HPV Cancer Cohort Consortium (HPVC3)

Mattias Johansson-JohanssonM@iarc.fr
Hilary Robbins-RobbinsH@iarc.fr

NIH intramural funds

In 2004, world experts met to evaluate the carcinogenicity of HPV[1]; they determined that for HPV16, the most carcinogenic of the HPV genotypes, data showed a causal role in anogenital cancers including the cervix, vulva, vagina, penis, and anus, and some head and neck tumors specifically in the oropharynx. Cancers of the oral cavity were included in the cancers considered to be caused by HPV16, however, the molecular and epidemiologic evidence in support of this association is considerably weaker than for oropharyngeal cancers. This has left the research community very skeptical and an additional investigation would certainly help in resolving this controversy. Lastly, HPV was reported to possibly have a role in cancers of the larynx, conjunctiva, and skin.

HPV. HPV is a phylogenetically diverse virus, with HPVs in different species having tropisms for different tissue types. As an example, HPV infections in the genus α mainly infect mucosal surfaces and are responsible for epithelial cancers occurring at the genital and head and neck sites; within this genus, HPV16 is the most important carcinogenic type accounting for half of cervix and higher proportions of HPV-associated cancers at other mucosal sites. HPV infections in the genus ß are responsible for cutaneous infections such as hand and foot warts and have been associated with skin cancers; current data do not implicate a specific HPV type in this genus being consistently associated with cancer risk.

Head and neck cancers. To date, one study prospectively evaluated the link between HPV and head and neck cancers: Mork et al used a Nordic cohort to evaluate HPV seropositivity in serum samples collected on average nine-years before diagnosis of head and neck cancer; they reported a 3.6-fold (non-significant), 14.4-fold (significant), and 2.4-fold (borderline significant) increased risk of oral (n=19 cases), oropharyngeal (n=26), and laryngeal (n=76) cancers associated with HPV16 seropositivity, respectively [2]. In this study, only antibodies against the L1 region of the HPV genome, which codes for the viral capsid, were investigated; this serologic biomarker serves as a marker of exposure to HPV infection. The limitation to this marker is that it is not anatomically site specific and likely reflects HPV exposure at the genital tract as well as the oral region. Because of collinear behaviors such as sexual activity and tobacco use, the most significant risk factor for head and neck cancer, the reported associations are potentially confounded; this is especially important for the weaker associations observed for oral cavity and laryngeal cancers. For this type of investigation, measurement of antibodies against the HPV16 oncogenes E6 and E7, considered to be validated markers of an HPV-associated invasive cancer, would be very interesting; these assays are now widely available.

Skin cancer. It remains unknown whether HPV is an etiologic factor causing skin cancers or merely a passenger virus. In many case-control studies to date (references), significant associations have been observed between HPV and cancer. To clarify the association, one prospective study evaluated the link between HPV and 54 cases of skin cancer: Casabonne et al found that the seroprevalence of many ß HPV types was higher among prevalent cases than among either incident cases or controls; among incident cases, seroprevalence was higher for specimens collected proximal to diagnosis (within 18 months of diagnosis) [3]. Despite this effort, it remains unknown whether HPV is involved in the etiology of skin cancer or whether the observed antibody response is instead a consequence of the cancer. Fortunately, standardized multiplex serologic assays have been recently developed for the L1, E6 and E7 regions of the ß HPV types, which allows for the evaluation of the association between these HPV types and cancers. Unlike head and neck cancers, the L1 assays are appropriate for investigating skin cancers because 1) we want to compare exposure between cases and controls, and 2) there is no concern about confounding by sexual activity and its correlates.

We propose to combine information from several cohorts in the NCI Cohort Consortium to prospectively evaluate the risk conferred by serologic evidence of HPV infection detected prior to diagnosis of squamous cell carcinoma of the oral cavity, oropharynx, larynx and skin.

N/A

We will conduct a nested case-control study comparing cases of cancers considered to be HPV-associated with cancer-free controls. Cases will be defined as individuals diagnosed with incident or prevalent squamous cell carcinoma of the oral cavity, oropharynx, larynx, and skin. Controls will be defined as individuals with no cancer, matched to cancer cases on age, gender, year of specimen collection, and cohort. For all individuals, serum will be tested for presence of antibodies to the L1, E6 and E7 viral proteins (HPV genus α for the head and neck cancers and genus ß for the skin cancers). Optimally, serial blood samples collected pre-diagnostically and at diagnosis will be collected and utilized for HPV antibody testing. To further enrich this work, questionnaire data on smoking and alcohol history, as well as sun exposure, will be utilized when available.

Only with prospective data will we be able to clarify the role of HPV infection in cancers where HPV associations have been reported. Head and neck cancers are considered rare and occur in <10/100,000 individuals. While skin cancer is a more common outcome (~90/100,000), many studies do not collect non-melanoma skin cancer as an endpoint, making this another difficult topic to evaluate. It is very unlikely that a single study will be able to adequately evaluate the link between HPV and cancers at the extra-cervical sites being considered in this proposal. With the availability of both biologic samples and questionnaire data from these pooled resources, we can address critical unresolved issues in the viral etiology of these important cancers.

50

N/A

N/A

N/A

No