Neoadjuvant versus perioperative immunotherapy in resectable NSCLC: A PD-L1-stratified Bayesian network meta-analysis.

BACKGROUND: Neoadjuvant immunochemotherapy (NIC) and perioperative immunochemotherapy (PIC) improve outcomes in resectable non-small cell lung cancer (NSCLC), but their relative effects across programmed death-ligand 1 (PD-L1) subgroups remain unclear because head-to-head trials are lacking. This study aimed to compare NIC, PIC, and platinum-based chemotherapy (CT) for event-free survival (EFS) and overall survival (OS), and to compare individual preoperative regimens for pathologic complete response (pCR) and major pathologic response (MPR), across PD-L1 subgroups. METHODS: PubMed, Embase, the Cochrane Library, and Web of Science were searched through 17 January 2026 for randomized trials. Bayesian random-effects network meta-analyses estimated hazard ratios (HRs) for EFS and OS and risk ratios (RRs) for pCR and MPR, with 95% credible intervals (CrIs). Analyses were conducted in R (rjags and gemtc); funnel plots were generated in Stata 17.0. RESULTS: Eight trials (10 publications; 3,608 patients) were included. Six publications were judged to have a low overall risk of bias, whereas four were judged to have some concerns. Among patients with PD-L1 <1%, PIC improved EFS versus CT (HR 0.74, 95% CrI 0.58-0.96), but neither PIC nor NIC improved OS (PIC: HR 0.93, 95% CrI 0.65-1.33; NIC: HR 1.32, 95% CrI 0.67-2.58). Among those with PD-L1 ≥1%, both PIC and NIC improved EFS (PIC: HR 0.51, 95% CrI 0.36-0.72; NIC: HR 0.44, 95% CrI 0.26-0.74) and OS (PIC: HR 0.57, 95% CrI 0.41-0.78; NIC: HR 0.47, 95% CrI 0.25-0.88) versus CT. In PD-L1 ≥50%, NIC was associated with longer EFS versus CT (HR 0.24, 95% CrI 0.06-0.87). In PD-L1 ≥1%, tislelizumab plus CT was associated with higher pCR versus CT (RR 11.07, 95% CrI 5.35-27.31), and camrelizumab plus CT was associated with higher MPR (RR 8.28, 95% CrI 3.58-21.94). CONCLUSION: Comparative estimates for neoadjuvant and perioperative immunochemotherapy differed by PD-L1 subgroup and outcome. Indirect comparisons and limited certainty preclude firm conclusions about the superiority of either strategy. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/, identifier CRD420261303366.