SEATTLE: Another flaw in prostate study - undetected cancer in controls

Medscape Urology (with video)
March 28, 2012

By Dr. Gerald Chodak

The latest headline appearing on the wire services reports an association between circumcision and the risk for developing or not developing prostate cancer in the future. This is based on a study published in Cancer, in which Wright and colleagues[1] looked at 2 case/control populations of about 1700 men.

Each of the 2 studies reported on whether or not the study cohort had been circumcised, their ages at first intercourse, and whether they had sexually transmitted diseases. The results showed about a 15% lower incidence of prostate cancer in the men who had undergone circumcision before beginning sexual activity. [And the statistical significance of the claim is quite marginal.]

These findings, and the headlines, have stimulated a lot of discussion about whether or not this lends support for circumcision and whether or not we should use it as a basis for recommending circumcision for newly born boys.

First, we need to realize that this is not the type of study that proves cause and effect, which means that the association may or may not be valid. Other important criticisms about the study should be mentioned. For example, the fact that men in the control group had not been diagnosed with prostate cancer does not mean that they do not have prostate cancer. It is unclear how many of them had undergone screening or prostate biopsies to determine whether microscopic disease was present. [Since they were obtained by a random telephone survey, probably very few.]

We know from autopsy studies that about 30% of the men may indeed have undetected prostate cancer. If that were the case here, it would certainly have an effect on the real difference in prostate cancer incidence in circumcised vs uncircumcised men.

At the end of the day, we have a lot of hype and publicity about a study that really does not prove whether or not circumcision reduces the likelihood of developing prostate cancer. There are other ongoing debates about the merits or risks associated with circumcision in newborn children that will not be resolved by the presence of these new data.

For now, all we can say is that this study found an interesting association, but the flaws in the study mean that the results are very questionable. We need better evidence before we can truly say that circumcision prevents prostate cancer. This study does not make that proof, nor does the authors' claim that it is proof of a cause and effect.

Earlier story