Abstract 6062: Viral genome structure correlates with patterns of somatic alteration in human papillomavirus associated head and neck squamous cell carcinoma

Abstract

Abstract Integration of viral DNA into the host genome has been implicated in human papillomavirus (HPV) associated tumorigenesis, and studies in HPV associated head and neck squamous cell carcinoma (HNSCC) attempting to classify viral integration status have reported associations among HPV status, viral load, and overall survival. However, few genomic studies have correlated HPV structure with commonly occurring somatic mutations in HNSCC. Here, we conducted hybrid-capture DNA sequencing using a panel of over 800 cancer related genomic targets, as well as probes specific for the full HPV16 and HPV18 genomes, in a novel cohort of 525 HNSCC cases. Paired reads were aligned to a human reference containing complete viral genomes, including all high-risk HPV types, which allowed for robust identification of HPV(+) tumors and specific HPV type. Variant calling and filtering, followed by pathway analysis of significantly mutated genes, defined the catalog of somatic driver alterations in HNSCC. Viral genome structure, integration status, and viral load were determined via initial copy number analysis of HPV-specific reads, followed by identification of viral/human breakpoints and indels with variant calling algorithms. Manual review was required in some cases due to variability in read coverage and alignment to HPV genome. We identified 252 HPV(+) tumors, the majority of which were HPV16 (n=228, 90%). Analysis of HPV16(+) tumors revealed five classes with variable viral loads: pure episome (EPI, n=54), episomal rearranged (EPI-R, n=51), episomal chimeric (EPI-C, n=22), integrated (INT, n=92), and integrated deleted (INT-A, n=12). Integrated tumors had at least one integration event (INT) and a few exhibited recurrent deletions in the viral genome (INT-A). Episomal samples had either complete HPV genomes (EPI), circular genomes with one or more indels (EPI-R), or circular genomes with HPV and human DNA (EPI-C). Interestingly, PIK3CA, the most frequently mutated gene in HPV(+) HNSCC, was mutated at a higher frequency in all integrated samples compared to all episomal types (44% vs 23%, p=0.005). We also noticed more deleterious mutations in innate immunity genes (HLA-A/B, B2M, TRAF3, BIRC3) in the episomal samples (16% vs 8%), which may allow cells with intact HPV genomes to evade anti-viral immune responses. Additionally, mutations in oxidative stress response genes that lead to the constitutive activation of NRF2 (NFE2L2, CUL3, KEAP1) were more frequent in integrated samples (11% vs 4%), which could confer radiotherapy resistance and a worse prognosis. Overall, using DNA sequencing and integrative genomic analysis in the largest cohort of HPV(+) HNSCC to date, we developed a novel classification method based on viral genome structure which associates with host somatic alterations that may contribute to variable outcomes in HNSCCs. Citation Format: Jeremiah Ray Holt, Xiaobei Zhao, Hyo Young Choi, Paul Little, Angela L. Mazul, Benjamin Wahle, Vonn Walter, Jose P. Zevallos, David Neil Hayes. Viral genome structure correlates with patterns of somatic alteration in human papillomavirus associated head and neck squamous cell carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6062.