Analysis Of Prognostic Factors And Establishment Of Prediction Model For Localized Renal Cell Carcinoma

Objective: In the SEER database,the cox proportional hazards model was used to analyze and screen the prognostic factors affecting localized renal cancer after surgery,and establish a predictive model for renal cancer after surgery.Materials and methods: This study included 34,022 patients with renal cancer surgery who were recorded in the US "Surveillance,Epidemiology,and End Results(SEER)" database from 2010 to 2015.Collect the clinicopathological data of the patients,including age,sex,diagnosis time,race,tumor size,tumor site,tumor grade,SEER stage,pathological type,treatment method,survival status,time of death and cause of death.In the Seer database,cox single-factor and multi-factor analysis were used to screen out predictive factors affecting renal cancer surgery,and R language was used to establish a predictive model to predict individualized tumor-specific survival rate(CSS)and overall survival after renal cancer surgery.Rate(OS).Finally,the consistency index(C-index)and calibration chart are used to verify the performance of the prediction model.Results: Among the prognostic factors that affect the prognosis of kidney cancer,the overall survival rate(OS)of kidney cancer affects the predictors,including the patient’s gender,age,pathological type,stage,histological grade,whether the lymph node metastasis and the number of metastases.Age,pathological type,histological grade,pathological stage,and positive number of lymph node metastasis are the factors affecting the prognosis of tumor-specific survival rate(CSS).In the prognosis analysis of renal cell carcinoma after surgery,3561 deaths and 1826 tumor-specific deaths occurred;the median follow-up time was 42 months(1-83 months).The prognosis of young kidney cancer is better than that of elderly kidney cancer.Among the pathological types,chromophobe cell carcinoma has a better prognosis than non-chromophobe cell carcinoma.Non-chromophobe cell carcinoma mainly includes clear cell renal carcinoma,papillary renal carcinoma,and renal epithelial carcinoma.In patients with lymph node metastasis after renal cell carcinoma surgery,the number of metastatic lymph nodes has a certain effect on the prognosis.Lymph node is an important factor affecting the prognosis.In the TNM staging based on OS analysis,the prognosis of lymph node metastasis is poor,especially the prognosis of stage III.It may be necessary to further refine the adjustment stage and further multi-center analysis.Therefore,it is not recommended to directly use stage I,II,III,IV for prognostic model establishment and prediction.In terms of constructing the prediction model,through these prediction factors,the OS and CSS prediction models are constructed.The C index and calibration chart suggest that the prediction model has a certain prediction ability.Conclusions: Based on the large sampler database,through screening and analysis of prognostic factors,gender,age,pathological type,histological grade,pathological stage,and number of positive lymph node metastases are significantly related to OS.For patients with CSS,age,pathological type,histological grade,pathological stage,whether lymph node metastasis and the number of positive metastases are independent risk factors.Through the above predictive factors,establish a predictive model for predicting postoperative renal cancer,and further analyze the model.After internal and external verification,the model’s efficacy and clinical practicality are acceptable,which is helpful for patient consultation,treatment decision-making and clinical trial design.Provide a reference for the establishment of renal cancer postoperative prediction model and precision medicine.