardized instrument used to evaluate subject for signs of peripheral neuropathy. NIS is the sum of scores of 37 items, from both the left and right side, where 24 items scored from 0 (normal) to 4 (paralysis), higher score indicated higher abnormality/impairment and 13 items scored from 0 (normal), 1 (decreased) and 2 (absent), higher score indicated higher impairment. NIS possible overall score ranged from 0 (no impairment) to 244 (maximum impairment), higher scores indicated increased impairment. Safety population included all randomized subjects who received at least one dose of SC study medication (either Tanezumab or matching placebo). ‘Overall number of subjects analysed’=subjects evaluable for this endpoint. ‘n’=subjects evaluable at specified time points for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 2, 4, 8, 16, 24, 32, 40, 48, 56, 64 and 80
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg NSAID
    Number of subjects analysed
    1000
    995
    994
    Units: units on a scale
    arithmetic mean (standard deviation)
        Baseline (n =1000, 995, 994)
    1.85 ( 4.30 )
    1.70 ( 4.45 )
    1.87 ( 4.47 )
        Change at Week 2 (n =971, 965, 951)
    -0.22 ( 2.15 )
    -0.13 ( 1.40 )
    -0.15 ( 1.67 )
        Change at Week 4 (n =986, 981, 977)
    -0.16 ( 2.06 )
    -0.17 ( 1.92 )
    -0.19 ( 2.32 )
        Change at Week 8 (n =988, 988, 982)
    -0.27 ( 2.28 )
    -0.22 ( 2.08 )
    -0.36 ( 2.55 )
        Change at Week 16 (n =989, 988, 982)
    -0.27 ( 2.32 )
    -0.31 ( 2.48 )
    -0.47 ( 3.06 )
        Change at Week 24 (n =989, 988, 982)
    -0.32 ( 2.39 )
    -0.35 ( 2.76 )
    -0.49 ( 3.08 )
        Change at Week 32 (n =989, 988, 982)
    -0.37 ( 2.46 )
    -0.40 ( 2.75 )
    -0.53 ( 3.17 )
        Change at Week 40 (n =989, 988, 982)
    -0.35 ( 2.42 )
    -0.43 ( 2.80 )
    -0.53 ( 3.32 )
        Change at Week 48 (n =989, 988, 982)
    -0.37 ( 2.46 )
    -0.49 ( 3.13 )
    -0.55 ( 3.21 )
        Change at Week 56 (n =989, 988, 982)
    -0.35 ( 2.48 )
    -0.52 ( 3.26 )
    -0.58 ( 3.22 )
        Change at Week 64 (n =989, 988, 982)
    -0.32 ( 2.60 )
    -0.47 ( 3.24 )
    -0.57 ( 3.20 )
        Change at Week 80 (n =989, 988, 982)
    -0.35 ( 2.53 )
    -0.47 ( 3.35 )
    -0.62 ( 3.28 )
    No statistical analyses for this end point

    Secondary: Number of Subjects With Anti-Tanezumab Antibodies

    		 Close Top of page
    End point title
    		 Number of Subjects With Anti-Tanezumab Antibodies [9]
    End point description
    Human serum anti-drug antibody (ADA) samples were analyzed for the presence or absence of anti-tanezumab antibodies by using a semi quantitative enzyme linked immunosorbent assay (ELISA). Safety population included all subjects treated with tanezumab or placebo SC. Here, “n” =subjects who were evaluable at specified time point for each arm, respectively.
    End point type
    Secondary
    End point timeframe
    Baseline, Weeks 8, 16, 32, 48, 56, 64 and 80
    Notes
    [9] - The end point is not reporting statistics for all the arms in the baseline period. It is expected all the baseline period arms will be reported on when providing values for an end point on the baseline period.
    Justification: No statistical analysis was planned for this endpoint.
    End point values
    		 Tanezumab 2.5 mg Tanezumab 5 mg
    Number of subjects analysed
    1002
    998
    Units: subjects
        Baseline (n=993, 985)
    116
    83
        Week 8 (n =923, 920)
    120
    93
        Week 16 (n =618, 616)
    98
    83
        Week 32 (n =512, 503)
    108
    81
        Week 48 (n =466, 444)
    96
    78
        Week 56 (n =436, 412)
    82
    66
        Week 64 (n =420, 390)
    69
    60
        Week 80 (n =414, 384)
    50
    42
    No statistical analyses for this end point

    Adverse events

    		 Close Top of page
    Adverse events information
    Timeframe for reporting adverse events
    Baseline up to Week 56
    Assessment type
    		 Non-systematic
    Dictionary used for adverse event reporting
    Dictionary name
    		 MedDRA
    Dictionary version
    21.1
    Reporting groups
    Reporting group title
    Tanezumab 2.5 mg
    Reporting group description
    		 Tanezumab (RN624 or PF-04383119) 2.5 milligram (mg) injection administered subcutaneously (SC) once every 8 weeks, from Baseline (Day 1) up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac extended release (ER), twice daily, from Baseline up to Week 56.

    Reporting group title
    NSAID
    Reporting group description
    		 Non-steroidal anti-inflammatory drug (NSAID) tablets (naproxen 500 mg, celecoxib 100 mg, or diclofenac ER 75 mg), administered orally, twice daily, from Baseline up to week 56 and placebo matched to tanezumab (RN624 or PF-04383119) injection administered SC once every 8 weeks, from Baseline up to Week 48.

    Reporting group title
    Tanezumab 5 mg
    Reporting group description
    		 Tanezumab (RN624 or PF-04383119) 5 mg injection administered SC once every 8 weeks, from Baseline up to Week 48 and oral placebo tablets matched to naproxen, celecoxib or diclofenac ER, twice daily, from Baseline up to week 56.

    Serious adverse events
    		 Tanezumab 2.5 mg NSAID Tanezumab 5 mg
    Total subjects affected by serious adverse events
         subjects affected / exposed
    51 / 1002 (5.09%)
    46 / 996 (4.62%)
    80 / 998 (8.02%)
         number of deaths (all causes)
    4
    1
    5
         number of deaths resulting from adverse events
    0
    0
    0
    Neoplasms benign, malignant and unspecified (incl cysts and polyps)
    Breast cancer
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cervix neoplasm
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Gastric cancer
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Invasive ductal breast carcinoma
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Lung adenocarcinoma
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
    Malignant melanoma
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Prostate cancer
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Rectal cancer
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Thyroid adenoma
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Thyroid neoplasm
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Vascular disorders
    Deep vein thrombosis
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Haematoma
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Hypertensive crisis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    1 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Phlebitis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    General disorders and administration site conditions
    Chest pain
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Non-cardiac chest pain
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Sensation of foreign body
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Immune system disorders
    Hypersensitivity
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Reproductive system and breast disorders
    Cervical cyst
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cystocele
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Ovarian cyst
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Uterine prolapse
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Respiratory, thoracic and mediastinal disorders
    Acute respiratory failure
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
    Chronic obstructive pulmonary disease
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 3
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cough
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pleural effusion
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pulmonary embolism
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
    Respiratory failure
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Psychiatric disorders
    Major depression
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Mental status changes
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Injury, poisoning and procedural complications
    Adjacent segment degeneration
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Ankle fracture
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Fall
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Forearm fracture
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Hip fracture
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Joint dislocation
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Limb crushing injury
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Meniscus injury
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    3 / 998 (0.30%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 3
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Road traffic accident
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Skin abrasion
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Skin laceration
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Ulna fracture
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cardiac disorders
    Acute myocardial infarction
         subjects affected / exposed
    1 / 1002 (0.10%)
    3 / 996 (0.30%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 3
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Aortic valve incompetence
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Atrial fibrillation
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Atrial flutter
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cardiac arrest
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
    Cardiac failure congestive
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Myocardial infarction
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 1
    0 / 0
    0 / 1
    Myocardial rupture
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
    Palpitations
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pericarditis
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Prinzmetal angina
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Ventricular tachycardia
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Nervous system disorders
    Brain stem infarction
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Carpal tunnel syndrome
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cerebrovascular accident
         subjects affected / exposed
    2 / 1002 (0.20%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 2
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Dizziness
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Loss of consciousness
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Presyncope
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Subarachnoid haemorrhage
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Syncope
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Transient ischaemic attack
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Blood and lymphatic system disorders
    Anaemia
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Eye disorders
    Cataract
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 2
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Entropion
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Retinal detachment
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Gastrointestinal disorders
    Colitis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Enteritis
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Gastric ulcer haemorrhage
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    1 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Gastrointestinal haemorrhage
         subjects affected / exposed
    0 / 1002 (0.00%)
    2 / 996 (0.20%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    2 / 2
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Large intestine polyp
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Obstructive pancreatitis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Hepatobiliary disorders
    Cholecystitis
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cholecystitis acute
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cholelithiasis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Skin and subcutaneous tissue disorders
    Rosacea
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Renal and urinary disorders
    Acute kidney injury
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    2 / 998 (0.20%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 2
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Ureterolithiasis
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Musculoskeletal and connective tissue disorders
    Arthralgia
         subjects affected / exposed
    4 / 1002 (0.40%)
    0 / 996 (0.00%)
    9 / 998 (0.90%)
         occurrences causally related to treatment / all
    0 / 4
    0 / 0
    5 / 10
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Arthritis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    1 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Back pain
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Intervertebral disc protrusion
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    2 / 998 (0.20%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 2
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Lumbar spinal stenosis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Musculoskeletal pain
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    1 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Osteoarthritis
         subjects affected / exposed
    9 / 1002 (0.90%)
    4 / 996 (0.40%)
    17 / 998 (1.70%)
         occurrences causally related to treatment / all
    2 / 10
    1 / 4
    2 / 18
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Osteonecrosis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    2 / 998 (0.20%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    1 / 2
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pain in extremity
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Rapidly progressive osteoarthritis
         subjects affected / exposed
    3 / 1002 (0.30%)
    0 / 996 (0.00%)
    11 / 998 (1.10%)
         occurrences causally related to treatment / all
    1 / 3
    0 / 0
    8 / 12
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Rotator cuff syndrome
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Spinal osteoarthritis
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Spondylolisthesis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Subchondral insufficiency fracture
         subjects affected / exposed
    1 / 1002 (0.10%)
    2 / 996 (0.20%)
    4 / 998 (0.40%)
         occurrences causally related to treatment / all
    1 / 1
    0 / 2
    4 / 4
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Thoracic spinal stenosis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Infections and infestations
    Appendicitis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Cellulitis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    2 / 998 (0.20%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 2
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Diverticulitis
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Endocarditis
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Escherichia pyelonephritis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Extradural abscess
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Influenza
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Intervertebral discitis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Localised infection
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Osteomyelitis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pelvic abscess
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    1 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Periorbital cellulitis
         subjects affected / exposed
    1 / 1002 (0.10%)
    0 / 996 (0.00%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 0
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pneumonia
         subjects affected / exposed
    1 / 1002 (0.10%)
    1 / 996 (0.10%)
    2 / 998 (0.20%)
         occurrences causally related to treatment / all
    0 / 1
    0 / 1
    0 / 2
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Pneumonia mycoplasmal
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Sepsis
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Testicular abscess
         subjects affected / exposed
    0 / 1002 (0.00%)
    1 / 996 (0.10%)
    0 / 998 (0.00%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 1
    0 / 0
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Wound infection
         subjects affected / exposed
    0 / 1002 (0.00%)
    0 / 996 (0.00%)
    1 / 998 (0.10%)
         occurrences causally related to treatment / all
    0 / 0
    0 / 0
    0 / 1
         deaths causally related to treatment / all
    0 / 0
    0 / 0
    0 / 0
    Frequency threshold for reporting non-serious adverse events: 5%
    Non-serious adverse events
    		 Tanezumab 2.5 mg NSAID Tanezumab 5 mg
    Total subjects affected by non serious adverse events
         subjects affected / exposed
    301 / 1002 (30.04%)
    264 / 996 (26.51%)
    318 / 998 (31.86%)
    Injury, poisoning and procedural complications
    Fall
         subjects affected / exposed
    65 / 1002 (6.49%)
    46 / 996 (4.62%)
    52 / 998 (5.21%)
         occurrences all number
    71
    48
    59
    Nervous system disorders
    Headache
         subjects affected / exposed
    56 / 1002 (5.59%)
    25 / 996 (2.51%)
    45 / 998 (4.51%)
         occurrences all number
    68
    30
    59
    Musculoskeletal and connective tissue disorders
    Arthralgia
         subjects affected / exposed
    131 / 1002 (13.07%)
    117 / 996 (11.75%)
    160 / 998 (16.03%)
         occurrences all number
    187
    148
    220
    Back pain
         subjects affected / exposed
    34 / 1002 (3.39%)
    34 / 996 (3.41%)
    55 / 998 (5.51%)
         occurrences all number
    37
    38
    59
    Infections and infestations
    Nasopharyngitis
         subjects affected / exposed
    57 / 1002 (5.69%)
    40 / 996 (4.02%)
    67 / 998 (6.71%)
         occurrences all number
    67
    50
    82
    Upper respiratory tract infection
         subjects affected / exposed
    57 / 1002 (5.69%)
    59 / 996 (5.92%)
    45 / 998 (4.51%)
         occurrences all number
    64
    70
    47

    More information

    		 Close Top of page

    Substantial protocol amendments (globally)
    Were there any global substantial amendments to the protocol? Yes
    Date
    Amendment
    15 May 2015
    Exclusion criteria updated to provide additional clarity regarding exclusion of subjects who have a history of heart block.

    Interruptions (globally)
    Were there any global interruptions to the trial? No

    Limitations and caveats
    Limitations of the trial such as small numbers of subjects analysed or technical problems leading to unreliable data.
    2 deaths occurred after end of study and are captured in adverse event section.
    For support, Contact us.
    The status and protocol content of GB trials is no longer updated since 1 January 2021. For the UK, as of 31 January 2021, EU Law applies only to the territory of Northern Ireland (NI) to the extent foreseen in the Protocol on Ireland/NI. Legal notice
    As of 31 January 2023, all EU/EEA initial clinical trial applications must be submitted through CTIS . Updated EudraCT trials information and information on PIP/Art 46 trials conducted exclusively in third countries continues to be submitted through EudraCT and published on this website.

    European Medicines Agency © 1995-Fri May 02 22:25:06 CEST 2025 | Domenico Scarlattilaan 6, 1083 HS Amsterdam, The Netherlands
    EMA HMA