Objectives: To investigate overall survival and prostate cancer-specific mortality in men with prostate cancer presenting with a PSA level <100 ng/mL at the time of diagnosis.
Patients: Five-thousand seven hundred and sixteen patients with prostate cancer and a recorded diagnostic PSA level extracted from the South Australian Prostate Cancer Clinical Outcomes Collaborative (SA-PCCOC) database. Men included were diagnosed between January 1998 and August 2013.
Methods: Patients were divided into groups according to diagnostic PSA level: <20, 20-≤100, 100-≤200 ng/mL, 200-≤500 ng/mL, and >500 ng/mL. Outcomes measured include overall survival and prostate cancer-specific mortality. Clinical stage, Gleason score and the presence of bony metastasis was evaluated to determine if they were prognostic factors in patients with PSA over 100 at diagnosis. Cox proportional hazards and competing risks regression were used to model overall survival and prostate cancer-specific mortality outcomes respectively.
Results: Of this cohort, 241 patients (4.2%) had a diagnostic PSA level >100 ng/mL. Patients with PSA >100 ng/mL have a significant reduction in five (29.1% vs 62.5% vs 87%) and ten-year (18.2% vs 36.7% vs 70.7%) overall survival when compared to men with diagnostic PSA 20-100 and <20 ng/mL respectively. In this group, prostate cancer-specific mortality was associated with Gleason score and metastases, but not PSA level at diagnosis. Overall survival was associated with PSA level, Gleason score and age. There was a linear increase in risk (overall survival) as PSA increased until 200 and no association thereafter. Models of overall survival and prostate cancer-specific mortality incorporating a risk stratification developed by Izumi et al. predicted overall survival but not prostate cancer-specific mortality. The use of this stratification did not improve model accuracy.
Conclusion: Only a small number of men (4.2%) with prostate cancer present with PSA >100 ng/mL at diagnosis. Overall survival at five and ten years was significantly poorer in patients with PSA >100 ng/mL. In this cohort of men presenting with PSA >100 at diagnosis, PSA level was not associated with prostate cancer-specific mortality. Gleason score and metastases are significant prognostic factors in this group of men.
Keywords: high PSA; mortality; prostate cancer; prostate- specific antigen; survival.
© 2016 The Authors BJU International © 2016 BJU International Published by John Wiley & Sons Ltd.