Barnett S. Kramer, MD, MPH
Associate Director for Disease Prevention
National Institutes of Health
Editor in Chief
Journal of the National Cancer Institute
Bethesda, Maryland
Alfred I. Neugut, MD, PhD
Professor of Medicine and Epidemiology
Head of Cancer Prevention and Control
Herbert Irving Comprehensive Cancer Center
Columbia University College of Physicians and Surgeons and Mailman School of Public Health
Co-Director
Cancer Prevention Program
Columbia Weill Cornell Cancer Centers
NewYork-Presbyterian Hospital
New York, New York
In 1985, prostate specific antigen (PSA) was licensed for the follow-up of patients with prostate cancer. Shortly after it was shown in a cross-sectional study to be capable of detecting early stage prostate cancer, it was rapidly adopted as a means of screening asymptomatic men over age 50 for prostate cancer. The immediate effect of this was to nearly triple the number of prostate cancers diagnosed annually in the US; even now, the number of prostate cancers diagnosed annually is more than double the 1984 figure. The potential for lowering the number of prostate cancer deaths annually (approximately 40,000 in the US) was, of course, highly attractive, and the number of radical prostatectomies performed in the US skyrocketed. More recently, radiation therapy is also used as an alternative to radical prostatectomy (RP), particularly in older men.
What impact has this actually had on prostate cancer mortality, the most important indicator of the efficacy of prostate cancer screening? It is remarkable to realize that recent data indicate that about two-thirds of older American men have had at least one PSA. Thus, such a substantial adoption of PSA screening, if efficacious, should have a substantial impact on mortality.
If one looks at the number of prostate cancer deaths that occur annually, it can be seen that they have fallen to approximately 30,000 at the present time. Some argue that this represents an effect of PSA screening. However, other changes have occurred in our approach to prostate cancer in the past 15 to 20 years--notably the introduction of more effective and more aggressive use of hormonal therapy for those with the disease-that may help to account for this reduction.
In fact, several randomized controlled trials have shown that adjuvant hormonal therapy, when added to local therapy for locally advanced prostate cancer, decreases prostate cancer mortality. Thus, whether the fall in annual prostate cancer deaths represents a true increase in cure rate (contributed to and/or caused by prostate cancer screening) or whether it represents primarily an increase in survival (in essence a treatment effect), leading to both a delayed death from prostate cancer and the opportunity for competing causes of mortality to intervene, remains an important question.
In the longer term, this question is likely to be resolved by two ongoing large-scale randomized trials of PSA screening for prostate cancer, the PLCO trial (Prostate, Lung, Colorectal, Ovarian), conducted by the National Cancer Institute (NCI) in the US and the ERSPC (European Randomized Study of Prostate Cancer) trial in the European Union. However, the results of these trials will not be available for at least another 5 years, and the issue of PSA screening remains a controversial one at the present time.
The question asked by many people, professional and lay alike, is why PSA screening would not work. It has obviously dramatically increased the yield of prostate cancer and down-staged those prostate cancers diagnosed. Lower stage prostate cancers do better with treatment, be it surgical, radiation, or hormonal, or even watchful waiting, as compared to those with late stage prostate cancer. Why is it such a difficult question to answer? Part of the answer stems from the high prevalence in older men of latent prostate cancer, cancers that are presumed to be relatively nonaggressive and indolent, and that are thought to progress infrequently to clinically significant and mortal prostate cancer. The increased yield observed with PSA screening presumably reflects the increased diagnosis of these types of prostate cancer, but with no necessarily concomitant improvement in mortality, and we already know that any efficacy of the treatments comes at a substantial cost of long-lasting morbidity, such as urinary incontinence and sexual impotence.
One critical aspect of this relates to the actual efficacy of RP in curing prostate cancer. Until recently, there have been very few sound data to support the benefits of radical prostatectomy in the management of early disease as compared to watchful waiting. A recent study by Holmberg and colleagues now helps to remedy this problem. In this randomized trial, conducted in Scandinavia, the investigators randomized 695 men with early stage, newly diagnosed prostate cancer (T1b, T1c, or T2) to either RP or watchful waiting. With a median follow-up of 6.2 years and follow-up out to 8 years, prostate cancer death occurred in 8.9% of the watchful waiting arm and in 4.6% of the RP arm (hazard ratio 0.50, 95% CI 0.27-0.91). Overall, the hazard ratio for overall mortality was 0.83 (0.57-1.2). Of note is that the difference in prostate cancer-specific mortality between the two arms began widening after a follow-up of about 5 years and the separation was still increasing up to 8 years. Furthermore, an accompanying article by Steineck and colleagues addressed the question of quality of life in the two arms of the study, and found much higher rates of erectile dysfunction and urinary leakage in the RP arm as compared to the watchful waiting arm. Overall quality of life was similar in the two management strategies.
How does this study affect thinking about the issue of prostate cancer screening? On one level, it does indeed confirm that there is a benefit in prostate cancer-specific mortality, albeit small, in early stage disease. However, several caveats are important. As in other circumstances, while the difference is statistically significant, in absolute terms, it is modest, and the benefits with longer follow-up are still being assessed. Furthermore, classification of cause of death is often subjective, and there is not a statistically significant difference out to 8 years in the more objective all-cause death rate. Some data in the paper suggest that the incidence of metastases in the two arms also showed a benefit for the RP arm; furthermore, there may be increasing benefits for the RP arm as time progresses. It does, however, suggest that the benefits do not really become substantial for five years, 10 years, or even more, thus making it difficult to observe an overall impact on mortality, particularly given the older age range of most individuals with prostate cancer.
Another consideration is that the absolute benefits are fairly small. For the differences observed, particularly given the high risk of permanent complications, RP may not be the ultimate answer or most desirable alternative for most men. Whether radiation therapy for many of these men would represent an equally effective and "less toxic" alternative remains a question.
Finally, the subjects for this study were recruited prior to the widespread use of PSA screening. Thus, the vast majority represents individuals with symptomatic or palpable tumors, a higher risk group than those with PSA-detected prostate cancer. If one adds a lead time of perhaps 5 to 10 years, which would be anticipated with PSA-detected prostate cancers, one can certainly anticipate the necessity to follow the treatment groups for a much longer time frame to observe this benefit. Thus, it may not be as useful a screening test in older individuals with shorter life spans. One study is currently in progress, the Prostate Intervention Versus Observation Trial (PIVOT), which evaluates RP versus watchful waiting in a primarily screen-detected group, and should provide some added information in this regard.
For the most part, the efficacy of prostate cancer screening, while still promising, must remain somewhat controversial pending the results of the two large randomized trials in progress. Their results will provide a welcome addition to medical knowledge for older men facing this decision.
References
Cancer Facts & Figures. American Cancer Society. Atlanta GA. 2003.
Holmberg L, Bill-Axelson A, Helgesen F, et al. A randomized trial comparing radical prostatectomy with watchful waiting in early prostate cancer.
N Eng J Med. 2002;347(11):781-789.
Steineck G, Helgesen F, Adolfsson J, et al. Quality of life after radical prostatectomy or watchful waiting.
N Eng J Med. 2002;347(11):790-796.
