Objective To analyze the factors related to abiraterone acetate(AA)resistance in patients with metastatic castration-resistant prostate cancer(mCRPC)and to construct a prediction model.Methods Clinical data of 81 mCRPC patients who received AA treatment in Tongji University Affiliated Tongji Hospital from January 2018 to December 2021 were retrospectively analyzed.AA resistance was defined as disease progression after AA treatment,including PSA progression and radiographic progression.The risk factors of AA resistance were analyzed with univariate and multivariate Cox regression analyses.The related factors for progression free survival(PFS)were evaluated with Log-rank test.A predictive model was established based on the regression coefficients from the multivariate Cox regression analysis.The predictive performance of the model was evaluated using time-dependent ROC curve.Nomogram was used to assess the risk of resistance at different time points after AA treatment.Results The median follow-up time was 9.9 months(5.9-17.9),the median PFS was 10.9 months(95％CI:5.7-16.0),and AA resistance developed in 64 cases(79.0％).Multivariate Cox regression showed that ECOG score(HR=2.121,P=0.019),NSE(HR=1.029,P=0.040),and LDH level(HR=1.004,P=0.042)were independent risk factors for AA resistance.Based on the Cox regression results,a prediction model was established:-0.05 × BMI+0.529 × lymph node metastasis+0.213 × HVD+0.421 × T stage+0.029 × NSE-0.004 × HB+0.004 × LDH-0.071 × ALB+0.752 × ECOG.Time-dependent ROC analysis showed that the area under curve(AUC)of the model for predicting resistance to AA after 6,12,and 24 months of treatment was 0.837,0.867,and 0.844,respectively.The AUC at 6 months was significantly higher than that of NSE(0.578),LDH(0.656),and ECOG score(0.673)(all P＜0.05).The AUC at 12 months was significantly higher than that of NSE(0.482)and ECOG score(0.605)(both P＜0.05).The AUC at 24 months was significantly higher than that of LDH(0.436)and ECOG score(0.732)(both P＜0.05).Conclusion The ECOG score,serum NSE level,and LDH level before treatment are independent risk factors for AA resistance.The established model in the study can effectively predict resistance to AA treatment at 6,12,and 24 months after treatment,and its predictive performance is superior to that of ECOG score,serum NSE level,and LDH level.
prostate cancercastration resistanceabirateronedrug resistanceprognostic model
维普
万方数据
