Abstract
OBJECTIVES: The role of HPV in oral cavity cancers was investigated using two markers of viral exposure. MATERIALS AND METHODS: HPV DNA and p16(INK4a) expression were evaluated in tumor tissue from a U.S. population-based sample of 122 invasive oral cavity cancer cases. RESULTS: HPV DNA was detected in 38 of 122 (31%) oral cavity tumors. Seven genotypes were detected including HPV 16, which was found in 22% of tumors. p16(INK4a) was expressed in 30% of tumors and was poorly correlated with HPV DNA detection (Kappa <0.1). Joint positivity for HPV 16 and/or 18 and p16(INK4a) was observed in only 7% of cases. When comparing cases diagnosed in 1993-1999 and in 2000-2004, positivity for HPV DNA 16/18 increased from 19% to 39% (p = 0.02) and joint HPV 16/18 - p16(INK4a) positivity increased from 0% to 12% (p = 0.01). For gingival tumors, HPV 16 and/or 18 positivity was 67% compared to 11-38% for other sites (p = 0.02); joint HPV 16/18 - p16(INK4a) positivity was 33% compared to 0-8% for other sites (p = 0.01). The association of HPV with gingival tumors and more recent diagnosis period remained after adjustment for age and stage (p < 0.05). Neither HPV DNA nor p16(INK4a) were associated with overall survival. CONCLUSIONS: Based on both HPV DNA and p16(INK4a), HPV is etiologically linked to a limited subset of oral cavity cancers. However, the role of HPV in oral cavity cancer may vary widely by subsite and may have increased over time, similar to trends observed for oropharyngeal cancer.