Background/Objectives: The aim of this study was to evaluate the performance of the Node-RADS scoring system for predicting lymph node metastasis in patients with non-small cell lung cancer (NSCLC) after neoadjuvant therapy and to assess its prognostic value for overall survival (OS) and event-free survival (EFS). Methods: A total of 247 patients with non-small cell lung cancer (NSCLC) from three centers who underwent surgery after neoadjuvant therapy were retrospectively enrolled. Post-treatment Node-RADS scores were reassessed by radiologists based on preoperative contrast-enhanced CT images. Logistic regression analysis was used to evaluate the predictive value of Node-RADS for postoperative pathological lymph node metastasis, while Cox regression analysis was performed to assess its associations with OS and EFS. Kaplan–Meier analysis was used to compare survival differences among different Node-RADS risk groups. Results: A total of 247 patients were included in this study, comprising 211 men and 36 women, with a mean age of 63.40 ± 7.58 years. Post-treatment Node-RADS score was significantly associated with both OS and EFS. In multivariable Cox regression analysis, Node-RADS remained independently associated with OS (HR = 1.79, 95% CI: 1.50–2.15, p < 0.001) and EFS (HR = 1.41, 95% CI: 1.23–1.62, p < 0.001). Using a Node-RADS score of 3 as the cutoff value, patients in the high-risk group had significantly worse OS and EFS than those in the low-risk group (both p < 0.01). For the prediction of lymph node metastasis, the inclusion of post-treatment Node-RADS markedly improved the discriminatory performance of the model, with an AUC of 0.769, a sensitivity of 46.2%, and a specificity of 87.6%. Conclusions: The Node-RADS score may provide useful imaging information for patient-level assessment of residual lymph node metastasis risk and survival stratification in patients with NSCLC after neoadjuvant therapy. These findings suggest that the scoring system may support patient-level post-treatment risk assessment.
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.