The use of early postoperative prostate-specific antigen to stratify risk in patients with positive surgical margins after radical prostatectomy

Abstract

Conclusions: The level of ultrasensitive PSA yields valuable information about the prostatectomy outcome already at the first month after the surgery and should aid risk stratification in patients with PSM. Patients not likely to experience subsequent disease progression may be spared the toxicity of immediate adjuvant radiotherapy. Background: It is well recognized that the presence of positive surgical margins (PSM) after radical prostatectomy (RP) adversely affects cancer specific outcomes and recent evidence from randomized trials supports the use of adjuvant radiotherapy in these cases. However, not all of the patients with PSM develop disease recurrence and the policy of adjuvant radiation could result in considerable over-treatment. We investigated the ability of early postoperative prostate specific antigen (PSA) and PSA decline rates to stratify the risk of disease progression during the first weeks after the surgery thereby allowing adequate time for planning eventual adjuvant therapy. Methods. We studied 116 consecutive patients with the finding of PSM after RP for localized prostate cancer between 2001 and 2012. No patients were treated with radiation or hormonal therapy. An intensive postoperative PSA monitoring using ultrasensitive assay started first at day 14 after the surgery, then at day 30, 60, 90 and 180, and subsequently in 3 monthly intervals. Biochemical recurrence (BCR) presented the failure of surgical treatment and it was defined as PSA ≥0.2 ng/ml. The ability of PSA decline parameters to predict BCR was assessed using Cox regression model and area under the curve (AUC) calculation. Results: Overall 55 (47%) patients experienced BCR during median follow-up of 31.4 months (range 6-69). Preoperative PSA, pathologic Gleason sum and pathologic grade failed to reveal any association with observation of BCR. Postoperative PSA levels achieved significant predictive accuracy already on day 30 (AUC 0.74). PSA >0.073 ng/ml at day 30 increased significantly the risk of BCR (HR 4.35, p < 0.001). Predictive accuracy was significantly exceeded on day 60 (AUC 0.84; p < 0.001), while further enhancements on day 90 (AUC 0.84) and 180 (AUC 0.91) were not significant.