Prevalence of human papillomavirus in epithelial ovarian cancer tissue. A meta‐analysis of observational studies

Abstract

The role of human papillomavirus (HPV) in the pathogenesis of ovarian cancer is controversial, and conflicting results have been published. We conducted a systematic review and meta-analysis to estimate the prevalence of HPV in epithelial ovarian cancer tissue. Observational studies published until 4 March 2013 were identified in PubMed and Embase. We adhered to MOOSE guidelines and included 22 studies (case-control, cross-sectional studies). A pooled estimate of the HPV prevalence with corresponding 95% confidence interval (CI) was calculated based on a random effect model. In a meta-regression analysis we examined the contribution of different factors to heterogeneity. Furthermore, publication bias was evaluated. The pooled HPV prevalence in ovarian cancer tissue was 15.5%, but wide variation was found (0-66.7%). After stratification by geographical region, publication year, tissue type and method of HPV detection, we found that the prevalence of HPV varied most markedly by geographical area, the prevalence being 45.6% (95% CI, 31.0-60.3) in Asia, 18.5% (95% CI, 8.5-28.6) in Eastern Europe, 1.1% (95% CI, -1.6 to 3.8) in Western Europe and zero in North America. A meta-regression analysis revealed that the difference between geographical regions could not be explained by HPV detection method or type of tissue. Great geographical variation exists in HPV prevalence in ovarian cancer tissue, which is not explained by different HPV detection methods. The results suggest that HPV is unlikely to play an important role in Western European and American women, but cannot reject a role of HPV in other populations.