Restricted Mean Survival Time Approaches Implications for Breast Cancer Data

This research assesses the utility of Restricted Mean Survival Time (RMST) as a viable alternative to conventional survival analysis techniques, including the Kaplan-Meier test and Cox Proportional Hazards model, especially when the proportional hazards assumption is not met. RMST offers an alternative approach to using hazard ratios in such situations. This study aimed to evaluate the impact of predictive factors on breast cancer survival. The Restricted Mean Survival Time (RMST) offers a clinically interpretable measure of average survival time within a defined period, providing a more detailed understanding of treatment effects and patient outcomes. Additionally, the study utilized the Cumulative Incidence Function (CIF) model and probability estimates for each time point to further analyze survival data. Using data from 4,420 breast cancer patients at Rizgary Hospital, this research quantifies the impact of various prognostic factors on survival outcomes. The study demonstrates that RMST effectively captures differences in survival probabilities across treatment groups, with patients undergoing surgery or receiving radiotherapy showing significantly higher survival rates than those who did not. Additionally, the analysis reveals that factors such as family history and tumor grade significantly influence survival outcomes, highlighting the heterogeneity of breast cancer. Statistical comparisons using RMST, alongside traditional tests like the Log-rank and Gehan-Wilcoxon tests, confirm significant differences in survival curves, particularly in early-stage survival, emphasizing the importance of timely interventions