Background: HPV-positive oropharyngeal squamous cell carcinoma (OPSCC) has increased in incidence over the last two decades.  The vast majority of HPV-positive OSPCC cases are associated with HPV-16 type 16, with a minority associated with other HPV types.  There is recent evidence of an association between HPV subtype and survival (Bratman et al.).  The purpose of this study is 1) to describe the prevalence of non-HPV-16 type cases in a cohort of HPV-positive OPSCC cases, and 2) determine the association between HPV subtype and survival.  In secondary analyses, we will compare HPV PCR genotyping with HPV DNA sequencing in a subset of tumor samples.

Methods: Our cases were identified from the Carolina Head and Neck Cancer Study (CHANCE), a population based case-control study. HPV subtyping using PCR was performed in all available tumor samples, and a subset of tumors underwent targeted DNA sequencing of the HPV genome.  PCR was used to determine the following HPV subtypes: HPV-16, HPV-6, HPV-11, HPV-18, HPV-26, HPV-31, HPV-33, HPV-35, HPV-39, HPV-58, HPV-59, and HPV-8.  Tumors that were classified as HPV-16-positive if they were only positive for HPV-16 and no other subtype, and as HPV-other-positive if they were positive for any other subtype.  Follow-up time was calculated from the date of diagnosis to date of death. Death was determined by linking cases from CHANCE to the National Death Index.  Kaplan-meier survival curves were constructed and 5-year survival was calculated with the lifetable method.

Results: Overall, 183 HPV-positive OPSCC cases were identified.  Approximately 65% (N = 161) were HPV-16-positive and 8.8% (N = 22) were HPV-other-positive, including 1 case of HPV-6, 4 cases of HPV-18, 1 case of HPV-12, 4 cases of HPV-33, 8 cases of HPV-35, 1 case of HPV-39, 1 case of HPV-58, 2 cases of HPV-59, and 1 case of HPV-82.  Although not significant (log-rank p-value: 0.142), 5-year survival for HPV-other-positive subtypes was clearly lower (56.25%) than HPV-16-positive (70.88%). Next generation sequencing of the HPV genome was performed in 64 cases, including 59 HPV-16 and 5 HPV-other subtype.  When HPV PCR genotyping was compared to sequencing, we demonstrated concordance of 100% for HPV16 cases and 3 of the 5 non-HPV16 subtypes cases.

Conclusions: This preliminary analysis suggests that non-HPV-16 subtype may be associated with worse survival compared to HPV-16 associated OPSCC. We also demonstrate that HPV subtypes 33 and 35 may be the most common non-HPV-16 subtypes in HPV-positive OPSCC.  Finally, we demonstrate excellent concordance between HPV PCR genotyping and DNA sequencing.  Larger sequencing studies are currently underway to further describe the HPV genome in HPV-positive OPSCC, including an analysis of HPV viral structure, integration events , and their effect on outcomes in HPV-positive OPSCC.