PURPOSE A precise oncologic approach for head and neck squamous cell carcinoma (HNSCC) is necessary. We performed a genomic profile-based umbrella trial for the patients with platinum-refractory recurrent and/or metastatic HNSCC.

METHODS In this multicenter, open-label, single-arm phase II trial, we performed targeted next-generation sequencing (NGS). Patients were assigned to each treatment arm on the basis of their matching genomic profiles: arm 1, alpelisib, a PIK3CA inhibitor; arm 2, poziotinib, an epidermal growth factor receptor/HER2 inhibitor; arm 3, nintedanib, an fibroblast growth factor receptor inhibitor; and arm 4, abemaciclib, a CDK4/6 inhibitor. If there was no matching target, patients were allocated to arm 5, duvalumab ± tremelimumab, anti-PD-L1/cytotoxic T-cell lymphocyte-4 inhibitor. When progressive disease (PD) occurred in arms 1-4, cross over to arm 5 was allowed. The primary end point was disease control rate (DCR) in arm 1 and overall response rate (ORR) in arms 2-5 by investigator assessment.

RESULTS Between October 2017 and August 2020, 203 patients were enrolled, including crossover. In arm 1, the ORR was 21.2% and DCR was 65.6%. The ORR was 0% for arm 2, 42.9% for arm 3, 0% for arm 4, and 15.6% for arm 5. In the case of PD with durvalumab, tremelimumab was added, and the ORR for durvalumab + tremelimumab was 2.2%. The median progression-free survival was 3.4, 3.2, 5.6, 1.6, and 1.7 months for each arm, respectively. The median overall survival was 12.4, 6.1, 11.1, 9.1, and 12.7 months, respectively. Overall, the toxicity profiles were manageable, and there were no treatment-related deaths.

CONCLUSION To our knowledge, this study is the first biomarker-driven umbrella trial for platinum-refractory HNSCC using matched molecular targeted agents. We found that NGS-based genomic phenotyping was methodologically feasible and applicable.