Background: The role of human papillomavirus (HPV) in colorectal cancer has been widely studied with conflicting results. We performed a systematic review and a meta-analysis to estimate the prevalence of HPV in colorectal adenocarcinomas and adenomas, and test the potential association.
Methods: The pooled HPV prevalence was estimated using a random effects model and the I(2) statistic was used to describe the amount of heterogeneity. Potential sources of heterogeneity were evaluated by meta-regression and stratified analyses. For the studies on adenocarcinomas including control tissue, random effects estimates of odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.
Results: Thirty-seven studies were included. Among the 2630 adenocarcinomas, the pooled HPV prevalence was 11.2% (95% CI, 4.9-19.6%) with substantial between-study heterogeneity (I(2)=97.2%). The HPV prevalence varied by geographical region with highest prevalence in South America (45.1%, 95% CI, 21.9-69.4%), Asia (39.2%, 95% CI, 20.3-60.0%) and the Middle East (32.2%, 95% CI, 1.1-79.3%), and by detection method with the highest HPV prevalence in PCR-based studies. In the eight case-control studies, the pooled HPV prevalence was 36.8% (95% CI, 21.3-53.8%) in adenocarcinomas and 1.6% (95% CI, 0.0-9.6%) in controls giving an OR of 6.0 (95% CI, 2.0-17.9%) for the association between HPV and colorectal cancer. Among the 415 adenomas, the pooled HPV prevalence was 5.1% (95% CI, 0.0-17.8%; I(2)=93.7%).
Conclusions: HPV may be associated with a subset of colorectal cancers. Future large-scale multicenter case-control studies with data on risk factors such as lifestyle and sexual behaviour are needed.
Keywords: Colorectal cancer; Human papillomavirus; Metaanalysis; Prevalence.
Copyright © 2014 Elsevier Ltd. All rights reserved.