HPV infection and oral carcinogenesis

Abstract

To the Editor,High risk human papilloma viruses (HPV) have been found in head and neck squamous cell carcinomas (HNSCC), particularly in oropharyngeal carcinomas (1). Nevertheless, there are some controversial aspects regarding this issue (2) such as whether the HPV infec-tion is a temporary or a persistent oral infection in these patients.Recently, Chuang et al. (3) have associated the presence of HPV-16 DNA in surveillance salivary rinses with a significant risk for recurrence in HNSCC.We hypothesized that the improved prognosis of many patients with HPV-related oropharyngeal carcinoma is due to the temporary nature of the infection, and there-fore in the absence of lesions no HPV genomic DNA is detected in oral smears.We have analyzed a longitudinal cohort of 47 patients who had been treated for a previous oral squamous cell carcinoma (OSCC) without any precancerous or cancer-ous oral lesion when the oral smear was collected. The group was composed of 29 men and 18 women with a mean age of 62.02 years (45-87 years). Primary OSCC sites were tongue (44.6%), floor of the mouth (31.9%), oropharynx (17%), gingiva (12.7%) and buccal muco-sa (8.5%). Tumours were classified as T1 (53.3%), T2 (37.7%), T3 (2.2%) and T4 (6.6%).A cytological brush (CytobrushTM) was vigorously swabbed over the oral cavity. Cytological samples were centrifuged at 1500 rpm for 15 minutes and washed with phosphate-buffered saline (PBS) to obtain the cellular pellet. DNA was extracted following the clas-sical proteinase-K, phenol/chloroform method and pre-cipitated with ethanol. The detection of HPV genomic DNA was performed using the Papitype kit (Progenie-Molecular S.L, Valencia. Spain). In brief, a polymerase-chain reaction (PCR) targeting the L1 region of the viral genome, harbouring an internal control to avoid false negatives, was carried out. Subsequently, and in HPV+ samples, typing was performed using a restriction frag-ment length polymorphism (RFLP).To our surprise, no HPV genomic DNA was detected in any of the samples. These results agree with other published data (4, 5) which indicate that the HPV infec-tion, when present in oral cancer patients, is temporary and probably related with the presence of a malignant or premalignant lesion.Conscious of the limitations of our study, we think, as other authors (3), that the presence of HPV genomic DNA in oral smears may be used as a biomarker in