To Develop A New Recurrence Risk Prediction Model For Non-muscular Invasive Bladder Cancer

Research background and purpose:Transurethral resection of bladder tumor(TURBT)combined with postoperative bladder perfusion is the standard treatment for non-muscular invasive bladder cancer(NMIBC),but the postoperative recurrence rate is high.There are few studies on the risk factors related to the relapse free survival(RFS)of NMIBC patients after the first TURBT.This paper analyzes the risk factors of RFS of NMIBC patients after the first TURBT,and develops a new recurrence risk prediction model for non-muscular invasive bladder cancer,which is helpful to screen high-risk groups of recurrence,take preventive measures and provide certain reference value for clinical practice.Materials and Methods:197 patients with NMIBC who were treated with TURBT for the first time in our hospital from December 2015 to May 2020 were selected and followed up for 24 to 77 months.They were divided into recurrent group(48 cases)and non-recurrent group(149 cases)according to whether they recurred.The COX risk prediction analysis was performed on the two groups of clinical data,and the statistically significant single factor index in the COX risk prediction analysis was included in the multifactor analysis to screen the risk factors that may affect the RFS after NMIBC,and then the relevant prediction model was constructed.Result:1.Univariate analysis showed that secondary electrotomy(hazard ratio(95%confidence interval)HR(95%CI)= 1.853(1.050-3.271),P =0.033),immediate bladder perfusion(HR(95%CI)= 0.229(0.114-0.462),P < 0.001),tumor stage(HR(95%CI)=4.927(2.498-9.717),P < 0.001),tumor grade(HR(95%CI)=5.010(2.492-10.070),P < 0.001),the number of tumors(HR(95%CI)=3.002(1.609-5.601),P= 0.001)were correlated with postoperative RFS in NMIBC patients.Gender(HR(95%CI)= 0.921(0.470-1.807),P = 0.811),age(HR(95%CI)= 1.936(0.987-3.797),P= 0.055),family history of tumor(HR(95%CI)= 1.002(0.360-2.794),P = 0.997),smoking history(HR(95%CI)= 0.930(0.515-1.681),P = 0.810),diabetes(HR(95%CI)= 1.988(0.891-4.435),P = 0.093),subsequent bladder perfusion therapy(HR(95%CI)= 1.241(0.491-3.133),P = 0.648),tumor size(HR(95%CI)= 1.533(0.651-3.606),P = 0,328)had no significant correlation with postoperative RFS in NMIBC patients.The ratio of neutrophil to lymphocyte(AUC =0.536,P =0.454),the ratio of neutrophil to lymphocyte(AUC =0.510,P =0.829),neutrophil multiplied by the ratio of hemoglobin to lymphocyte(AUC value=0.534,P=0.478)and the index of systemic immune inflammation(AUC=0.514,P=0.769)showed no statistical significance.2.Multivariate analysis showed that immediate bladder perfusion(HR(95%CI)=0.369(0.177-0.770),P = 0.008),tumor stage(HR(95%CI)= 2.503(1.176-5.327),P =0.017),tumor grade(HR(95%CI)= 2.839(1.266-6.368),P = 0.011)were significantly correlated with postoperative RFS in NMIBC patients,and were independent predictors of postoperative RFS in NMIBC patients.3.The correlative risk factors selected by COX regression analysis were used to construct the postoperative RFS column chart of NMIBC.At the same time,the calibration curve was used to evaluate the performance of the postoperative RFS column chart of NMIBC,and the consistency of the postoperative RFS was 0.77,indicating that the prediction efficiency was good.Conclusion:1.T1 stage of tumor,high grade of tumor and no immediate intravesical infusion chemotherapy were independent risk factors for RFS after the first TURBT in NMIBC patients.2.The RFS of NMIBC patients after the first TURBT is closely related to tumor stage,tumor grade,and immediate intravesical perfusion chemotherapy.The predictive model combining immediate intravesical perfusion,secondary resection,tumor stage,tumor grade,and tumor number has better predictive efficacy than the EORTC risk prediction mode.