Neoadjuvant versus perioperative chemo-immunotherapy according to pathological response in resectable NSCLC: a reconstructed individual patient data meta-analysis.

Neoadjuvant chemo-immunotherapy transformed early-stage non-small cell lung cancer (NSCLC) treatment. However, the prognostic value of different pathological responses and the impact of adjuvant immunotherapy within a chemo-immunotherapy perioperative strategy remains unclear. We estimated time-to-event outcomes by graphical reconstruction of event-free survival (EFS) curves by pathological response (pCR, MPR, no-MPR) reported in early-stage NSCLC neoadjuvant/perioperative chemo-immunotherapy trials. MPR 1-10% subgroup, previously unreported, was retrieved by removing patients achieving pCR from the MPR group. Survival analysis by pathological response and comparison between neoadjuvant/perioperative strategies within subgroups were assessed. A statistically significant EFS difference according to pathological response was found, showing a prognostic gradient shifting from pCR (good), MPR 1-10% (intermediate) and no-MPR (poor). There was no difference between neoadjuvant/perioperative strategies within subgroups, however a trend for EFS benefit with perioperative and neoadjuvant chemo-immunotherapy was observed in MPR 1-10% and no-MPR patients, respectively. In conclusio, a pathological response-based algorithm could better tailor early-stage NSCLC treatment.