HPV in oral squamous cell carcinoma vs head and neck squamous cell carcinoma biopsies: a meta-analysis (1988-2007)

In the literature, there exists a wide range of human papillomavirus (HPV) DNA prevalence for head and neck squamous cell carcinoma (HNSCC), especially in relation to methods of viral detection and the lesion site. We estimated the pooled prevalence of HPV DNA in biopsies of HNSCC generically grouped versus oral squamous cell carcinoma (OSCC) in relation to the method of viral DNA detection, with the primary end point of verifying if these two variables (specification of tumour site and method of HPV DNA identification) influence the datum on HPV assay.

By means of MEDLINE/PubMED/Ovid databases, we selected studies examining paraffin-embedded (PE) biopsies of HNSCC and OSCC. According to the inclusion criteria, 62 studies were analyzed. The following data were abstracted: sample size, HPV DNA prevalence, methods of detection [PCR and in situ hybridization (ISH)] and HPV genotypes. After testing the heterogeneity of the studies by the Cochran Q test, metanalysis was performed using the random effects model.

The pooled prevalence of HPV DNA in the overall samples (Sigma: 4852) was 34.5%, in OSCC it was 38.1% and in the not site-specific HNSCC was 24.1%. With regard to the detection method, PCR-based studies reported a higher prevalence rate than ISH-based rates (34.8, versus 32.9%) especially in the OSCC subgroup (OSCC PCR based: 39.9%).

These findings support the assumption that a correct distinction of HNSCC by site, together with the use of more sensitive HPV DNA detection methods, should be considered as essential prerogatives in designing future investigations into viral prevalence in head and neck tumors.