Predictors of prostate-specific antigen biochemical recurrence in patients undergoing primary whole-gland prostate cryoablation.

Abstract

Cryoablation has been proven as a less invasive, safe, and effective treatment for localized prostate cancer. We attempted to identify the predictors of biochemical recurrence after prostate cryoablation for localized prostate cancer in this study. We reviewed 114 patients who underwent primary whole-gland prostate cryoablation for localized prostate cancer from October 2008 to March 2013. The perioperative parameters included age >70 years, initial prostate-specific antigen (PSA), preoperative prostate volume, Gleason score, T stage, D'Amico risk group, postoperative PSA nadir, time to PSA nadir, and PSA biochemical recurrence, defined by Phoenix definition (nadir plus 2 ng/mL). Receiver operating characteristic (ROC) analysis was used for the best cutoff value of PSA nadir for PSA biochemical recurrence. The parameters were analyzed in binary logistic regression and Kaplan-Meier analysis for PSA biochemical recurrence. A total of 31.6 % (N = 36) patients had PSA biochemical recurrence during the median follow-up of 34.87 ± 16.49 months. ROC analysis revealed that the best cutoff value for biochemical recurrence prediction was when the PSA nadir = 0.3 ng/mL. On multivariate analysis and Kaplan-Meier analysis, the D'Amico high-risk group [hazard ratio (HR) 6.552; p = 0.014], PSA nadir >0.3 ng/mL (HR 34.062; p < 0.001), and time to PSA nadir <3 months (HR 4.144; p = 0.021) were statistically significant for PSA biochemical recurrence. The D'Amico high-risk group, postoperative PSA nadir >0.3 ng/mL, and time to PSA nadir <3 months predict biochemical recurrence in primary whole-gland prostate cryoablation.