Risk stratification in acute myeloid leukemia (AML) is currently based on pretreatment characteristics. It remains to be established whether relapse risk can be better predicted through assessment of minimal residual disease (MRD). One proposed marker is the Wilms tumor gene WT1, which is overexpressed in most patients with AML, thus providing a putative target for immunotherapy, although in the absence of a standardized assay, its utility for MRD monitoring remains controversial. Nine published and in-house real-time quantitative polymerase chain reaction WT1 assays were systematically evaluated within the European LeukemiaNet; the best-performing assay was applied to diagnostic AML samples (n = 620), follow-up samples from 129 patients treated with intensive combination chemotherapy, and 204 normal peripheral blood (PB) and bone marrow (BM) controls. Considering relative levels of expression detected in normal PB and BM, WT1 was sufficiently overexpressed to discriminate > or = 2-log reduction in transcripts in 46% and 13% of AML patients, according to the respective follow-up sample source. In this informative group, greater WT1 transcript reduction after induction predicted reduced relapse risk (hazard ratio, 0.54 per log reduction; 95% CI, 0.36 to 0.83; P = .004) that remained significant when adjusted for age, WBC count, and cytogenetics. Failure to reduce WT1 transcripts below the threshold limits defined in normal controls by the end of consolidation also predicted increased relapse risk (P = .004). Application of a standardized WT1 assay provides independent prognostic information in AML, lending support to incorporation of early assessment of MRD to develop more robust risk scores, to enhance risk stratification, and to identify patients who may benefit from allogeneic transplantation.