Objective: Accurately predicting response during neoadjuvant chemoimmunotherapy for resectable non-small cell lung cancer remains clinically challenging. In this study, we investigated the effectiveness of blood-based tumor mutational burden (bTMB) and a deep learning (DL) model in predicting major pathologic response (MPR) and survival from a phase 2 trial.
Methods: Blood samples were prospectively collected from 45 patients with stage IIIA (N2) non-small cell lung cancer undergoing neoadjuvant chemoimmunotherapy. An integrated model, combining the computed tomography-based DL score, bTMB, and clinical factors, was developed to predict tumor response to neoadjuvant chemoimmunotherapy.
Results: At baseline, bTMB were detected in 77.8% (35 of 45) of patients. Baseline bTMB ≥11 mutations/megabase was associated with significantly greater MPR rates (77.8% vs 38.5%, P = .042), and longer disease-free survival (P = .043), but not overall survival (P = .131), compared with bTMB <11 mutations/megabase in 35 patients with bTMB available. The developed DL model achieved an area under the curve of 0.703 in all patients. Importantly, the predictive performance of the integrated model improved to an area under the curve of 0.820 when combining the DL score with bTMB and clinical factors. Baseline circulating tumor DNA (ctDNA) status was not associated with pathologic response and survival. Compared with ctDNA residual, ctDNA clearance before surgery was associated with significantly greater MPR rates (88.2% vs 11.1%, P < .001) and improved disease-free survival (P = .010).
Conclusions: The integrated model shows promise as a predictor of tumor response to neoadjuvant chemoimmunotherapy. Serial ctDNA dynamics provide a reliable tool for monitoring tumor response.
Keywords: blood-based tumor mutational burden; circulating tumor DNA; deep learning; neoadjuvant chemoimmunotherapy; non–small cell lung cancer.
Copyright © 2024 The American Association for Thoracic Surgery. Published by Elsevier Inc. All rights reserved.