BACKGROUND. We aim to establish the normal pattern of prostate specific antigen (PSA) change with age to provide a baseline from which disease progression might be identified in prostate cancer patients included in active surveillance programs. METHODS. In a community-based study, PSA values were determined at baseline and after 2.1 and 4.2 years in men without prostate cancer. A bivariate multilevel growth curve model was used to estimate the pattern of change of PSA with age. RESULTS. The final model showed that PSA was related to age only. The future PSA of an individual can be predicted based on his age and known history of PSA. The model was also used to calculate PSA doubling time for men with different PSA values at different ages. CONCLUSIONS. This method establishes normal PSA levels by age using PSA history in men without prostate cancer. The model provides baseline data from which disease progression might be detected.

, , ,
doi.org/10.1002/pros.20293, hdl.handle.net/1765/71325
The Prostate
Department of Urology

Bosch, R., Tilling, K., Bohnen, A., & Donovan, J. (2006). Establishing normal reference ranges for PSA change with age in a population-based study: The Krimpen study. The Prostate, 66(4), 335–343. doi:10.1002/pros.20293