Screening serum levels of prostate-specific antigen (PSA) is now a major strategy for early detection of prostate cancer (PC). Quantification of the lead time thus obtained is important both for understanding the development of PC and for evaluating the advantages and disadvantages of widespread screening. In our study, 1,233 randomly selected men living in Stockholm in 1988 were invited to participate in an early detection (ED) program, in which suspicious findings provided by digital rectal examination (DRE), transrectal ultrasonography (TRUS) and/or a PSA value >/=10.0 ng/mL were followed up by biopsy. The cumulative incidence (Kaplan-Meier) of PC in the 946 participants (ED) during 12 years of follow-up was compared to that of an age-matched, randomly selected reference population (RP) of 657 men for whom PSA values (from frozen serum samples) could also be obtained. The PC incidence in men in the RP with PSA values >/=3.0 ng/mL reached the corresponding level for the ED group after 10.6 years (the "catch-up" point). After 12 years of follow-up, the estimated median lead time for men with PSA values in this interval was 4.5 years in the ED population, compared to 7.8 years in the RP. With 20 years of follow-up, the estimated median lead time of the RP was enhanced to 10.7 years. The lead time in connection with PC was influenced by the initial PSA level (although with large variations), length of follow-up and sensitivity of the ED procedure employed. The ED program described here was not associated with major overdetection.
Copyright 2003 Wiley-Liss, Inc.