Download PDF

Clinical Trial Results:
Five year observational follow-up of the IVAN trial cohort: a study of function and morphology

Summary
EudraCT number	2015-002608-97
Trial protocol	GB
Global end of trial date	01 Dec 2017

Results information
Results version number	v1(current)
This version publication date	24 Mar 2019
First version publication date	24 Mar 2019
Other versions

Trial Information
Subject Disposition
Baseline Characteristics
End Points
Adverse Events
More Information

Collapse all Expand all

Trial information

Top of page

Sponsor protocol code

14206UC-AC

ISRCTN number

ISRCTN92166560

US NCT number

WHO universal trial number (UTN)

Sponsor organisation name

Belfast Health & Social Care Trust

Sponsor organisation address

Research Office, 2nd Floor King Edward Building, Royal Hospitals, Grosvenor Road, Belfast, United Kingdom, BT12 6BA

Public contact

Usha Chakravarthy, Belfast Health and Social Care Trust, +44 02890632527, u.chakravarthy@qub.ac.uk

Scientific contact

Usha Chakravarthy, Belfast Health and Social Care Trust, +44 02890632527, alison.murphy@belfasttrust.hscni.net

Is trial part of an agreed paediatric investigation plan (PIP)

Does article 45 of REGULATION (EC) No 1901/2006 apply to this trial?

Does article 46 of REGULATION (EC) No 1901/2006 apply to this trial?

Analysis stage

Final

Date of interim/final analysis

19 Feb 2019

Is this the analysis of the primary completion data?

Yes

Primary completion date

01 Dec 2017

Global end of trial reached?

Yes

Global end of trial date

01 Dec 2017

Was the trial ended prematurely?

Main objective of the trial

To investigate the care of IVAN participants in the NHS, and changes to their vision, since the end of the IVAN trial in order to better inform future NHS strategies for treatment of wet Age-related Macular Degeneration. To compare outcomes by original randomised allocations at 5-7 years after randomisation.

Protection of trial subjects

Participants were in an observational study of usual care during a period after IMP was discontinued and the trial was closed. This was a follow up study of long term outcomes and participants were not put at any risk.

Background therapy

Not applicable- see above

Evidence for comparator

Not applicable- see above

Actual start date of recruitment

26 May 2016

Long term follow-up planned

Independent data monitoring committee (IDMC) involvement?

Number of subjects enrolled per country

Country: Number of subjects enrolled

United Kingdom: 532

Worldwide total number of subjects

532

EEA total number of subjects

532

Number of subjects enrolled per age group

In utero

Preterm newborn - gestational age < 37 wk

Newborns (0-27 days)

Infants and toddlers (28 days-23 months)

Children (2-11 years)

Adolescents (12-17 years)

Adults (18-64 years)

From 65 to 84 years

384

85 years and over

129

Subject disposition

Top of page

Recruitment details

All participants who did not die or withdraw from the IVAN trial were invited to participate in the follow up study by either attending a single research visit in person, for eye examinations and a questionnaire, or completing a questionnaire by post. They were also be offered the option to withdraw from the study.

Screening details

Of the 610 patients recruited to the IVAN trial, 532 were included in the follow-up study; 66 died or withdrew during the IVAN trial, 7 were at a single site that did not participate in the extended IVAN follow up and 5 withdrew during the extended follow up. 124 of the 532 participants had died by the time of follow up.

Period 1 title

Follow-up cohort (overall period)

Is this the baseline period?

Yes

Allocation method

Not applicable

Blinding used

Not blinded

Blinding implementation details

The participants were initially randomised to one of four treatment regimens in the original IVAN trial, and participants, clinical and trial personnel were masked. At the start of this study, after exiting the trial, participants were treated in the NHS. Although health professionals and participants have not been masked to ongoing care, ophthalmologists managing participants are extremely unlikely to know the original experimental allocation

Are arms mutually exclusive

Follow-up cohort

Arm description

All participants in the extended follow-up study Of the 532 participants in the extended follow-up study, 124 had died and 199 agreed to attend a specific research visit. Data were collected passively for non-attenders (124 deceased patients and 209 who did not wish to attend a research visit)

Arm type

Overall

Investigational medicinal product name

No investigational medicinal product assigned in this arm

IVAN randomised allocation: Ranibizumab

Arm description

This is an observational follow up of the original participants who were enrolled in the IVAN trial. The participants were initially randomised to one of four treatment regimens in the original IVAN trial and after exiting the trial, participants were treated in the NHS. The follow up study looked at the long term outcomes of these patients and is purely observational in that there was data collection but no intervention. This arm is the treatment allocation of the original IVAN trial.

Arm type

IVAN trial allocation (reference group)

Investigational medicinal product name

Lucentis

Investigational medicinal product code

Other name

Ranibizumab

Pharmaceutical forms

Injection

Routes of administration

Intravitreal use

Dosage and administration details

Drug dose during IVAN trial: 0.5 mg

IVAN randomised allocation: Bevacizumab

Arm description

Arm type

IVAN trial allocation

Investigational medicinal product name

Avastin

Investigational medicinal product code

Other name

Bevacizumab

Pharmaceutical forms

Injection

Routes of administration

Intravitreal use

Dosage and administration details

Drug dose during IVAN trial: 1.25 mg

IVAN randomised allocation: Continuous

Arm description

Arm type

IVAN trial allocation (reference group)

Investigational medicinal product name

No investigational medicinal product assigned in this arm

IVAN randomised allocation: Discontinuous

Arm description

Arm type

IVAN trial allocation

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Intralesional macular atrophy (ILMA) absent in study eye

Arm description

Intralesional macular atrophy (ILMA) absent in study eye at extended follow up. Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA)) during extended follow up.

Arm type

ILMA (reference)

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Intralesional macular atrophy (ILMA) present in study eye

Arm description

Intralesional macular atrophy (ILMA) present in study eye at extended follow up Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA)) during extended follow up.

Arm type

ILMA

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Geographic atrophy (GA) absent in study eye

Arm description

Geographic atrophy (GA) absent in study eye at extended follow up Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA)) during extended follow up.

Arm type

GA (reference)

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Geographic atrophy (GA) present in study eye

Arm description

Geographic atrophy (GA) present in study eye at extended follow up Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA)) during extended follow up.

Arm type

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Developed/worsened intralesional macular atrophy (ILMA)

Arm description

Defined as the development (incident) or expansion of the area (worsened) of ILMA. Expansion of the area of ILMA was quantified and was classified as worsened when it increased by >20%

Arm type

ILMA

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Did not develop/worsen intralesional macular atrophy (ILMA)

Arm description

Defined as the development (incident) or expansion of the area (worsened) of ILMA. Expansion of the area of ILMA was quantified and was classified as worsened when it increased by >20%

Arm type

ILMA

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Developed/worsend geographic atrophy (GA)

Arm description

Defined as the development (incident) or expansion of the area (worsened) of GA. Expansion of the area of GA was quantified and was classified as worsened when it increased by >20%.

Arm type

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Did not develop/worsen geographic atrophy (GA)

Arm description

Defined as the development (incident) or expansion of the area (worsened) of GA. Expansion of the area of GA was quantified and was classified as worsened when it increased by >20%.

Arm type

Investigational medicinal product name

No investigational medicinal product assigned in this arm

Number of subjects in period 1	Follow-up cohort	IVAN randomised allocation: Ranibizumab	IVAN randomised allocation: Bevacizumab	IVAN randomised allocation: Continuous	IVAN randomised allocation: Discontinuous	Intralesional macular atrophy (ILMA) absent in study eye	Intralesional macular atrophy (ILMA) present in study eye	Geographic atrophy (GA) absent in study eye	Geographic atrophy (GA) present in study eye	Developed/worsened intralesional macular atrophy (ILMA)	Did not develop/worsen intralesional macular atrophy (ILMA)	Developed/worsend geographic atrophy (GA)	Did not develop/worsen geographic atrophy (GA)
Started	532	272	260	269	263	122	346	309	159	243	138	142	311
Completed	532	272	260	269	263	122	346	309	159	243	138	142	311

Baseline characteristics

Top of page

Reporting group title

Follow-up cohort

Reporting group description

Reporting group values	Follow-up cohort	Total
Number of subjects	532	532
Age categorical
Age at IVAN exit
Units: Subjects
Adults (18-64 years)	19	19
From 65-84 years	384	384
85 years and over	129	129
Age continuous
Age at IVAN exit
Units: years
arithmetic mean (standard deviation)	79.4 ( 7.4 )	-
Gender categorical
Units: Subjects
Female	320	320
Male	212	212
Drug
Drug allocation during IVAN trial
Units: Subjects
Ranibizumab	272	272
Bevacizumab	260	260
Drug frequency
Treatment regimen during IVAN trial
Units: Subjects
Continuous	269	269
Discontinuous	263	263
Angina at IVAN entry
Units: Subjects
Yes	72	72
No	460	460
Dyspnoea at IVAN entry
Units: Subjects
Yes	97	97
No	433	433
Missing	2	2
Dyspnoea at IVAN exit
Units: Subjects
Yes	106	106
No	424	424
Missing	2	2
Myocardial infarction at IVAN entry
Units: Subjects
Yes	37	37
No	495	495
Transient ischemic attack at IVAN entry
Units: Subjects
Yes	26	26
No	477	477
Missing	29	29
Stroke at IVAN entry
Units: Subjects
Yes	8	8
No	524	524
DVT/PE at IVAN entry
DVT = deep vein thrombosis; PE =pulmonary embolism
Units: Subjects
Yes	30	30
No	501	501
Missing	1	1
Current or past smoker at IVAN entry
Units: Subjects
Yes	331	331
No	197	197
Missing	4	4
Systolic BP at IVAN entry
BP = Blood pressure
Units: mmHg
arithmetic mean (standard deviation)	142.4 ( 19.2 )	-
Diastolic BP at IVAN entry
BP = Blood pressure
Units: mmHg
arithmetic mean (standard deviation)	77.0 ( 10.2 )	-
Systolic BP at IVAN exit
BP = Blood pressure Data missing for 8 patients (3 Attenders, 1 Survivor, passive data collection only, 4 Deceased, passive data collection only)
Units: mmHG
arithmetic mean (standard deviation)	138.5 ( 18.9 )	-
Diastolic BP at IVAN exit
BP = Blood pressure Data missing for 8 patients (3 Attenders, 1 Survivor, passive data collection only, 4 Deceased, passive data collection only)
Units: mmHg
arithmetic mean (standard deviation)	74.3 ( 10.2 )	-

Subject analysis set title

Attenders

Subject analysis set type

Sub-group analysis

Subject analysis set description

Participants who attended a research visit as part of the IVAN follow-up study. 196/199 (98.5%) have at least one ophthalmology appointment as part of usual care since IVAN exit Length of follow up during usual care since IVAN entry (years, Median IQR): 6.8 (6.1, 7.4) Length of follow up during usual care since IVAN exit (years, Median IQR): 4.8 (4.2, 5.4) Length of follow up to research appointment since IVAN entry (years, Median IQR): 7.3 (6.8, 7.8) Length of follow up to research appointment since IVAN exit (years, Median IQR): 5.3 (4.8, 5.8)

Subject analysis set title

Survivor, passive data collection only

Subject analysis set type

Sub-group analysis

Subject analysis set description

Non-attenders (passive data collection only) - survivors 198/209 (94.7%) have at least one ophthalmology appointment as part of usual care since IVAN exit Length of follow up during usual care since IVAN entry (years, Median IQR): 6.3 (4.3, 7.4) Length of follow up during usual care since IVAN exit (years, Median IQR): 4.4 (2.3, 5.4)

Subject analysis set title

Deceased, passive data collection only

Subject analysis set type

Sub-group analysis

Subject analysis set description

Non-attenders (passive data collection only) - deceased 112/124 (90.3%) have at least one ophthalmology appointment as part of usual care since IVAN exit Length of follow up during usual care since IVAN entry (years, Median IQR): 3.8 (3.0, 4.8) Length of follow up during usual care since IVAN exit (years, Median IQR): 1.7 (1.1, 2.8)

Subject analysis sets values	Attenders	Survivor, passive data collection only	Deceased, passive data collection only
Number of subjects	199	209	124
Age categorical
Age at IVAN exit
Units: Subjects
Adults (18-64 years)	9	10	0
From 65-84 years	166	144	74
85 years and over	24	55	50
Age continuous
Age at IVAN exit
Units: years
arithmetic mean (standard deviation)	76.5 ( 7.0 )	79.6 ( 7.5 )	83.8 ( 5.4 )
Gender categorical
Units: Subjects
Female	114	138	68
Male	85	71	56
Drug
Drug allocation during IVAN trial
Units: Subjects
Ranibizumab	105	105	62
Bevacizumab	94	104	62
Drug frequency
Treatment regimen during IVAN trial
Units: Subjects
Continuous	101	103	65
Discontinuous	98	106	59
Angina at IVAN entry
Units: Subjects
Yes	19	25	28
No	180	184	96
Dyspnoea at IVAN entry
Units: Subjects
Yes	25	39	33
No	173	169	91
Missing	1	1	0
Dyspnoea at IVAN exit
Units: Subjects
Yes	31	41	34
No	168	168	88
Missing	0	0	2
Myocardial infarction at IVAN entry
Units: Subjects
Yes	7	14	16
No	192	195	108
Transient ischemic attack at IVAN entry
Units: Subjects
Yes	7	11	8
No	180	188	109
Missing	12	10	7
Stroke at IVAN entry
Units: Subjects
Yes	3	2	3
No	196	207	121
DVT/PE at IVAN entry
DVT = deep vein thrombosis; PE =pulmonary embolism
Units: Subjects
Yes	8	9	13
No	191	199	111
Missing	0	1	0
Current or past smoker at IVAN entry
Units: Subjects
Yes	113	128	90
No	83	81	33
Missing	3	0	1
Systolic BP at IVAN entry
BP = Blood pressure
Units: mmHg
arithmetic mean (standard deviation)	140.6 ( 17.7 )	143.1 ( 19.6 )	144.0 ( 20.7 )
Diastolic BP at IVAN entry
BP = Blood pressure
Units: mmHg
arithmetic mean (standard deviation)	76.7 ( 9.0 )	78.3 ( 10.3 )	75.0 ( 11.6 )
Systolic BP at IVAN exit
BP = Blood pressure Data missing for 8 patients (3 Attenders, 1 Survivor, passive data collection only, 4 Deceased, passive data collection only)
Units: mmHG
arithmetic mean (standard deviation)	137.3 ( 17.2 )	140.2 ( 18.9 )	137.6 ( 21.5 )
Diastolic BP at IVAN exit
BP = Blood pressure Data missing for 8 patients (3 Attenders, 1 Survivor, passive data collection only, 4 Deceased, passive data collection only)
Units: mmHg
arithmetic mean (standard deviation)	75.4 ( 9.3 )	74.9 ( 10.4 )	71.6 ( 10.9 )

End points

Top of page

Reporting group title

Follow-up cohort

Reporting group description

Reporting group title

IVAN randomised allocation: Ranibizumab

Reporting group description

Reporting group title

IVAN randomised allocation: Bevacizumab

Reporting group description

Reporting group title

IVAN randomised allocation: Continuous

Reporting group description

Reporting group title

IVAN randomised allocation: Discontinuous

Reporting group description

Reporting group title

Intralesional macular atrophy (ILMA) absent in study eye

Reporting group description

Reporting group title

Intralesional macular atrophy (ILMA) present in study eye

Reporting group description

Reporting group title

Geographic atrophy (GA) absent in study eye

Reporting group description

Reporting group title

Geographic atrophy (GA) present in study eye

Reporting group description

Reporting group title

Developed/worsened intralesional macular atrophy (ILMA)

Reporting group description

Defined as the development (incident) or expansion of the area (worsened) of ILMA. Expansion of the area of ILMA was quantified and was classified as worsened when it increased by >20%

Reporting group title

Did not develop/worsen intralesional macular atrophy (ILMA)

Reporting group description

Defined as the development (incident) or expansion of the area (worsened) of ILMA. Expansion of the area of ILMA was quantified and was classified as worsened when it increased by >20%

Reporting group title

Developed/worsend geographic atrophy (GA)

Reporting group description

Defined as the development (incident) or expansion of the area (worsened) of GA. Expansion of the area of GA was quantified and was classified as worsened when it increased by >20%.

Reporting group title

Did not develop/worsen geographic atrophy (GA)

Reporting group description

Defined as the development (incident) or expansion of the area (worsened) of GA. Expansion of the area of GA was quantified and was classified as worsened when it increased by >20%.

Subject analysis set title

Attenders

Subject analysis set type

Sub-group analysis

Subject analysis set description

Subject analysis set title

Survivor, passive data collection only

Subject analysis set type

Sub-group analysis

Subject analysis set description

Subject analysis set title

Deceased, passive data collection only

Subject analysis set type

Sub-group analysis

Subject analysis set description

Primary: Development of intralesional macular atrophy (ILMA)

Top of page

End point title

Development of intralesional macular atrophy (ILMA)

End point description

ILMA = intralesional macular atrophy (defined as atrophy lying within the footprint of the neovascular lesion); Primary morphology outcome Study eyes with ILMA at IVAN exit (baseline) were excluded from this analysis. Missing data imputed using multiple imputation methods. We had intended to examine the impact of HRM type (well or ill defined). However, review of the OCT data did not support dichotomising participant eyes based on HRM. HRM was therefore modelled as present vs. absent. ILMA= Intralesional macular atrophy, GA= Geographic atrophy, FU=Follow up, CNV= Choroidal neovascularisation, HRM= Hyperreflective material, PED= Pigment Epithelial Detachment, SRF= Subretinal fluid, nAMD= neovascular age-related macular degeneration, CI=Confidence interval

End point type

Primary

End point timeframe

Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA))

End point values	Follow-up cohort	IVAN randomised allocation: Ranibizumab	IVAN randomised allocation: Bevacizumab	IVAN randomised allocation: Continuous	IVAN randomised allocation: Discontinuous	Intralesional macular atrophy (ILMA) absent in study eye	Intralesional macular atrophy (ILMA) present in study eye	Geographic atrophy (GA) absent in study eye	Geographic atrophy (GA) present in study eye	Developed/worsened intralesional macular atrophy (ILMA)	Did not develop/worsen intralesional macular atrophy (ILMA)	Developed/worsend geographic atrophy (GA)	Did not develop/worsen geographic atrophy (GA)
Number of subjects analysed	319 ^[1]	167 ^[2]	152 ^[3]	157 ^[4]	162 ^[5]	122	197 ^[6]	225 ^[7]	94 ^[8]	197 ^[9]	122 ^[10]	84 ^[11]	225 ^[12]
Units: Patients
Developed ILMA	197	101	96	59	63	0	197	109	88	197	0	78	109
Did not develop ILMA	122	66	56	59	63	122	0	116	6	0	122	6	116

Notes
[1] - Exclusions: a) No images available (n=64). b) ILMA at baseline (n=149).
[2] - Exclusions: a) No images available (n=27). b) ILMA at baseline (n=78).
[3] - Exclusions: a) No images available (n=37). b) ILMA at baseline (n=71).
[4] - Exclusions: a) No images available (n=35). b) ILMA at baseline (n=77).
[5] - Exclusions: a) No images available (n=29). b) ILMA at baseline (n=72).
[6] - Exclusions: ILMA at baseline (n=149)
[7] - Exclusions: a) ILMA at baseline (n=84).
[8] - Exclusions: a) ILMA at baseline (n=65).
[9] - Exclusions: a) ILMA at baseline (n=46).
[10] - Exclusions: a) ILMA at baseline (n=16).
[11] - Exclusions: a) ILMA at baseline (n=58).
[12] - Exclusions: a) ILMA at baseline (n=86).

Effect of age at IVAN exit (per 10 years)

Statistical analysis description

Multivariable model adjusted for: Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.84

Confidence interval

level

95%

sides

2-sided

lower limit

1.3

upper limit

2.62

Effect of gender (male)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.3

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.76

Confidence interval

level

95%

sides

2-sided

lower limit

0.45

upper limit

1.28

Effect of IVAN trial drug allocation (Bevacizumab)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) present in study eye v Intralesional macular atrophy (ILMA) absent in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.519

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.18

Confidence interval

level

95%

sides

2-sided

lower limit

0.71

upper limit

1.98

Effect of IVAN trial treatment regimen (Discont.)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug allocation during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.879

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.96

Confidence interval

level

95%

sides

2-sided

lower limit

0.57

upper limit

1.61

Effect of injection rate during EFU (per year)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) present in study eye v Intralesional macular atrophy (ILMA) absent in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.021

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.88

Confidence interval

level

95%

sides

2-sided

lower limit

0.79

upper limit

0.98

Effect of >50% classic CNV at IVAN entry

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.173

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.66

Confidence interval

level

95%

sides

2-sided

lower limit

0.36

upper limit

1.2

Effect of retinal thickness at IVAN exit(per 10µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10mm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.068

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.42

Confidence interval

level

95%

sides

2-sided

lower limit

0.97

upper limit

2.07

Effect of subretinal fluid (SRF) at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.676

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.88

Confidence interval

level

95%

sides

2-sided

lower limit

0.48

upper limit

1.6

Effect of PED at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.359

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.66

Confidence interval

level

95%

sides

2-sided

lower limit

0.28

upper limit

1.6

Effect of PED at EFU (at fovea height <85µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence ^[13]

P-value

= 0.115 ^[14]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.92

Confidence interval

level

95%

sides

2-sided

lower limit

0.48

upper limit

1.75

Notes
[13] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[14] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent at fovea)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence ^[15]

P-value

= 0.115 ^[16]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.56

Confidence interval

level

95%

sides

2-sided

lower limit

1.28

upper limit

5.14

Notes
[15] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[16] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) present in study eye v Intralesional macular atrophy (ILMA) absent in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence ^[17]

P-value

= 0.115 ^[18]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.55

Confidence interval

level

95%

sides

2-sided

lower limit

0.45

upper limit

5.37

Notes
[17] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[18] - p-value is for overall effect of categorised PED at EFU

Effect of HRM at EFU

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) present in study eye v Intralesional macular atrophy (ILMA) absent in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.833

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.07

Confidence interval

level

95%

sides

2-sided

lower limit

0.57

upper limit

1.99

Effect of ILMA FE at IVAN exit (ILMA present)

Statistical analysis description

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence ^[19]

P-value

= 0.937 ^[20]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.39

Confidence interval

level

95%

sides

2-sided

lower limit

0.5

upper limit

3.88

Notes
[19] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[20] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of ILMA FE at IVAN exit (no nAMD lesion)

Statistical analysis description

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence ^[21]

P-value

= 0.937 ^[22]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.04

Confidence interval

level

95%

sides

2-sided

lower limit

0.6

upper limit

1.8

Notes
[21] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[22] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of GA in FE at IVAN exit

Statistical analysis description

Comparison groups

Intralesional macular atrophy (ILMA) present in study eye v Intralesional macular atrophy (ILMA) absent in study eye

Number of subjects included in analysis

319

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.106

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.82

Confidence interval

level

95%

sides

2-sided

lower limit

0.8

upper limit

9.91

Primary: Final distance visual acuity (DVA) in study eye

Top of page

End point title

Final distance visual acuity (DVA) in study eye

End point description

DVA was the primary visual function outcome

End point type

Primary

End point timeframe

DVA in study eye (most recent reading during extended FU, including research appointment for attenders).

Follow-up cohort

IVAN randomised allocation: Ranibizumab

IVAN randomised allocation: Bevacizumab

IVAN randomised allocation: Continuous

IVAN randomised allocation: Discontinuous

Intralesional macular atrophy (ILMA) absent in study eye

Intralesional macular atrophy (ILMA) present in study eye

Geographic atrophy (GA) absent in study eye

Geographic atrophy (GA) present in study eye

Developed/worsened intralesional macular atrophy (ILMA)

Did not develop/worsen intralesional macular atrophy (ILMA)

Developed/worsend geographic atrophy (GA)

Did not develop/worsen geographic atrophy (GA)

Number of subjects analysed

508 ^[23]

262 ^[24]

246 ^[25]

257 ^[26]

251 ^[27]

113 ^[28]

315 ^[29]

289 ^[30]

139 ^[31]

243

138

141 ^[32]

310 ^[33]

Units: Letters

median (inter-quartile range (Q1-Q3))

57 (29.5 to 71.5)

57.5 (30.0 to 72.0)

57.0 (28.0 to 71.0)

53.0 (30.0 to 73.0)

58.0 (29.0 to 70.0)

66.3 (47.0 to 74.2)

50.1 (27.6 to 67.6)

60.9 (39.4 to 71.5)

38.5 (24.3 to 64.4)

52.0 (24.0 to 69.0)

64.5 (36.0 to 76.0)

41.0 (22.0 to 63.0)

63.0 (34.0 to 73.0)

Notes
[23] - Exclusions: No VA reading since IVAN exit (n=24)
[24] - Exclusions: No VA reading since IVAN exit (n=10)
[25] - Exclusions: No VA reading since IVAN exit (n=14)
[26] - Exclusions: No VA reading since IVAN exit (n=12)
[27] - Exclusions: No VA reading since IVAN exit (n=12)
[28] - Exclusions: Less than 5 VA readings since IVAN exit (n=9)
[29] - Exclusions: Less than 5 VA readings since IVAN exit (n=31)
[30] - Exclusions: Less than 5 VA readings since IVAN exit (n=20)
[31] - Exclusions: Less than 5 VA readings since IVAN exit (n=20)
[32] - Exclusions: No VA reading since IVAN exit (n=1)
[33] - Exclusions: No VA reading since IVAN exit (n=1)

Effect of trial drug allocation on DVA

Statistical analysis description

The effects of the randomized allocations on DVA in the study eye at the most recent recorded visit was assessed using linear regression. These analyses were adjusted for centre size (seven strata as per the IVAN trial). Excluding patients who do not have a DVA reading since IVAN exit (n=24)

Comparison groups

IVAN randomised allocation: Bevacizumab v IVAN randomised allocation: Ranibizumab

Number of subjects included in analysis

508

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.6429

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-1

Confidence interval

level

95%

sides

2-sided

lower limit

-5.5

upper limit

3.4

Effect of trial treatment regimen on DVA

Statistical analysis description

Comparison groups

IVAN randomised allocation: Continuous v IVAN randomised allocation: Discontinuous

Number of subjects included in analysis

508

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.9917

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

0.02

Confidence interval

level

95%

sides

2-sided

lower limit

-4.4

upper limit

4.4

Effect of ILMA on DVA

Statistical analysis description

A linear random effects methods model was fitted to DVA readings between IVAN exit visit and the date of the image used to grade ILMA. From this model, participants' DVA values were predicted for the date of the image used to grade ILMA. A linear regression model was then fitted to estimate associations between the presence of ILMA on predicated DVA. Analyses were adjusted for visual function at IVAN exit. Excluding patients with no SE images (n=64) and patients with <5 DVA readings (n=40)

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

428

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.0206

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-5

Confidence interval

level

95%

sides

2-sided

lower limit

-9.2

upper limit

-0.8

Effect of GA on DVA

Statistical analysis description

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

428

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.0179

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-5.2

Confidence interval

level

95%

sides

2-sided

lower limit

-9.5

upper limit

-0.9

Primary: Survival

Top of page

End point title

Survival

End point description

Survival was the primary safety outcome

End point type

Primary

End point timeframe

Time to death since IVAN entry. Times were censored at most recent clinical review.

Follow-up cohort

IVAN randomised allocation: Ranibizumab

IVAN randomised allocation: Bevacizumab

IVAN randomised allocation: Continuous

IVAN randomised allocation: Discontinuous

Intralesional macular atrophy (ILMA) absent in study eye

Intralesional macular atrophy (ILMA) present in study eye

Geographic atrophy (GA) absent in study eye

Geographic atrophy (GA) present in study eye

Developed/worsened intralesional macular atrophy (ILMA)

Did not develop/worsen intralesional macular atrophy (ILMA)

Developed/worsend geographic atrophy (GA)

Did not develop/worsen geographic atrophy (GA)

Number of subjects analysed

532

272

260

269

263

122

346

309

159

243

138

142

311

Units: Years

median (inter-quartile range (Q1-Q3))

6.2 (4.3 to 7.2)

6.2 (4.4 to 7.2)

6.2 (4.2 to 7.3)

6.0 (4.3 to 7.1)

6.4 (4.4 to 7.3)

6.6 (4.7 to 7.3)

6.3 (4.8 to 7.3)

6.5 (5.0 to 7.3)

6.1 (4.4 to 7.1)

6.3 (5.1 to 7.3)

6.6 (4.7 to 7.3)

6.0 (4.4 to 7.0)

6.5 (5.0 to 7.3)

Effect of trial drug allocation on survival

Statistical analysis description

The effects of the randomized allocations on survival since IVAN entry was assessed using Cox proportional-hazards regression. These analyses were adjusted for centre size (seven strata as per the IVAN trial).

Comparison groups

IVAN randomised allocation: Ranibizumab v IVAN randomised allocation: Bevacizumab

Number of subjects included in analysis

532

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.3789

Method

Regression, Cox

Parameter type

Hazard ratio (HR)

Point estimate

1.17

Confidence interval

level

95%

sides

2-sided

lower limit

0.82

upper limit

1.67

Effect of trial treatment regimen on survival

Statistical analysis description

Comparison groups

IVAN randomised allocation: Continuous v IVAN randomised allocation: Discontinuous

Number of subjects included in analysis

532

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.5115

Method

Regression, Cox

Parameter type

Hazard ratio (HR)

Point estimate

0.89

Confidence interval

level

95%

sides

2-sided

lower limit

0.62

upper limit

1.27

Secondary: Development of geographic atrophy (GA)

Top of page

End point title

Development of geographic atrophy (GA)

End point description

GA = Geographic atrophy (defined as atrophy lying outside the footprint of the neovascular lesion) Study eyes with GA at IVAN exit (baseline) were excluded from this analysis. Missing data imputed using multiple imputation methods. ILMA= Intralesional macular atrophy, GA= Geographic atrophy, FU=Follow up, CNV= Choroidal neovascularisation, HRM= Hyperreflective material, PED= Pigment Epithelial Detachment, SRF= Subretinal fluid, nAMD= neovascular age-related macular degeneration, CI=Confidence interval

End point type

Secondary

End point timeframe

Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA))

End point values	Follow-up cohort	IVAN randomised allocation: Ranibizumab	IVAN randomised allocation: Bevacizumab	IVAN randomised allocation: Continuous	IVAN randomised allocation: Discontinuous	Intralesional macular atrophy (ILMA) absent in study eye	Intralesional macular atrophy (ILMA) present in study eye	Geographic atrophy (GA) absent in study eye	Geographic atrophy (GA) present in study eye	Developed/worsened intralesional macular atrophy (ILMA)	Did not develop/worsen intralesional macular atrophy (ILMA)	Developed/worsend geographic atrophy (GA)	Did not develop/worsen geographic atrophy (GA)
Number of subjects analysed	448 ^[34]	233 ^[35]	215 ^[36]	223 ^[37]	225 ^[38]	122	326 ^[39]	309	139 ^[40]	229 ^[41]	137 ^[42]	139 ^[43]	309 ^[44]
Units: Patients
Developed GA	139	78	61	62	77	6	133	0	139	87	6	139	0
Did not develop GA	309	155	154	161	148	116	193	309	0	142	131	0	309

Notes
[34] - Exclusions: a) No images available (n=64). b) GA at baseline (n=20).
[35] - Exclusions: a) No images available (n=27). b) GA at baseline (n=12).
[36] - Exclusions: a) No images available (n=37). b) GA at baseline (n=8).
[37] - Exclusions: a) No images available (n=35). b) GA at baseline (n=11).
[38] - Exclusions: a) No images available (n=29). b) GA at baseline (n=9).
[39] - Exclusions: a) GA at baseline (n=20).
[40] - Exclusions: GA at baseline (n=20)
[41] - Exclusions: a) GA at baseline (n=14).
[42] - Exclusions: a) GA at baseline (n=1).
[43] - Exclusions: a) GA at baseline (n=3).
[44] - Exclusions: a) GA at baseline (n=2).

Effect of age at IVAN exit (per 10 years)

Statistical analysis description

Multivariable model adjusted for: Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.1

Confidence interval

level

95%

sides

2-sided

lower limit

1.49

upper limit

2.96

Effect of gender (male)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.613

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.89

Confidence interval

level

95%

sides

2-sided

lower limit

0.56

upper limit

1.41

Effect of IVAN trial drug allocation (Bevacizumab)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.136

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.7

Confidence interval

level

95%

sides

2-sided

lower limit

0.44

upper limit

1.12

Effect of IVAN trial treatment regimen (Discont.)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug alliocation during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.079

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.5

Confidence interval

level

95%

sides

2-sided

lower limit

0.95

upper limit

2.37

Effect of injection rate during EFU (per year)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.045

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.9

Confidence interval

level

95%

sides

2-sided

lower limit

0.81

upper limit

0.998

Effect of >50% classic CNV at IVAN entry

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.518

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.2

Confidence interval

level

95%

sides

2-sided

lower limit

0.7

upper limit

2.06

Effect of retinal thickness at IVAN exit(per 10µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.318

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.18

Confidence interval

level

95%

sides

2-sided

lower limit

0.86

upper limit

1.61

Effect of subretinal fluid (SRF) at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) present in study eye v Geographic atrophy (GA) absent in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.67

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.88

Confidence interval

level

95%

sides

2-sided

lower limit

0.49

upper limit

1.57

Effect of PED present at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.812

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.08

Confidence interval

level

95%

sides

2-sided

lower limit

0.58

upper limit

2.01

Effect of PED at EFU (at fovea height <85µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence ^[45]

P-value

= 0.003 ^[46]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.08

Confidence interval

level

95%

sides

2-sided

lower limit

0.55

upper limit

2.11

Notes
[45] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[46] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent at fovea)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence ^[47]

P-value

= 0.003 ^[48]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.81

Confidence interval

level

95%

sides

2-sided

lower limit

1.51

upper limit

5.21

Notes
[47] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[48] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence ^[49]

P-value

= 0.003 ^[50]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.28

Confidence interval

level

95%

sides

2-sided

lower limit

0.81

upper limit

6.46

Notes
[49] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[50] - p-value is for overall effect of categorised PED at EFU

Effect of ILMA FE at IVAN exit (ILMA present)

Statistical analysis description

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence ^[51]

P-value

= 0.974 ^[52]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.07

Confidence interval

level

95%

sides

2-sided

lower limit

0.51

upper limit

2.22

Notes
[51] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[52] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of ILMA FE at IVAN exit (no nAMD lesion)

Statistical analysis description

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence ^[53]

P-value

= 0.974 ^[54]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.98

Confidence interval

level

95%

sides

2-sided

lower limit

0.59

upper limit

1.62

Notes
[53] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[54] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of GA in FE at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion)

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

448

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.001

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

3.14

Confidence interval

level

95%

sides

2-sided

lower limit

1.56

upper limit

6.32

Secondary: Development/worsening of intralesional macular atrophy (ILMA)

Top of page

End point title

Development/worsening of intralesional macular atrophy (ILMA)

End point description

ILMA = intralesional macular atrophy (defined as atrophy lying within the footprint of the neovascular lesion) This outcome was defined as the development (incident) or expansion of the area (worsened) of ILMA. Expansion of the area of ILMA was quantified and was classified as worsened when it increased by >20%. The outcome was defined to include worsening of ILMA so that the analyses did not exclude study eyes with ILMA at baseline. Missing data imputed using multiple imputation methods. We had intended to examine the impact of HRM type (well or ill defined). However, review of the OCT data did not support dichotomising participant eyes based on HRM. HRM was therefore modelled as present vs. absent. ILMA= Intralesional macular atrophy, GA= Geographic atrophy, FU=Follow up, CNV= Choroidal neovascularisation, HRM= Hyperreflective material, PED= Pigment Epithelial Detachment, SRF= Subretinal fluid, nAMD= neovascular age-related macular degeneration, CI=Confidence interval

End point type

Secondary

End point timeframe

Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA))

End point values	Follow-up cohort	IVAN randomised allocation: Ranibizumab	IVAN randomised allocation: Bevacizumab	IVAN randomised allocation: Continuous	IVAN randomised allocation: Discontinuous	Intralesional macular atrophy (ILMA) absent in study eye	Intralesional macular atrophy (ILMA) present in study eye	Geographic atrophy (GA) absent in study eye	Geographic atrophy (GA) present in study eye	Developed/worsened intralesional macular atrophy (ILMA)	Did not develop/worsen intralesional macular atrophy (ILMA)	Developed/worsend geographic atrophy (GA)	Did not develop/worsen geographic atrophy (GA)
Number of subjects analysed	381 ^[55]	204 ^[56]	177 ^[57]	187 ^[58]	194 ^[59]	122	259 ^[60]	273 ^[61]	108 ^[62]	243	138	96 ^[63]	275 ^[64]
Units: Patients
Developed/worsened	243	130	113	120	123	0	243	142	101	243	0	89	144
Did not develop/worsen	138	74	64	67	71	122	16	131	7	0	138	7	131

Notes
[55] - Exclusions: No images available since IVAN exit (n=64). Worsening unknown (n=87)
[56] - Exclusions: No images available since IVAN exit (n=27). Worsening unknown (n=41)
[57] - Exclusions: No images available since IVAN exit (n=37). Worsening unknown (n=46)
[58] - Exclusions: No images available since IVAN exit (n=35). Worsening unknown (n=47)
[59] - Exclusions: No images available since IVAN exit (n=29). Worsening unknown (n=40)
[60] - Exclusions: Worsening unknown (n=87)
[61] - Exclusions: Worsening unknown (n=36)
[62] - Exclusions: Worsening unknown (n=51)
[63] - Exclusions: Worsening unknown (n=46)
[64] - Exclusions: Worsening unknown (n=36)

Effect of age at IVAN exit (per 10 years)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.74

Confidence interval

level

95%

sides

2-sided

lower limit

1.28

upper limit

2.36

Effect of gender (male)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.432

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.83

Confidence interval

level

95%

sides

2-sided

lower limit

0.53

upper limit

1.31

Effect of IVAN trial drug allocation (Bevacizumab)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.539

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.15

Confidence interval

level

95%

sides

2-sided

lower limit

0.73

upper limit

1.82

Effect of IVAN trial treatment regimen (Discont.)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.872

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.04

Confidence interval

level

95%

sides

2-sided

lower limit

0.65

upper limit

1.67

Effect of injection rate during EFU (per year)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.089

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.92

Confidence interval

level

95%

sides

2-sided

lower limit

0.83

upper limit

1.01

Effect of >50% classic CNV at IVAN entry

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.13

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.68

Confidence interval

level

95%

sides

2-sided

lower limit

0.41

upper limit

1.12

Effect of retinal thickness at IVAN exit(per 10µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.311

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.18

Confidence interval

level

95%

sides

2-sided

lower limit

0.86

upper limit

1.61

Effect of subretinal fluid (SRF) at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.497

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.83

Confidence interval

level

95%

sides

2-sided

lower limit

0.48

upper limit

1.43

Effect of PED at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.835

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.93

Confidence interval

level

95%

sides

2-sided

lower limit

0.48

upper limit

1.86

Effect of PED at EFU (at fovea height <85µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence ^[65]

P-value

= 0.115 ^[66]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.02

Confidence interval

level

95%

sides

2-sided

lower limit

0.57

upper limit

1.81

Notes
[65] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[66] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent at fovea)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence ^[67]

P-value

= 0.115 ^[68]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.94

Confidence interval

level

95%

sides

2-sided

lower limit

1.07

upper limit

3.54

Notes
[67] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[68] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit HRM present at EFU ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence ^[69]

P-value

= 0.115 ^[70]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.21

Confidence interval

level

95%

sides

2-sided

lower limit

0.35

upper limit

4.19

Notes
[69] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[70] - p-value is for overall effect of categorised PED at EFU

Effect of HRM at EFU

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.625

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.17

Confidence interval

level

95%

sides

2-sided

lower limit

0.63

upper limit

2.16

Effect of ILMA FE at IVAN exit (ILMA present)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence ^[71]

P-value

= 0.937 ^[72]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.87

Confidence interval

level

95%

sides

2-sided

lower limit

0.39

upper limit

1.91

Notes
[71] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[72] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of ILMA FE at IVAN exit (no nAMD lesion)

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence ^[73]

P-value

= 0.937 ^[74]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.97

Confidence interval

level

95%

sides

2-sided

lower limit

0.6

upper limit

1.58

Notes
[73] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[74] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of GA in FE at IVAN exit

Statistical analysis description

Comparison groups

Developed/worsened intralesional macular atrophy (ILMA) v Did not develop/worsen intralesional macular atrophy (ILMA)

Number of subjects included in analysis

381

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.325

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.57

Confidence interval

level

95%

sides

2-sided

lower limit

0.64

upper limit

3.84

Secondary: Development/worsening of geographic atrophy (GA)

Top of page

End point title

Development/worsening of geographic atrophy (GA)

End point description

GA = Geographic atrophy (defined as atrophy lying outside the footprint of the neovascular lesion) This outcome was defined as the development (incident) or expansion of the area (worsened) of GA. Expansion of the area of GA was quantified and was classified as worsened when it increased by >20%. The outcome was defined to include worsening of GA so that the analyses did not exclude study eyes with ILMA at baseline. Missing data imputed using multiple imputation methods. We had intended to examine the impact of HRM type (well or ill defined). However, review of the OCT data did not support dichotomising participant eyes based on HRM. HRM was therefore modelled as present vs. absent. ILMA= Intralesional macular atrophy, GA= Geographic atrophy, FU=Follow up, CNV= Choroidal neovascularisation, HRM= Hyperreflective material, PED= Pigment Epithelial Detachment, SRF= Subretinal fluid, nAMD= neovascular age-related macular degeneration, CI=Confidence interval

End point type

Secondary

End point timeframe

Determined using most recent image available (color fundus photography, blue light autofluorescence (AF), optical coherence tomography (OCT) or dye angiography fluorescein/indocyanine green (FFA))

End point values	Follow-up cohort	IVAN randomised allocation: Ranibizumab	IVAN randomised allocation: Bevacizumab	IVAN randomised allocation: Continuous	IVAN randomised allocation: Discontinuous	Intralesional macular atrophy (ILMA) absent in study eye	Intralesional macular atrophy (ILMA) present in study eye	Geographic atrophy (GA) absent in study eye	Geographic atrophy (GA) present in study eye	Developed/worsened intralesional macular atrophy (ILMA)	Did not develop/worsen intralesional macular atrophy (ILMA)	Developed/worsend geographic atrophy (GA)	Did not develop/worsen geographic atrophy (GA)
Number of subjects analysed	453 ^[75]	236 ^[76]	217 ^[77]	226 ^[78]	227 ^[79]	122	331 ^[80]	309	144 ^[81]	233 ^[82]	138	142	311
Units: Patients
Developed/worsened	142	79	63	64	78	6	136	0	142	89	7	142	0
Did not develop/worsen	311	157	154	162	149	116	195	309	2	144	131	0	311

Notes
[75] - Exclusions: No images available since IVAN exit (n=64). Worsening unknown (n=15)
[76] - Exclusions: No images available since IVAN exit (n=27). Worsening unknown (n=9)
[77] - Exclusions: No images available since IVAN exit (n=37). Worsening unknown (n=6)
[78] - Exclusions: No images available since IVAN exit (n=35). Worsening unknown (n=8)
[79] - Exclusions: No images available since IVAN exit (n=29). Worsening unknown (n=7)
[80] - Exclusions: Worsening unknown (n=15)
[81] - Exclusions: Worsening unknown (n=15)
[82] - Exclusions: Worsening unknown (n=10)

Effect of age at IVAN exit (per 10 years)

Statistical analysis description

Multivariable model adjusted for: Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.02

Confidence interval

level

95%

sides

2-sided

lower limit

1.44

upper limit

2.83

Effect of gender (male)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.77

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.93

Confidence interval

level

95%

sides

2-sided

lower limit

0.59

upper limit

1.48

Effect of IVAN trial drug allocation (Bevacizumab)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.233

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.76

Confidence interval

level

95%

sides

2-sided

lower limit

0.49

upper limit

1.19

Effect of IVAN trial treatment regimen (Discont.)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug allocation during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.098

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.46

Confidence interval

level

95%

sides

2-sided

lower limit

0.93

upper limit

2.27

Effect of injection rate during EFU (per year)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Did not develop/worsen geographic atrophy (GA) v Developed/worsend geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.067

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.91

Confidence interval

level

95%

sides

2-sided

lower limit

0.82

upper limit

Effect of age >50% classic CNV at IVAN entry

Statistical analysis description

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.448

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.21

Confidence interval

level

95%

sides

2-sided

lower limit

0.71

upper limit

2.05

Effect of retinal thickness at IVAN exit(per 10µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.316

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.17

Confidence interval

level

95%

sides

2-sided

lower limit

0.86

upper limit

1.6

Effect of subretinal fluid (SRF) at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.517

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.83

Confidence interval

level

95%

sides

2-sided

lower limit

0.47

upper limit

1.46

Effect of PED present at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.886

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.04

Confidence interval

level

95%

sides

2-sided

lower limit

0.58

upper limit

1.9

Effect of PED at EFU (at fovea height <85µm)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence ^[83]

P-value

= 0.003 ^[84]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.04

Confidence interval

level

95%

sides

2-sided

lower limit

0.54

upper limit

2.03

Notes
[83] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[84] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent at fovea)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence ^[85]

P-value

= 0.003 ^[86]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.57

Confidence interval

level

95%

sides

2-sided

lower limit

1.41

upper limit

4.69

Notes
[85] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[86] - p-value is for overall effect of categorised PED at EFU

Effect of PED at EFU (absent)

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion) GA in FE at IVAN exit

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence ^[87]

P-value

= 0.003 ^[88]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.19

Confidence interval

level

95%

sides

2-sided

lower limit

0.79

upper limit

6.04

Notes
[87] - PED categorised: 1: Present at fovea height ≥ 85µm (Reference category) 2: Present at fovea height <85µm 3: Absent at fovea 4: PED absent
[88] - p-value is for overall effect of categorised PED at EFU

Effect of ILMA FE at IVAN exit (ILMA present)

Statistical analysis description

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence ^[89]

P-value

= 0.974 ^[90]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.09

Confidence interval

level

95%

sides

2-sided

lower limit

0.53

upper limit

2.24

Notes
[89] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[90] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of ILMA FE at IVAN exit (no nAMD lesion)

Statistical analysis description

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence ^[91]

P-value

= 0.974 ^[92]

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

1.01

Confidence interval

level

95%

sides

2-sided

lower limit

0.61

upper limit

1.67

Notes
[91] - ILMA in FE at IVAN exit categorised: 1: nAMD lesion, ILMA absent (reference category) 2: nAMD lesion, ILMA present 2: No nAMD lesion
[92] - p-value is for overall effect of categorised ILMA in FE at IVAN exit

Effect of GA in FE at IVAN exit

Statistical analysis description

Multivariable model adjusted for: Age at IVAN exit (per 10 years) Gender Drug and treatment regimen during IVAN trial Injection rate during EFU (per year) >50% classic CNV at IVAN entry Retinal thickness at IVAN exit (per 10µm) SRF present at IVAN exit PED present at IVAN exit PED at EFU (at fovea: height ≥ 85µm, <85µm, PED absent at fovea, PED absent) ILMA at IVAN exit in FE (nAMD lesion ILMA absent, nAMD lesion ILMA present, No nAMD lesion)

Comparison groups

Developed/worsend geographic atrophy (GA) v Did not develop/worsen geographic atrophy (GA)

Number of subjects included in analysis

453

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.006

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

2.48

Confidence interval

level

95%

sides

2-sided

lower limit

1.29

upper limit

4.76

Secondary: Low luminance acuity (LLA) in study eye

Top of page

End point title

Low luminance acuity (LLA) in study eye

End point description

The association between IVAN treatment allocations and ILMA/GA and LLA was estimated only for attenders because LLA was only measured at the additional research visit. LLA was also not measured during the IVAN trial and therefore these models were not adjusted for baseline values. Sensitivity analyses adjusted for BCVA at IVAN exit visit.

End point type

Secondary

End point timeframe

Extended follow-up clinic appointment

Follow-up cohort

IVAN randomised allocation: Ranibizumab

IVAN randomised allocation: Bevacizumab

IVAN randomised allocation: Continuous

IVAN randomised allocation: Discontinuous

Intralesional macular atrophy (ILMA) absent in study eye

Intralesional macular atrophy (ILMA) present in study eye

Geographic atrophy (GA) absent in study eye

Geographic atrophy (GA) present in study eye

Developed/worsened intralesional macular atrophy (ILMA)

Did not develop/worsen intralesional macular atrophy (ILMA)

Developed/worsend geographic atrophy (GA)

Did not develop/worsen geographic atrophy (GA)

Number of subjects analysed

199 ^[93]

105 ^[94]

94 ^[95]

101 ^[96]

98 ^[97]

86 ^[98]

113 ^[99]

165 ^[100]

34 ^[101]

93 ^[102]

98 ^[103]

28 ^[104]

167 ^[105]

Units: Letters

median (inter-quartile range (Q1-Q3))

38.0 (17.0 to 56.0)

40.0 (18.0 to 56.0)

30.5 (13.0 to 55.0)

31.0 (15.0 to 56.0)

44.0 (20.0 to 55.0)

53.5 (26.0 to 61.0)

27.0 (13.0 to 50.0)

43.0 (20.0 to 57.0)

12.0 (4.0 to 29.0)

27.0 (13.0 to 51.0)

50.5 (26.0 to 60.0)

12.0 (4.0 to 27.5)

43.0 (20.0 to 57.0)

Notes
[93] - Attenders only
[94] - Attenders only
[95] - Attenders only
[96] - Attenders only
[97] - Attenders only
[98] - Attenders only
[99] - Attenders only
[100] - Attenders only
[101] - Attenders only
[102] - Attenders only
[103] - Attenders only
[104] - Attenders only
[105] - Attenders only

Effect of ILMA on LLA

Statistical analysis description

Linear regression was used to analyse the effect of ILMA in study eyes on low luminance visual acuity at extended follow-up.

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

199

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-15.5

Confidence interval

level

95%

sides

2-sided

lower limit

-21.6

upper limit

-9.4

Effect of GA on LLA

Statistical analysis description

Linear regression was used to analyse the effect of GA in study eyes on low luminance visual acuity at extended follow-up.

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

199

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-18.5

Confidence interval

level

95%

sides

2-sided

lower limit

-26.6

upper limit

-10.4

Secondary: Low luminance acuity (LLA) in study eye, adjusted for BCVA at IVAN exit (sensitivity analysis)

Top of page

End point title

Low luminance acuity (LLA) in study eye, adjusted for BCVA at IVAN exit (sensitivity analysis)

End point description

Sensitivity analyses adjusting for BCVA at IVAN exit visit.

End point type

Secondary

End point timeframe

Extended follow-up clinic appointment

Follow-up cohort

IVAN randomised allocation: Ranibizumab

IVAN randomised allocation: Bevacizumab

IVAN randomised allocation: Continuous

IVAN randomised allocation: Discontinuous

Intralesional macular atrophy (ILMA) absent in study eye

Intralesional macular atrophy (ILMA) present in study eye

Geographic atrophy (GA) absent in study eye

Geographic atrophy (GA) present in study eye

Developed/worsened intralesional macular atrophy (ILMA)

Did not develop/worsen intralesional macular atrophy (ILMA)

Developed/worsend geographic atrophy (GA)

Did not develop/worsen geographic atrophy (GA)

Number of subjects analysed

199 ^[106]

105 ^[107]

94 ^[108]

101 ^[109]

98 ^[110]

86 ^[111]

113 ^[112]

165 ^[113]

34 ^[114]

93 ^[115]

98 ^[116]

28 ^[117]

167 ^[118]

Units: Letters

median (inter-quartile range (Q1-Q3))

38.0 (17.0 to 56.0)

40.0 (18.0 to 56.0)

30.5 (13.0 to 55.0)

31.0 (15.0 to 56.0)

44.0 (20.0 to 55.0)

53.5 (26.0 to 61.0)

27.0 (13.0 to 50.0)

43.0 (20.0 to 57.0)

12.0 (4.0 to 29.0)

27.0 (13.0 to 51.0)

50.5 (26.0 to 60.0)

12.0 (4.0 to 27.5)

43.0 (20.0 to 57.0)

Notes
[106] - Attenders only
[107] - Attenders only
[108] - Attenders only
[109] - Attenders only
[110] - Attenders only
[111] - Attenders only
[112] - Attenders only
[113] - Attenders only
[114] - Attenders only
[115] - Attenders only
[116] - Attenders only
[117] - Attenders only
[118] - Attenders only

Effect of ILMA on LLA

Statistical analysis description

Linear regression was used to analyse the effect of ILMA in study eyes on low luminance visual acuity at extended follow-up; adjusting for BCVA at IVAN exit

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

199

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-11

Confidence interval

level

95%

sides

2-sided

lower limit

-16.7

upper limit

-5.3

Effect of GA on LLA

Statistical analysis description

Linear regression was used to analyse the effect of ILMA in study eyes on low luminance visual acuity at extended follow-up; adjusting for BCVA at IVAN exit

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

199

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

< 0.001

Method

Regression, Linear

Parameter type

Mean difference (final values)

Point estimate

-17.2

Confidence interval

level

95%

sides

2-sided

lower limit

-24.4

upper limit

-10.1

Secondary: Quality of life (EQ5D)

Top of page

End point title

Quality of life (EQ5D)

End point description

End point type

Secondary

End point timeframe

This analysis was restricted to participants who were willing to compete the EQ-5D-5L and who had data at baseline and follow up.

Follow-up cohort

IVAN randomised allocation: Ranibizumab

IVAN randomised allocation: Bevacizumab

IVAN randomised allocation: Continuous

IVAN randomised allocation: Discontinuous

Intralesional macular atrophy (ILMA) absent in study eye

Intralesional macular atrophy (ILMA) present in study eye

Geographic atrophy (GA) absent in study eye

Geographic atrophy (GA) present in study eye

Developed/worsened intralesional macular atrophy (ILMA)

Did not develop/worsen intralesional macular atrophy (ILMA)

Developed/worsend geographic atrophy (GA)

Did not develop/worsen geographic atrophy (GA)

Number of subjects analysed

273 ^[119]

147 ^[120]

126 ^[121]

139 ^[122]

134 ^[123]

95 ^[124]

172 ^[125]

203 ^[126]

64 ^[127]

133 ^[128]

107 ^[129]

53 ^[130]

205 ^[131]

Units: EQ5D state score

median (inter-quartile range (Q1-Q3))

0.8 (0.6 to 0.9)

0.8 (0.7 to 1.0)

0.7 (0.6 to 0.9)

0.8 (0.6 to 0.9)

0.7 (0.6 to 0.9)

0.8 (0.7 to 1.0)

0.7 (0.6 to 1.0)

0.8 (0.6 to 0.9)

Notes
[119] - Exclusions: Participants who did not opt to complete the questionnaire (n=259)
[120] - Exclusions: Participants who did not opt to complete the questionnaire (n=125)
[121] - Exclusions: Participants who did not opt to complete the questionnaire (n=134)
[122] - Exclusions: Participants who did not opt to complete the questionnaire (n=130)
[123] - Exclusions: Participants who did not opt to complete the questionnaire (n=129)
[124] - Exclusions: Participants who did not opt to complete the questionnaire (n=27)
[125] - Exclusions: Participants who did not opt to complete the questionnaire (n=174)
[126] - Exclusions: Participants who did not opt to complete the questionnaire (n=106)
[127] - Exclusions: Participants who did not opt to complete the questionnaire (n=95)
[128] - Exclusions: Participants who did not opt to complete the questionnaire (n=110)
[129] - Exclusions: Participants who did not opt to complete the questionnaire (n=31)
[130] - Exclusions: Participants who did not opt to complete the questionnaire (n=89)
[131] - Exclusions: Participants who did not opt to complete the questionnaire (n=106)

Effect of trial drug allocation on EQ5D

Statistical analysis description

For EQ-5D-5L, no transformation could be applied to the data to satisfy the assumptions for linear regression. Therefore, the index score was categorised (EQ-5D-5L score, 1 (no problem in any dimension), 0.80 to 0.99 (a moderate problem in only one dimension), 0.5 to 0.79 (at least two moderate problems in any dimension) and <0.5 (at least one extreme problem)) and ordinal logistic regression was used. These analyses were adjusted for centre size (seven strata as per the IVAN trial).

Comparison groups

IVAN randomised allocation: Ranibizumab v IVAN randomised allocation: Bevacizumab

Number of subjects included in analysis

273

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.2419

Method

Regression, Logistic

Parameter type

Log odds ratio

Point estimate

1.3

Confidence interval

level

95%

sides

2-sided

lower limit

0.8

upper limit

Effect of trial treatment regimen on EQ5D

Statistical analysis description

Comparison groups

IVAN randomised allocation: Continuous v IVAN randomised allocation: Discontinuous

Number of subjects included in analysis

273

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.3887

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.8

Confidence interval

level

95%

sides

2-sided

lower limit

0.5

upper limit

1.3

Effect of ILMA on EQ5D

Statistical analysis description

Comparison groups

Intralesional macular atrophy (ILMA) absent in study eye v Intralesional macular atrophy (ILMA) present in study eye

Number of subjects included in analysis

267

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.0413

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

0.6

Confidence interval

level

95%

sides

2-sided

lower limit

0.4

upper limit

Effect of GA on EQ5D

Statistical analysis description

Comparison groups

Geographic atrophy (GA) absent in study eye v Geographic atrophy (GA) present in study eye

Number of subjects included in analysis

267

Analysis specification

Pre-specified

Analysis type

equivalence

P-value

= 0.9794

Method

Regression, Logistic

Parameter type

Odds ratio (OR)

Point estimate

Confidence interval

level

95%

sides

2-sided

lower limit

0.6

upper limit

1.7

Adverse events

Top of page

Timeframe for reporting adverse events

The AE reporting period for the study began upon completion of the consent form for the clinic appointment until the patient leaves the clinic appointment.

Adverse event reporting additional description

Only patients who attended a clinic appointment as part of the IVAN follow-up study are included in the adverse events reporting group. Adverse event that occurred during the clinic appointment were recorded. Any adverse events that occurred between the end of the IVAN trial and consent to the follow-up were considered medical history.

Assessment type

Systematic

Dictionary name

N/A

Dictionary version

N/A

Reporting group title

Attenders

Reporting group description

Serious adverse events	Attenders
Total subjects affected by serious adverse events
subjects affected / exposed	0 / 199 (0.00%)
number of deaths (all causes)	0
number of deaths resulting from adverse events	0

Frequency threshold for reporting non-serious adverse events: 0%

Non-serious adverse events	Attenders
Total subjects affected by non serious adverse events
subjects affected / exposed	1 / 199 (0.50%)
Eye disorders
Vitreous floaters
Additional description: This is almost certainly an event that occurred historically but which was noted at the research visit attended by the participant.
subjects affected / exposed	1 / 199 (0.50%)
occurrences all number	1

More information

Top of page

Were there any global substantial amendments to the protocol? Yes

17 Dec 2015

SUBSTANTIAL AMENDMENT 1: Section 8 of the protocol (safety reporting) was updated substantially in line with the Sponsor's policy and a table added detailing the terms and definitions of adverse events. New wording was inserted into section 4.4.2 on secondary outcomes explaining that Best Corrected Visual Acuity would be measured with both standard and low luminance visual acuity.

08 Mar 2016

SUBSTANTIAL AMENDMENT 2: Grounds of Non-Acceptance were received for protocol v5.0 from the MHRA therefore the safety sections were updated further in line with their recommendations and protocol v6.0 was submitted for approval.

21 Mar 2016

SUBSTANTIAL AMENDMENT 3: A change was made to the Patient Information Leaflet clarifying that research nurses could telephone participants if the reply slip had not been returned after two weeks. The change was made in response to feedback from research nurses and the Macular Disease Society highlighting the importance of verbal communication in this elderly, visually impaired group of participants.

Were there any global interruptions to the trial? No

Limitations of the trial such as small numbers of subjects analysed or technical problems leading to unreliable data.

None reported

Select Country:	Select Age Range:
Select Trial Status:	Select Trial Phase:
Select Gender:
Select Date Range: to
Select Rare Disease:
IMP with orphan designation in the indication
Orphan Designation Number:
Results Status:
Clear advanced search filters

Clinical trials

Clinical Trial Results:
Five year observational follow-up of the IVAN trial cohort: a study of function and morphology

Trial information

Trial information

Subject disposition

Subject disposition

Baseline characteristics

Baseline characteristics

End points

End points

Primary: Development of intralesional macular atrophy (ILMA)

Primary: Development of intralesional macular atrophy (ILMA)

Primary: Final distance visual acuity (DVA) in study eye

Primary: Final distance visual acuity (DVA) in study eye

Primary: Survival

Primary: Survival

Secondary: Development of geographic atrophy (GA)

Secondary: Development of geographic atrophy (GA)

Secondary: Development/worsening of intralesional macular atrophy (ILMA)

Secondary: Development/worsening of intralesional macular atrophy (ILMA)

Secondary: Development/worsening of geographic atrophy (GA)

Secondary: Development/worsening of geographic atrophy (GA)

Secondary: Low luminance acuity (LLA) in study eye

Secondary: Low luminance acuity (LLA) in study eye

Secondary: Low luminance acuity (LLA) in study eye, adjusted for BCVA at IVAN exit (sensitivity analysis)

Secondary: Low luminance acuity (LLA) in study eye, adjusted for BCVA at IVAN exit (sensitivity analysis)

Secondary: Quality of life (EQ5D)

Secondary: Quality of life (EQ5D)

Adverse events

Adverse events

More information

Substantial protocol amendments (globally)

Interruptions (globally)

Limitations and caveats

More information

Austria
Belgium
Bulgaria
Croatia
Cyprus
Czechia
Denmark
Estonia
Finland
France
Germany
Greece
Hungary
Iceland
Ireland
Italy
Latvia
Liechtenstein
Lithuania
Luxembourg
Malta
Netherlands
Norway
Poland
Portugal
Romania
Slovakia
Slovenia
Spain
Sweden
United Kingdom
Outside EU/EEA

Clinical trials

Clinical Trial Results: Five year observational follow-up of the IVAN trial cohort: a study of function and morphology

Trial information

Trial information

Subject disposition

Subject disposition

Baseline characteristics

Baseline characteristics

End points

End points

Primary: Development of intralesional macular atrophy (ILMA)

Primary: Development of intralesional macular atrophy (ILMA)

Primary: Final distance visual acuity (DVA) in study eye

Primary: Final distance visual acuity (DVA) in study eye

Primary: Survival

Primary: Survival

Secondary: Development of geographic atrophy (GA)

Secondary: Development of geographic atrophy (GA)

Secondary: Development/worsening of intralesional macular atrophy (ILMA)

Secondary: Development/worsening of intralesional macular atrophy (ILMA)

Secondary: Development/worsening of geographic atrophy (GA)

Secondary: Development/worsening of geographic atrophy (GA)

Secondary: Low luminance acuity (LLA) in study eye

Secondary: Low luminance acuity (LLA) in study eye

Secondary: Low luminance acuity (LLA) in study eye, adjusted for BCVA at IVAN exit (sensitivity analysis)

Secondary: Low luminance acuity (LLA) in study eye, adjusted for BCVA at IVAN exit (sensitivity analysis)

Secondary: Quality of life (EQ5D)

Secondary: Quality of life (EQ5D)

Adverse events

Adverse events

More information

Substantial protocol amendments (globally)

Interruptions (globally)

Limitations and caveats

More information

Clinical Trial Results:
Five year observational follow-up of the IVAN trial cohort: a study of function and morphology