Screening for prostate cancer and its effect on mortality: The Tyrol study

In the early 90s, a remarkable increase in the incidence of prostate cancer in many countries, particularly in the United States,1 was observed. This observation can be attributed to the widespread use of prostate-specific antigen (PSA), which was first approved for the detection of recurrent disease in patients with established prostate cancer in 1986. Thereafter, the potential of this test for early diagnosis of prostate cancer was soon recognized. From 1984 until 1994, PSA was increasingly used for diagnostic purposes. In 1984, PSA testing was used in 5.1% and in 1994 in 60.6% of all newly diagnosed prostate carcinomas.2 It has been shown that a great number of cancers detected by PSA testing are clinically significant and potentially curable.3-6 However, the introduction of PSA testing in prostate cancer screening programs has also led to controversy surrounding several distinct issues, including the sensitivity and specificity of the screening test; treatment of early prostate cancer and, indeed, whether some cancers will do equally well if left untreated; and the side effects of therapy, particularly radical prostatectomy. The two most common cancer screening programs-Papanicolaou smears for cervical cancer and mammographic examination for breast cancer-came into common use and acceptance through widely different mechanisms: the results of randomized trials of mammographic screening for breast cancer, and the observation of the decrease in incidence and mortality from cervical cancer after the policy to introduce cervical cancer screening to populations. The present study reports the incidence and mortality rates of prostate cancer in the Federal State of Tyrol, Austria, where regular PSA testing has been made freely available to the population since 1993, and where use of the test has been high. This population is also characterized by being particularly stable. Prostate-specific antigen testing was not freely available in the rest of Austria, although it will have been used, probably evolving in a similar manner to the use in many Western countries. A comparison of the mortality rates between Tyrol and the rest of Austria allowed evaluation of the outcome of this natural experiment.