Valid surrogate end-points in nonsmall cell lung cancer peri-operative systemic therapy trials: a systematic review

Recurrence-, progression-, disease- and event-free survival are often selected as the primary end-points for trials assessing peri-operative systemic therapy for nonsmall cell lung cancer (NSCLC). As overall survival (OS) has increased, these surrogates, which we hereafter collectively term "recurrence-/progression-free survivals (RPFS)", have become more attractive end-points.

This systematic review, without meta-analysis, was conducted in accordance with the guidelines established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (UMIN000052558). Electronic databases were searched on 11 October 2023. English-language articles presenting a randomised controlled trial assessing neoadjuvant and/or adjuvant systemic therapy for NSCLC were included. The main outcome was the weighted Spearman's rank correlation coefficient (r) between the hazard ratios (HRs) of OS (HRos) and RPFS (HRrpfs). The weight assigned to each study was determined using the inverse variance of the log HRos. Pathological subtype and driver mutations were not questioned.

We identified 31 trials with a total of 15 776 patients. The weighted correlation coefficient was 0.86 based on the raw data from the 31 trials. After reciprocal duplication, the weighted correlation coefficient was 0.91 (p<0.001 for unweighted r). Subgroup analyses showed the correlation was 0.98 for trials with immune checkpoint inhibitors but was 0.54 for molecular targeted therapy CONCLUSIONS: We hope that our data will justify the use of the HRs of recurrence-, progression-, disease- and event-free survival as primary end-points in peri-operative immune checkpoint inhibitor regimens for NSCLC.