Analysis Of Factors Influencing Pathological Results Of High-risk Prostate Cancer Radical Prostatectomy And Establishment Of A Prognostic Predictive Model

Objective: To study the influencing factors of pathological results after radical prostatectomy of high-risk prostate cancer,and establish a nomogram predictive model for predicting postoperative pathological results,evaluating prognosis,and providing reference for neoadjuvant therapy.Research method: Retrospectively analyze the case data of radical prostatectomy in our hospital from January 2017 to January 2020,collecting preoperative factors such as age,body mass index,free prostate specific antigen,total prostate specific antigen,fPSA / tPSA,prostate volume,PSA density,percentage of positive puncture needles,biopsy nerve invasion,clinical T staging,biopsy Gleason score,surgical methods and postoperative clinical data such as pathological staging,pelvic lymph node dissection results,margin status.The favorable pathological results were defined as pT2,no pelvic lymph node metastasis(pN0),negative surgical margin(NSM),SPSS26.0 software was used as the dependent variable,and the related influencing factors were used as independent variables for single factor binary logistics regression analysis,step by step multi-factor binary logistics regression analysis of influencing factors and pathological results with P value <0.05,according to logistics regression results,draw a nomogram prediction model with Rstudio software,a calibration curve with bootstrap self-sampling and ROC curve,calculate the area under the curve.Results: Univariate analysis showed that fPSA,tPSA,PSAD,percentage of puncture-positive needles,biopsy nerve invasion,cT,and biopsy GS were all P<0.05.Multivariate analysis showed that tPSA,percentage of puncture-positive needles,cT,and biopsy GS were independent Influencing factors of pathological results,the pathological outcome prediction model was established according to the following factors.The calibration curve of the model was well fitted,with an mean absolute error of 0.021,a C-index of 0.915,sensitivity and specificity of 83.3% and 86.9%,and an area under the curve of 0.916.Conclusion: tPSA,GS,cT stage and percentage of positive needles are the independent risk factors for pathological results after high-risk laparoscopic prostatectomy.The prediction model has a good ability to predict the pathological outcome and can assess the prognosis,providing a reference for the neoadjuvant treatment of high-risk PCa.