To Örebro University

Background: Most localized prostate cancers are believed to have an indolent course. Within 15 yr of diagnosis, most deaths among men with prostate cancer (PCa) can be attributed to other competing causes. However, data from studies with extended follow-up are insufficient to determine appropriate treatment for men with localized disease.

Objective: To investigate the long-term natural history of untreated, early-stage PCa.

Design, setting, and participants: We conducted a population-based, prospective-cohort study using a consecutive sample of 223 patients with untreated, localized PCa from a regionally well-defined catchment area in central Sweden. All subjects were initially managed with observation. Androgen deprivation therapy was administered when symptomatic tumor progression occurred.

Outcome measurements and statistical analysis: Based on >30 yr of follow-up, the main outcome measures were: progression-free, cause-specific, and overall survival, and rates of progression and mortality per 1000 person-years.

Results and limitations: After 32 yr of follow-up, all but 3 (1%) of the 223 men had died. We observed 90 (41.4%) local progression events and 41 (18.4%) cases of progression to distant metastasis. In total, 38 (17%) men died of PCa. Cause-specific survival decreased between 15 and 20 yr, but stabilized with further follow-up. All nine men with Gleason grade 8-10 disease died within the first 10 yr of follow-up, five (55%) from PCa. Survival for men with well-differentiated, nonpalpable tumors declined slowly through 20 yr, and more rapidly between 20 and 25 yr (from 75.2% [95% confidence interval, 48.4-89.3] to 25% [95% confidence interval, 22.0-72.5]). It is unclear whether these data are relevant for tumors detected by elevated prostate-specific antigen levels.

Conclusions: Although localized PCa most often has an indolent course, local progression and distant metastasis can develop over the long term, even among patients considered low risk at diagnosis.