Precision medicine aims to deliver the right therapy to the right patient, but our current models of risk stratification are often too simplistic to meet that goal. Many models treat genetic variants as universally „good“ or „bad“ risk factors, without accounting for obvious modifiers like patient age.
In our latest study, now published in the Journal of the European Hematology Association Hemasphere, we analyzed data from over 3,000 pediatric and adult patients with acute myeloid leukemia (AML). Using machine learning, we stratified patients into six age groups, developed models to accurately predict treatment response and overall survival, thereby autonomously identifying age-dependent prognostic effects of individual genetic variants.
Key finding: No genetic alteration showed uniform prognostic value across age groups. Age fundamentally altered the impact of each mutation.
This underscores the urgent need to move beyond one-size-fits-all models in clinical routine. Context matters! To realize the promise of precision medicine, we must move towards data-driven models that reflect the complexity of disease biology and patient characteristics.
Read the full paper here: https://onlinelibrary.wiley.com/doi/10.1002/hem3.70132
