Impact of Treatment Strategies on Survival and Within Multivariate Predictive Model for Renal Cell Carcinoma Based on the SEER Database: A Retrospective Cohort Study.

Abstract

Background: This project aims to shed light on how various treatment approaches affect RCC patients' chances of survival and create a prediction model for them.

Methods: Data from the Surveillance, Epidemiology, and End Results database were used in this investigation. OS and RCSS after radiation, chemotherapy, and surgery were investigated using the Kaplan-Meier approach. Fourteen factors, including gender, age, race, and others, were subjected to univariate and multivariate COX analyses. Predicting RCSS at three, five, or ten years is the main goal. Predicting OS at three, five, or ten years is the secondary endpoint. Cox analyses, both univariate and multivariate, were used to identify prognostic factors. Furthermore, a nomogram was developed to precisely forecast patient survival rates at 3-, 5-, and 10-year intervals. DCA, calibration curves, and ROC were used to assess the nomogram's efficacy.

Results: Kaplan-Meier analysis revealed that PN was associated with better survival compared to RN for tumors ≤10 cm. Cox analysis identified 10 independent prognostic factors. These variables included gender, age, race, histological type, histological grade, AJCC stage, N stage, T stage, M stage, and surgical type. Based on these variables, a nomogram for OS and RCSS prediction was created.

Conclusion: PN is advised over RN for RCC patients whose tumors are less than 10 cm in diameter since it offers more advantages. The combined nomogram model, which is based on clinicopathological characteristics, therapy data, and demographic variables, may be used to predict the survival of RCC patients and perform prognostic and survival analysis with accuracy.