Heterogeneity of progression-free survival surrogacy by sex in randomized trials testing immunotherapy in non-small cell lung cancer

Background: The surrogacy of progression-free survival (PFS) for overall survival (OS) at the trial-level in randomized clinical trials (RCTs) testing immune checkpoint inhibitors (ICIs) in patients with advanced non-small cell lung cancers (NSCLC) is influenced by several clinical-pathological factors. However, potential heterogeneity of PFS surrogacy according to patients’ sex has never been investigated. Methods: RCTs testing ICIs as monotherapy or combined with chemotherapy in patients with advanced NSCLC reporting hazard ratios (HRs) for PFS and OS according to patients’ sex were included. The main objective was to assess sex-based heterogeneity in the trial-level association between PFS (surrogate endpoint) and OS (reference endpoint), overall and in subgroups defined by treatment type (ICIs as monotherapy vs ICIs plus chemotherapy). We used the coefficient of determination (R2) to quantify surrogacy. Results: Twenty RCTs, for a total of 7528 male and 3008 female patients, were included. Overall, the association between OS-HR and PFS-HR was moderate: the R2 from a model adjusted by the type of treatment administered in the experimental arm was 0.69 (95% confidence interval [CI] = 0.34 to 0.88). Sex-disaggregated analysis showed heterogeneity in PFS surrogacy: the association was strong in male patients (adjusted R2 = 0.77; 95% CI = 0.56 to 0.89), but poor in female (adjusted R2 = 0.31, 95% CI = 0.03 to 0.63). Consistent results were obtained in subgroups analyses by treatment type, and in cross-validation analysis. Conclusions: In RCTs testing ICIs alone or combined with chemotherapy in patients with advanced NSCLC, PFS is a robust surrogate endpoint for OS in male patients but not in female.