Prognostic Factors for Biochemical Recurrence More than 10 Years after Radical Prostatectomy

Some patients with long postoperative intervals of undetectable prostate specific antigen are still at risk for biochemical recurrence. Our aims were to identify prognostic factors for late biochemical recurrence, including cancer family history, and evaluate cancer specific mortality.

We identified 10,310 patients after radical prostatectomy without neoadjuvant or adjuvant therapy between 1979 and 2015 in the prospective German Familial Prostate Cancer database. A subgroup of 2,480 patients with more than 10 years of followup (median 12.8) had undetectable prostate specific antigen. Biochemical recurrence, defined as prostate specific antigen 0.2 ng/ml or greater, developing at more than 10 years was defined as late biochemical recurrence. Multiple proportional hazards regression with forward selection was applied to determine prognostic factors for late biochemical recurrence.

The Kaplan-Meier estimated biochemical recurrence rate at 10, 15 and 20 years was 34.3%, 44.0% and 52.7%, respectively. Of 2,480 patients with undetectable prostate specific antigen 10 years postoperatively 249 subsequently had biochemical recurrence, of whom 12 died of prostate cancer. The factors associated with late biochemical recurrence were age at surgery (HR 1.04 per year, p = 0.027), prostate specific antigen at diagnosis (HR 1.02 per ng/ml, p = 0.020), pathological Gleason score (categorical 2-6 vs 7 [3 + 4], 7, 7 [4 + 3] and 8-10, p = 0.002) and pathological tumor stage pT3a or greater (HR 1.50, p = 0.065).

From years 10 to 15 and 10 to 20 postoperatively the biochemical recurrence rate increased by 9.7% and 18.4%, respectively. In contrast to a family history of prostate cancer, age at surgery, prostate specific antigen at diagnosis, pathological tumor stage and pathological Gleason score were prognostic factors for late biochemical recurrence. Patients with late biochemical recurrence are still at risk for death from prostate cancer.