Analysis of independent risk factors for early castrate-resistant after endocrine therapy in elderly patients with prostate cancer / 解放军医学杂志

ABSTRACT

[Abstract] Objective To analyze the independent risk factors for castrate-resistant prostate cancer (CRPC) after conservative endocrine therapy in elderly prostate cancer patients. Methods Totally 90 elderly prostate cancer patients undergoing endocrine therapy from July 2013 to July 2015 were included into this study as research objects. The basic data and pathological, and laboratory indexes were collected from them. The incidence of CRPC in 2 years after discharge was recorded by follow-up, the Cox risk regression model was used to analyze the high-risk factors of CRPC. The application value of the risk factors to predict CRPC were analyzed by receiver operating characteristic curve and the exponential equation was established to predict the occurrence of CRPC in elderly patients with prostate cancer for 2 years. Results Totally 90 patients were followed up for 24 months in average, of whom 2 cases were lost to follow-up, 26 developed CRPC, the incidence of CRPC was 29.55%. The Cox risk regression model showed that T stage (OR=3.823, 95%CI 2.137-6.839, P<0.001), Gleason score (OR=8.045, 95%CI 3.501-18.487, P<0.001), prostatic specific antigen (PSA) (OR=2.983, 95%CI 1.407-6.324, P=0.004) and prostate cancer gene 3 (PCA3) (OR=1.998, 95%CI 1.263-3.161, P=0.003) were the independent risk factors for prostate cancer progressing to early CRPC in elderly prostate cancer patients. The exponential equation for predicting CRPC model was 0.367X1+0.642X2+0.409X3 +0.815X4 according to the results of multivariate analysis (X1 T stage, X2 Gleason score, X3 PSA value, X4 PCA3 value), the AUC was 0.855 (β=0.056, 95%CI 0.745-0.965, P<0.001). The sensitivity was 0.919, the specificity 0.857, the Youden index 0.776, and the corresponding index of prognosis 1.325. Conclusion The T stage, Gleason score, PSA and PCA3 are independent risk factors for prostate cancer progressing to CRPC after endocrine therapy in elderly prostate cancer patients, can be used comprehensively to establish a model of predicting CRPC for improving the accuracy of judgments.