Roles of Restricted Mean Survival Time and Restricted Mean Time Lost in Evaluating Immune Checkpoint Inhibitor...

Chemo-immunotherapy is the current standard of care for extensive-stage small cell lung cancer (ES-SCLC), but interpreting hazard ratios (HR) from Cox models can be challenging when immune checkpoint inhibitors (ICI) produce early crossing or delayed separation of Kaplan-Meier curves. Restricted mean survival time (RMST) and restricted mean time lost (RMTL) have emerged as alternative metrics that do not rely on the proportional hazards (PH) assumption. We conducted a systematic review and meta-analysis using reconstructed individual patient data (IPD) from phase III trials comparing ICIs to standard chemotherapy in the first-line setting for ES-SCLC. Kaplan-Meier curves were digitized using the IPDfromKM R package version 4.3.2. to reconstruct pseudo-IPD, from which HRs, RMSTs, and RMTLs were derived. Seven trials comprising 1,766 patients were included. The pooled HR for progression-free survival (PFS) was 0.67 [95% confidence interval (CI), 0.59-0.76] with an RMST gain of 1.84 months and RMTL reduction of 1.84 months. The pooled HR for overall survival (OS) was 0.73 (95% CI, 0.68-0.79) with an RMST gain of 1.98 months and RMTL reduction of 1.97 months. PH violations were more frequently observed in PFS than OS. Although HRs, RMSTs, and RMTLs were generally consistent, discrepancies in some trials underscore the value of RMST and RMTL as complementary, clinically intuitive measures. Incorporating RMST and RMTL into future ES-SCLC trials may improve the interpretability of treatment effects beyond conventional Cox model estimates.

This first IPD meta-analysis evaluating RMST and RMTL as alternative endpoints in ES-SCLC demonstrates that these measures generally align with HR, with some exceptions. Given their clinical interpretability and ability to quantify survival benefit, RMST and RMTL should be considered as endpoints in future ES-SCLC trials.