Prevalence of human papillomaviruses in semen: a systematic review and meta-analysis

Study question: What is the prevalence of human papillomavirus (HPV) in semen?

Summary answer: HPV is present in the semen of asymptomatic men, with a pooled prevalence in a random effects meta-analysis of populations seeking fertility evaluation/treatment of 16%, versus 10% in other populations.

What is known already: The main risk of donor insemination (DI) is known to be contamination with an infectious agent. HPV is the necessary cause of cervical cancer, and plays an etiologic role in other anogenital cancers. Although it is known to be prevalent and sexually transmitted, donor semen specimens are not tested for the presence of HPV.

Study design, size, duration: A systematic review and meta-analysis of studies published between January 1980 and June 2013 were performed. Variables collected included characteristics of study populations, method of semen preparation, HPV DNA detection and genotyping, HPV types targeted and proportion of HPV positivity.

Participants/materials, setting, methods: Two investigators independently assessed the studies for inclusion in the review and abstracted the data, while others reviewed the extracted data in detail. Studies were included if they provided data on HPV DNA prevalence in semen and PCR-based methods were used. For the meta-analysis, reports were separated according to the study populations, creating two distinct subgroups: populations seeking fertility evaluation/treatments, and other populations. Data were analysed using a random effects model for each subpopulation.

Main results and the role of chance: The literature search identified 285 studies, and in the 27 studies that were included the HPV DNA prevalence in 4029 semen samples varied from 0 to 100%. The three studies focusing on sperm donors identified HPV DNA in 26.3, 7.5 and 16.0% of semen samples. HPV-16 was the most common type overall. The pooled prevalence in a random effects meta-analysis of seven studies focusing on infertile populations was 16% [95% confidence interval (CI): 10-23%] versus 10% (95% CI: 7-14%) in 11 reports focusing on other populations.

Limitations, reasons for caution: First, despite defining clinically relevant subgroups, substantial heterogeneity remained. Secondly, although we retrieved data from reports in English or French only, after reviewing the five reports in other languages only two more could have been added and, as their prevalence estimates were similar to those of studies included in our review, we do not believe that exclusion of these reports biased our results or conclusions.

Wider implications of the findings: HPV DNA can be found in donor semen and preliminary studies confirm genome activity. For this reason, and although the exact consequences of insemination with HPV-infected semen (cervical infections/lesions, impact on success rate of DI) remain to be clarified, we believe that HPV-infected sperm should be considered a health risk unless well-designed studies prove otherwise. The development and validation of adequate sperm washing techniques before DI appears to be a promising option.

Study funding/competing interest(s): C.L. and P.M. have no conflicts of interests relevant to the submitted work. H.T. has served as a consultant and on advisory boards and has received speaker fees and travel assistance from Merck-Frosst Canada, Glaxo SmithKline Pharmaceuticals, Belgium and Gilead Sciences. F.C. has received grants through his institution from Merck and Roche, as well as honoraria from Merck and Roche for lectures on HPV. M.-H.M. has received grants though her institution from Merck and Qiagen and lecture honoraria from Merck and GSK for conferences on HPV and best practices in cervical cancer prevention.

Trial registration number: N/A.