Purpose: The recent FDA approval of a highly-effective vaccine that prevents infection with two of the most common cancer-causing types of human papillomavirus (HPV) requires that we seriously consider the appropriate target population. Using both a dynamic model of HPV transmission and a first-order microsimulation model of cervical carcinogenesis, we assessed the comparative benefits and cost-effectiveness of vaccinating one versus both genders.

Methods: We developed a dynamic model of sexual transmission of HPV between men and women using population statistics from Brazil. Model inputs, including sexual mixing and transmission per partnership, were adjusted to achieve consistent visual fits with epidemiological data on HPV prevalence and cervical disease from Brazil. Data were obtained from behavioral surveys, epidemiological studies, and cancer registries in Brazil. Estimates of reduction in HPV incidence among women under different scenarios of vaccine coverage for men and women were derived using the dynamic model and then applied to our previously-developed microsimulation disease model to generate estimates of long-term benefits and cost-effectiveness of vaccinating both men and women and women alone. We assumed the vaccine prevents 90% of infection with types 16 and 18, and that vaccine-induced immunity is lifelong.

Results: The models predicted that vaccinating both men and women could reduce a woman's lifetime risk of cervical cancer by an additional 1% to 80%, depending on the level of coverage, compared to vaccinating women alone. When vaccine coverage for women is low, covering men has the potential to substantially decrease cancer risk; however, at high coverage rates for women, only nominal health gains are achieved when vaccinating men. For example, increasing coverage of men from 10% to 100% provides nearly 5 times greater benefit when coverage is low (25%) than when coverage is high (75%) for women. Similar trends were apparent in terms of cost-effectiveness; using a cost-effectiveness threshold of GDP per capita in Brazil, including men in vaccination programs becomes cost-effective only when coverage of women decreases below 75%.

Conclusions: For countries interested in introducing an HPV vaccination program, decisions regarding vaccinating men in addition to women should take into consideration the level of coverage that may be achieved for women. Non-linear effects involving HPV transmission dynamics may be important to consider when predicting long-term effects of an HPV vaccination program.