Objectives: Efficacy, safety, pharmacokinetics (PK), and immunogenicity of the biosimilar candidate SB8 was compared to its reference product bevacizumab (BEV) in patients with metastatic or recurrent nonsquamous non-small cell lung cancer. Methods: Patients were randomized (1:1) in a phase III, double-blind study to receive intravenous SB8 or BEV 15 mg/kg with paclitaxel/carboplatin every 3 weeks for 24 weeks, followed by SB8 or BEV maintenance monotherapy. The primary endpoint was best overall response rate (ORR) by 24 weeks. Secondary endpoints included survival outcomes, safety, PK, and immunogenicity. Results: 763 patients (SB8, n = 379; BEV, n = 384) were randomized; baseline characteristics were well balanced. Best ORR in the FAS was 47.6% and 42.8%, and best ORR in the PPS was 50.1% and 44.8% for SB8 and BEV, respectively. The risk ratio of best ORR was 1.11 (90% CI, 0.975−1.269), and the risk difference in best ORR was 5.3% (95% CI, −2.2%-12.9%). Median survival outcomes were comparable between SB8 and BEV: progression-free survival was 8.50 vs 7.90 months, respectively (HR [95% CI], 0.99 [0.83-1.18]; p = 0.9338); overall survival was 14.90 vs 15.80 months, respectively (HR [95% CI], 1.03 [0.83-1.28]; p = 0.7713); and duration of response was 7.70 vs 7.00 months, respectively (HR [95% CI], 1.05 [0.81-1.37]; p = 0.6928). Severity and incidence of treatment-emergent adverse events, PK, and immunogenicity were comparable between SB8 and BEV. Conclusion: This study demonstrated equivalence between SB8 and BEV in terms of best ORR risk ratio, with comparable safety, PK, and immunogenicity.