Circumcised men are less likely than uncircumcised men to have penile human papillomavirus (HPV) infection, and female partners of men who engage in risky sexual behavior have a reduced likelihood of having cervical cancer if the man is circumcised. According to a multinational study conducted by the International Agency for Research on Cancer (IARC), the odds that circumcised men had penile HPV infection were about 60% lower than the odds that uncircumcised men had this diagnosis.¹ Among monogamous women who had a partner reporting six or more lifetime sexual partners, the odds of having cervical cancer were reduced by about 60% if the partner had been circumcised.

Recognizing that cervical cancer is the second most common cancer among women worldwide and that nearly all cases may be caused by HPV, the IARC Multicenter Cervical Cancer Study Group sought to investigate how circumcision affects the risks of genital HPV infection in men and cervical cancer in their partners. The researchers conducted a total of seven case-control studies between 1985 and 1993 in Spain, Colombia, Brazil, Thailand and the Philippines. They used a standardized questionnaire to interview women who had newly diagnosed, histologically confirmed cervical carcinoma in situ or invasive cervical cancer, as well as age-matched women without these forms of cancer. The investigators also enrolled each partner with whom these women had had regular intercourse for at least six months before the study. These men were administered a similar questionnaire and were tested for penile HPV infection.

The analyses were based on 1,913 male respondents--977 partners of women with cervical cancer and 936 partners of control women. Roughly half of the men were aged 38-56, about one-quarter were 37 or younger and the remainder were 57 or older. The women were distributed about equally among three age-groups--36 or younger, 37-48, and 49 or older. Nineteen percent of the men reported that they were circumcised.

Medical examination of about two-fifths of the men showed that 95% had reported their circumcision status correctly. Circumcised men had a marginally significantly higher educational level than did uncircumcised men; they also were less likely to report genital washing after intercourse but more likely to have good genital hygiene. The two groups of men were no different in the following characteristics: age, age at first sex, number of lifetime partners, whether they had had sex with a prostitute, whether they had used a condom with their regular partner, the frequency of condom use with prostitutes and their current partner's number of lifetime partners.

Of the 847 uncircumcised men who had valid results in tests for penile HPV infection, 20% were positive for the virus, whereas only 6% of the 292 circumcised men with valid test results had the infection. The presence of HPV was consistently less prevalent among circumcised men than among uncircumcised men when the two groups were stratified according to the various characteristics. After the analysts made adjustments for potentially confounding factors (including study location), logistic regression analysis showed that the odds of penile HPV infection for circumcised men were about 60% lower than those for uncircumcised men, whether the circumcision status was self-reported or medically confirmed (odds ratio, 0.4 for each). Furthermore, men who had had six or more lifetime partners were more likely than those who had had five or fewer partners to have a diagnosis of penile HPV infection (odds ratio, 2.0).

Analysis of the effect of circumcision status on the prevalence of cervical cancer suggested that circumcision may reduce the likelihood of cervical cancer in the men's current partners (odds ratio, 0.7); however, this result was not statistically significant. Circumcision status did not affect the odds that women had cervical cancer, regardless of women's age, age at first sex, educational level, number of lifetime partners or condom use. When the researchers limited their analysis to the 1,420 monogamous women, they found that the odds of having cervical cancer were reduced by about 60% among women with a partner who reported six or more lifetime sexual partners and had been circumcised (0.4).

The investigators also studied the effect of circumcision on the risk of cervical cancer according to males' level of risky sexual behavior. They classified males who reported six or more lifetime partners and an age at first sex of below 17 as having a high risk, those reporting five or fewer partners and an age at first sex of at least 17 as having a low risk, and the remainder as having an intermediate risk. Circumcision of low-risk men did not reduce women's risk of cervical cancer (odds ratio, 1.6). However, circumcision was linked to reductions in the risk of cervical cancer as the men's sexual behavior got riskier (odds ratios for women whose circumcised partners had intermediate and high risk were 0.5 and 0.2, respectively).

The authors of an accompanying editorial point out that the strengths of this study are its size, the HPV detection method used (polymerase chain reaction assay) and the generalizability of the findings because of the multiple study locations.² They note, however, that some confounding factors are difficult to measure accurately and control for, such as frequency of genital washing and genital hygiene. Furthermore, although the investigators focused on monogamous women, they cannot exclude the possibility that women with cervical cancer had been infected with HPV by an earlier, unreported partner; hence, the association between circumcision and the risk of cervical cancer may have been underestimated.

The researchers suggest that circumcision may reduce the risk of acquiring and transmitting HPV and hence the risk of cervical cancer. They propose that circumcision reduces the vulnerability of the penis to HPV infection: In uncircumcised men, the inner surface of the foreskin offers a portal of entry for HPV when it is exposed, by way of tiny ulcers and abrasions that occur during intercourse. The authors of the editorial emphasize that circumcision itself does not protect against cervical cancer: The protective effect relates only to a reduction in the likelihood of genital infection with oncogenic HPV in men. Regarding other potential health benefits of circumcision, the investigators recommend that "further study is needed to determine whether routine circumcision can reduce the risks of HIV and HPV infections and other sexually transmitted diseases."--T. Lane

REFERENCES

1. Castellsagué X et al., Male circumcision, penile human papillomavirus infection, and cervical cancer in female partners, New England Journal of Medicine, 2002, 346(15):1105-1112.

2. Adami H-O and Trichopoulos D, Cervical cancer and the elusive male factor, editorial, New England Journal of Medicine, 2002, 346(15):1160-1161.