Study Between Preoperative Magnetic Resonance Imaging Positivity And Biochemical Recurrence After Radical Prostatectomy

Objective:To evaluate whether preoperative magnetic resonance imaging(MRI) findings as a predictor of biochemical recurrence after radical prostatectomy(RP). And to evaluate whether preoperative magnetic resonance imaging, when converted with conventional factors, can increase the prediction efficiency.Methods:The data from 115 diagnosed localized prostate cancer patients with RP from 2006 to 2013 were collected retrospectively. Biochemical recurrence was defined as increase or persistence of serum PSA levels>0.2 ng/ml after surgery. In this study, according to the European Urogenital Radiology guidelines, the patients were devides into two groups base on MRI score for the probability of the presence of prostate cancer using a five-point scale of diagnostic confidence level, 5 points, 4 points and 3 points patients were given to preoperative MRI positive group(87 cases), 2 points and 1 point patients were given to preoperative MRI negative group(28 cases). The univariate and multivariate cox proportional hazard regression models were used to analyze the risk factors associated with biochemical recurrence. TWO protocols as follows were subjected to receiver operating characteristic(ROC) cruve analysis using the three risk factors from the multivariate analysis : preoperative MRI(T2WI)+biopsy Gleason score+clinical stage(protocol 1), biopsy Gleason score+clinical stage(protocol 2). Calculate the accuracy for each protoco, and apply Z test for comparison.Results:The survival time of MRI positive group was shorter than the MRI negative group, preoperative PSA level of the positive groups was higher than the negative group(P<0.05), the age of two groups showed no significant difference. In MRI positive group, the proportion of preoperative PSA>20 ng/ml, clinical stage T3, pathological Gleason score≥8, SVI and endocrine therapy were higher than negative group(P<0.05). Preoperative MRI positivity, PSA level at diagnosis, biopsy GS, clinical stage, pathological GS, SVI and LNM were associated with biochemical recurrence in the univariate analysis(P<0.05). However, in the multivariate analysis only preoperative MRI positivity, biopsy GS, and clinical stage were independent prognostic factors(P<0.05).The areas under the ROC curve with each protocol were displayed : protocol 1(0.806), and protocol 2(0.763),(both >0.7), while the differences between protocol 1 and 2 were not significant(Z=0.658,P>0.05)Conclusions:Preoperative MRI positivity can predict biochemical recurrence after RP successfully, which can better predict the biochemical recurrence of prostate cancer after RP, and improve the prognosis of patients.When MRI was added with conventional factors, the predictive improvement was not.