Frequency of DNA repair gene mutations and impact on prognosis in HNSCC.

Abstract

e18016 Background: Next generation sequencing (NGS) has introduced the opportunity for targeted therapies and guidance regarding disease course and prognosis. The mutational landscape of squamous cell cancers of the head and neck (HNSCC) remains incompletely described. DNA repair gene (DRG) mutations are targeted by PARP inhibitors. This study presents the prevalence of DRG mutations in tumor and blood samples of patients with HNSCC and their outcome data. Methods: In this retrospective study, demographic and outcome data were collected and analyzed with regard to the presence or absence of mutated DRG (BRCA1, BRCA2, ATM, BRIP1, BARD1, CDK12, CHEK1, CHEK2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D, RAD51L, APC, ARID1A and MLL3). Mutated DRG were detected in tumor tissue (tDNA) by Foundation Medicine and/or in blood (ctDNA) by Guardant 360. All 18 genes were analyzed by FM, but only six are included in the G360 panel (BRCA1, BRCA2, ATM, APC, ARID1A and CDK12). Results: Our analysis included 170 HNSCC patients. Of these, 138 underwent NGS via ctDNA, 146 via tDNA and 114 via both methods. Sixty-five patients (47%) had at least one tDNA DRG mutation, 54 patients (37%) had at least one ctDNA DRG mutation and 96 patients (56%) had at least one DRG mutation detected by either method. No significant association was found between DRG mutations and age, gender, race, HPV status, tobacco/alcohol use or stage at diagnosis. Subsite analyses revealed that laryngeal primaries were associated with higher prevalence of DRG mutations detected via tDNA (p = 0.05), ctDNA (p = 0.03) or either method (p = 0.01). Oropharyngeal primaries correlated with a lower prevalence of DRG mutations detected via tDNA (p = 0.03) or via either method (p = 0.02) but were less significant when mutations were detected via ctDNA alone (p = 0.08). Mutated DRG detected via ctDNA correlated significantly with stage at time of ctDNA collection (p = 0.03), presence/absence of cancer at last visit (p = 0.05) and with stage at last visit (p = 0.05). Two-year survival and overall survival (OS) measured from the time of ctDNA collection correlated significantly with mutated ctDNA DRG. The relationship between ctDNA DRG mutations and OS remained statistically significant in a Cox proportional hazards regression model when adjusted for age, tobacco use, tumor site, nodal stage at diagnosis, and previous treatment with chemotherapy, radiation or combined chemoradiation therapy in a multivariate analysis model (p = 0.05). No similar correlation was found between tDNA mutations in DRG and prognosis. Conclusions: A significant proportion of patients with HNSCC were found to have mutations in DRG. Patients with laryngeal disease were most likely to have DRG mutations, whereas those with oropharyngeal disease were less likely. Patients with DRG mutations in ctDNA, but not tDNA, had significantly worse prognoses with a lower likelihood of overall survival and higher disease burden at last visit.