Accurate Testing Needed to Identify Variety of HPVs in Head and Neck Cancers

PET/CT image of patient with stage IVA HPV-positive squamous cell carcinoma of the head and neck

High-risk human papillomaviruses may be present more frequently than previously estimated in a variety of head and neck cancers, supporting the use of a combination of methods to detect high-risk strains, according to the results of a study published recently in JAMA Otolaryngology-Head & Neck Surgery.

Assessment of HPV strains is an important aspect of defining effective treatments for patients with head and neck cancers based on their individual tumor characteristics. Currently there are multiple methods employed to detect HPV, with no one method defined as the optimal choice.

Thomas E. Carey, PhD, codirector of the head and neck oncology program at the University of Michigan Comprehensive Cancer Center, and colleagues conducted a study to define optimal assessment strategies for high-risk HPVs in formalin-fixed paraffin-embedded head and neck tissue specimens. According to the researchers, “availability of fresh-frozen tumor tissue or access to serologic assays is rare,” making fixed tumor from a biopsy the most readily accessible.

They compared the use of p16 immunohistochemical staining, in situ hybridization (ISH), and polymerase chain reaction-MassArray (PCR-MA), with L1 PGMY-PCR and sequencing to resolve method discordance. The study had 338 tissue samples from patients with oropharyngeal, nasopharyngeal, and oral cavity tumor tissue specimens treated from 2001 to 2011.

“There was a much higher incidence of high-risk HPV involvement in head and neck cancer at multiple sites, including the observation that in our population the proportion of oropharyngeal cancers with high-risk HPV was between 80% and 90%, which rivals cervical cancer for HPV involvement,” Carey said.

Looking at all of the methods examined, 86% of oropharyngeal tumors were positive for HPV: 83% by PCR-MA, 83% by p16, 73% by ISH. The majority of tumors testing positive had HPV16 alone; however, 6% contained other high-risk HPVs including HPV18, HPV33, HPV35, and HPV39. Two cases contained multiple types.

Carey and colleagues also found that 50% of nasopharynx tumors were positive for HPV using p16, and 26% of oral cavity tumors were positive for HPV using one or more methods.

“It was also surprising and important to observe that the heterogeneity of high-risk HPV types was greater in nasopharynx and oral cavity than in oropharyngeal sites,” Carey said. “Many investigators have been biased to the notion that only HPV16 is important.”

Among oropharyngeal cancers the researchers found HPV16, HPV18, HPV39, and HPV59. In the oral cavity cancers, HPV16, HPV31, HPV33, HPV35, and HPV66 were all present, with two cases presenting with multiple HPV types.

Carey and colleagues also calculated the sensitivity and specificity of each assessment method. PCR-MA had 99.5% sensitivity and 100% specificity; p16, 94.2% sensitivity and 85.5% specificity; and ISH, had 82.9% sensitivity and 81% specificity.

The current HPV vaccines only vaccinate against HPV16 and HPV18. As the vaccine begins to take effect and reduce infection by HPV16 and HPV18, there may be an increase in infection by other HPV types, Carey said.

“We don’t yet know whether the other types are more or less dangerous and vigilant surveillance is necessary,” Carey said.