Tumor genomic alterations and their exploratory association with treatment response in recurrent head and neck cancer patients

Abstract While the genetic landscape of treatment-naïve head and neck squamous cell carcinoma (HNSCC) is well characterized, the molecular features of recurrent disease remain less defined. We investigated the genetic alterations in recurrent HNSCC and their associations with overall survival (OS), progression-free survival (PFS), and treatment response. We performed a retrospective analysis of 86 patients with recurrent HNSCC treated at the University of Maryland Medical Center. Clinical data were extracted from electronic health records, and targeted DNA sequencing of tumor samples was performed by Tempus. Genomic profiles were compared with primary HNSCC profiles from The Cancer Genome Atlas (TCGA). Associations between mutations and clinical outcomes were evaluated using Cox proportional hazards models. TP53 mutations were most prevalent (55%) and significantly correlated with worse OS (HR: 1.77, p = 0.033) and poor treatment response ( p = 0.0067). CDKN2A (29%) and TERT (26%) mutations occurred more frequently in recurrent tumors compared to primary TCGA tumors. While CDKN2A mutations showed better PFS (HR: 0.57, p = 0.045), TERT mutations showed association with poorer PFS (HR: 1.66, p = 0.045) and reduced response to immune therapy ( p = 0.0854), although the latter did not reach statistical significance. Notably, although in small number, FGF4 (9%) mutations were associated with significantly poor survival (HR: 2.16, p = 0.049) and poorer outcomes following surgery and/or radiotherapy ( p = 0.0124). However, contradictory to other mutated genes, PIK3CA mutations (14%) correlated with better OS (HR: 0.37, p = 0.032). Recurrent HNSCC exhibit distinct genomic alterations associated with survival and therapeutic response. These findings are exploratory and hypothesis-generating. Prospective validation in larger, independent cohorts is necessary before these associations can inform treatment decisions.