A0081143

Pfizer Inc.

235 E 42nd Street, New York, United States, NY 10017

Pfizer ClinicalTrials.gov Call Center, Pfizer Inc., 001 800-718-1021, ClinicalTrials.gov_Inquiries@pfizer.com

Pfizer ClinicalTrials.gov Call Center, Pfizer Inc., 001 800-718-1021, ClinicalTrials.gov_Inquiries@pfizer.com

No

No

No

Final

22 Sep 2014

Yes

24 Jul 2013

Yes

24 Jul 2013

No

To compare the efficacy of pregabalin (300-600 milligram (mg)/day flexible dose) and gabapentin (1200-1800 mg/day flexible dose), both administered three times daily (TID), as adjunctive therapy in subjects with refractory partial seizures.

This study was conducted in compliance with the ethical principles originating in or derived from the Declaration of Helsinki and in compliance with all International Conference on Harmonization Good Clinical Practice (GCP) Guidelines. In addition, all local regulatory requirements were followed; in particular, those affording greater protection to the safety of study subjects.

12 Feb 2008

No

Yes

Portugal: 1

Romania: 32

Slovakia: 26

Spain: 4

Bulgaria: 86

Croatia: 15

Costa Rica: 32

China: 54

El Salvador: 24

Guatemala: 23

India: 84

Pakistan: 51

Peru: 19

Serbia: 31

482

164

0

0

0

0

0

0

474

8

0

Overall Period

Randomised - controlled

Yes

Pregabalin

PD-144,723

Capsule

Oral use

Pregabalin was initiated at 150 mg/day [50 mg capsules orally three times a day (TID)] for 1 Week followed by pregabalin 300 mg/day (100 mg TID) orally TID up to Week 5 in the Titration Phase (TP). Subjects who had adequate seizure control (adequate: >=50% reduction in seizures) with acceptable tolerability with pregabalin 300 mg/day in TP, continued same dose until the end of TP (Week 9) and then entered the maintenance phase (MP) on this dose. If seizure control was inadequate, the pregabalin dose was escalated to 450 mg/day orally (150 mg TID) from Weeks 5 through 9. If the subjects had adequate seizure control with acceptable tolerability on this dose, then they entered the MP on this dose. If there was inadequate seizure control at the end of Week 9, the dose was escalated a final time to 600 mg/day (200 mg TID) at which time they entered the MP.

Gabapentin

Capsule

Oral use

Gabapentin was initiated at 300 mg/day [100 mg capsules orally three times a day (TID)] followed by gabapentin 600 mg/day (200 mg TID) on Day 3. At the end of Week 1 (Day 7), the dose was increased to 1200 mg/day (400 mg TID) and remained on this dose for the next 4 weeks). Subjects who had adequate seizure control (>=50% reduction in seizure frequency) with acceptable tolerability during this initial 5-week period, remained on this dose until the end of theTP (Week 9), then entered the MP on this dose. If seizure control was inadequate, the gabapentin dose was escalated to 1500 mg/day (500 mg TID) during Weeks 5 through 9. If they had adequate seizure control with acceptable tolerability on this dose, they entered the MP on this dose. If there was inadequate seizure control at the end of Week 9, the dose was escalated a final time to 1800 mg/day (600 mg TID), at which time they entered the MP.

Pregabalin

Gabapentin

Pregabalin

Gabapentin

The seizures were recorded by the subjects, by a family member, by a caregiver, or by a legal guardian and documented in a daily seizure diary. Subject’s 28-day seizure frequency of all partial seizure was assessed during double blind (TP + MP) phase compared with baseline. Total partial seizure is defined as the total number of (simple partial seizure + complex partial seizure + secondary generalized tonic clonic seizure [SGTC]). Full analysis set (FAS) included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 baseline and post baseline primary efficacy evaluation.

Primary

6 weeks Baseline, 21 weeks through End of MP for 27 weeks

Rank analysis of covariance (ANCOVA) model was used to derive p-value with treatment as main effect and cluster as cofactor, percent change in 28-day seizure counts between Baseline and treatment periods as dependent variable. Median differences and 95% confidence interval (CI) were based on Hodges-Lehmann estimation.

Gabapentin v Pregabalin

478

Pre-specified

0

2-sided

Subjects who had at least 50% reduction in seizure frequency from Baseline to double-blind treatment (TP + MP) were considered as 50% responders. If percent change from baseline <= -50 then responder rate = 1 (yes) otherwise responder rate = 0 (no). FAS included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 Baseline and post Baseline primary efficacy evaluation. n= is the number of subjects that can be analyzed for each treatment group.

Secondary

6 weeks Baseline, 21 weeks through End of MP for 27 weeks

The odds ratio and its 95% CI are calculated by exponentiating the log odds ratio and 95% CI that correspond to the treatment contrast in the logistic regression model with treatment as fixed effect and Baseline term and country as covariate. All statistical tests for secondary efficacy parameters were two sided and performed at significance level of α = 0.05. The above statistical analysis applies to All Partial Seizures - FAS Population.

Gabapentin v Pregabalin

478

Pre-specified

0.92

2-sided

Subjects who had at least 75% reduction in seizure frequency from Baseline to double-blind treatment (TP + MP) were considered as 75% responders. If percent change from baseline <= -75 then 75% responder rate = 1 (yes) otherwise responder rate = 0 (no). Total partial seizure is defined as the total number of (simple partial seizure + complex partial seizure + SGTC). FAS included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 Baseline and post Baseline primary efficacy evaluation. n= is the number of subjects that can be analyzed for each treatment group.

Secondary

6 weeks Baseline, 21 weeks through End of MP for 27 weeks

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

478

Pre-specified

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

478

Pre-specified

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

478

Pre-specified

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

478

Pre-specified

Seizure free for 28 days was defined as subjects who have not experienced any seizure (simple partial, complex partial and SGTC) for at least 28 consecutive days from their last seizure until the end of the MP. Same subject could be seizure free for a specific type of seizure but not necessarily for the other types of seizure. FAS included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 Baseline and post Baseline primary efficacy evaluation. n= is the number of subjects that can be analyzed for each treatment group.

Secondary

6 weeks Baseline, 21 weeks through End of MP for 27 weeks

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

371

Pre-specified

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

371

Pre-specified

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

371

Pre-specified

p-value was calculated from the 2-sided Fisher’s exact test.

Pregabalin v Gabapentin

371

Pre-specified

Change in SGTC = (Proportion of SGTC/All Partial Seizure rate during at the double-blind phase) - (Proportion of SGTC/All Partial Seizure rate at Baseline). Negative values indicate reduction from baseline. SGTC population included all subjects who had at least 1 SGTC seizure during either Baseline or double-blind phase. n= number of subjects evaluable at specific time points for each arm group, respectively.

Secondary

6 weeks Baseline, 21 weeks through End of MP for 27 weeks

SGTC Responder is defined as a subject who shows reduction from Baseline to double-blind phase in proportion of 28-Day SGTC Seizure Rate to 28-Day All Partial Seizure Rate. SGTC population included all subjects who had at least 1 SGTC seizure during either Baseline or double-blind phase. n is the number of subjects analyzed for SGTC. Twenty-six subjects were not included in the n because they did not have a post Baseline all partial seizure, but they were included in the analysis by seizure type.

Secondary

6 weeks Baseline, 21 weeks through End of MP for 27 weeks

p-value is calculated using Fisher Exact Test.

Pregabalin v Gabapentin

202

Pre-specified

HADS: subject rated questionnaire with 2 subscales. Hospital Anxiety and Depression Scale - anxiety (HADS-A) assesses state of generalized anxiety (anxious mood, restlessness, anxious thoughts, panic attacks); Hospital Anxiety and Depression Scale - depression (HADS-D) assesses state of lost interest and diminished pleasure response (lowering of hedonic tone). Each subscale comprised of 7 items with range 0 (no presence of anxiety or depression) to 3 (severe feeling of anxiety or depression). Total score 0 to 21 for each subscale; higher score indicates greater severity of anxiety and depression symptoms. FAS included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 baseline and post baseline primary efficacy evaluation. n= number of subjects evaluable at specific time points for each arm group, respectively.

Secondary

Baseline, Week 21

ANCOVA model was used to calculate least square (LS) mean estimates of the treatment difference along with 95% CI, with main effects of treatment and and country as covariates.

Gabapentin v Pregabalin

478

Pre-specified

0.22

2-sided

Standard error of the mean

0.38

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country as covariates.

Pregabalin v Gabapentin

478

Pre-specified

0.09

2-sided

Standard error of the mean

0.32

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and combined center, and for post-baseline assessments baseline was included as a covariate.

Gabapentin v Pregabalin

478

Pre-specified

0.28

2-sided

Standard error of the mean

0.35

ANCOVA model was used to calculate LS mean estimates of the treatment1difference along with 95% CI, with main effects of treatment and combined center, and for post-baseline assessments baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

-0.17

2-sided

Standard error of the mean

0.29

Subject-rated 12-item questionnaire to assess constructs of sleep over past week; 7 subscales: sleep disturbance, snoring, awakened short of breath, sleep adequacy, somnolence (range: 0-100); sleep quantity (range: 0-24), optimal sleep (yes/no), and 9 item index measures of sleep disturbance provide composite scores: sleep problem summary, overall sleep problem. Except adequacy, optimal sleep and quantity, higher scores= more impairment. Scores transformed (actual raw score [RS] minus lowest possible score divided by possible RS range*100); total score range: 0-100; higher score= more intensity of attribute. FAS included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 baseline and post baseline primary efficacy evaluation. n= number of subjects evaluable at specific time points for each arm group, respectively.

Secondary

Baseline, Week 21

ANCOVA model was used to calculate least square (LS) mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

3.25

2-sided

Standard error of the mean

2.07

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Gabapentin v Pregabalin

478

Pre-specified

-0.32

2-sided

Standard error of the mean

1.68

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

1.19

2-sided

Standard error of the mean

2.82

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

1.96

2-sided

Standard error of the mean

2.3

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

4.03

2-sided

Standard error of the mean

2.52

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

-1.93

2-sided

Standard error of the mean

2.11

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

-0.03

2-sided

Standard error of the mean

0.14

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

0.02

2-sided

Standard error of the mean

0.2

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

-2.37

2-sided

Standard error of the mean

2.45

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

-0.66

2-sided

Standard error of the mean

2.05

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

2.98

2-sided

Standard error of the mean

1.9

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

2.06

2-sided

Standard error of the mean

1.66

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

3.45

2-sided

Standard error of the mean

1.61

ANCOVA model was used to calculate LS mean estimates of the treatment difference along with 95% CI, with main effects of treatment and country; for post-baseline, baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

0.34

2-sided

Standard error of the mean

1.28

MOS-SS: subject-rated 12 item questionnaire to assess constructs of sleep over past week. It included 7 subscales: sleep disturbance, snoring, awakened short of breath, sleep adequacy, somnolence, sleep quantity and optimal sleep. Subjects responded whether their sleep was optimal or not by choosing yes or no. Percentage of subjects with optimal sleep are reported. FAS included all randomized subjects who had received at least 1 blinded dose of study drug and had at least 1 baseline and post baseline primary efficacy evaluation. n= number of participants evaluable at specific time points for each arm group, respectively.

Secondary

Baseline, Week 21

Logistic regression model was used to estimate odds ratio and corresponding 95% CI of treatment difference, with treatment as fixed effect and for post-baseline assessments baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

0.811

2-sided

Logistic regression model was used to estimate odds ratio and corresponding 95% CI of treatment difference, with treatment as fixed effect and for post-baseline assessments baseline was included as a covariate.

Pregabalin v Gabapentin

478

Pre-specified

0.653

2-sided

Baseline to end of treatment

The same event may appear as both an adverse event (AE) and a serious adverse event (SAE). However, what is presented are distinct events. An event may be categorized as serious in one subject and as nonserious in another subject, or one subject may have experienced both a serious and nonserious event during the study.

16.1

Gabapentin

Pregabalin