R0 resection following neoadjuvant PD-1 inhibition with paclitaxel-cisplatin is associated with improved overall survival in locally advanced and recurrent oral squamous cell carcinoma: multicenter retrospective cohort study

Background Locally advanced oral squamous cell carcinoma (LA-OSCC) has poor prognosis with standard therapy; recurrent disease is palliative. We compared neoadjuvant PD-1 inhibition plus paclitaxel-cisplatin followed by R0 resection ± radiotherapy versus non-surgical treatment in treatment-naive and recurrent LA-OSCC.Methods This multicenter retrospective study (2021–2024) included 178 stage III–IVb LA-OSCC patients (43 recurrent, 135 treatment-naive) receiving neoadjuvant chemoimmunotherapy. The primary endpoint was overall survival (OS), analyzed using multivariable time-dependent Cox regression, propensity score matching (PSM), inverse probability of treatment weighting (IPTW), and landmark analysesResults The overall response rate was 66.3%; the pathological complete response rate among 117 surgical patients was 35.0%. Responders had better OS than non-responders (HR 0.19, 95%CI 0.11–0.32, p < 0.001). Surgery was associated with improved OS versus non-surgical management (HR 0.23, 95%CI 0.13–0.39, p < 0.001), and the association remained consistent after multivariable adjustment (HR 0.29, 95%CI 0.16–0.52), PSM (47 pairs; HR 0.34, 95%CI 0.17–0.66), IPTW (HR 0.36, 95% CI 0.18–0.74), and 6-month landmark analysis (HR 0.22, 95%CI 0.13–0.39). Radiotherapy was also associated with improved OS (HR 0.43, 95%CI 0.21–0.87, p = 0.018); postoperative radiotherapy was associated with longer OS than surgery alone (HR 0.43, 95%CI 0.16–0.98, p = 0.047), and surgery was associated with improved OS in recurrent disease (HR 0.47, 95%CI 0.21–0.99, p = 0.049).Conclusion In treatment-naive and recurrent LA-OSCC, neoadjuvant chemoimmunotherapy followed by response-directed R0 resection and risk-adapted radiotherapy was associated with improved OS. These retrospective associations warrant confirmation in prospective randomized trials.