Pu-Gen An, Jie Zhang, X. Hu, Zi-Qi Zhang, Tong Zhang, Wenjie Wu
PURPOSE: The outcomes of salvage surgery for previously irradiated recurrent head and neck squamous cell carcinoma (HNSCC) remain suboptimal. This phase II trial evaluated the effects of preoperative tislelizumab (an anti-programmed cell death protein 1 monoclonal immunoglobulin G4 antibody) plus chemotherapy followed by salvage surgery and adjuvant tislelizumab in this setting. PATIENTS AND METHODS: Eligible patients (n = 34) with resectable recurrent HNSCC after radiotherapy received preoperative tislelizumab (200 mg), albumin-bound paclitaxel (260 mg/m2), and cisplatin (60-75 mg/m2) every 3 weeks for 2 cycles, followed by salvage surgery and 6 cycles of adjuvant tislelizumab. The primary endpoint was major pathologic response (MPR). Secondary endpoints included pathologic complete response (pCR), the objective response rate (ORR), 2-year event-free survival (EFS), 2-year overall survival (OS), and safety. RESULTS: The ORR was 35.3% (12/34). Of 26 surgical patients, R0 resection was achieved in 19 (73.1%). The MPR rate was 19.2% (5/26), with a pCR rate of 15.4% (4/26). At a median follow-up of 32 months, 2-year EFS was 39.6% and 2-year OS was 54.8%. All MPR patients remained disease free. Grade 1 to 2 adverse events were common; one grade 3 hyperglycemia occurred. High baseline B-cell receptor (BCR) repertoire diversity and clonal abundance (top 1%/10%) correlated with poor prognosis, with top 1% clonality showing strong prognostic power (AUC = 0.910; P = 0.006). CONCLUSIONS: Preoperative chemoimmunotherapy followed by surgery and adjuvant immunotherapy was feasible with encouraging survival in previously irradiated recurrent HNSCC. Baseline BCR repertoire characteristics may serve as a noninvasive prognostic biomarker.