How to use PCa prevalence data to guide chemoprevention

Patients need to understand that, based on their age, prevalence of prostate cancer can be predicted.

Data from the Prostate Cancer Prevention Trial (PCPT) have demonstrated that if men over the age of 55 years are biopsied, despite having a normal digital rectal exam and a PSA of less than 4.0 ng/mL, nearly a quarter will be found to have prostate cancer (N Engl J Med 2003; 349:215-24). It is not surprising that patients seen in the office increasingly request information on how they may prevent the development of prostate cancer.

This article discusses the impact of the prevalence of prostate cancer on chemoprevention, and provides guidance when patients ask about their likelihood of being diagnosed with the disease and their ability to prevent it.

"Prevention" is a term used to describe intervention to eliminate the cancer prior to its inception; "treatment" is the management of established cancer. Prevention may include changes in lifestyle, diet, and the use of natural or synthetic substances (chemoprevention) to reduce the risk of development of the malignancy.

Several ongoing phase I, II, and III trials are examining a variety of substances thought to be useful in the prevention of prostate cancer, including the 5-alpha-reductase inhibitors finasteride (Proscar) and dutasteride (Avodart), which are being studied in the PCPT, and the Reduction by Dutasteride in Prostate Cancer Events (REDUCE) trials, respectively; vitamin E and/or selenium, under study in the Selenium and Vitamin E Cancer Prevention Trial (SELECT); lycopene; soybean; beta-carotene; and others.

Most of these trials enroll men of middle age or older with normal DREs and normal PSA levels with the presumption that they do not have prostate cancer at enrollment. This may be erroneous study design because 25% of the PCPT placebo group who had normal PSA and who were diagnosed with prostate cancer were unlikely to develop their cancer solely during the trial period. PCPT, which was powered to find 6% of cancers in the control population, found cancer incidence to be four times higher. This occurred because the design of the trial did not take into consideration the prevalence of prostate cancer in the targeted population.

Incidence and prevalence

Simply stated, the incidence of prostate cancer is the number of cases diagnosed in a given time period. Prevalence is all the cases that exist within that population, diagnosed or occult. While good data exist regarding prostate cancer incidence according to age, race, geographic region, and other parameters, data on the true prevalence of the disease are much harder to come by.

The autopsy incidence of pathologic findings has always been a convenient way to identify disease prevalence. The notion that all men will eventually develop prostate cancer if they live long enough arose from the work of Franks (J Pathol Bacteriol 1954; 68:603-16). Franks studied the prostate glands of dead men of various ages and found that 100% of the glands from those over the age of 90 years contained cancer. Several other studies noted that, even when the glands of men under the age of 50 years were examined, a surprisingly high percentage already showed early signs of carcinoma (J Natl Cancer Inst 1980; 65:311-6).

These anecdotal reports in the literature prompted our group to look at this issue in a systematic fashion with particular attention directed at early lesions in Caucasian and African-American men.