New data from Uganda indicate that male circumcision—which has already been shown to reduce men's risk of becoming infected with human papillomavirus (HPV), HIV and other STIs—confers partial protection against HPV infection on female partners.
New data from Uganda indicate that male circumcision—which has already been shown to reduce men's risk of becoming infected with human papillomavirus (HPV), HIV and other STIs—confers partial protection against HPV infection on female partners.1 At the end of a two-year randomized trial, 28% of women whose male partner had been circumcised tested positive for a high-risk (i.e., potentially cancer-causing) type of HPV, compared with 39% of women whose partner was uncircumcised (prevalence risk ratio, 0.7).
The data come from a randomized controlled trial conducted in Rakai, Uganda, in 003–2007 to assess the effects of circumcision on transmission of HIV and other STIs. Researchers recruited uncircumcised men aged 15–49 who wished to undergo the procedure and randomly assigned them to be circumcised immediately (intervention group) or after 24 months (control group). In addition, they invited the female partners of male participants who were married or in a long-term relationship to take part in the study. At baseline, 12 months and 24 months, female partners provided information on their social and demographic characteristics, sexual risk behaviors and health, including symptoms of genital-tract infection. They also provided self-administered vaginal swabs to be tested for 14 high-risk HPV types. If a woman tested positive for an HPV type she had not previously had, the infection was assumed to have occurred midway between the last negative test for that type and the new positive test; infections were classified as single or multiple, depending on whether more than one HPV genotype was detected. The infection was considered cleared if a previously detected HPV type was no longer present at a subsequent visit.
Because having HIV increases one's risk of HPV infection (and vice versa), participants who were HIV-positive or who seroconverted during the trial were excluded from the analysis. Women who had not enrolled at the same time as their partner or had not provided initial vaginal swabs were also excluded, yielding a sample consisting of 648 women in the intervention group and 597 in the control group. Data from 84–87% of female participants were available at the 12- and 24-month follow-ups. At baseline, most women in the intervention and control groups (82–85%) were in monogamous unions, while the remainder were in polygynous ones. Most were Catholic (60–61%), had a primary education (72–74%) and had not used condoms in the past year (81–85%). Fewer than 1% had had transactional sex, about one-third reported alcohol use with sex and 20–23% had ever used voluntary HIV counseling and testing services. Almost half of women in the two groups (46–47%) reported having had at least one STI symptom in the past year. Patterns of sexual behavior differed between the women and their partners, as 43–46% of men but only 4% of women reported having had more than one sexual partner in the past year.
At baseline, the proportion of women in the intervention group who tested positive for high-risk HPV (35%) was similar to that in the control group (37%). At 12 months, however, prevalence was lower in the intervention group (34%) than in the control group (41%; prevalence risk ratio, 0.8); after 24 months, those proportions were 28% and 39%, respectively (0.7). The prevalence of low-risk HPV types was also lower among women in the intervention group than among those in the control group (33% vs. 43%; prevalence risk ratio, 0.8).
Similarly, the incidence of infection with any high-risk HPV type was lower in the intervention group than in the control group—25 vs. 34 cases per 100 person-years during the first 12 months (incidence rate ratio, 0.7) and 21 vs. 27 cases per 100 person-years during the full 24 months (0.8). Moreover, the incidence of infection with multiple high-risk strains was lower among women in the intervention group than among those in the control group, both during the first year (7 vs. 11 cases per 100 person-years) and over the full length of the study (9 vs. 13 per 100).
A regression analysis examining the relationship between infection with high-risk HPV at 24 months and women's social and demographic traits suggested that a number of subgroups benefited from the intervention. Women in the intervention group were less likely than those in the control group to become infected if they were aged 15–19 (incidence rate ratio, 0.5) or 25–29 (0.7), were in a monogamous relationship (0.8) or had at least a secondary education (0.6). They also benefited from circumcision if, in the past year, they had had only one sex partner (0.8), had not had any nonmarital relationships (0.8), had not used condoms (0.8), had not had sex while using alcohol (0.7) and had not experienced STI symptoms (0.7–0.8). After adjustment for women's age, condom use, alcohol use with sex, number of sex partners in the past year and polygyny, the overall incidence of HPV was lower among women in the intervention group than among those in the control group (0.8).
Finally, clearance of high-risk HPV was more common in the intervention group than in the control group. After adjustment for enrollment characteristics, 66% of infections among women in the intervention group cleared, compared with 59% of those among controls.
The investigators conclude that the apparent reduction in HPV risk among women whose partner had been circumcised was most likely due to "a reduction of penile HPV carriage ... [leading to] decreased female incidence and increased clearance, probably by lowering the chances of re-infection." Although these results were obtained among a sample of HIV-negative, partnered individuals and may not be applicable to other populations, the researchers contend that their data strengthen the case for using circumcision to reduce transmission of STIs in resource-poor settings. Because the procedure provides only partial protection against HPV, however, the investigators point out that "the promotion of safe sex practices is also important."
—H. Ball
REFERENCE
1. Wawer MJ et al., Effect of circumcision on HIV-negative men on transmission of human papilloma-virus to HIV-negative women: a randomised trial in Rakai, Uganda, Lancet, 2011, 377(9761):209–218.