Front Line Diabetic Retinopathy What Not to Miss and Why

Transcription

1 Front Line Diabetic Retinopathy What Not to Miss and Why David M Brown MD FACS Clinical Professor of Ophthalmology Blanton Eye Institute, Houston Methodist Hospital Baylor College of Medicine Retina Consultants of Houston Houston, TX USA 1 Relevant Disclosures Consultant/ Grant Support: Regeneron, Bayer, Senju, Allergan, Optos, Zeiss, Heidelberg, OHR, Biotime, Gemini, Genentech/Roche, Novartis, Apellis, Regenxbio, Chengdu Kanghong Biotechnology Co-patent holder on OPTOS de-warping algorithms Genentech/ Roche & OPTOS provided support for Trex DME & DAVE DMB had full control of the presentation 2 Case Study 1 57 Year Old Female Blurry Vision OS Diabetic for 15 years- Hemoglobin A1c = 9.2 Refraction- Plano OD-2/ OS- 2/

2 ETDRS Standardized Eye Chart Outcomes Measure for Clinical Trials Early Treatment Diabetic Retinopathy Study Vision Gain 7 8 ETDRS: Vision Loss Macular Laser for DME Standard of care No impact on underlying disease progression Reduces risk of vision loss, but few patients experience visual improvement Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study report number 1. Early Treatment Diabetic Retinopathy Study research group. - Arch Ophthalmol - 1-DEC-1985; 3(12): Focal Laser Grid Laser Courtesy of Dr Donald D Amico Anti-VEGF Key Studies: Ranibizumab RISE/RIDE: 2 parallel phase III, multicenter, double-masked, sham-injection controlled, randomized studies Assessed safety and efficacy of intravitreal ranibizumab for the treatment of DME In the third year, patients who had received sham therapy were eligible to switch to treatment with ranibizumab

3 Sustained Improvement With Ongoing Anti-VEGF Therapy Mean change in BCVA from baseline (pooled) Effects of Treatment Delay Mean change in BCVA from baseline (pooled) Severe VA loss (15 letters) significantly reduced Rapid improvements in vision & anatomy maintained for 3 years Brown D et al, RISE and RIDE Research Group. Ophthalmology. 213;12: With crossover to 1 year of.5 mg ranibizumab therapy at third year, original sham treatment group s visual gains were lower than those seen in first year of ranibizumab-treated groups (2.8 vs.6 and 11.1 letters) Delayed treatment reduced magnitude of VA benefits of anti-vegf therapy Brown D et al. RISE and RIDE Research Group. Ophthalmology. 213;12: year-old Male With DME Treated With Monthly Ranibizumab Baseline Snellen = 2/ Month 36 Snellen = 2/5 August, 212 First & only FDA approved medication for DME Changed the standard of care for the first time in > 25 years Open Label Extension RISE/RIDE OLE Maintenance of Vision Boyer D, AAO Poster Boyer D, AAO Poster

4 8 Frequent Visits Multiple Injections 2/8 2/32 Long-Term Management Following RIDE & RISE Treatment Frequency Beyond 3 Years Patients (N = 5) %: injections Change VA Months Mean.5 injections Annualized 3.8 injections Anti-VEGF Injections Number of Ranibizumab Injections During OLE (mean follow-up: 1.1 months) OLE Ranibizumab Exposure Anti-VEGF Key Studies: Aflibercept VIVID-DME and VISTA-DME: assessed safety and efficacy of aflibercept vs laser alone in the treatment of DME Treatment groups: intravitreal aflibercept monthly, every 2 months (after 5 initial monthly injections), or laser photocoagulation 21 Boyer D, AAO Poster Korobelnik JF, Brown DM, et al. Ophthalmology. 21 Jul 8. Study Sites Japan Mean Change in BCVA Primary Analysis (LOCF) Europe Austria, Czech Republic, Denmark, France, Germany, Hungary, Italy, Poland, Spain USA ETDRS letters VIVID VISTA Control 2q 2q8 *P<.1 vs control *P<.1 vs control 11.7*.3* 1.6.5*.* VIVID 73 Centers Patients Australia VISTA 5 Centers 61 Patients Week Compared with baseline Primary analysis (LOCF): excludes patients who received rescue treatment; P values are not confirmatory VIVID FAS: control: n=132; IAI 2q: n=136; IAI 2q8: n=135; VISTA FAS: control: n=15; IAI 2q: n=15; IAI 2q8: n=

5 Proportion of Patients Gaining 15 Letters at Week 18 Primary Analysis (LOCF) Proportion of Patients Losing 15 Letters VIVID VISTA Proportion of patients % 8% 6% % 2% % 18.9% VIVID *P<.1 vs control 1.2%* 2.2%* 2.9%* 13.6% VISTA *P<.1 vs control Control 35.8%* 2q 2q8 Proportion of Patients 5% % 3% 2% %.6% 9.1%.7%.%.6%.7% Laser 2q 2q8 Compared with baseline Primary analysis (LOCF): excludes patients who received rescue treatment; P values are not confirmatory VIVID FAS: control: n=132; IAI 2q: n=136; IAI 2q8: n=135; VISTA FAS: control: n=15; IAI 2q: n=15; IAI 2q8: n=151 % 25 VIVID FAS- Laser: n=132; 2q: n=136; 2q8: n=135 VISTA FAS- Laser: n=15; 2q: n=15; 2q8: n=151 Compared to baseline; LOCF 26 Long-Term Management Following VISTA Treatment Frequency Beyond 3 Years July 29, 21 Diabetic Macula Edema Number of Patients % 23% 27% 75% None 1 to 5 6 to 11 to to 2 >2 Number of PRN Injections 12% Year 1 Weighted Mean =.5 Year 2 Weighted Mean = 3. Cumulative Weighted Mean = 7.7 8% 5% 27 Diabetic Retinopathy Clinical Research Network Comparative Effectiveness Study of Aflibercept, Bevacizumab, or Ranibizumab for DME Supported through a cooperative agreement from the Baseline Randomization Randomly Assigned Eyes (one per participant): N = 66 Aflibercept (2. mg) N = 22 Bevacizumab (1.25 mg) N = 218 Ranibizumab (.3 mg) N = 218 National Eye Institute; National Institute of Diabetes and Digestive and Kidney Diseases; National Institutes of Health, Department of Health and Human Services EY1231, EY1229, EY18817 One Year N = 28 (93%) N = 26 (9%) N = 26 (9%) One Year Excluding Deaths 9% 97% 96%

6 *P values adjusted for baseline visual acuity and multiple evaluations. 1. Diabetic Retinopathy Clinical Research Network. N Engl J Med. 215;372(13): Wells JA et al. Ophthalmology. 216 Feb 27. [Epub ahead of print] 52 Week Treatment Group Comparison*: Aflibercept vs Bevacizumab P=.69 Aflibercept vs Ranibizumab P=.69 Ranibizumab vs Bevacizumab P=.69 Week Treatment Group Comparison*: Aflibercept vs Bevacizumab P=.51 Aflibercept vs Ranibizumab P=.51 Ranibizumab vs Bevacizumab P=.31 *P values adjusted for baseline visual acuity and multiple evaluations. 1. Diabetic Retinopathy Clinical Research Network. N Engl J Med. 215;372(13): Wells JA et al. Ophthalmology. 216 Feb 27. [Epub ahead of print] 52 Week Treatment Group Comparison*: Aflibercept vs Bevacizumab P<.1 Aflibercept vs Ranibizumab P=.3 Ranibizumab vs Bevacizumab P=.21 Week Treatment Group Comparison*: Aflibercept vs Bevacizumab P=.2 Aflibercept vs Ranibizumab P=.18 Ranibizumab vs Bevacizumab P=.18 Visual acuity gains varied based on anti-vegf at the primary endpoint of 1 year and were maintained through 2 years 1,2 Overall Mean Change in Visual Acuity Letter Score from Baseline to 2 Years Visual acuity gains varied by baseline visual acuity Mean Change in Visual Acuity Letter Score from Baseline to 2 Years by Baseline Visual Acuity Mean Change From Baseline in Visual Acuity Letter Score Aflibercept Bevacizumab Ranibizumab Primary Endpoint: 52 Week Treatment Week Treatment Group Comparison*: Group Comparision*: Aflibercept vs Bevacizumab P=.2 Aflibercept vs Bevacizumab P<.1 Aflibercept vs Ranibizumab P=.7 Aflibercept vs Ranibizumab P=.3 Ranibizumab vs Bevacizumab P=.11 2 Ranibizumab vs Bevacizumab P= Week Mean Change in Visual Acuity Letter Score /32 2/ (51% of participants) Week /5 or Worse (9% of participants) Week Aflibercept Bevacizumab Case Study 2 35 Year Old Male- Failed DPS test Just need glasses Diabetic for 5 years- Hemoglobin A1c = Unknown Doesn t want dilation-- Need to work tonight Refraction ish 2/25 OU

7 Diabetic Retinopathy 25% of patients with diabetes have some retinopathy 5,, in US Leading cause of visual loss & new-onset blindness 2 6 Pathophysiology of Diabetic Retinopathy Capillary pericyte loss Endothelial cell loss Nonfunctional acellular capillaries Capillary basement membrane thickening Microaneurysm formation Neovascularization Normal vessels Diabetic vessels 37 Frank RN. Etiologic mechanisms in diabetic retinopathy. In: Ryan SJ, ed. Retina, Schachat AP and Murphy RP, eds vol. 2 Medical Retina, St. Louis, 199, Mosby Photos copyright acknowledgement to publication. 38 Retinopathy Progression and HbA1c Diabetic Retinopathy: More Prevalent With Longer Disease Duration and Higher A1C Levels N = 333 Patients With Diabetic Retinopathy (%) < y -8 y >8 y Duration of Diabetes <6.3 >7.5 A1C 39 DCCT Research Group. N Engl J Med. 1993;329: Chart-based analysis of study patients with known type 2 diabetes. A1C = glycosylated hemoglobin. Tapp RJ, et al. Diabetes Care. 23;26: Prevalence of Proliferative Retinopathy Type I DM 15 years duration: 3% Treatments > 9% Effective at Preventing Blindness Earlier Tx = Better Outcomes Type II DM Receiving insulin: 15 years duration: 15% 2% Not receiving insulin: 15 years duration: 5% % Redrawn from Klein R, Klein BEK, et al. Arch Ophthalmol 2:52 526, 198 in Frank RN. Etiologic mechanisms in diabetic retinopathy. In Ryan SJ, ed: Retina, Schachat AP and Murphy RP, eds. vol. 2 Medical Retina, St. Louis, 199, Mosby, p

8 3 5 6 Protocol S Five-Year Outcomes of Panretinal Photocoagulation vs Intravitreous Ranibizumab for Proliferative Diabetic Retinopathy: Randomized Clinical Trial DRCR.net Study Design Randomized, multi-center clinical trial (55 Sites) Study eye(s) meeting all of the following criteria (a participant can have 2 study eyes): PDR No history of PRP Best corrected visual acuity letter score 2 (~Snellen equivalent 2/32 or better) Eyes with or without central-involved DME were eligible Primary Objective: Compare the efficacy and safety of PRP with that of intravitreous ranibizumab (.5-mg in.5 ml) for proliferative diabetic retinopathy (PDR) Diabetic Retinopathy Clinical Research Network (DRCR.net): 8

9 Five-Year Outcomes of Panretinal Photocoagulation vs Intravitreal Ranibizumab for Proliferative Diabetic Retinopathy PRP Baseline then additional PRP if size or amount of NV increased Ranibizumab Q through W2 with deferral option at W16 & W2 if NV resolved After W2 = Q if NV improved or worsened, deferral if NV resolved or stable after 2 consecutive injections Mean Number of Injections 5-Year Completers Only Ranibizumab Group (N = 117) PRP Group (N = 123) Year Year Year Year Year Cumulative Through 5 Years Gross JG, et al. JAMA Ophthalmol. 218;136(): Visual Acuity Visual Acuity at 5-Years Ranibizumab (N = 117) PRP (N = 123) Mean letter score 8 81 ~Snellen Equivalent, Mean 2/25 2/25 Median letter score (25 th, 75 th percentile) ~Snellen Equivalent, Median 8 (89, 78) 2/2 (2/16, 2/32) 8 (89, 77) 2/2 (2/16, 2/32) Mean Change in VA Letter Score Mean Changes in VA From Baseline Over Time - Overall Cohort 15 5 Adjusted Mean Difference at 5 Years: +.6 letters 95% Confidence Interval: (-2.3, +3.5), P =.68 Ranibizumab PRP N = 191 Visit Week N = 117 N = 23 N = 123 Outlying values were truncated to 3 SD from the mean Mean Change in Cumulative Visual Field Total Point Score ( ) - Overall Cohort Mean Change in Cumulative Visual Field Total Point Score ( ) - Overall Cohort Mean Change in Cumulative Visual Field Total Point Score (db) Ranibizumab PRP 52 N = 81 N = 1 Visit Week N = 86 N = 38 Outlying values were truncated to 3 SD from the mean Mean Change in Cumulative Visual Field Total Point Score (db) Year Adjusted Mean Difference: 28 db 95% Confidence Interval (-9, 8), P =. Ranibizumab PRP N = 81 N = 1 Visit Week N = 86 N = 38 Outlying values were truncated to 3 SD from the mean

10 DR Adverse Events: Over 5 Years Ranibizumab (N = 117) PRP (N = 123) Any Retinal detachment, % 6% 15% Retinal Detachment involving Center of the Macula, % 1% % Neovascular Glaucoma, % 3% % Neovascularization of the Iris, % 3% 1% Vitreous Hemorrhage, % 8% 6% Vitrectomy, % 11% 19% Adjusted Difference (95% CI) -9% (-1%, -%) -3% (-7%, %) -2% (-6%, 2%) 1% (-1%, 3%) 2% (-6%, 11%) -7% (-1%, -1%) CLARITY: 22 UK Centers Phase 2b Trial PRP versus aflibercept for PDR Included treatment naïve PDR (53%) and patients with PDR previously treated with PRP (7%) Different from Protocol S which only included treatment naïve Excluded patients with DME Different from Protocol S which included both PDR with and without DME CLARITY: Study Population CLARITY (N=232) Adults with type 1 or 2 diabetes mellitus; active PDR (defined as treatment naive OR persistent retinal neovascularization after initial PRP requiring additional PRP); no baseline macular edema; and BCVA letter score 5 (~Snellen equivalent 6/2 or better) Randomized 1:1 Aflibercept arm PRP arm (n=116) (n=116) Treatment-naive Active PDR Treatment-naive Active PDR PDR post initial PRP PDR post initial PRP n=6 (52%) n=56 (8%) n=63 (5%) n=53 (6%) Week 52: Primary endpoint Primary outcome: Change in BCVA from baseline to Week 52 in the study eye of the aflibercept group relative to the PRP group CLARITY VA Results At 52-weeks (primary outcome): BCVA difference between groups = letters (P<.1) Laser group lost 2.9 letters Aflibercept group gained 1.3 letters At 12-weeks (secondary outcome): BCVA difference between groups = 2.3 letters Laser group lost.9 letters Aflibercept group gained 1.5 letters Other Findings in CLARITY: Presence of Macular Edema at Week 52 CST and cube volume increased significantly in the PRP group compared to aflibercept At 52-weeks: 89% of patients in the aflibercept group did not have CME 71% of patients in the PRP group did not have CME Proportion of patients (%) (Pearson s chi-squared test comparing multiple categories between arms) 89% P=.7 9% n=93 n=9 Aflibercept No macular edema Central macular edema 3% n=3 71% 21% n=7 n=22 n=8 PRP (n=) (n=5 1 ) 8% Non-central macular edema Case Study 3 2 Year Old School Teacher- Annual Exam Diabetic for 12 years- Hemoglobin A1c = 8.1 Refraction /2 OU 6

11 61 62 Classification of Diabetic Retinopathy

12 Microaneurysms, Dot/ Blot Hemorrhages IRMA, Venous Beading, Soft Exudates (CWS) Hard Exudates NVE NVD Vitreous Heme Fibrosis Standard Fields Modified for DRS Modified for ETDRS Modified for DRS 199s- Modified for ETDRS 69 The ETDRS DR Severity Scale is a well established and validated scale for evaluation of DR severity DR severity is graded based on 7-field fundus photography, a standardized method for detection of DR First applications of standardized retinal imaging to clinical trials 3-degree overlapping stereo photographs: 9 3% ETDRS Diabetic Retinopathy Severity Scale Illustrated Using Images from RIDE/RISE DRSS = Highly Predictive of PDR Development, 12 DR absent 1, 15, 2 DR questionable 35 Mild NPDR 3 Moderate NPDR 7 Moderately Severe NPDR 53 Severe NPDR 6, 61 Mild PDR 65 Moderate PDR 71 High-risk PDR 75 High-risk PDR 81 Advanced PDR 85 Advanced PDR Eyes Progressing to PDR 8% 66% 71% 8% 52% 5 Years 26% 3 Years 1 Year Level 35 Level 3 Level 7 Level 53 Mild NPDR Moderate NPDR Moderately Severe NPDR Severe NPDR Baseline Severity of DR ETDRS Report 12. Ophthalmology 1991 N at 1 year = 1958; 3 years = 176; 5 years =

13 Incidence of DME Increases With DR Severity Progressive Drop in Health-Related QOL When DRSS >3 9 Patients With DME, % % Mild NPDR (n = 228) 39% Moderate NPDR (n = 11) 77% Severe NPDR (n = 7) NEI-VFQ-25 Scores Driving Difficulty Score NEI-VFQ-25 Composity Score / 2/<2 2/2 31/<31 31/31 37/<37 37/37 3/<3 3/3 7/<7 7/7 DR Severity (ETDRS Scale) 53/<53 53/53 6/<6+ 6+/6+ Worse QOL Ranibizumab: Slowing Progression of PDR in RIDE/RIDE Time to Development of PDR Event Crossover: sham to ranibizumab Pharmaceutical Dosing Can Slow Progression to PDR Cumulative probabilities calculated using the Kaplan-Meier method. Progression was defined by (1) progression from NPDR (DR severity level < 6) at baseline to PDR (DR severity level 6) at a later time point, (2) need for PRP laser, (3) vitreous hemorrhage (AE or slit lamp grade at baseline to > at a later time point, () cases identified by ophthalmoscopy, (5) vitrectomy, (6) iris neovascularization AE, or (7) retinal neovascularization AE. Case Study: 3-Step DR Improvement from RIDE/RISE in a Ranibizumab-Treated Patient With PDR at Baseline Pharmaceutical Dosing Can Improve DR Severity Screening High risk PDR (71A) Month 36 Mild NPDR (35E) 6 Level Improvement 13

14 DR Severity Improvements During RIDE/RISE DRSS Improvements 3 Years DR Severity Improvements During VISTA/VIVID 2-Step DRSS Improvements 3 Years Patients, % Sham/Crossover (n = 257).3 mg RBZ (n = 25).5 mg RBZ (n = 252) 1/3 rd Month Month Month Month 6 Month 12 Month 2 Month Month 36 2-Step Improvement 3-Step Improvement IAI 2q8 IAI 2q Control VIVID *P<.1 versus control 17.% 7.8%.3% * IAI 2q % *.% 2.%.% 6.% 8.%.% * Proportion of Patients IAI 2q8 Control VISTA *P<.5 versus control **P<.1 versus control % 3.% **.% 2.%.% 6.% 8.%.% Proportion of Patients In RIDE/RISE the Benefit of Ranibizumab on DR was Greatest in Patients With Baseline Moderately Severe or Severe NPDR (Level 7/53) 8 Sham Ranibizumab.3 mg Ranibizumab.5 mg 2-Step DR Severity Improvement at Month 2 78% 81% ETDRS Diabetic Retinopathy Severity Scale Illustrated Using Images from RIDE/RISE, 12 DR absent 1, 15, 2 DR questionable 35 Mild NPDR 3 Moderate NPDR 7 Moderately Severe NPDR 53 Severe NPDR Patients, % 6 2 1% 36% 31% 16% % 12% 7% 35/ /53 Without Prior PRP 6, 61 Mild PDR 65 Moderate PDR 71 High-risk PDR 75 High-risk PDR 81 Advanced PDR 85 Advanced PDR Baseline DR Severity Level n = DR, diabetic retinopathy; PRP, pan-retinal photocoagulation. >75% of Ranibizumab-Treated Patients With Moderately Severe or Severe NPDR (Level 7/53) had a 2-Step DR Improvement by Month 12 Patients, % Sham/ Crossover (n = 86) Ranibizumab.3 mg (n = 88) Ranibizumab.5 mg (n = 7) Month 2-Step Improvement a 3-Step Improvement b Management of NPDR without DME PANORAMA a All sham vs ranibizumab comparisons for 2-step improvement, P <.1. b All sham vs ranibizumab comparisons from month 12 on for 3-step improvement, P.2. 1

15 PANORAMA Study Design Inclusion & Exclusion Criteria Phase 3, Double-masked, Randomized, Study of the Efficacy & Safety of Intravitreal Aflibercept in patients with moderately severe to severe NPDR (DRSS Level 7 and 53) N=2 Sham N=133 + after 3 initial monthly doses and 1 q8 interval *after 5 initial monthly doses, flexible treatment schedule after week 52 **Patients were stratified by baseline DRSS level Group 1 IAI 2 mg Q16 weeks + N=135 Week 2 and Week 52 Primary Endpoint Proportion of patients improving 2 steps on DRSS Groups 1 & 2 combined Group 2 IAI 2 mg Q8 weeks* N=13 Follow up through Week Wykoff, CC. Key points from the Phase 3 PANORAMA Study. (July 218) American Society of Retina Specialists, Annual Meeting, Vancouver, BC, CA. Inclusion Moderately severe to severe NPDR (DRSS levels 7 or 53), confirmed by the central reading center, in whom PRP could be safely deferred for 6 months BCVA ETDRS letter score of 69 letters ( Snellen equivalent of 2/) Exclusion Presence of DME threatening the center of the macula Evidence of retinal neovascularization Any prior treatment with: Focal or grid laser photocoagulation or PRP Systemic or intravitreal anti-vegf agents Intraocular steroids Current ASNV, vitreous hemorrhage, or traction retinal detachment HbA1c >12% or HbA1c 12% with uncontrolled diabetes mellitus Uncontrolled blood pressure History of cerebrovascular accident or myocardial infarction within 6 months of study start Dosing Schedule Week: BL SHAM* O O O O O O O O O O Group 1* X X X O X O X O X O Group 2* X X X X X X X X X + X=active injection, O=sham injection 1 dose difference between Group 1 & 2 through week 2 +Group 2 (Q8) group continues PRN through Week based on DRSS level *Patients progressing to PDR/ASNV or CI-DME were eligible for rescue treatment (IAI or laser) at the discretion of the investigator. Data for patients receiving rescue treatment was censored from the time of rescue. Baseline Demographics Sham Group 1 Group 2 All IAI Total N (FAS, SAF) Age (years (SD)) 55.8 (.31) 55. (11.13) 55.8 (.19) 55.6 (.66) 55.7 (.53) Women # (%) 6 (8.1%) 6 (.%) 53 (39.6%) 113 (2.%) 177 (.%) Race # (%) White 7 (8.5%) 99 (73.3%) (77.6%) 23 (75.5%) 3 (77.1%) Black or African American 13 (9.8%) 16 (11.9%) 12 (9.%) 28 (.%) 1 (.2%) Asian (3.%) 12 (8.9%) 7 (5.2%) 19 (7.1%) 23 (5.7%) Other 9 (6.8%) 8 (5.9%) 11 (8.2%) 19 (7.1%) 28 (7.%) Hemoglobin A1C (%) 8.5 (1.5) 8.6 (1.69) 8. (1.6) 8.5 (1.66) 8.5 (1.62) Duration of Diabetes (years (SD)) 15.5 (9.3) 13.7 (8.61) 1. (9.69) 13.8 (9.15) 1. (9.2) Diabetes Type (92.5%) 121 (89.6%) 12 (92.5%) 25 (91.1%) 368 (91.5%) Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8 Baseline Disease Characteristics & Disposition N (FAS/SAF) ETDRS BCVA (letters) Mean (SD) Snellen Equivalent CRT(microns) Mean (SD) Sham Group 1 Group 2 All IAI Total 82.7 (6.3) 2/ (38.1) 82.2 (6.63) 2/ (3.3) 82.3 (5.15) 2/ (31.59) 82.3 (5.93) 2/ (32.9) 82. (5.96) 2/ (3.82) Diabetic Retinopathy Severity Score (DRSS) Level 7 99 (7.%) 2 (75.6%) 1 (75.%) 23 (75.5%) 32 (75.1%) Level 53 3 (25.6%) 33 (2.%) 33 (2.6%) 66 (2.5%) (2.9%) Number of Injections Treatment Experience # Active Injections (out of for Group 1 and 5 for Group 2) Number of Patients who Completed at Week (89.5%) 129 (95.6%) 132 (98.5%) 261 (97.%) 38 (9.5%) Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8 Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8 Sham n=133, Group 1 n=135, Group 2 n=13, All IAI n=269 15

16 Proportion of Patients with 2-Step Improvement from Baseline in DRSS Mean Change in Best Corrected Visual Acuity Proportion of Patients ** ** * ETDRS letters /133 83/135 7/13 157/269 Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8 Sham n=133, Group 1 n=135, Group 2 n=13, All IAI n=269 Week Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8 Sham n=133, Group 1 n=135, Group 2 n=13, All IAI n=269 Nominal p =.57 All IAI vs. sham Nominal p =.19 Group 1 vs. sham Nominal p =.139 Group 2 vs. sham Mean Change in Central Retinal Thickness Safety Events Through Week 2 μm Sham All IAI N (FAS/SAF) Ocular TEAEs ( 3%) Conjunctival hemorrhage 5 (3.8%) 32 (11.9%) Vitreous floaters 1 (.8%) 1 (5.2%) Diabetic retinal edema 2 (15.%) 11 (.1%) Eye pain 2 (1.5%) 11 (.1%) Diabetic retinopathy (3.%) 1 (.%) Week Nominal p <.1 All vs. sham Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8 Sham n=133, Group 1 n=135, Group 2 n=13, All IAI n=269 Non Ocular Events Patients with 1 APTC, n (%) 2 (1.5%) 1 (.%) Deaths 3 (2.3%) One Serious ocular AE of iris neovascularization occurred in 1 patient One ocular AE of vitreal cells occurred in 1 patient, which was considered mild Proportion of Patients Developing a Vision Threatening Complication (VTC) or Center Involved (CI)-DME through Week 2 Sham Group 1 Group 2 All IAI N (SAF) % Patients Developing PDR/ASNV or CI-DME % Developing PDR/ASNV 3/133 (25.6%) 18/133 (13.5% % Developing CI-DME + 18/125 (1.%) 5/135 (3.7%) 2/135 (1.5%) 3/128 (2.3%) 7/13 (5.2%) 1/13 (.7%) 6/128 (.7%) 12/269 (.5%) 3/269 (1.1%) 9/256 (3.5%) PANORAMA Week 52 Outcomes Proportion of Patients PDR/ASNV or PDR/ASNV CI-DME + CI-DME + Group 1: 3 monthly doses followed by 1 Q8 interval then Q16, Group 2: 5 monthly doses then Q8. Exploratory analysis. VTC = PDR/ASNV PDR/ASNV: Proliferative Diabetic Retinopathy/Anterior Segment Neovascularization; CI-DME: Central involved DME + CI-DME evaluable set excluded patients who, at baseline, both had CRT>3μm and qualitative evidence of CI-DME as assessed by the reading center. 16

17 Proportion of Patients with 2-Step Improvement from Baseline in DRSS Proportion of Patients with 2-Step Improvement from Baseline in DRSS Proportion of Patients % 9% 8% 8% 7% 65% 6% 5% p<.1 vs. sham % 3% 2% 15% % % Sham Group 1 Q16 Group 2 Q8 Aflibercept Improves Diabetic Retinopathy and Reduces Vision Threatening Complications in Phase 3 Trial; October 25, Proportion of Patients Developing a Vision Threatening Complication (VTC) or Center Involved (CI)-DME 5% % 1% Front Line Diabetic Retinopathy What Not to Miss and Why Proportion of Patients % 35% 3% 25% 2% 15% % 5% % 11% p<.1 vs. sham DME PDR NPDR % Sham Group 1 Q16 Group 2 Q8 Aflibercept Improves Diabetic Retinopathy and Reduces Vision Threatening Complications in Phase 3 Trial; October 25,