Association of human papillomavirus genotype and phylogenic clade with oropharyngeal cancer outcomes.

ABSTRACT

BACKGROUND: Human papillomavirus (HPV)+ oropharynx cancer (OPC) has a more favorable prognosis than HPV-negative disease, but the impact of specific HPV genotype and phylogenic clade on patient outcomes is not well understood and has profound implications for treatment de-intensification. METHODS: The objective of this single-institution cohort study was to investigate the association of HPV genotype (16 vs high-risk non-16) and clade (A9 vs A7) with OPC outcomes. The primary endpoints were overall survival (OS) and event-free survival (EFS) in patients with M0 disease treated with curative intent. RESULTS: The cohort included 598 patients (87% HPV16, 98% A9). Compared to those with HPV16 OPC, individuals with non-HPV16 OPC had a higher age, comorbidity index, and proportion of non-whites, HIV+ patients, T4 tumors, and stage IV disease (AJCC 7th edition). Non-HPV16 genotype was associated with worse OS in univariate (HR = 2.17, 95% CI, 1.24-3.80, P = .0066), but not in multivariate analysis (HRadj = 0.84, 95% CI, 0.43-1.62, P = .5921). A7 clade was associated with worse OS in univariate (HR = 4.42, 95% CI, 1.60-12.30, P = .0041), but not in multivariate analysis (HRadj = 2.39, 95% CI, 0.57-9.99, P = .2325). Neither HPV genotype (HR = 1.60, 95% CI, 0.99-2.60, P = .0566) nor phylogenic clade (HR = 2.47, 95% CI, 0.91-6.72, P = .0761) was associated with EFS. CONCLUSION: Non-HPV16 genotype and A7 clade were associated with worse OS and trended toward worse EFS in univariate analyses. The survival differences were more pronounced by phylogenic clade than by HPV16 status, suggesting that the former may be a more useful classification for future studies. However, neither HPV16 status nor phylogenic clade was prognostic when adjusting for patient and tumor covariates, raising the question as to whether possible differences in outcomes are related to distinct clinical profiles rather than inherent viral properties.