Does PSA screening offer sufficient benefit?

Many if not all physicians have seen the downside of cancer screening — the anxiety related to possible cancer, complications from the confirmatory tests required to follow up on an abnormal screening test. I firmly believe that more screening/tests/labs are not necessarily better and may, in fact, be worse. That is also what a recent study on prostate cancer screening found (abstract here).

Investigators were able to determine — again using the SEER database since 1986 when PSA screening emerged — the frequency at which we over-diagnose prostate cancer using PSA screening. The authors found that 20 men with an elevated PSA had to be treated to benefit one man with prostate cancer, which would have limited or shortened his life. The senior author Dr H. Gilbert Welch is quoted in the Forbes article, saying, "In general, the benefits of screenings have been vastly oversold and the harms have been totally ignored ... We have sold this idea that it is always better to find disease early — and that is just plain not true."

The accompanying editorial is a bit more mild, stating that the benefit of PSA screening is up for discussion, which is where I think I also fall on this issue. I think it is clear that we are over-diagnosing and committing many men to surgery, radiation or hormonal therapy they likely do not need. But the question is, can we do better? I had a recent blog entry about PSA velocity being unhelpful, for example. Clearly, widespread screening cannot be recommended, but I do think the current recommendations to have an involved conversation with one's primary care provider about the risks and benefits of PSA screening before committing to the test, and limiting the testing to patients with a reasonable life expectancy to benefit from early detection, make the most sense. For all of PSA's foibles, I'm not sure it's right to not use it at all.