Comparative Efficacy and Safety of First-Line Immune Checkpoint Inhibitors Plus Chemotherapy with or Without Bevacizumab in Advanced Non-Squamous Non-Small Cell Lung Carcinoma.

First-line chemoimmunotherapy (I + C) is the standard of care for advanced non-squamous non-small cell lung cancer (NSCLC) without oncogenic mutation. Bevacizumab has been shown to enhance the efficacy of chemotherapy in non-squamous NSCLC, yet its added value when combined with I + C (I + C + B) remains unclear. To address this gap, we conducted a real-world comparative study and a network meta-analysis to evaluate I + C + B versus I + C in this setting. This retrospective study included patients with advanced EGFR/ALK-negative non-squamous NSCLC treated with first-line I + C + B or I + C. Propensity score matching (PSM) was employed to balance baseline characteristics between groups. Efficacy endpoints were progression-free survival (PFS) and overall survival (OS). Subgroup analyses examined outcomes by PD-L1 expression, age, metastases, and chemotherapy, among other factors. In parallel, a network meta-analysis of four randomized trials ( = 2026) indirectly compared I + C + B against I + C for PFS, OS, and safety outcomes. A total of 277 patients were included, with 167 (60.3%) receiving I + C + B and 110 (39.7%) receiving I + C. Before PSM, the I + C + B regimen significantly prolonged PFS versus I + C (hazard ratio [HR] = 0.69, 95% CI 0.52-0.92, = 0.010), with this benefit maintaining post-matching (HR = 0.70, 95% CI 0.49-0.99, = 0.045). However, OS did not differ significantly between groups in either the pre-PSM (HR = 0.93, 95% CI: 0.67-1.30; = 0.665) or matched analyses (HR = 0.84, 95% CI: 0.54-1.29; = 0.421). Subgroup analyses suggested greater PFS benefit from I + C + B among PD-L1-negative, older patients, those with brain metastases or multiple metastatic sites, and in patients receiving specific chemotherapy doublets. The network meta-analysis confirmed a PFS advantage for I + C + B over I + C (HR = 0.84, 95% CI: 0.71-0.98) without an OS benefit (HR = 0.95, 95% CI: 0.79-1.14). Toxicity was higher with I + C + B; rates of grade 3-5 adverse events, serious adverse events, and treatment discontinuation were all significantly increased compared to I + C. In the first-line treatment of advanced EGFR/ALK-negative non-squamous NSCLC, adding bevacizumab to I + C improved PFS but did not translate into an OS gain. Although PFS benefits were observed in certain subgroups, these were accompanied by significantly increased treatment-related toxicities. Our findings suggest that no clear subgroup has been identified where the benefit outweighs the risks, necessitating extreme clinical caution.