Therapeutic resistance remains the principal problem in acute myeloid leukemia (AML). We used area under receiver operator characteristic curves (AUC) to quantify our ability to predict therapeutic resistance in individual patients where AUC=1.0 denotes perfect prediction and AUC=0.5 denotes a coin flip, using data from 4,601 patients with newly diagnosed AML given induction therapy with 3+7 or more intense standard regimens in MRC/NCRI, HOVON, SWOG, and MD Anderson Cancer Center studies. Age, performance status, white blood cell count, secondary disease, cytogenetic risk, and FLT3-ITD/NPM1 mutation status were each independently associated with failure to achieve complete remission despite no early death (“primary refractoriness”). However, the AUC of a bootstrap-corrected multivariable model predicting this outcome was only 0.78, indicating only fair predictive ability. Removal of FLT3-ITD and NPM1 information only slightly decreased the AUC (0.76). Prediction of resistance, defined as primary refractoriness or short relapse-free survival (RFS), was even more difficult. Our ability to forecast resistance based on routinely available pre-treatment covariates provides a rationale for continued randomization between standard and new therapies and supports further examination of genetic and post-treatment data to optimize resistance prediction in AML.

A B S T R A C T PurposeYounger patients with acute myeloid leukemia (AML) harboring NPM1 mutations without FLT3-internal tandem duplications (ITDs; NPM1-positive/FLT3-ITD-negative genotype) are classified as better risk; however, it remains uncertain whether this favorable classification can be applied to older patients with AML with this genotype. Therefore, we examined the impact of age on the prognostic significance of NPM1-positive/FLT3-ITD-negative status in older patients with AML. Patients and MethodsPatients with AML age Ն 55 years treated with intensive chemotherapy as part of Southwest Oncology Gorup (SWOG) and UK National Cancer Research Institute/Medical Research Council (NCRI/MRC) trials were evaluated. A comprehensive analysis first examined 156 patients treated in SWOG trials. Validation analyses then examined 1,258 patients treated in MRC/NCRI trials. Univariable and multivariable analyses were used to determine the impact of age on the prognostic significance of NPM1 mutations, FLT3-ITDs, and the NPM1-positive/FLT3-ITD-negative genotype. ResultsPatients with AML age 55 to 65 years with NPM1-positive/FLT3-ITD-negative genotype treated in SWOG trials had a significantly improved 2-year overall survival (OS) as compared with those without this genotype (70% v 32%; P Ͻ .001). Moreover, patients age 55 to 65 years with NPM1-positive/FLT3-ITD-negative genotype had a significantly improved 2-year OS as compared with those age Ͼ 65 years with this genotype (70% v 27%; P Ͻ .001); any potential survival benefit of this genotype in patients age Ͼ 65 years was marginal (27% v 16%; P ϭ .33). In multivariable analysis, NPM1-positive/FLT3-ITD-negative genotype remained independently associated with an improved OS in patients age 55 to 65 years (P ϭ .002) but not in those age Ͼ 65 years (P ϭ .82). These results were confirmed in validation analyses examining the NCRI/MRC patients. ConclusionNPM1-positive/FLT3-ITD-negative genotype remains a relatively favorable prognostic factor for patients with AML age 55 to 65 years but not in those age Ͼ 65 years.

• The FAB type (ie, M0-M7) does not provide prognostic information for cases of "AML, NOS" in the 2008 WHO classification.