Development of a blood test-based predictive scoring tool for treatment response in chronic myeloid leukemia

The European LeukemiaNet recommendations for the management of chronic myeloid leukemia (CML) consider achievement of major molecular response (MMR) within 12 months to be an optimal response (OR). However, no currently available tool can predict treatment response at diagnosis. This study aimed to develop a predictive scoring tool for OR and subsequent deep molecular response (DMR) using peripheral blood parameters at diagnosis. A retrospective analysis was conducted in 535 patients with CML from the CML Cooperative Study Group database. Patients were categorized into OR (MMR within 12 months; n = 355) and non-OR groups (MMR after 13 months; n = 180). Logistic regression analysis identified white blood cell count, platelet count, eosinophil percentage, and imatinib use as significant predictors of OR. These variables were used to construct a novel score. The score was significantly higher in the non-OR group and showed superior predictive accuracy for OR compared with existing prognostic scores. Furthermore, lower scores correlated with higher rates of DMR achievement. Notably, the score effectively predicted OR achievement among patients with low/intermediate ELTS risk treated with imatinib. This new scoring tool may facilitate individualized treatment selection in CML, guiding upfront tyrosine kinase inhibitor selection and advancing precision medicine.