THE PSA TEST: YOUR BEST FRIEND OR HOMER’S CIRCE?
The “Rudy & Torre push (for) prostate tests” by Susan Ferraro of the New York Daily News (14 June, 2002) is a well intentioned step forward in the battle against prostate cancer. Unfortunately, Messrs. Giuliani and Torre may have been misinformed. Early detection of prostate cancer by the PSA test has not been proven to increase survival. In fact, no randomized clinical trials have been completed that definitively establish the efficacy of radical prostatectomy or radiation therapy for the treatment of localized prostate cancer. Improved survival associated with PSA screening is primarily due to earlier detection of asymptomatic prostate cancer, with unchanged time of death (known as lead time bias).
Given the availability of definitive treatments for prostate cancer, the benefits of early detection and value of screening are intuitively appealing ideas. However, such has not been documented to be the case and the medical community may be promising more than they can deliver. By way of example of the potential harm of widespread PSA screening, a prospective analysis (Marshall. JAMA, 274:607, 1995) of the evaluation of PSA detection of prostate cancer from one of an infinite number of studies (Gann et al. JAMA, 273:289, 1995) demonstrated a positive predictive value of 13%, i.e., a positive PSA test in a man with prostate cancer. This meant that 87% of men who had a positive PSA test would not have cancer, but would be subjected to the psychological and physical trauma associated with investigation of their elevated level of PSA. Of those patients with localized prostate cancer, who elect to have a radical prostatectomy, 25-35% will have a recurrence within five years, while coping with incontinence (up to 30%) and impotence (up to 60%).
It is ironic with all that has been written about how big a problem prostate cancer is, nowhere do we find these sobering facts. For instance, while a 50-year old man has an approximately 40% chance of developing asymptomatic (non-clinically significant) prostate cancer within the next 25 years, the chance that he will experience any clinical symptoms and be diagnosed with prostate cancer is 10% and the chance that he will die from prostate cancer is 3%. While a 50-year old man’s chances of dying from non-prostate cancer are 20%, his chances of dying from cardiovascular disease are 50%. How big a problem is prostate cancer?
In the end, heightened fear brought forth by the introduction and aggressive marketing of a non specific screening test and media attention focused at high profile celebrity cases has misled the public into believing that the PSA test has only a positive impact and that prostate cancer more often than not is an aggressive, fatal disease. And how many men are aware that failed treatment of early cancers has converted slow growing to more threatening tumours (Wheeler et al. Cancer, 71:3783, 1993)?
Given the foregoing facts and in keeping with the first ethical principle of medicine – primum non nocere (first do no harm), in some cases maybe the best treatment is no treatment.
Richard J. Ablin, Ph.D. Robert Benjamin Ablin Foundation for Cancer Research Mahwah, NJ 07430