Purpose: Following radical retropubic prostatectomy for prostate cancer, if the serum prostate specific antigen fails to become undetectable, occult micrometastatic disease is suspected. We assessed the natural history of disease progression, and predictors of recurrence and survival in this group of patients.
Materials and methods: We identified 303 men treated with radical retropubic prostatectomy for prostate cancer between 1990 and 1999, who had a detectable prostate specific antigen between 60 and 120 days postoperatively. Systemic recurrence-free and cancer specific survival were estimated using the Kaplan-Meier method, and analyzed using Cox proportional hazards models.
Results: Clinical and pathological features were more adverse among men whose postoperative prostate specific antigen was detectable. These men had poorer systemic recurrence-free survival and cancer specific survival compared to men with an undetectable postoperative prostate specific antigen, and even men whose prostate specific antigen subsequently became detectable. These differences persisted after multivariate adjustment for preoperative prostate specific antigen, specimen Gleason score, seminal vesicle and margin status. With a median followup of 8.5 years, 50 systemic recurrences and 26 deaths from cancer were observed. Gleason score and the prostate specific antigen doubling time were multivariate predictors of systemic recurrence, while Gleason score, margin status and seminal vesicle invasion were predictors of death from cancer.
Conclusions: A detectable prostate specific antigen immediately following radical retropubic prostatectomy confers an increased risk of progression and death, but only in a subset of patients, who may be identified on the basis of pathological features and prostate specific antigen doubling time. In future such patients may be suitable for trials of systemic therapy.