Letters

Neonatal circumcision and penile cancer

EDITOR,--As chairman of the American Academy of Pediatrics' Task Force on Circumcision¹ and a reviewer on the topic for the American Cancer Society,² I am amazed at the claim by Paul M Fleiss and Frederick Hodges that no link exists between circumcision and penile cancer.³ In an incredible example of selective medical amnesia the authors ignore evidence that has accumulated in the past 64 years. They discredit Wolbarst's study in 1932 without mentioning the data.⁴ Wolbarst reviewed 1103 cases of penile cancer in the United States and found that all occurred in uncircumcised men and none in Jewish men, even though 33 cases would have been expected because Jewish men constituted 3% of the population. Wolbarst cited confirming figures from Europe and noted that Moslems, who also circumcise boys, are protected against cancer. This landmark study served as the stimulus for five future series that were published in refereed journals in the next 50 years² but are apparently unknown to Fleiss and Hodges (although Hodges, a known lay anticircumcisionist, says that he is a medical historian).

In a classic review in 1935 Dean analysed 120 cases of penile cancer reviewed at what is now Memorial Sloan-Kettering Cancer Center in New York City.⁵ All of these cases occurred in uncircumcised men. None occurred in Jewish men, although more than a third of other inpatients with cancer at the facility were Jewish. Dean concluded that "prophylactic treatment of cancer of the penis consists in circumcising all male infants." In four later, separate series 139 cases in Illinois were reviewed in 1946; 100 in Roswell Park, New York, in 1972; 156 in Michigan in 1973; and 77 in Cleveland in 1986. Of these 592 cases at five institutions renowned for treating cancer, none occurred in men who had been circumcised as newborn infants, although most newborn male infants in the United States were circumcised.

Since 1935, about 50 000 cases of penile cancer have been reported in the United States (annually, about 750-1000 cases and 200 deaths), only 10 of which occurred in men who had been circumcised as newborn infants. This yields a ratio of 5000:1 for the incidence of penile cancer in uncircumcised to circumcised men--overwhelming evidence.

As Kochen and McCurdy pointed out, the low incidence of penile cancer in the United States is misleading because it has been calculated by combining circumcised men (in whom the incidence is essentially zero) with uncircumcised men (in whom it is 2.2/100 000).⁶

Director Regional Perinatal Screening Program, Kaiser Permanente Medical Care Program, 3900 Broadway, Oakland, CA 94611, USA

Edgar J Schoen 

1. Schoen EJ, Anderson G, Bohon C, Hinman F Jr, Poland RL, Wakeman EM, American Academy of Pediatrics. Report of the task force on circumcision [published erratum appears in Pediatrics 1989;84:761.] Pediatrics 1989;84:388-91.
2. Schoen EJ. The relationship between circumcision and cancer of the penis. CA Cancer J Clin 1991;41:306-9. [Medline]
3. Fleiss PM, Hodges F. Neonatal circumcision does not protect against penile cancer. BMJ 1996:312:779-80. (23 March.)
4. Wolbarst AL. Circumcision and penile cancer. Lancet 1932;i:150-3.
5. Dean AL Jr. Epithelioma of the penis. J Urol 1935;33:252-83.
6. Kochen M, McCurdy S. Circumcision and the risk of cancer of the penis: a life-table analysis. Am J Dis Child 1980;134:484-6. [Abstract]