PET/CT predict pathological response to neoadjuvant nivolumab in resectable non-small cell lung cancer.

Abstract

BACKGROUND: This study aimed to evaluate the predictive value of 18F-FDG PET/CT for pathological complete response (pCR) in patients with resectable non-small cell lung cancer (NSCLC) receiving neoadjuvant nivolumab-based therapy. METHODS: This retrospective study included 125 patients with stage II–III NSCLC who received three cycles of neoadjuvant nivolumab plus chemotherapy between June 2019 and October 2024. All patients underwent ¹⁸F-FDG PET/CT imaging at baseline and after neoadjuvant therapy prior to surgery. Metabolic parameters including SUVmax, SUVmean, SUVpeak, metabolic tumor volume (MTV), and total lesion glycolysis (TLG)were measured. The diagnostic performance of these parameters for predicting pCR was evaluated using receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA). Multiple logistic regression was used to identify independent predictive factors. An XGBoost model integrating significant imaging and clinicopathological predictors was developed. RESULTS: The pCR rate was 38.4% (48/125). Post-therapy metabolic parameters and their Δ% were significantly lower in the pCR group (n = 48) compared to the non-pCR group (n = 77) (P < 0.001). A ΔSUVmax (%) cut-off of -69.8% predicted pCR with an AUC of 0.832 (sensitivity 93.8%, specificity 61.0%). DCA confirmed the clinical net benefit of models using post-therapy or Δ% parameters. Multivariate logistic regression identified squamous cell carcinoma (SCC) histology, PD-L1 expression ≥ 1%, and lower post-SUVmax as independent predictors of pCR (P < 0.05). The XGBoost model, integrating post-SUVmax, squamous histology, and PD-L1 expression (≥ 1%), demonstrated superior discriminative ability with an AUC of 0.884. CONCLUSION: Post-therapy PET/CT metabolic parameters and their percentage changes (Δ%) are robust, non-invasive biomarkers for predicting pCR. An integrated prediction model combining imaging and clinicopathological data shows high performance.