A predictive model for therapy failure in chronic myeloid leukemia patients receiving tyrosine kinase inhibitor therapy

Although tyrosine kinase inhibitor (TKI) therapy has markedly improved the survival of people with chronic-phase chronic myeloid leukemia (CML), 20-30% of people still experienced therapy failure. Data from 1,955 consecutive subjects with chronic-phase CML diagnosed by the European LeukemiaNet (ELN) recommendations from 1 center receiving initial TKI imatinib or a second-generation (2G-) TKI therapy were interrogated to develop a clinical prediction model for TKI therapy failure. This model was subsequently validated in 3,454 subjects from 76 other centers. Using the predictive clinical co-variates associated with TKI therapy failure, we developed a model that stratified subjects into low-, intermediate- and high-risk subgroups with significantly different cumulative incidences of therapy failure (p < 0.001). There was good discrimination and calibration in the external validation dataset, and the performance was consistent with that of the training dataset. Our model had the better prediction discrimination than the Sokal and ELTS scores did, with the greater time-dependent area under the receiver-operator characteristic curve (AUROC) values and a better ability to re-defined the risk of therapy failure. Our model could help physicians estimate the likelihood of initial imatinib or 2G-TKI therapy failure in people with chronic-phase CML.