Data Analysis Of Risk Factors For Biochemical Recurrence After Radical Prostatectomy

Objective:Analyze the risk factors for biochemical recurrence(BCR)after radical prostatectomy(RP)and establish a predictive model to guide the management of biochemical recurrence in prostate cancer patients after surgery.Methods:Retrospectively analyze the clinical data of 400 patients who underwent RP in the Department of Urology of our hospital from January 2014 to June 2022.Based on whether the patients had BCR after surgery,the 400 patients were divided into a recurrence group and a non-recurrence group.The preoperative serological test results,imaging results,and postoperative pathological information of the patients were compared between the two groups.Univariate and multivariate COX survival regression analyses were performed using biochemical recurrence as the dependent variable and the patient’s clinical data as independent variables to determine the independent risk factor indicators affecting the prognosis of biochemical recurrence after radical prostatectomy.The Kaplan-Meier method was used to plot survival curves of the risk factors for recurrence prognosis.R4.2.0 software was used to plot the risk factor nomogram affecting recurrence prognosis and to construct a recurrence risk prediction model(training set)and risk scoring formula.In addition,this study collected 320 patients from other hospital as a validation set for model construction according to the inclusion and exclusion criteria.The concordance index(C-index),time-dependent receiver operating characteristic(ROC)curve and calibration curve were used as reference indicators to verify the reliability of the training set model.P<0.05 was considered statistically significant.Results:1.Results of univariate and multivariate COX regression analysis in the training set:Out of 400 patients,92 experienced biochemical recurrence,with an incidence rate of 23%.The median follow-up time was 62.236(95%CI:59.178～75.945)months.Univariate COX regression results showed that total prostate-specific antigen(TPSA)≥24.23 ng/ml,Gleason score>7 points,prostate-specific antigen density(PSAD)≥0.726 ng/ml/cc,tumor composition containing ’intraductal carcinoma of the prostate(IDC-P)’,seminal vesicle invasion and positive surgical margins were associated with postoperative BCR(P<0.05).Multivariate COX regression results showed that Gleason score>7 points(P=0.006),PSAD≥0.726 ng/ml/cc(P=0.048),tumor composition containing ’IDC-P’(P=0.002),seminal vesicle invasion(P=0.005)and positive surgical margins(P=0.004)were independent risk factors for biochemical recurrence after radical prostatectomy.2 Establishment of prediction model and external validation of the model:The risk score calculation formula for the training set model is total risk score = 96.31214 ×[1.0383×(Gleason score>7 points)+0.9278×(tumor composition=containing intraductal carcinoma)+0.9646×(PSAD≥0.726 ng/ml/cc)+0.7383×(seminal vesicle invasion=yes)+0.7235×(surgical margin=positive)].External validation of the validation set model found that the general data of the two models were consistent and comparable;the 1-year,3-year,and 5-year time-dependent ROC AUCs of the training set and validation set models were all within the range of 0.7-0.9,indicating that both models had good predictive ability.The C-index of the internal validation of the training set model was 0.821(95%CI:0.775,0.867)(P<0.001),and the C-index of the external validation of the validation set model was 0.745(95%CI:0.689,0.802)(P<0.001),indicating that both models had high consistency.Conclusion:This study indicates that Gleason score>7 points,tumor components containing "IDC-P",preoperative PSAD≥0.726 ng/ml/cc,seminal vesicle invasion,and positive surgical margins are independent risk factors for biochemical recurrence after radical prostatectomy.A predictive model based on these factors has good predictive ability and reliability and can be used as a method for predicting biochemical recurrence.It is recommended to communicate with patients who have the above-mentioned risk status indicators to develop a stringent follow-up protocol and actively intervene in treatment to minimize the occurrence of recurrence.For patients with a lower risk of biochemical recurrence,the follow-up interval can be appropriately extended to avoid overtreatment and improve the postoperative quality of life of patients.