Abstract
Background: Neoadjuvant immunochemotherapy improves pathological response in resectable non-small cell lung cancer (NSCLC), but the need and intensity of postoperative adjuvant therapy across different pathological response rate (PRR) strata remain uncertain. Methods: In this single-center retrospective cohort, 105 patients with resectable NSCLC received neoadjuvant platinum-based chemotherapy with or without PD-1/PD-L1 inhibitors followed by R0 resection. PRR was defined as 1-residual viable tumor fraction and categorized as 0-60%, 60-90%, and ≥90% (major pathological response, MPR). Postoperative strategies included no further therapy, chemotherapy alone, or immunotherapy ± chemotherapy. Event-free survival (EFS) was analyzed using Kaplan-Meier estimates, multivariable Cox models, and restricted cubic spline-based treatment × PRR interaction. Results: Deeper PRR was associated with lower ypT/ypN stage and improved EFS. In the PRR 0-60% subgroup, immunotherapy-containing adjuvant regimens were associated with better EFS, whereas chemotherapy alone did not outperform observation. In the PRR 60-90% and MPR strata, EFS curves for different postoperative strategies largely overlapped, and in MPR patients, hazard ratios were close to 1. Interaction modeling suggested that the absolute 3-year EFS benefit of immunochemotherapy peaked at intermediate PRR (≈60-80%) and diminished as PRR approached ≥90%. Conclusions: The robustness of these findings was further confirmed through a sensitivity analysis focusing on a homogeneous cohort of clinical stage II-III patients receiving adjuvant therapy. Among NSCLC patients treated with neoadjuvant systemic therapy, PRR is a strong prognostic marker and modulates the benefit of postoperative immunotherapy. These data support a response-adapted strategy, with adjuvant immunotherapy intensification in low-PRR patients and potential de-escalation or surveillance alone in MPR patients, warranting validation in prospective PRR-stratified trials.