Prognostic Analysis Of Renal Cell Carcinoma With Venous Invasion And Nomogram For Survival Prediction

Objective To investigate the prognostic factors of survival of patients with renal cell carcinomas(RCCs)and venous invasion,and establish nomograms to predict the survival time and survival probabilities of patients who developed RCCs with venous invasion.Methods The clinical data of 222,157 patients with renal malignancies diagnosed between 1987-2015 were downloaded from the SEER(Surveillance Epidemiology and End Results)database,and 16,703 cases of renal cell carcinoma with venous tumour thrombus were obtained after screening.The clinical data of 202 patients(Most of them come from central China)with renal tumours and venous tumour thrombus admitted to the department of urology of Tongji Hospital(TJH)from 2009 to 2019 were searched,and 99 cases of renal cell carcinomas with tumour thrombus pathologically confirmed were screened out,of which 53 had complete follow-up information,and comparative analysis was performed between cases of TJH and SEER database.For cases with complete information in the SEER database,propensity score matching at a ratio of 1:1 was performed based on whether nephrectomy and thrombectomy was performed.The matched data set was divided into two parts,a training set and a validation set,at a ratio of 8: 2.Univariate and multivariate Cox regression analysis were performed on the clinical data of the training set cases for OS(overall survival)and CSS(cancer-specific survival),and independent prognostic factors were screened out to establish OS-nomogram and CSS-nomogram.The training set,validation set and Tongji Hospital case were used for internal validation and external validation of OS-nomogram and CSS-nomogram,respectively.The reliability of OS-nomogram and CSS-nomogram for predicting the survival of the training set,validation set and cases of Tongji Hospital was evaluated by the concordance index(C-index),calibration curve,and the time-dependent ROC curve.The ROC curve was used to compare the nomogram’s ability to predict survival with prognostic models based on TNM staging.Results 16,703 cases of renal cell carcinomas with tumour thrombus were screened from the SEER database,and 99 cases were screened from Tongji Hospital.The average age of SEER cases was 64.75(±11.66)years old,and the average age of Tongji Hospital cases was 54.48(±13.43)years old;cases of Tongji Hospital are more advanced than SEER cases in TNM stage.The median follow-up time for SEER cases was 35 months;At the last follow-up,10590 patients died,excluding 135 cases of unknown cause,7817(74.8%)died of renal cell carcinomas.Fifty-three cases of Tongji Hospital were followed up,with a median follow-up time of 14 months.24patients died at the last follow-up,and all died of renal cell carcinomas.OS and CSS in SEER cases at 1,3,and 5 year were 77.9%,58.6%,47.3%and 81.0%,64.1%,54.9%,respectively;and median OS and CSS were 54 months and 79 months,respectively.The 1,2 and 3 year OS of Tongji Hospital cases were the same as CSS,which were 70.0%,49.8%,and 34.8%,respectively;The median OS and CSS were both 25 months.Survival differences between the non-nephrectomy with thrombectomy group and the nephrectomy with thrombectomy group were reduced in the 2920 cases obtained after the propensity score matching.The univariate regression analysis of the training set showed that the significant influencing factors of OS were age,gender,year of diagnosis,histologic type,renal tumour diameter,nuclear grade,T stage,N stage,M stage,level of tumour thrombus,surgical approach,radiotherapy and chemotherapy(Including chemotherapy,targeted therapy and immunotherapy);The significant influencing factors of CSS were the year of diagnosis,histologic type,diameter of renal tumour,nuclear grade,T stage,N stage,M stage,level of tumour thrombus,surgical approach,radiotherapy and chemotherapy.Multivariate Cox regression analysis revealed 9 key prognostic factors for OS:age(p<0.001),year of diagnosis(p<0.001),diameter of renal tumour(p=0.001),histologic type(p<0.001),and nuclear grade(p<0.001),N stage(p<0.001),M stage(p<0.001),surgical approach(p<0.001),chemotherapy(p<0.001).There are 8 key prognostic factors for CSS:year of diagnosis(p=0.010),diameter of renal tumour(p<0.001),histologic type(p<0.001),nuclear grade(p<0.001),N stage(p<0.001),M Staging(p<0.001),surgical approach(p<0.001),chemotherapy(p<0.001).The C-indexes for OS-nomogram predicting training set,validation set,and Tongji cohort were 0.765(95%CI:0.753-0.777),0.747(95%CI:0.722-0.772),0.804(95%CI:0.708-0.900);and for the CSS-nomogram predicting training set,validation set,and Tongji case set were 0.779(95%CI:0.767-0.791),0.770(95%CI:0.745-0.795),0.821(95%CI:0733.-0.909).The Bootstrap method(repeated 1000 times)was used to validate the nomogram,and the calibration curves were plotted.The calibration curves are all close to a straight line with a slope of 1.In the ROC curve,nomograms got greater AUCs than the TNM staging models in predicting the 3-year OS and CSS.The all three validations have proved that the nomograms established based on key prognostic factors have reliable accuracy in predicting the survival of both Chinese and American patients who developed RCCs with venous tumour thrombus.Conclusion Compared with patients with renal cell carcinomas and venous tumour thrombus in the United States,patients in central China are younger,more advanced in cancer stage,have worse prognosis,and the proportion of death from kidney cancer is higher.The level of tumour thrombus does not predict the survival of all patients with tumour thrombus well.The development of a new T3 stage may need to consider patients who have not undergone thrombectomy.Nomograms based on key prognostic factors can predict the survival of patients in both America and central China with a reliable accuracy.