Assessing HPV Prevalence and Types in Cancer Lesions

Michigan Cancer Surveillance Program

With the development, approval, and use of a new vaccine designed to prevent specific human papillomavirus (HPV) infections, the potential to reduce the incidence of cervical neoplasia has been greatly increased. A number of issues, however, are unknown. One critical issue is the lack of population-based information on the HPV types present in pre-invasive and invasive lesions that are known to be HPV-associated. The other is information on HPV types by cancer site and what variations, if any, exist by demographic factors such as age and sex.

A feasibility study was undertaken initially in Michigan, Florida, Kentucky, and Louisiana and subsequently in Hawaii, Iowa, and California to explore the potential for using population-based registries to assess HPV prevalence and HPV types within selected cancer lesions. The purposes of this feasibility study are to—

• Develop an infrastructure for systematic monitoring of HPV types in cervical cancer cases.
• Use the infrastructure to conduct a pilot study on HPV typing on all or a sample of cervical and other HPV-associated (penile, vulvar, vaginal, anal, and oropharyngeal) cancer cases from participating population-based cancer registries (estimated total N = 2,400) for the years 2004–2005. This pilot study will test the feasibility of the new infrastructure and provide the baseline distribution of HPV types associated with cancer cases in selected areas prior to the implementation of the HPV vaccine in 2006.

A standard protocol was developed, and each registry worked with laboratories in its region to track down tissue blocks from selected cancer cases. The laboratory procedures required were difficult, but laboratories did an excellent job overall. All participating registries were successful in selecting sample cases, locating suitable tissue blocks, and negotiating with laboratories for tissue preparation and shipment to CDC. All major project goals were met: more than 3,000 tissue samples were collected, representing six cancer sites.

While the baseline information on HPV types by cancer site is an extremely important result of the project, the demonstrated ability to conduct such a study is also an important finding. In the future, similar data will need to be assessed to determine if HPV prevalence and HPV type distribution shift following HPV vaccination. Population-based cancer registries demonstrated a new potential study design by coordinating this study successfully across multiple states. Central cancer registries can be used to address, on a population basis, other important issues associated with cancer diagnosis and prognosis that require access to tumor tissue samples.