STUDY DESIGN AND PATIENTS

This study adheres to the tenets of the Declaration of Helsinki. Institutional Review Board/Ethics Committee approval was obtained at each study site before study initiation (Supplementary Table S1). Informed consent was obtained from all patients before enrollment. Randomization and treatment masking were not performed in GALE because it was a prospective, open-label extension. The GALE study design and methodology have been described. In brief, GALE is a phase 3, multicenter, open-label extension study evaluating long-term safety and efficacy of pegcetacoplan over 36 months in patients with GA secondary to AMD. Patients who completed the 24-month phase 3 OAKS or DERBY trials or completed the phase 1b study (APL2-103) evaluating pegcetacoplan safety (ClinicalTrials.gov identifier: NCT03777332) were eligible to enroll in the GALE open-label extension study. Of note, patients with either nonsubfoveal or subfoveal GA were eligible to enroll in the preceding studies. ^, The analyses described herein were based on data through the first 12 months of GALE.

PEGCETACOPLAN TREATMENT GROUPS AND PROJECTED SHAM

Pegcetacoplan 15 mg/0.1 mL intravitreal injection was administered either PM or PEOM in the study eye of all patients in GALE. In OAKS and DERBY, patients were randomly assigned in a 2:2:1:1 ratio to receive PM, PEOM, sham monthly (SM), or sham every other month (SEOM). Patients who received PM or PEOM in OAKS, DERBY, or the phase 1b study continued the same treatment regimen in GALE (PM-PM and PEOM-PEOM, respectively). Patients who were in the SM or SEOM arms crossed over to receive active pegcetacoplan intravitreal treatment in GALE at the same interval (SM-PM and SEOM-PEOM, respectively).

Because all patients in GALE received active pegcetacoplan treatment, a projected sham, calculated from the prior 24-month GA growth rate of sham-observed patients in OAKS and DERBY averaged across four 6-month segments, served as the comparator for the first 12 months of GALE (months 24-36). This approach is based on robust evidence that prior 2-year GA growth rate is a strong indicator of subsequent 2-year growth rate, and that the rate of GA growth follows a linear trend over this time. ^, Moreover, in OAKS and DERBY, the GA growth rate in sham-observed eyes was linear at approximately 1 mm ² every 6 months (or 2 mm ² every 12 months) over the 24-month study period, matching the projected sham rate of GA lesion growth ( Figure 1 ).

Mean rate of change in GA area growth over 36 months in the sham groups and untreated fellow eyes ^a in patients with bilateral GA. ^b eAMD, exudative age-related macular degeneration; GA, geographic atrophy. ^a For inclusion in this analysis, untreated fellow eyes needed to have an absence of eAMD in their medical history, a baseline GA area between 2.5 and 17.5 mm ² , the presence of any pattern of hyperautofluorescence in the junctional zone of GA, and GA not confluent with any peripapillary atrophy. ^b The GA growth rate in untreated fellow eyes was observed to be linear over 36 months and nearly identical to the GA growth rate in eyes of sham-observed patients (months 0-24) and projected sham (months 24-36).

The validity of the projected sham comparator in GALE was further confirmed through comparison with the average GA growth rate in treatment-naive fellow eyes from patients with bilateral GA at baseline whose fellow nonstudy eye met OAKS and DERBY study inclusion. These untreated fellow eyes needed to have the following characteristics at baseline in OAKS or DERBY: an absence of exudative AMD (eAMD) in their medical history, a baseline GA area between 2.5 and 17.5 mm ² , the presence of any pattern of hyperautofluorescence in the junctional zone of GA, and GA not confluent with any peripapillary atrophy. The GA growth rate in untreated fellow eyes was observed to be linear over 36 months, and this trend was nearly identical to the growth rate of sham-observed patients (months 0-24) and projected sham (months 24-36), corroborating the use of projected sham in the efficacy analyses ( Figure 1 ).

OUTCOMES

The objective of GALE is to evaluate safety and efficacy of long-term pegcetacoplan treatment in this chronic retinal disorder. The prespecified primary endpoint includes incidence and severity of ocular and systemic (nonocular) treatment-emergent adverse events (TEAEs). A secondary endpoint is the mean rate of change in GA area using a piecewise linear slope model. An additional prespecified endpoint is the total number of scotomatous points as assessed by microperimetry. Outcomes were evaluated in the first 12 months of GALE which represent data from OAKS and DERBY baseline through month 36.

STATISTICAL ANALYSIS

The safety population included all patients who enrolled in GALE and received ≥ 1 injection of pegcetacoplan in GALE. Descriptive statistics were used to summarize ocular and nonocular TEAEs in each treatment group in the first 12 months of GALE. The incidence of ocular TEAEs of interest included new-onset eAMD, intraocular inflammation (IOI), ischemic optic neuropathy (ION), and infectious endophthalmitis; these were evaluated in an integrated population of the total OAKS, DERBY, and GALE patients across 36 months.

Events of new-onset eAMD were investigator-determined and included Medical Dictionary for Regulatory Activities (MedDRA) preferred terms of choroidal neovascularization and neovascular AMD, and only patients without a history of eAMD in their study eye were eligible for the analysis of new-onset eAMD. Patients who developed eAMD in the study eye during OAKS or DERBY were eligible to participate in GALE. Across the phase 3 studies of pegcetacoplan (OAKS, DERBY, and GALE), determination of the eAMD diagnosis and decision to treat with anti-vascular endothelial growth factor (anti-VEGF) remained at the sole discretion of the investigator, not a reading center. Consistent with the previously published protocol in OAKS and DERBY for monitoring of eAMD, all investigator-determined eAMD events were reported regardless of reading center confirmation. If eAMD was suspected by an investigator, the protocol recommended multimodal imaging (spectral domain optical coherence tomography [OCT], fundus fluorescein angiography, and, if available, OCT-angiography) and submission to the central reading center. The reading center (DARC, now Voiant) subsequently assessed the images. In GALE, investigators were permitted to treat immediately with anti-VEGF according to local standard of care (in contrast to OAKS and DERBY where treatment was required to be on-label dosing of ranibizumab or aflibercept) and on the same day as pegcetacoplan; anti-VEGF was administered at least 30 minutes before pegcetacoplan per protocol. Since reading center assessments were not available in real time and the protocol-recommended fundus fluorescein angiography was not always conducted due in part to the COVID-19 pandemic, discrepancies between the reading center and physician determinations were noted. The reported rates of new-onset eAMD reflect solely physician determinations per protocol and are described herein. Best-corrected visual acuity (BCVA) was measured using the Early Treatment Diabetic Retinopathy Study (ETDRS) method by certified staff.

The efficacy endpoint of mean rate of change in GA area was assessed in the modified full analysis set and in patient subgroups defined by GA location (nonsubfoveal or subfoveal) at antecedent study baseline (i.e., the primary pivotal studies, OAKS and DERBY). The modified full analysis set included patients who received ≥ 1 pegcetacoplan injection in GALE, were in the intention-to-treat set in the preceding OAKS or DERBY study and were not previously enrolled in the phase 1b safety set. The difference between pegcetacoplan-treated groups and sham/projected sham in the mean rate of change in GA area from antecedent study baseline through month 36 was determined with a rate-of-change analysis using a piecewise linear slope model. Mean rate of change in GA area (based on fundus autofluorescence imaging) was estimated from baseline to month 36 with knots at months 12 and 24, and the mean rate of change in GA area with projected sham from months 24 to 36 was calculated by averaging the mean rate of change in each 6-month period from OAKS and DERBY baseline through month 24 in the sham-observed group. Comparisons between pegcetacoplan and sham were assessed up to month 24 and comparison versus projected sham was assessed between months 24 and 36.

The number of scotomatous points on microperimetry was evaluated in patients who were in the modified intent-to-treat set in OAKS, including patients who did and did not enroll in GALE, using the Macular Integrity Assessment device (iCare, Padova, Italy). Microperimetry testing was performed using a test grid with 20-degree macular coverage and 68 test loci. ^, The test was performed with the grid centered on the anatomic fovea, with a 4-2 threshold projection strategy. Microperimetry was not required or assessed in DERBY. Comparisons between pegcetacoplan and sham were assessed up to month 24 and between pegcetacoplan and sham crossover between months 24 and 36. The change from OAKS baseline in the number of scotomatous points was analyzed using a mixed-effect model for repeated measures including covariates for treatment group, time, presence of eAMD in the fellow eye, antecedent study baseline GA area, antecedent study baseline number of scotomatous points, and interaction terms including time × treatment and time × antecedent study baseline number of scotomatous points.

PATIENT CHARACTERISTICS

Of the 792 patients who enrolled in GALE, 790 were included in the safety population and 780 were included in the modified full analysis set ( Figure 2 ). Over the first 12 months of GALE, 6305 pegcetacoplan injections were administered among 790 patients. Of note, 696/790 (88.1%) patients received ≥ 75% of their planned pegcetacoplan injections (≥ 9 of 12 injections in the PM-PM and SM-PM groups and 5 of 6 injections in the PEOM-PEOM and SEOM-PEOM groups). In total, 18 025 total pegcetacoplan injections were administered across the integrated OAKS, DERBY, and 12-month GALE population (n = 1111).

Study design and patient disposition. AE, adverse event; PEOM, pegcetacoplan every other month; PM, pegcetacoplan monthly; SEOM, sham every other month; SM, sham monthly. ^a Efficacy analyses in GALE were performed in the analysis set from antecedent study baseline (month 0) through the first 12 months of GALE (month 36). Ocular and nonocular treatment-emergent AEs in the GALE safety set were reported for the first 12 months of GALE (months 24-36). ^b The enrolled set included all patients who provided written informed consent and were enrolled in GALE. ^c The safety set included all patients who were enrolled in GALE and received ≥ 1 injection of pegcetacoplan in GALE. Ten patients enrolled in GALE from the phase 1b APL2-103 study (9 in PM-PM, 1 in PEOM-PEOM) were included in the safety set. ^d The modified full analysis set included all patients who were in the intention-to-treat population in OAKS and DERBY, were not enrolled in the phase 1b APL2-103 study, and received ≥ 1 injection of pegcetacoplan in GALE. ^e The integrated OAKS, DERBY, and GALE safety set included all patients who received ≥ 1 injection of pegcetacoplan in OAKS, DERBY, or GALE. AEs of interest including exudative age-related macular degeneration, intraocular inflammation, infectious endophthalmitis, and ischemic optic neuropathy were reported for this data set. ^f Percentages calculated based on total number of patients in each group in the safety set.

The baseline demographic and clinical characteristics of the GALE study population at OAKS and DERBY baseline have been reported. Demographic and study eye characteristics of patients at GALE baseline are summarized across treatment groups in Table 1 . At GALE baseline, the majority of patients were White and female, and the mean age was 79.6 years (median: 80.0; range: 62.0-98.0). In the study eye, mean GA area was 11.78 mm ² (range: 2.9-40.9 mm ² ), and 76.6% (605/790) of patients had GA area ≥ 7.5 mm ² . At baseline of the preceding trials, 36.3% (287/790) of patients had nonsubfoveal GA and 63.7% (503/790) had subfoveal GA in their study eye. At GALE baseline, 28.2% (223/790) of patients had nonsubfoveal GA and 71.8% (567/790) of patients had subfoveal GA in their study eye. The baseline characteristics of the untreated fellow eyes (n = 407) were similar to those of the study eyes of patients in the SM-PM and SEOM-PEOM groups ( Table 1 ). In the fellow eye, mean GA area was 12.36 mm ² (range: 3.10-27.80 mm ² ), 82.8% (337/407) of patients had GA area ≥ 7.5 mm ² , and 30.5% (124/407) of patients had nonsubfoveal GA.

SAFETY

In total, 727/790 (92.0%) patients completed the first 12 months of treatment in GALE. Reasons for study discontinuation included withdrawal of consent (n = 26), death (n = 21), adverse events (n = 11), loss to follow-up (n = 4), and physician’s decision (n = 1). Table 2 shows the most common ocular TEAEs, defined as occurring in ≥ 3% of patients in ≥ 1 treatment group, including vitreous floaters, ocular discomfort, new-onset eAMD, increased intraocular pressure, conjunctival hemorrhage, cataract, retinal hemorrhage, posterior capsule opacification, and blepharitis. Two patients had an ocular TEAE that led to study discontinuation: 1 patient in the PM-PM group discontinued due to vitritis and 1 patient in the SEOM-PEOM group discontinued because of retinal vein occlusion. These 2 events occurred in the first 6 months of GALE and were previously reported. There were no additional ocular TEAEs that led to study discontinuation in the subsequent 6 months of GALE. The most common nonocular TEAE was COVID-19, occurring in 106 (13.4%) patients (Supplementary Table S2).

TABLE 2

Study Eye Ocular TEAEs in the First 12 Months of GALE ^a

	PM-PM (n = 250)	PEOM-PEOM (n = 268)	SM-PM (n = 129)	SEOM-PEOM (n = 143)
Any ocular TEAE, n (%)	98 (39.2)	74 (27.6)	61 (47.3)	55 (38.5)
Most common ocular TEAEs, n (%) b
Vitreous floaters	11 (4.4)	6 (2.2)	13 (10.1)	8 (5.6)
Ocular discomfort	9 (3.6)	7 (2.6)	8 (6.2)	10 (7.0)
New-onset eAMD c	17 (7.9)	5 (2.0)	7 (5.6)	4 (2.9)
Intraocular pressure increased	12 (4.8)	14 (5.2)	5 (3.9)	2 (1.4)
Conjunctival hemorrhage	8 (3.2)	7 (2.6)	12 (9.3)	6 (4.2)
Cataract	13 (5.2)	5 (1.9)	5 (3.9)	5 (3.5)
Retinal hemorrhage	8 (3.2)	6 (2.2)	3 (2.3)	2 (1.4)
Posterior capsule opacification	5 (2.0)	3 (1.1)	5 (3.9)	2 (1.4)
Blepharitis	0	1 (0.4)	7 (5.4)	1 (0.7)
Serious ocular TEAEs, n (%)	2 (0.8)	2 (0.7)	1 (0.8)	1 (0.7)
Ischemic optic neuropathy	1 (0.4) d	0	0	0
Vitritis	1 (0.4) e	0	0	0
Blindness (transient)	0	1 (0.4) d	0	0
Visual impairment	0	1 (0.4) d	0	0
Retinal artery occlusion	0	0	0	1 (0.7) d
Infectious endophthalmitis	0	0	1 (0.8) f	0

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Ocular Features of VEXAS Syndrome: A Systematic Review and Meta-analysis
2. Chlorhexidine Versus Povidone-Iodine for Intravitreal Injection Antisepsis: A Systematic Review and Meta-Analysis
3. Choroidal Layer Analysis Under Flat Irregular Pigment Epithelial Detachments in Central Serous Chorioretinopathy
4. Reply to Comment on: Dropped Lens Fragments Retrieval Using Endoscopy-Assisted Pars Plana Vitrectomy in Patients With Corneal Opacity

Stay updated, free articles. Join our Telegram channel

Join