Efficacy of intravitreal pegcetacoplan in geographic atrophy- results from the DERBY and OAKS trials.pptx

These are Efficacy of intravitreal pegcetacoplan in geographic atrophy- results from the DERBY and OAKS trials.pptx

1. Jeffrey Heier, Charles Wykoff, Rishi Singh, Nathan Steinle, David Boyer, Jordi Monés, Giovanni Staurenghi, Frank G. Holz, Caleb Bliss, Pascal Deschatelets, Federico Grossi, Cedric Francois, Ramiro Ribeiro 30 September 2021 Retina Society 2021, Chicago, USA Efficacy of intravitreal pegcetacoplan in geographic atrophy: results from the DERBY and OAKS trials

2. 2 Dr. Heier is a consultant for 4DMT, Adverum, Aerie, Aerpio, Aldeyra, Allegro, Alzheon, Annexon, Apellis , Aprea, Asclepix, Aviceda, BVT, Dark Horse, DTx, Eloxx, Galimedix, Genentech, Graybug, Gyroscope, Iveric Bio, jCyte, Kanghong, LensGen, NGM, Novartis, Ocular Therapeutix, OcuTerra, Oxurion, Palatin, Regeneron, Regenxbio, Stealth, Thea, Version, Vinci, and Voyant. Dr. Heier receives research funding from: Apellis , Asclepix, Bayer, Genentech, Graybug, Gyroscope, Hemera, Iveric , Kanghong, Kodiak, NGM, Notal Vision, Novartis, Regeneron, Regenxbio, and Stealth. Studies funded by Apellis Pharmaceuticals Disclosures

3. 3 Dysregulation of the complement cascade has been implicated in GA pathogenesis All 3 complement pathways end in the central cleavage of C3 Pegcetacoplan is a pegylated, highly- selective peptide that binds C3, preventing its cleavage Inhibition of C3 blocks steps in the complement cascade needed for opsonization, inflammation, and formation of MAC Introduction C5a C5b C5 C3 Inflammation Cell death, secretion, lysis or proliferation C3b LECTIN Polysaccharides on microorganisms CLASSICAL Antigen- antibody complexes ALTERNATIVE Pathogen cell surfaces and nonspecific/spontaneous activation C3a C3 convertase C5 convertase MAC C5b, C6, C7, C8, C9 Phagocytosis Activation of cell removal and antigen uptake by APCs Opsonisation C3 amplification loop Pegcetacoplan Figure adapted from Ricklin D, et al. Immunol Rev 2016;274(1):33- 58. APC=antigen- presenting cell; GA=geographic atrophy; MAC=membrane attack complex. Liao DS, et al. Ophthalmology 2020;127:186–95.

4. 4 *P<0.1 was the predefined threshold for statistical significance in FILLY. AMD=age-related macular degeneration; GA=geographic atrophy; LS=least squares; M=Month; PEOM=pegcetacoplan every other month; PM =pegcetacoplan monthly; SE=standard error. Phase 3 DERBY & OAKS objective: to assess the efficacy and safety of multiple intravitreal injections of pegcetacoplan in patients with GA secondary to AMD Introduction and objective M6 PEOM (n=78) M12 PM (n=84) Phase 2 FILLY Results Change from baseline in square root GA lesion size (mm) Baseline M2 Sham (n=80, pooled) 0.1 0.2 0.3 0.4 Primary endpoint Month 12 LS mean (±SE) change from baseline in square root GA lesion (mm) 20% (every other month) reduction p=0.067 vs sham* 29% (monthly) reduction p=0.008 vs sham* 20% Liao DS, et al. Ophthalmology 2020;127:186–95. 29%

5. 5 Global phase 3 program: Design of studies AMD=age-related macular degeneration; EOM=every other month; GA=geographic atrophy. Double masked Patients with GA secondary to AMD ~600 patients at ~200 sites globally in 2 studies (1258 enrollees total) Randomized 2:2:1:1 Primary endpoint at 12 months Change in total area of GA lesions based on fundus autofluorescence End of study at 24 months Protocol study number, APL-2 303 (DERBY); NCT03525600 Protocol study number, APL-2 304 (OAKS); NCT03525613 Pegcetacoplan 15 mg/0.1 mL monthly Pegcetacoplan 15 mg/0.1 mL EOM Sham monthly Sham EOM

6. 6 Key inclusion and exclusion criteria AMD=age-related macular degeneration; BCVA=best- corrected visual acuity; CNV=choroidal neovascularisation; DA=disc area; ETDRS=Early Treatment Diabetic Retinopathy Study; GA=geographic atrophy; RPE=retinal pigment epithelium. Key inclusion criteria Age ≥60 years BCVA ≥24 letters ETDRS (20/320 Snellen equivalent) GA lesion requirements: Total size: ≥2.5 and ≤17.5 mm 2 Foveal and extrafoveal GA allowed If multifocal, at least 1 focal lesion must be ≥1.25 mm 2 (0.5 DA) Presence of perilesional hyperautofluorescence Key exclusion criteria GA secondary to a condition other than AMD, such as Stargardt disease in either eye Ocular history of or active CNV in the study eye, including presence of RPE tear (assessed by reading center) Ocular history of active CNV in the fellow eye is not exclusionary

7. 7 Key endpoints *Patients must meet following criteria: (a) able to detect fixation target, (b) total elapsed time to complete 68 -point exam <30 min, (c) reliability test ratio <20%, (d) willing and able to undertake microperimetry in investigator’s opinion. BCVA=best-corrected visual acuity; FAF=fundus autofluorescence; LL- BCVA=low-luminance BCVA; GA=geographic atrophy. Primary Change from baseline to Month 12 in total area of GA lesion(s) in the study eye (in mm 2 ) based on FAF Secondary (prespecified analyses to be conducted at 24 months) BCVA, LL-BCVA, low- luminance deficit Reading speed Microperimetry (OAKS study only*) — Macular Integrity Assessment (MAIA) device National Eye Institute Visual Functioning Questionnaire 25- Item Version (NEI VFQ- 25) Functional Reading Independence Index (FRI) composite score

8. These analyses were performed on the modified intention- to-treat (mITT) population. The mITT population was defined as all randomized patients who received Key demographics and baseline study eye characteristics OAKS Characteristic PM (N=202) PEOM (N=205) Sham Pooled (N=206) Age, mean (SD) 78.8 (7.24) 78.1 (7.74) 78.6 (7.26) Female, n (%) 125 (61.9%) 117 (57.1%) 133 (64.6%) Male, n (%) 77 (38.1%) 88 (42.9%) 73 (35.4%) Geographic region US, n (%) 147 (72.8%) 142 (69.3%) 147 (71.4%) ROW, n (%) 55 (27.2%) 63 (30.7%) 59 (28.6%) Caucasian, n (%) 185 (91.6%) 189 (92.2%) 187 (90.8%) GA lesion size (mm 2 ), mean (SD) 8.18 (3.893) 8.29 (3.904) 8.20 (3.722) Square root GA lesion size (mm), mean (SD) 2.78 (0.682) 2.80 (0.674) 2.79 (0.649) GA lesion size, n (%) <7.5 mm 2 101 (50.0%) 99 (48.3%) 104 (50.5%) GA lesion location, n (%) Extrafoveal 86 (42.6%) 74 (36.1%) 60 (29.1%) GA lesion focality, n (%) Unifocal 59 (29.2%) 62 (30.2%) 68 (33.0%) Intermediate/large drusen, n (%) >20 93 (46.0%) 104 (50.7%) 103 (50.0%) NL- BCVA (ETDRS letters), mean (SD) 61.0 (15.30) 58.2 (17.03) 57.5 (16.57) at least 1 injection of pegcetacoplan or sham and have baseline and at least 1 post -baseline value of GA lesion area in the study eye. GA=geographic atrophy; mm=millimeters; n=number of patients; NL-BCVA=normal luminance best- corrected visual acuity; PM=pegcetacoplan monthly; PEOM=pegcetacoplan every other month; ROW=rest of world; SD=standard deviation; US=United States. 8

9. DERBY Characteristic PM (N=201) PEOM (N=200) Sham Pooled (N=194) Age, mean (SD) 78.7 (6.91) 79.2 (7.07) 78.6 (7.29) Female, n (%) 118 (58.7%) 120 (60.0%) 122 (62.9%) Male, n (%) 83 (41.3%) 80 (40.0%) 72 (37.1%) Geographic region US, n (%) 142 (70.6%) 122 (61.0%) 122 (62.9%) ROW, n (%) 59 (29.4%) 78 (39.0%) 72 (37.1%) Caucasian, n (%) 187 (93.0%) 185 (92.5%) 187 (96.4%) GA lesion size (mm 2 ), mean (SD) 8.36 (4.182) 8.22 (3.886) 8.26 (4.260) Square root GA lesion size (mm), mean (SD) 2.80 (0.723) 2.79 (0.677) 2.78 (0.734) GA lesion size, n (%) <7.5 mm 2 99 (49.3%) 98 (49.0%) 94 (48.5%) GA lesion location, n (%) Extrafoveal 72 (35.8%) 81 (40.5%) 73 (37.6%) GA lesion focality, n (%) Unifocal 54 (26.9%) 53 (26.5%) 66 (34.0%) Intermediate/large drusen, n (%) >20 78 (38.8%) 78 (39.0%) 98 (50.5%) NL- BCVA (ETDRS letters), mean (SD) 59.5 (17.40) 58.9 (15.97) 59.1 (16.85) These analyses were performed on the modified intention- to-treat (mITT) population. The mITT population was defined as all randomized patients who received Key demographics and baseline study eye characteristics at least 1 injection of pegcetacoplan or sham and have baseline and at least 1 post -baseline value of GA lesion area in the study eye. GA=geographic atrophy; mm=millimeters; n=number of patients; NL-BCVA=normal luminance best- corrected visual acuity; PM=pegcetacoplan monthly; PEOM=pegcetacoplan every other month; ROW=rest of world; SD=standard deviation; US=United States. 9

10. 10 16% (every other month) reduction p=0.0052 vs sham 22% (monthly) reduction p=0.0003 vs sham Sham (n=206, pooled) Pegcetacoplan monthly and every other month met the primary endpoint in OAKS LS means estimated from a mixed- effects model for repeated measures (MMRM). The modified intention -to-treat population was used for the analysis, defined as all randomized patients who received at least 1 injection of pegcetacoplan or sham and have baseline and at least 1 post -baseline value of GA lesion area in the study eye. M6 M8 PEOM (n=205) GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. M12 PM (n=202) 2 0.5 1 1.5 2.5 LS mean change (±SE) from baseline in GA lesion (mm²) Baseline M2 M4 M10

11. Pegcetacoplan did not meet the primary endpoint in DERBY 11% (every other month) reduction p=0.0750 vs sham 12% (monthly) reduction p=0.0528 vs sham LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 11 Baseline M2 M4 M6 M8 PEOM (n=200) M10 M12 PM (n=201) 2 0.5 1 1.5 2.5 Sham (n=194, pooled) LS mean change (±SE) from baseline in GA lesion (mm²)

12. Pegcetacoplan reduced lesion growth in a prespecified analysis of OAKS and DERBY combined 14% (every other month) reduction p=0.0012 vs sham 17% (monthly) reduction p<0.0001 vs sham Baseline M2 M4 M6 M8 PEOM (n=405) M10 M12 PM (n=403) 2 0.5 1 1.5 2.5 Sham (n=400, pooled) LS mean change (±SE) from baseline in GA lesion (mm²) LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 12

13. OAKS DERBY Sham (n=60, pooled) M6 M8 PEOM (n=74) M10 M12 PM (n=86) 0.5 1 1.5 2 2.5 3 Baseline M2 M4 21% (every other month) reduction p=0.0159 vs sham 35% (monthly) reduction p<0.0001 vs sham Pegcetacoplan reduced lesion growth in patients with extrafoveal lesions in a prespecified analysis 0.5 1 1.5 2 3 16% (monthly) reduction P=0.0712 vs sham 25% (every other month) reduction p=0.0028 vs sham Sham (n=73, pooled) M6 M8 PEOM (n=81) M10 M12 PM (n=72) Baseline M2 M4 16% 25% 21% 2.5 35% LS mean change (±SE) from baseline in GA lesion (mm²) LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 13

14. 26% (monthly) reduction p<0.0001 vs sham Pegcetacoplan reduced lesion growth in patients with extrafoveal lesions in a prespecified combined analysis 23% (every other month) reduction p=0.0002 vs sham 3.0 2 LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 14 1.5 1 0.5 2.5 Baseline M2 M4 M6 M8 PEOM (n=155) M10 M12 PM (n=158) Sham (n=133, pooled) LS mean change (±SE) from baseline in GA lesion (mm²)

15. Pegcetacoplan reduced lesion growth in patients with foveal lesions in OAKS in a prespecified analysis 0.5 1 1.5 Baseline M2 M4 Sham (n=146, pooled) M6 M8 PEOM (n=131) M10 M12 PM (n=116) 16% (every other month) reduction p=0.0179 vs sham 16% (monthly) reduction p=0.0324 vs sham 0.5 1 7% (monthly) reduction p=0.3611 vs sham 1% (every other month) increase p=0.8993 vs sham OAKS DERBY 2 2 16% 1.5 16% 1% 7% Baseline M2 Sham (n=121, pooled) M4 M6 PEOM (n=119) M8 M10 PM (n=129) M12 LS mean change (±SE) from baseline in GA lesion (mm²) LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 15

16. 9% (every other month) reduction p=0.1127 vs sham 12.0% (monthly) reduction p=0.0280 vs sham Pegcetacoplan reduced lesion growth in patients with foveal lesions in a prespecified combined analysis 2 LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 16 0.5 1 1.5 Baseline M2 M4 M6 M8 PEOM (n=250) M10 M12 PM (n=245) Sham (n=267, pooled) LS mean change (±SE) from baseline in GA lesion (mm²)

17. 0.5 1 1.5 2 p=0.0023 vs sham Prespecified subgroup analysis: United States sites 0.5 1 1.5 2 22% (monthly) reduction 19% (monthly) reduction p=0.0169 vs sham OAKS DERBY 2.5 2.5 13% (every other month) reduction 13% (every other month) reduction p=0.0769 vs sham p=0.1097 vs sham Baseline M2 M4 Sham (n=147, pooled) M6 M8 PEOM (n=142) M10 M12 PM (n=147) Baseline M2 M4 Sham (n=122, pooled) M6 M8 PEOM (n=122) M10 M12 PM (n=142) 13% 19% 13% 22% LS mean change (±SE) from baseline in GA lesion (mm²) LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 17

18. 0.5 1 1.5 2 Prespecified subgroup analysis: Rest of World sites 0.5 1 2 21% (monthly) reduction 3% (monthly) increase p=0.0420 vs sham p=0.7799 vs sham OAKS DERBY 2.5 2.5 25% (every other month) reduction 7% (every other month) reduction p=0.0069 vs sham p=0.4468 vs sham 21% 25% 1.5 Baseline M2 Sham (n=72, pooled) M6 M8 PEOM (n=78) M10 M12 PM (n=59) Sham (n=59, pooled) M6 M8 PEOM (n=63) M10 M12 PM (n=55) Baseline M2 M4 3% 7% M4 LS mean change (±SE) from baseline in GA lesion (mm²) LS means estimated from a mixed- effects model for repeated measures. The modified intention- to-treat population was used for the analysis. GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error. 18

19. 19 Patients with bilateral GA were included in this analysis In addition, for a subject to be included, the fellow eye had to meet the following criteria: Absence of CNV in the medical history Baseline GA lesion size between 2.5 and 17.5 mm 2 Presence of any pattern of hyperautofluorescence in the junctional zone of GA GA not confluent with peripapillary atrophy Analysis of fellow eye vs study eye lesion growth CNV=choroidal neovascularization; GA=geographic atrophy.

20. 20 Pegcetacoplan reduced lesion growth in an analysis of the study eye vs untreated fellow eye, supporting primary analysis 0.5 1 1.5 2 OAKS AND DERBY PEOM (n=211) PM (n=177) Study Fellow 16% slower growth vs fellow eye p < 0.0001 LS Mean change (±SE) from baseline in GA lesion(s) (mm 2 ) Baseline M6 M12 0.5 1 1.5 2 Study Fellow 11% slower growth v s fellow eye p = 0.0011 0.5 1 1.5 2 4% faster growth vs fellow eye p = 0.2666 Study Fellow Baseline M6 M12 Baseline M6 M12 Sham pooled (n=195) In DERBY and OAKS, study eye vs fellow eye comparison was prespecified; statistical modeling was performed post -hoc. LS means estimated from a mixed-effects model for repeated measures. The modified intention-to-treat population was used for the analysis. In addition, patients must have bilateral GA and a fellow eye that meets the following key characteristics at baseline: absence of CNV in the medical history; baseline GA lesion size between 2.5 and 17. 5 mm 2 and have at least one study eye or fellow eye at measurement at Month 6 or Month 12. In the FILLY analysis, all bilateral GA patients are included. CNV=choroidal neovascularization; GA=geographic atrophy; LS=least square; M=month; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly; SE=standard error.

21. 21 Post-hoc covariate analysis: Background Question: What is the real effect size of pegcetacoplan? Apellis has taken a multi- pronged approach to understand inconsistent results across trials, including a significant investigation of study operations, regional differences, as well as imbalances in baseline characteristics that impact lesion growth We undertook a post-hoc analysis to examine the potential contribution of baseline characteristic imbalances on the diverging results. The 8 most relevant variables related to GA were investigated for imbalance, and analyses were re- run adjusting for the imbalanced variables: We are presenting initial findings from this investigation GA=geographic atrophy; PEOM=pegcetacoplan every other month; PM=pegcetacoplan monthly. Study eye focality Imbalanced in DERBY ( favoring sham ) Study eye lesion location Imbalanced in OAKS ( favoring sham ) Study eye lesion size Study eye pseudodrusen Study eye low luminance deficit Imbalanced in FILLY ( favoring PM ) Region GA laterality Study eye intermediate/large drusen Imbalanced in DERBY ( favoring sham ) and FILLY (favo ring PEOM )

22. 22 Converging treatment effect of pegcetacoplan across OAKS, DERBY, in FILLY in covariate- adjusted post- hoc analysis Sham (n=205, pooled) PEOM (n=205) PM (n=202) 0.5 1 1.5 2 2.5 Baseline M2 M4 M6 M8 M10 M12 18% (every other month) reduction 26% (monthly) reduction LS means estimated from a mixed- effects model for repeated measures. The mITT population was used for the analysis. SE=standard error; PM, pegcetacoplan monthly; PEOM, pegcetacoplan every other month. LS Mean Change (±SE) from Baseline in GA Lesion (mm²) 0.5 1 18% 26% 1.5 2 16% (monthly) reduction 15% (every other month) reduction DERBY OAKS FILLY Sham (n=194, pooled) PEOM (n=200) PM (n=201) Baseline M2 M4 M6 M8 M10 M12 Sham (n=80, pooled) PEOM (n=78) PM (n=84) 0.5 1 1.5 2 2.5 2.5 Baseline M2 M4 M6 M8 M10 M12 18% (every other month) reduction 25% (monthly) reduction 15% 16% 18% 25%

23. 23 Pegcetacoplan monthly and every other month met the primary endpoint in OAKS Pegcetacoplan monthly and every other month did not meet the primary endpoint in DERBY Pegcetacoplan demonstrated greater efficacy in patients with extrafoveal lesions at baseline OAKS, DERBY, and FILLY all show consistent efficacy of pegcetacoplan in treated study eyes versus untreated fellow eyes In a post-hoc analysis, after correcting for disparities in baseline characteristics, OAKS, DERBY, and FILLY results are more convergent. Investigation is still ongoing The pegcetacoplan GA development program includes over 1,500 patients across OAKS, DERBY, and FILLY, collectively demonstrating slowing of GA progression by pegcetacoplan monthly and every other month Conclusions GA=geographic atrophy.