PURPOSE: To evaluate the impact of intravitreal aflibercept (EYLEA, Regeneron Pharmaceuticals, Tarrytown, NY) versus laser on progression of diabetic retinopathy (DR) severity in Intravitreal Aflibercept Injection in Vision Impairment due to DME (VIVID-DME) and Study of Intravitreal Aflibercept Injection in Patients with Diabetic Macular Edema (VISTA-DME). DESIGN: Secondary and exploratory analyses of 2 phase 3, randomized, controlled studies. PARTICIPANTS: All patients with a baseline Diabetic Retinopathy Severity Scale (DRSS) score based on fundus photograph (full analysis), patients who progressed to proliferative DR (PDR) (safety analysis) in VIVID-DME (n = 403) and VISTA-DME (n = 459), or both. METHODS: We randomized patients with diabetic macular edema (DME) to intravitreal aflibercept 2 mg every 4 weeks (2q4), intravitreal aflibercept 2 mg every 8 weeks after 5 initial monthly doses (2q8), or macular laser photocoagulation at baseline and sham injections at every visit. MAIN OUTCOME MEASURES: Proportions of patients with 2-step or more and 3-step or more improvements from baseline in DRSS score, who progressed to PDR, and who underwent panretinal photocoagulation (PRP). RESULTS: Among patients with an assessable baseline DRSS score, most showed moderately severe or severe nonproliferative DR. The proportions of patients treated with 2q4, 2q8, and laser with a 2-step or more improvement in DRSS score at week 100 were 29.3%, 32.6%, and 8.2%, respectively, in VIVID-DME and 37.0%, 37.1%, and 15.6%, respectively, in VISTA-DME; the proportions with a 3-step or more improvement in DRSS score were 7.3%, 2.3%, and 0%, respectively, and 22.7%, 19.9%, and 5.2%, respectively. Fewer patients in the 2q4 and 2q8 groups versus the laser group progressed to PDR at week 100 in VISTA-DME (1.5% and 2.2% vs. 5.3%) and VIVID-DME (3.2% and 2.0% vs. 12.3%). The proportions of patients who underwent PRP were 2.9%, 0.7%, and 4.5%, respectively, in VIVID-DME and 1.9%, 0.7%, and 5.2%, respectively, in VISTA-DME. The most frequent serious ocular adverse event at week 100 was cataract (pooled intravitreal aflibercept, 1.7% of patients; laser, 3.5% of patients). CONCLUSIONS: These analyses demonstrate the benefit of intravitreal aflibercept over laser with respect to DR progression, suggesting a benefit on DME, and on underlying DR.< Réduire