Editorial

Abstract

Early detection of prostate cancer is one of the great controversies in medicine. In this issue of the Journal, Fowler (1) and Meigs (2) and their colleagues demonstrate that prostate-specific antigen (PSA) testing for the purpose of early detection of prostate cancer is performed as part of routine health maintenance by most physicians. PSA testing in the preceding year was reported by more than half of the men older than 70 years of age. Rates were highest in men with urinary symptoms consistent with benign prostatic hyperplasia (BPH). Additionally, the majority of primary care physicians and urologists believe that surgery and radiation improve survival compared with watchful waiting in men with at least a 10-year life expectancy. Are these patterns and beliefs about good clinical practice substantiated by evidence? The rationale for screening and early treatment is that untreated prostate cancer may grow locally, metastasize, or cause death, and that early detection and appropriate treatment will delay or prevent these events. These assumptions, while plausible, are unproved. The disparity between incidence and mortality indicates that many more men die with prostate cancer than from it. For a 50-year-old man with a life expectancy of 25 years, the lifetime risk of microscopic prostate cancer is approximately 42%, whereas the risk of clinically evident prostate cancer is only 10%, and that of fatal prostate cancer is only 3% (3). And in the minority of men in whom early intervention would be necessary, there is accumulating evidence that surgery or radiation may not improve survival. About one out of four asymptomatic men will have an abnormal PSA measurement or rectal examination. Two thirds of these tests are false-positive results. Although often considered part of health maintenance, routine digital rectal examinations do not reduce prostate cancer mortality (4). PSA testing detects a reservoir of latent asymptomatic disease that also may not benefit from, and might be harmed by, surgery or radiation (4 – 6). In addition, 20% of men with known prostate cancer have normal PSA levels. While PSA testing is common in men in their 70s and 80s, these men generally have life expectancies of less than 10 years and are unlikely to benefit from testing. In men with symptomatic BPH, the prevalence of prostate cancer is no greater than in ageand racematched asymptomatic men (4). Because PSA levels increase with prostate enlargement, as many as 50% of men with BPH will have elevated PSA levels. Routine PSA testing in men with symptomatic BPH results in more biopsies, complications, costs, and anxiety. For example, about 400 men out of 1,000 with moderate symptoms of BPH will have an abnormal PSA level or rectal exam. Of these 400 men, only 30 will have prostate cancer. Because initial testing will find prevalent cancers, detection rates and predictive values in subsequent years will be lower. The other 370 men have undergone unnecessary diagnostic procedures including prostate ultrasound and biopsy. The risks of these tests are relatively minor but bleeding or infection occur in as many as 40%. Modifications to PSA testing, such as measurement of “free and complexed PSA,” slightly improve specificity. But they decrease sensitivity, require additional blood tests, and have not been demonstrated to improve survival. It is not known if surgery or radiation improves survival or quality of life in men found to have cancer by PSA testing. Because the lead time associated with PSA testing is at least 5 years, outcomes after 15 to 20 years of follow-up for PSA-detected cancers are likely to be similar to the 10-year results reported in men with palpable tumors. The long-term outcomes in men with palpable, clinically localized prostate cancer treated with observation and delayed palliative hormone therapy (watchful waiting) are excellent and approach that of age-matched controls (7– 10). Prostate cancer specific survival in men treated with watchful waiting is approximately 90% after 10 years of follow-up; surgery and radiation are unlikely to substantially improve this rate. The chance that men will remain free of symptomatic progression requiring palliative treatment is 70% at 5 years and 40% at 10 years. In men Am J Med. 1998;104:602– 604. From the Department of Medicine, Minneapolis Veterans Affairs Medical Center, Veterans Affairs Coordinating Center of the Cochrane Review Group in Prostate Diseases and Urologic Malignancies, Minneapolis VAMC/VISN 13 Center for Chronic Disease Outcomes Research, Minneapolis, Minnesota. Supported in part by the Department of Veterans Affairs Cooperative Studies Program and the Health Services Research and Development Program, the National Cancer Institute, and the Agency for Health Care Policy and Research. Requests for reprints should be addressed to Timothy J. Wilt, MD, MPH, Section of General Internal Medicine, Minneapolis VA Medical Center (111-0), 1 Veterans Drive, Minneapolis, Minnesota 55417. Manuscript submitted March 12, 1998 and accepted March 17, 1998.