Without question, PSA screening for prostate cancer in asymptomatic patients does them both harm and good; the difficulty in quantifying how much harm versus good has historically been the source of disagreements among primary care physicians and urologists over how much screening we ought to be doing, or if we should be screening men at all. In 2012, the U.S. Preventive Services Task Force took the position (which it partially reversed in 2018) that the way to prevent harm from PSA screening was to generally stop doing it. But those who believe that selective testing saves lives that otherwise would have been lost to prostate cancer argue that too much testing isn't the problem, it's too much treatment. For every potentially fatal tumor identified by PSA testing, we also "overdiagnose" numerous low-grade, indolent prostate cancers that should perhaps not be called "cancer" at all but are nonetheless treated or at least monitored, exposing patients to harm with very little likelihood of benefit.
It's instructive to compare the typical evaluation for a positive prostate cancer screening test with a positive breast cancer screening test. If breast surgeons diagnosed breast cancer the way urologists diagnose prostate cancer, they would not only biopsy the area of the breast corresponding to an abnormality on a mammogram or ultrasound, they would also systematically biopsy 12 to 20 additional normal-appearing areas of the breast to make sure that no cancer is missed. If that sounds crazy, that's because it is. Multiparametric MRI is increasingly being used for targeted prostate biopsy and active surveillance of low-risk prostate cancer, but whether MRI-targeted biopsy can safely replace systematic prostate biopsy remains an unanswered question.
Unanswered, that is, until last week, when the New England Journal of Medicine published the results of a randomized trial comparing MRI-targeted versus systematic biopsy in men with a PSA level of 3 ng/mL or higher. The researchers found that men in the systematic biopsy group were twice as likely as those in the MRI-targeted group to be diagnosed with an "insignificant" cancer (as judged by pathologists) but slightly less likely to be diagnosed with a clinically significant cancer. In other words, the cost of reducing prostate cancer overdiagnosis is that a small number of clinically significant cancers that would only have been diagnosed with systematic biopsy are missed and not caught until later in the disease course. Granted, pathology does not correlate perfectly with tumor behavior, and it may not predict clinical prognosis since many men have comorbid medical conditions that are more likely to cause death than prostate cancer. But I think these findings make sense; whether and how they will affect academic or community urology practices remains to be seen. As a family physician, would I feel more comfortable with doing PSA screening if I knew that our local urologists performed MRI-targeted rather than systematic prostate biopsies? Absolutely.