The true test of a screening program rests on the clinical outcomes of patients undergoing screening. Numerous publications have focused on the relative merits of various parameters, such as prostate specific antigen (PSA) density, PSA velocity and free:bound ration in order to identify men with clinically significant prostate cancer. Unfortunately, these articles are irrelevant if screening for PSA does not result in a significant improvement in men’s health. Some clinicians justify screening for PSA solely on the basis that the test can detect organ-confined disease. Unfortunately, this evidence is insufficient. Screening cannot be justified unless patients who are screened have better health outcomes than those who are not. Clinical trials and epidemiological studies support the efficacy of breast, cervical and colorectal cancer screening. Unfortunately, similar evidence is lacking for prostate cancer screening. In fact, observational studies of screening by digital rectal examination alone reported no benefit.

The significance of stage shift

Indirect evidence suggests the possibility that screening with PSA may be beneficial. Before 1986, the diagnosis of localized prostate cancer accounted for the increase in incident cases noted during the two previous decades. Only modest increases in regional and distant-stage disease were detected during this period. After 1986, the stage-specific incidence began to increase exponentially for all stages within the United States, with exception of distant-stage disease. From 1986 (the beginning of the PSA era) to 1991, the incidence of localized disease increased 75%, while the incidence of regional and unstaged disease rose 144% and 161%, respectively. The incidence of distant-stage disease remained essentially unchanged.

Since 1991, the age-adjusted incident rates for distant-stage disease have fallen dramatically, so that they are now approximately half the value they were at the start of the decade. Decreasing rates of distant-stage disease most probably reflect the widespread use of PSA testing. While the decreasing rates of distant-stage disease are a significant indicator that early detection may subsequently lead to a decrease in prostate cancer mortality, this fact alone is not sufficient to demonstrate the efficacy of aggressive prostate and treatment.

Lessons from medical history

Screening for disease can produce significant stage shifts and still not result in significant reductions in mortality rates. In the 1960s, for example, a number of organizations within the United States advocated lung cancer screening because data from case series suggested a shift in stage and an increase in patient survival. Considerable controversy surrounded the conduct of clinical trials until these trials demonstrated the lack of efficacy of the screening actually increased mortality, especially in populations with significant incidence of disease.

The recent Japanese and Canadian experiences with neuroblastoma screening are also an example of a cancer screening test yielding a favorable stage shift and increased survival without decreasing morality. Studies of screening for catecholamine metabolites have led to the concept that there are two kinds of neuroblastoma: a benign form that will never progress and an aggressive, usually fatal form that is unlikely to be found at a curable stage with screening. Neuroblastoma screening has caused a significant number of children to undergo unnecessary treatment.