Sex difference in survival benefits of immune checkpoint inhibitors in advanced non-small cell lung cancer: A meta-analysis

Background: Whether sex influences the survival benefit of immune checkpoint inhibitors (ICIs) in advanced non-small cell lung cancer (NSCLC) remains uncertain. Methods: We systematically searched PubMed, Embase, Cochrane Library, Web of Science, and Scopus up to April 1, 2026, for randomized controlled trials (RCTs) comparing ICIs with chemotherapy and reporting overall survival (OS) or progression-free survival (PFS) stratified by sex. A random-effects model was used for meta-analysis. Results: Sixty-three publications (reporting 42 phase 3 RCTs) involving 25,827 patients (18,772 male; 7,055 female) were included. Compared with chemotherapy, ICIs significantly improved both OS and PFS in males (OS: HR 0.74 [95% CI 0.71–0.78]; PFS: HR 0.59 [95% CI 0.54–0.64]) and females (OS: HR 0.83 [95% CI 0.77–0.90]; PFS: HR 0.76 [95% CI 0.66–0.86]). Pooled treatment-by-sex interaction analyses demonstrated significantly greater OS (interaction HR = 0.90 [95% CI 0.84–0.97]; P = 0.01) and PFS (interaction HR = 0.82 [95% CI 0.73–0.92]; P = 0.0006) benefits in males than in females. In subgroup analyses, sex‑based differences in OS reached statistical significance in patients receiving ICI monotherapy (P = 0.02), including those treated with anti‑PD‑1 monotherapy (P = 0.04) and those with PD‑L1 expression ≥50% (P = 0.02). However, these differences did not persist after the False Discovery Rate (FDR) correction (adjusted P = 0.07, 0.10, and 0.07, respectively). In contrast, sex‑based differences in PFS remained statistically significant after FDR correction in the ICI monotherapy (adjusted P = 0.001), anti‑PD‑1 monotherapy (adjusted P = 0.001), and PD‑L1 ≥ 50% (adjusted P = 0.004) subgroups. No significant sex‑based differences in either OS or PFS were observed among patients treated with ICI plus chemotherapy. Conclusions: Although ICIs are superior to chemotherapy in both sexes, the PFS benefit of ICI monotherapy appears to be attenuated in females relative to males. Given the imbalances in baseline confounders coupled with the underrepresentation of females in the included trials, these findings should be interpreted with caution and warrant confirmation in future RCTs designed specifically to investigate sex‑based differences.