Macular Atrophy in the HARBOR Study for Neovascular Age-Related Macular Degeneration.

PURPOSE: To evaluate macular atrophy (MA) presence in the 24-month HARBOR study (NCT00891735) for neovascular age-related macular degeneration (AMD). DESIGN: Post hoc analysis of a phase 3 multicenter, prospective, randomized, double-masked, active treatment-controlled clinical trial. PARTICIPANTS: Evaluable subjects (N = 1095) with subfoveal choroidal neovascularization (CNV) secondary to neovascular AMD treated with ranibizumab 0.5 mg or 2.0 mg monthly or pro re nata (PRN). METHODS: Fluorescein angiograms (FAs) and color fundus photographs at baseline and months 3, 12, and 24 were retrospectively graded by masked graders for MA: well-defined areas of depigmentation with increased choroidal vessel visibility, diameter ≥250 µm, corresponding to flat areas of well-demarcated staining on FA, excluding atrophy associated with retinal pigment epithelium tears. Atrophy immediately within, adjacent, and nonadjacent to CNV lesions was included. MAIN OUTCOME MEASURES: Macular atrophy incidence, best-corrected visual acuity (BCVA). RESULTS: At baseline, MA was detected in 11.2% (123/1095) of study eyes. At month 24, 29.4% (229/778) of eyes without baseline atrophy had detectable MA. Eyes with and without baseline MA had significant mean BCVA gains from baseline at month 24 (letters [95% confidence interval]: +6.7 [4.1-9.3]; +9.1 [8.0-10.2], respectively). Among eyes with and without MA at month 24, mean month 24 BCVA was 62.0 [60.3-63.7] and 64.7 [63.2-66.3] letters, respectively. Baseline risk factors for month 24 MA presence included intraretinal cysts (hazard ratio [HR], 2.45 [1.76-3.42]) and fellow eye atrophy (HR, 2.02 [1.42-2.87]); subretinal fluid was associated with a lower MA risk (HR, 0.50 [0.33-0.74]). Ranibizumab dose was not associated with MA development. Monthly versus PRN treatment trended toward an association with MA (HR, 1.29 [0.99-1.68]), but was not statistically significant. CONCLUSIONS: New MA was detected in 29% of study eyes after 24 months of treatment. Clinically significant BCVA gains were achieved with MA present over 24 months. Baseline subretinal fluid absence, intraretinal cyst presence, and fellow eye atrophy presence were associated with month 24 MA presence. With existing data, the benefits of ranibizumab for neovascular AMD outweighed the risk of MA development over 24 months in HARBOR, although outcomes >2 years were not evaluated.

Patterns of Early and Delayed Visual Response to Ranibizumab Treatment for Neovascular Age-Related Macular Degeneration.

IMPORTANCE: Understanding the range of temporal responses to ranibizumab is critical for the assessment of individualized treatment regimens for neovascular age-related macular degeneration. OBJECTIVE: To examine patterns of visual and anatomical response to ranibizumab treatment. DESIGN, SETTING, AND PARTICIPANTS: This study is a retrospective subanalysis of HARBOR (a phase 3, double-masked, multicenter, randomized, active treatment-controlled study of the efficacy and safety of 0.5 mg and 2.0 mg ranibizumab administered monthly or on an as-needed basis (PRN) in patients with subfoveal neovascular age-related macular degeneration). A total of 1097 patients with neovascular age-related macular degeneration were randomized to intravitreal ranibizumab, 0.5 or 2.0 mg, administered monthly or as needed (PRN) with monthly monitoring. Of the 1097 patients, 1057 were included in the analysis for early responders (best-corrected visual acuity [BCVA] obtained at baseline and month 3), and 988 patients were included in the analysis for delayed responders (BCVA obtained at baseline, month 3, and month 12). The HARBOR study began July 7, 2009, with the primary 12-month end point completed on August 5, 2011, ongoing to 24 months. Data analysis for the subgroup was performed from January 4, 2013, through December 17, 2015. INTERVENTIONS: Patients were categorized based on BCVA outcomes as early 15-letter responders (gained ≥15 letters from baseline at month 3) or delayed 15-letter responders (did not gain ≥15 letters from baseline at month 3 but did so at month 12). MAIN OUTCOMES AND MEASURES: Changes from baseline in BCVA and central foveal thickness (CFT). RESULTS: In total, 266 early and 135 delayed 15-letter responders were identified. In the 0.5-mg monthly, 0.5-mg PRN, 2.0-mg monthly, and 2.0-mg PRN treatment groups, 63 (24.0%) of 263, 65 (24.6%) of 264, 68 (25.7%) of 265, and 70 (26.4%) of 265 patients were early responders, respectively, and 40 (16.3%) of 246, 31 (12.6%) of 247, 35 (14.1%) of 248, and 29 (11.7%) of 247 patients were delayed responders, respectively. By month 12, early vs delayed responders in the PRN treatment groups received 7.5 vs 7.4 ranibizumab injections, respectively (P = .84). More than 80% of early responders receiving PRN treatment maintained 15-letter or greater gains at month 24. At baseline, early vs delayed responders had worse BCVA (49.8 vs 55.4 letters; P < .001) and greater CFT (374.9 vs 339.0 µm; P = .02), although anatomical results were comparable by month 3 (CFT, 187.7 vs 188.9 µm). CONCLUSIONS AND RELEVANCE: Improvement of 15 letters or more from baseline occurred in 266 (25.2%) of 1057 patients within 3 months of beginning ranibizumab treatment, whereas an additional 135 (13.7%) of 988 patients achieved this gain by 12 months. The 2 cohorts had similar anatomical temporal response patterns. PRN treatment with monthly monitoring was effective in maintaining early vision gains and allowing delayed vision gains. These results suggest that vision improvement can continue in some patients after macular edema resolves and CFT decreases stabilize.

Cerebrovascular accidents in patients treated for choroidal neovascularization with ranibizumab in randomized controlled trials.

PURPOSE: To analyze cerebrovascular accidents (CVAs) pooled from large, randomized, controlled clinical trials of ranibizumab treatment for neovascular age-related macular degeneration. METHODS: Events in five trials (FOCUS, MARINA, ANCHOR, PIER, and SAILOR) were analyzed using a standard safety monitoring process. Exact methods, stratified by study, were used to test for treatment differences based on odds ratios. A stepwise logistic regression model was fit to classify subjects' risk for CVA based on medical history. Treatment differences in CVA rates at 1 year or 2 years were evaluated within risk groups using stratified exact methods. RESULTS: Pooled 2-year CVA rates were <3%; odds ratios (95% confidence intervals) for CVA risk were 1.2 (0.4-4.4) for ranibizumab 0.3-mg versus control, 2.2 (0.8-7.1) for 0.5 mg versus control, and 1.5 (0.8-3.0) for 0.5-mg versus 0.3-mg ranibizumab. No substantial increased risk of CVA for 0.5 mg versus 0.3 mg was identified in pooled analyses or any of the individual trials. In pooled analyses, the difference between 0.5-mg ranibizumab and control was larger (7.7 [1.2-177]) among high-risk CVA patients. CONCLUSION: This analysis provided some evidence, although not definitive, of a potential increased risk of CVA with ranibizumab versus control or with 0.5-mg versus 0.3-mg ranibizumab. Continued monitoring for CVA within clinical trials seems warrented.