Construction of a predictive model for recurrence after plasma kinetic resection of bladder tumors in patients with non-muscle invasive bladder cancer

HE Lei; WANG Ke; LI Tianmin

doi:10.3969/j.issn.1674-4136.2024.05.013

Abstract：Objective This study aimed to construct a predictive model for assessing the risk of recurrence in patients with non-muscle invasive bladder cancer following plasma kinetic resection of bladder tumors,thereby aiding clinicians in personalized risk evaluation and management.Methods A retrospective analysis of 106 patients with non-muscle invasive bladder cancer treated in Suzhou Municipal Hospital from March 2015 to March 2019 was conducted.General clinical data,including age,gender,smoking history,diabetes history,cardiovascular disease history,hypertension history,number of tumors,diameter,grade,body mass index,Prognostic Nutritional Index(PNI),albumin(ALB)levels,lymphocyte count(LYM),platelet count(PLT),platelet distribution width(PDW),and mean platelet volume(MPV)levels,were collected.Receiver operating characteristic(ROC)curves were plotted for these variables in relation to postoperative recurrence,and patients were divided into two groups based on the optimal cutoff values of significant factors to plot survival curves for postoperative recurrence.Cox proportional hazards regression analysis was used to identify independent risk factors for recurrence,and a Nomogram prediction model was constructed based on these findings.Internal validation and decision curve analysis were performed on the prediction model.Results ROC curve analysis of the continuous variables showed the areas under the curve(AUC)forPNI,tumor diameter,histological grade,PLT,MPV,BLCA-4,BTA,NMP22,and CEA were 0.965,0.636,0.687,0.994,0.670,0.997,0.995,0.632,and 0.872,respectively.The optimal cutoff values were 40.50％,2.49 cm,—,251.50× 109/L,12.55 fL,143.03 ng/mg,7.32 U/mg,6.99 pg/mg,and 1.96 ng/mg,respectively.Cox univariate analysis revealed PNI,tumor diameter,histological grade,PLT,MPV,BLCA-4,BTA,and CEA as independent risk factors(P＜0.05).Multivariate Cox regression analysis identified PNI,tumor diameter,histological grade,PLT,MPV,BLCA-4,BTA,NMP22,and CEA as risk factors for postoperative recurrence(P＜0.05).The validation results indicated that the Nomogram prediction model performed well.Conclusions This study successfully constructed and validated a predictive model for the risk of postoperative recurrence in patients with non-muscle invasive bladder cancer.With its high accuracy and potential for clinical application,the model can provide significant decision-making support for personalized management and subsequent treatment of bladder cancer patients.