For many years, the purported benefits and harms of male circumcision have been debated in the medical literature and society at large, with no clear consensus to date. One of the more commonly postulated but never well-substantiated benefits is protection against sexually transmitted disease. Castellsagué and associates, in concert with the International Agency for Research on Cancer Multicenter Cervical Cancer Study Group, reported on a pooled analysis of seven retrospective, case control studies of cervical cancer from five countries (Spain, Colombia, Brazil, Thailand, and the Philippines).

Women with newly diagnosed cervical cancer were matched with control subjects from the same hospital populations. Circumcision of male partners was verified by a clinician in the earliest studies. These examinations verified that self-reported circumcision status was reliable; therefore, self-reported status was used in the later studies. Only husbands and stable male partners were evaluated in the studies. A stable partner was defined as one having regular intercourse with the female patient for at least six months.

The investigators screened for penile human papillomavirus (HPV) infection by polymerase chain reaction (PCR) assay of HPV DNA from a urethral swab and an external swab of the glans penis. Of the 1,896 women with cervical cancer, 1,329 had a stable partner. Of these partners, 984 were interviewed, and 610 had valid PCR assays. Through interview of the male partner, investigators obtained information about education level, age at first intercourse, lifetime number of sexual partners, and frequency of genital washing after intercourse.

The overall rate of reported circumcision was 19.3 percent. Circumcised men were more likely to have a secondary education or higher and were also more likely to have good genital hygiene on clinical examination compared with uncircumcised men. HPV DNA was detected in 19.6 percent of uncircumcised men and in 5.5 percent of circumcised men. There was no significant decrease overall in the risk of cervical cancer for women with circumcised partners. In the subgroup of men who reported having six or more lifetime sexual partners, there was a significant decrease in cervical cancer rates in their female partner (adjusted odds ratio: 0.42).

The authors conclude that circumcision is associated with a lower rate of penile HPV infection and a reduced risk of cervical cancer in female partners of men with higher numbers of lifetime partners.

editor's note: The sound bite that will hit the evening news from this study—“Circumcision prevents HPV infection and cervical cancer!”—needs so many qualifiers that it has little scientific rigor. The initial association between circumcision and cervical cancer was lessened with each of the confounding variables included in the statistical analysis (e.g., education level, number of partners) and there is no guarantee that other socioeconomic or clinical variables that were not tracked may explain the association entirely. Retrospective, case control studies have led the medical community astray before with findings that were not borne out in prospective, controlled trials (which would obviously not be feasible in this situation). The lack of a significant association overall between circumcision and cervical cancer is a less dramatic result than the 58 percent reduction in cancer for the high-risk subgroup, but it is likely the more reliable take-home point.—b.z.