Real-World Application of a Machine Learning Pipeline for Overall Survival in Chronic Lymphocytic Leukemia.

We developed a machine learning model and validated it using a large-scale, real-world cohort to predict overall survival in patients with Chronic Lymphocytic Leukemia (CLL). The model captured key survival trends between treatment eras. Individual-level uncertainty was quantified, showing good alignment with observed outcomes. Our findings highlight the potential of predictive modeling to support personalized risk assessment in CLL.