In 2014, about 37,000 people in the United States will be diagnosed with oral cancer, 20% of whom will die from the cancer. Although tobacco use is among the strongest risk factors for oral cancers, the drastic decline in the prevalence of cigarette smoking since 1975 (from ~40% to <20%) has caused only moderate change in the incidence of oral cancer. This suggests that risk factors besides smoking also play important roles. The similarity between cervical cancer and oral cancer, both of which are squamous cell carcinomas, has prompted investigations into the role of human papilloma virus (HPV) in causing oral cancer. Oral cancer can be divided into oropharyngeal squamous cell carcinoma (OPSCC) and oral cavity (non-oropharyngeal) squamous cell carcinoma (OCSCC). Notwithstanding a demonstrated link between HPV and OPSCC, with HPV detected at a rate of ~60% in OPSCC patients, conclusions for OCSCC are controversial, for reasons unclear. Our understanding about the causality and prevalence of HPV in OCSCC is derived primarily from surveys conducted using traditional HPV detection methods that were designed to detect a limited number of high/low risk (HLR) HPV types in cervical cancer. By contrast, our preliminary analyses have used non-selective metagenomics analysis to provide a comprehensive map of HPV communities at different body sites in healthy human subjects. Of 109 HPV types detected, most were invisible by the commercial HPV detection kits that target cervical HPV types, and do not belong to the HLR HPV types. Furthermore, 85% of the oral HPV types belonged to beta and gamma groups, whereas 60% of vaginal HPV types belong to the alpha group. These data suggest that the failure to link HPV with OCSCC might be attributed to the difficulty of using traditional methods to detect oral HPV. Thus, we hypothesize that specific oral HPV types from the oral cavity might be associated with OCSCC. To test the hypothesis, our case-control study will conduct metagenomics analysis to examine whether unique oral HPV types are linked to OCSCC. A bioinformatics pipeline developed in our preliminary study will be used to recognize the HPV reads from the sequencing data, and further assign reads to HPV types. Significant differences in the presence or absence of HPV sequences in case and control samples will be assessed by ?2-test. A multivariate regression model will be established to identify the association of HPV types and OCSCC status. Knowledge gained from this study would help us re-evaluate the impact of oral HPV in OCSCC. If our hypothesis is correct, we will conduct a new prospective study to examine whether the pre-diagnostic presence of these HPV types predicts subsequent development of OCSCC using samples collected in existing prospective studies. Pre-diagnostic HPV types informative of late risk of OCSCC, if identified, could be used in clinical practice for more efficient screening and early detection of OCSCC. Furthermore, these HPV types could be incorporated into an existing HPV vaccine to extend the spectrum of protection beyond cervical cancer by also preventing OCSCC.
Although tobacco use is among the strongest risk factors for oral cancers, the drastic decline in the prevalence of cigarette smoking since 1975 (from ~40% to <20%) has caused only a moderate change in the incidence of oral cancer, suggesting that some other risk factors, such as human papillomavirus, also play important roles. In this study, we plan to examine whether non-cervical human papillomavirus types are associated with oral cavity squamous carcinoma using metagenomic analysis that is capable of detecting all of known HPV types. Knowledge gained from this study would help us re-evaluate the risk degree of HPV in oral cavity squamous cell carcinoma and conduct a new study to identify HPV-Markers that could potentially be used in clinical practice for more efficient screening and earlie detection of oral cavity squamous cell carcinoma.
