Integrating clinicopathologic factors and immune repertoire sequencing to optimize decision-making in neoadjuvant chemoimmunotherapy for HNSCC: A pooled analysis.

Abstract

e18065 Background: Neoadjuvant chemoimmunotherapy (NACIT) improves survival in locally advanced head and neck squamous cell carcinoma (HNSCC), but optimal patient selection and treatment decisions remain challenging. Clinicopathologic factors combined with Immune repertoire sequencing (IRS), which can directly reflect the activation of the adaptive immune system and its capacity for tumor recognition, are promising to identify prognostic factors guiding clinical decision at different stages and explore underlying mechanisms. Methods: Patients with locally advanced oral or oropharyngeal carcinoma across 3 studies treated with tislelizumab, albumin-bound paclitaxel, and cisplatin were included. Clinicopathologic factors labeled with baseline, preoperative, and postoperative timepoints, integrated with whole exome sequencing (WES) mutation features and IRS data quantifying indexes (d50, inverse Simpson, clonal proportion) as well as clonotypes, were analyzed. Key endpoints were overall survival (OS), event-free survival (EFS), and major pathological response (MPR). Multivariate logistic regression and Cox proportional hazards models identified independent prognostic factors, while correlation analyses explored associations between immune repertoire indices and clinical characteristics. Results: A total of 101 patients were included with a median follow-up of 38 months, ranging from 2 to 55 months. Pathological assessment revealed 51 (50.5%) cases of MPR and 32 (31.7%) cases of pCR. The 3-year OS and EFS rates were 69.4% and 60.1%, respectively. Clinically, performance status (PS) was found as a robust independent baseline predictor across all endpoints and timepoints. De-escalated share similar prognosis with radical surgery but initial surgical margin significantly affected survival. Biologically, the IRS model based on specific T-cell and B-cell clonotypes demonstrated high predictive accuracy for prognostic prediction. High immune repertoire diversity and clonality were significantly correlated with better PS and primary tumor. WES data showed limited prognostic utility in this cohort. Conclusions: Clinically, performance status can reflect antitumor immune activity, while compromised initial surgical margins warn against excessive de-escalation surgery. Biologically, B-cell clonotypes are critical for predicting NACIT response, and the correlation between higher immune diversity and better physical status as well as primary tumor supports immunotherapy in advance and strategies focused on rejuvenating systemic immunity.