Multivariate Analysis Of Prognostic Factors In Renal Cell Carcinoma And Predicton Of Progression After Surgery For Patients With Clear Cell Renal Cell Carcinoma

Renal cell carcinoma (RCC) is the most common malignancy in adult kidney, represent over85%of renal cell carcinomas, and accounts for3%of all human malignancy. The incidence of RCC has improved in recent years and had a trend of making younger in the incident age. Furthermore, approximately25%of patients will present with metastatic disease. Currently, there are several factors can be used as prognostic predictors for patients with RCC including tumor stage, Fuhrman grade, tumor necrosis, tumor size and some molecular markers. Prognostic models which including several most prognostic factors have been statistically proposed by some researchers. These models can estimate risk-stratification for RCC, get the individual survival and use the convenient formula or graph instead of data. However, the established models are based on the foreign samples and have the uncertainty on the Chinese samples.This study aims to analysis the clinical and pathological features for RCC patients, establish and validate a prognostic model based on a Chinese population. This model will help improve individualized prognostication, design clinical trials and postoperative follow-up decision making.Part1. Multivariate analysis of prognostic factors in renal cell carcinomaObjective:To discuss the effect of prognostic factors in RCC patients for provide scientific proof for improving the survival of RCC patients.Methods:The clinicopathologic bariables of749RCC patients with definite pathological diagnosis in the second hospital of Tianjin Medical University from January2006to December2010were analyzed retrospectively.613cases (81.8%) have been followed up. The Kaplan-Meier method and log-rank test were used for univariate comparisons of survival. Multivariate analysis was done by the COX regression model.Results:The median follow-up was40months (range,4-72months). The overall survival rates were98.5%at1year,89.0%at3years,83.0%at5years. Univariate analysis showed that the following features were associated with survival:the founding way of cancer, cachexia, haemoglobin, platelet count, tumor size, Fuhrman grade, histologic tumor necrosis, renal capsular invasion, microvessle invasion, urinary collecting system invasion, tumor stage (2010), regional lymph node status, adjuvant therapy. Multivariate analysis showed that the following features were independent associated with survival:tumor size (P=0.001), Fuhrman grade (P=0.022), histologic tumor necrosis (P<0.001), tumor stage (P=0.004), regional lymph node status (P=0.034).Conclusions:In patients with RCC, there are several factors affecting the survival. Tumor stage, Fuhrman grade, regional lymph node status, histologic tumor necrosis are significant independent predictor of survival. Histologic subtypes may not be associated with survival. As clinic workers, understanding these risk factors could guide the therapy and improve the survival rate of RCC patients.Part2. Establishment and validation of prediction of progression after surgery for patients with clear cell renal cell carcinomaObjective:To propose a prediction of progression after surgery for patients with clear cell renal cell carcinoma (ccRCC) and validate the accuracy of this model.Methods:The clinicopathologic bariables of623nonmetastatic ccRCC patients with definite pathological diagnosis in the second hospital of Tianjin Medical University from January2006to December2010were reviewed retrospectively.507cases (81.4%) have been followed up. The Kaplan-Meier method and log-rank test were used for univariate comparisons of survival. Multivariate analysis was done by the COX regression model. The accuracy of the model was assessed by the c-index.Results:The median follow-up was36months (range,4-72months). The recurrence free survival (RFS) rates were96%at1year,86%at3years,79%at5years. Univariate analysis showed that the following features were associated with RFS for ccRCC patients:age at surgery, tumor discovery, cachexia, haemoglobin, platelet count, tumor size, Fuhrman grade, histologic tumor necrosis, renal capsular invasion, microvessle invasion, urinary collecting system invasion, tumor stage (2010), regional lymph node status, adjuvant therapy. Multivariate analysis showed that the following features were independent associated with RFS:tumor stage, Fuhrman grade, histologic tumor necrosis, regional lymph node status, tumor size and haemoglobin. These factors were assigned and a algorithm model was developed. Three groups which based on risk for recurrence after surgery were stratified into: low-risk (score0-3), intermediate-risk (score4-6), high-risk (score=7). These RFS rates were93.2%,66.4%,22.5%respectively, and the difference between three groups has statistics significance. In this sample, the accuracy of Leibovich model was0.765. Validated this current model internally (C=0.689).Conclusions:In patients with ccRCC, tumor stage, Fuhrman grade, histologic tumor necrosis, regional lymph node status, tumor size and haemoglobin are significant independent predictor of progression after surgery. Established a prediction of progression after surgery for ccRCC patients and had a preferably accuracy. In this sample, the accuracy of this model was less than the Leibovich model. This model needs to improve and needs the different population, larger sample, longer follow-up clinical research.