Certain strains of human papillomavirus (HPV) have been shown to be etiologically related to the development of uterine cervical and other genital cancers, but their role in the development of malignancies at other sites is less well established. Previous studies have shown HPV DNA in tumors of the head and neck, but its prevalence has varied depending on the detection methods and the types of tumor and/or tissue examined. This study was undertaken to estimate the frequency of HPV DNA in squamous cell carcinoma (SCC) at different sites of the esophagus, head and neck and to compare the clinical behavior of HPV positive and negative tumors.

DNA was extracted from frozen tissue of 167 SCCs of the esophagus, head and neck. The DNA was screened for HPV sequences by polymerase chain reaction with two sets of consensus primers, one to a conserved region in the L1 gene (MY09/MY11) and the other to a conserved region in the E1 open reading frame (IU/IWDO). The products were run on agarose gels, detected by ethidium bromide staining, and then the gels were subjected to Southern blot analysis and hybridized with probes specific to HPV 6, 16, and 18. All tumors found to be HPV positive with the consensus primers were amplified with type specific primers, and in selected cases the presence of HPV DNA was confirmed by restriction enzyme digestion of the tumor DNA with conventional Southern blot analysis.

Overall, HPV sequences were found in 25 of 167 tumors (15%), but HPV was detected most frequently in tumors in Waldeyer's tonsillar ring. In that area, 9 of 15 (60%) were HPV positive. No HPV DNA was detected in 11 esophageal SCCs, 7 tumors of the pharynx/hypopharynx, or 6 pyriform sinus carcinomas. HPV DNA was detected in the following tumor sites: 1 of 28 (3.6%) in the larynx, 1 of 10 (10%) in the oral cavity, 5 of 39 (12.8%) in the tongue, 2 of 15 (13.5%) in the floor of the mouth, 3 of 21 (14.3%) supraglottic, and 1 of 7 (14.3%) in the lip. A high incidence of HPV DNA was also found in metastatic tumors located in cervical lymph nodes for which no primary site was clinically identified (3 of 8, 37.5%). With respect to age, gender, and tobacco and alcohol consumption, analysis of clinical data obtained by retrospective review showed no difference between patients with HPV DNA in their tumors and those in which no HPV was detected. However, HPV positive patients had larger tumors (P = 0.09) and a higher incidence of lymph node metastasis (P = 0.003). In spite of the higher stage of disease at presentation in HPV positive patients, there was no significant difference in 3‐year survival rates between HPV positive patients and HPV negative patients (43.1% vs. 48.8%, respectively). Median follow‐up was 27 months.

In the head and neck, HPV‐associated SCC had site specificity with the viral DNA frequently found in tumors in Waldeyer's tonsillar ring. Patients with HPV positive tumors presented with a higher stage of disease than patients with HPV negative tumors, but there was no significant difference in the 3‐year survival rates between these two groups of patients. Cancer 1997; 79:595‐604. © 1997 American Cancer Society.