Objective To evaluate the literature-pooled functional, anatomical and safety outcomes of aflibercept biosimilars compared with reference aflibercept for neovascular age-related macular degeneration (nAMD).Methods and analysis Systematic review and meta-analysis. Medline, Embase and CENTRAL were searched from inception to 2 June 2025. Phase 3 randomised controlled trials (RCTs) comparing aflibercept biosimilars with the reference product in patients with nAMD were included. Two independent reviewers conducted screening, data extraction, risk of bias (RoB 2) assessment and certainty of evidence assessment (Grading of Recommendations, Assessment, Development and Evaluation), with a third reviewer resolving discrepancies. Primary outcomes included mean difference (MD) change in best-corrected visual acuity (BCVA) and retinal thickness over time, risk ratio (RR) for proportion of participants gaining >15 Early Treatment Diabetic Retinopathy Study (ETDRS) letters and RR for adverse events. Meta-regression was used to evaluate the temporal stability of continuous outcomes.Results Six RCTs involving 2044 participants (1026 biosimilar, 1018 reference) were included. Five aflibercept biosimilars were evaluated, representing more than 50% of the eight biosimilars currently approved worldwide. Meta-regression revealed no significant difference in MD change in BCVA over time between biosimilar and reference aflibercept (slope: 0.0321 letters/week; p=0.1013). For MD change in retinal thickness, an initial anatomical advantage for reference aflibercept at week 0 (intercept: 9.58 µm; p=0.0449) was not sustained over time (slope: –0.1685 µm/week; p=0.2303). The pooled RR for gaining >15 ETDRS letters from baseline was 1.19 (95% CI 0.98 to 1.45; p=0.079). No statistically significant differences were observed across 15 safety outcomes.Conclusion This systematic review and meta-analysis found no statistically significant differences in functional, anatomical or safety outcomes betwee