Efficacy and Safety of Brolucizumab for Diabetic Macular Edema The KINGFISHER Randomized Clinical Trial

IMPORTANCE Despite the effectiveness of existing anti-vascular endothelial growth factor (VEGF) therapies, a need remains for further treatment options to improve response rates and/or reduce injection or monitoring frequency in patients with diabetic macular edema (DME).OBJECTIVE To evaluate the efficacy and safety of brolucizumab vs aflibercept dosed every 4 weeks in participants with DME.DESIGN, PARTICIPANTS, AND SETTING This 52-week, double-masked, phase 3 randomized clinical trial included treatment-naive adults and adults who had previously received anti-VEGF therapy. Data were collected from September 2019 to March 2020, and data were analyzed from April 2020 to February 2021. INTERVENTION Brolucizumab, 6 mg, intravitreal injection every 4 weeks or aflibercept, 2 mg, intravitreal injection every 4 weeks.MAIN OUTCOMES AND MEASURES Participants were randomized 2:1 to brolucizumab, 6 mg, or aflibercept, 2 mg. The primary end point was change from baseline in best-corrected visual acuity at week 52. Secondary end points were the proportion of participants with a 2-step improvement or greater from baseline in Diabetic Retinopathy Severity Scale score, the proportion of eyes with absence of both subretinal fluid and intraretinal fluid, change from baseline in central subfield thickness, and safety at week 52.RESULTS A total of 517 participants were randomized to brolucizumab (n = 346) or aflibercept (n = 171); 299 (57.8%) were male, and the mean (SD) age was 60.7 (10.2) years. Brolucizumab was noninferior to aflibercept in best-corrected visual acuity (Early Treatment Diabetic Retinopathy Study letter score) change from baseline at week 52 (brolucizumab, 12.2-letter improvement; aflibercept, 11.0-letter improvement; difference, 1.1; 95% CI, -0.6 to 2.9; noninferiority margin, 4; P < .001). Brolucizumab was superior to aflibercept for the proportion of eyes without subretinal and intraretinal fluid (brolucizumab, 144 of 346 [41.6%]; aflibercept, 38 of 171 [22.2%]; difference, 20.0%; 95% CI, 12.5to 28.6; P < .001) and mean central subfield thickness change from baseline at week 52 (brolucizumab, -237.8 mu m; aflibercept, -196.5 mu m; difference, -41.4; 95% CI, -58.9 to -23.8; P < .001). Incidence of intraocular inflammation was 4.0% (14 of 346) in the brolucizumab arm and 2.9% (5 of 171) in the aflibercept arm, incidence of retinal vasculitis was 0.9% (3 of 346) and 0.6% (1 of 171), respectively, and incidence of retinal vascular occlusion was 0.3% (1 of 346) and 0.6% (1 of 171). One participant in the brolucizumab arm had retinal artery occlusion. CONCLUSIONS AND RELEVANCE In these study participants with DME, no clinically meaningful differences in visual outcomes were noted between the brolucizumab and aflibercept arms; some superior anatomic improvements were noted in the brolucizumab arm. No new safety concerns were identified.