Accurate diagnosis and classification of leukemias are the bases for the appropriate management of patients. The diagnostic accuracy and efficiency of present methods may be improved by the use of microarrays for gene expression profiling. We analyzed gene expression profiles in 937 bone marrow and peripheral blood samples from 892 patients with all clinically relevant leukemia subtypes and from 45 nonleukemic controls by U133A and U133B GeneChip arrays. For each subgroup, differentially expressed genes were calculated. Class prediction was performed using support vector machines. Prediction accuracy was estimated by 10-fold cross-validation and was assessed for robustness in a 100-fold resampling approach using randomly chosen test sets consisting of one third of the samples. Applying the top 100 genes of each subgroup, an overall prediction accuracy of 95.1% was achieved that was confirmed by resampling (median, 93.8%; 95% confidence interval, 91.4%-95.8%). In particular, acute myeloid leukemia (AML) with t(15;17), AML with t(8;21), AML with inv(16), chronic lymphatic leukemia (CLL), and pro–B-cell acute lymphoblastic leukemia (pro–B-ALL) with t(11q23) were classified with 100% sensitivity and 100% specificity. Accordingly, cluster analysis completely separated all 13 subgroups analyzed. Gene expression profiling can predict all clinically relevant subentities of leukemia with high accuracy.