Circumcision and risk for prostate cancer: an unresolved debate

For more than 70 years there has been an ongoing if sporadic debate about whether circumcision (in infancy or later in life) has any impact on risk for prostate cancer … or not. People tend to have strong opinions about this issue — for a multitude of reasons. And their underlying reasons for their mindsets can clearly color those opinions.

The last time this topic came up was in association with a paper published by Wright et al. in 2012. However, a newly published paper by Spence et al. in BJU International (along with an associated media release from the University of Montreal) has once again brought this issue back into the media.

Spence et al. postulate, based on the data they collected in Montreal, that circumcision prior to age 35 is indeed associated with a reduction in risk of prostate cancer, and that (in their study) this was particularly the case among the small subset of patients of historic African ethnicity. The authors further include statements about the perception that prostate cancer occurs with less frequency among men from families of Jewish and Muslim religious persuasion.

The problem, however, is that none of the available data on this topic are truly compelling. We are dealing with “associations” — and those associations come with  varying degrees of statistical significance. The strength of these associations clearly depend on the type of data collected and the rigor with which they are assessed. So, for example, as most readers of this blog will understand, the fact that someone has not been diagnosed with prostate cancer provides no evidence whatsoever about whether they do or don’t actually have cancer in their prostate. All it tells us is that no one has diagnosed such a condition.

The value of the data in many of the studies in this arena actually depend upon the likelihood that a man who is or is not circumcised will subsequently be diagnosed with prostate cancer (anything up to 75 years later) … and that, in and of itself, depends on a whole bunch of other factors. For example, just how many men in Israel and the Palestinian areas of the Middle East actually had a PSA test and were appropriately evaluated for risk of prostate cancer in the past 5 years? Did their doctors carefully note whether those men were circumcised or not; whether they classified themselves as Jewish, Muslim, or of some other religious persuasion; and whether they were actually first-generation immigrants to the area in which they now lived? Do we have complete details about their sexual proclivities and behaviors over the ensuring period from birth to time of circumcision and on to time of diagnosis? (There are probably at least a dozen more truly critical factors that would be relevant to an accurate analysis of any correlation between circumcision and diagnosis with prostate cancer, and many of those critical factors are things that many men wouldn’t be completely honest about … as in answers to questions like, “When did you stop beating your wife?)

Thus, to add context to this debate, we need to understand that:

  • It is quite certainly true that the risk of a diagnosis of prostate cancer today is much higher in men of historic African ethnicity than it is among those of historic Caucasian ethnicity (including men of Caucasian Jewish ethnicity) — but we still have no clear idea of precisely why this is the case.
  • It is not actually clear whether men of Semitic ethnicity really do or don’t have a lower risk for a diagnosis of prostate cancer than men of non-Semitic Caucasian ethnicity.
  • Male members of Ashkenazi Jewish communities are at high risk for certain genetic abnormalities (e.g., BRCA1 and BRCA2 gene mutations) that do increase risk for a diagnosis of prostate cancer.
  • It is not clear at all whether being of the Islamic faith is in any way associated with a reduction in risk for prostate cancer (compared to the risk of prostate cancer among men of any other religious faith).
  • Circumcision quite certainly does seem to be associated with a reduction in risk for diagnosis with certain types of sexually transmitted infection, but … there is no compelling evidence that such infections are associated with an increase in risk for prostate cancer.

The media, of course, find the whole subject of circumcision to be very good for “sales”. So papers like the one by Spence et al. get a great deal of media coverage. And one can hardly blame the University of Montreal’s PR department for wanting to ensure whatever visibility they can obtain for the university.

The truth, unfortunately, is that the available data don’t really help anyone at all. Circumcision of male children is a tradition based on ethnicity, sociocultural mores, and religious beliefs. Before the development of antibiotics and modern medicine, it may well have also had significant medical merit (although there is no actual proof that this was ever the case either). Whether circumcision is a good or a bad thing for a specific individual, and the degree of medical risk associated with the actual procedure as compared to the degree of medical risk associated with not having the procedure, are both unknown. In the vast majority of cases, circumcision is best considered (rightly or wrongly) to be a rite of passage that has far more to do with cultural attitudes than medical evidence: the “patient” only rarely has any control over the issue.

Since the 17th Century, jurors hearing criminal cases in Scotland have had an alternative to the two verdicts of “guilty” or “not guilty” that are used almost exclusively in most legal systems. It is possible for a jury in Scotland to find that a prosecutor’s case against an accused defendant is “not proven” (although this verdict is not used that often today). The “New” Prostate Cancer InfoLink would argue that such a verdict is highly appropriate to the case of whether circumcision does or does not actually protect against risk for prostate cancer. We find the case to be “not proven”. We also suspect that a 70-year-long, randomized clinical trial to try to resolve this question (along with collection of all sorts of associated and highly necessary evidence related to sexual behaviors during the course of the study) is neither possible nor practicable!

2 Responses


    Sitemaster, your review notes that there are a number of other explanations possible for the difference in apparent prostate cancer incidence between Jewish/Muslim and non-Semitic populations. I’m thinking one of them could be the role of pork in the diet. It is suspected that pork, a red meat, is associated with increased prostate cancer risk, and there is some evidence in research (e.g., below). Pork is essentially forbidden for Jewish and Muslim men, but is widely consumed by others.

    Several years ago the Prostate Cancer Research Program (under the Congressionally Directed Medical Research Program) funded research comparing prostate cancer outcomes for African Americans in North Carolina (especially around Wake County, where tobacco use and pork consumption are especially prominent), where the risk of prostate cancer among African American men was unusually high, and Louisiana, where the risk among African American men was unusually low — below that of Caucasians. One suspect was pork — a favorite for all in North Carolina, but not much eaten in Louisiana, where seafood is a prominent element in the diet.

    Results of this North Carolina-Louisiana Prostate Cancer Project were reported at the PCRP’s IMPACT conference on March 10, 2011, in a plenary session at 1:30-2:30 PM, moderated by Thomas Laveist, with presentations by Dr. James Mohler and Dr. Elizabeth Fontham. (I’m providing this detail as there may be a transcript somewhere or an online recording, and follow-up research publications are contained in a number of papers captured in with “mohler j [au] AND fontham e [au]” as authors.) Here is a link to one I just checked that seems especially relevant.

    Another factor that may amplify the risk of pork is cooking at too high a temperature, thereby generating cancer-causing elements.

    I love pork barbecue and pork generally but have been avoiding it almost entirely for a decade and a half now.

  2. Dear Jim:

    Do you listen to jazz?

    My bet would be that I could get a similar associative correlation based on the same data that suggested that men who hear a lot of jazz music, because they live in that area of Louisiana (compared to those who do not, because they live in Wake County, NC) are at a significantly reduced risk for a diagnosis of prostate cancer.

    Do I think that that is a likely explanation of the facts? Of course not, but I bring it to readers’ attention because it points out just how dangerous these sorts of epidemiological association studies can be.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: