Given the different time scales of the interventions, this approach would require that a chosen glipizide phenotype be normalized across all participants and compared against a similarly normalized metformin phenotype. For this hypothetical example, there would be 85% power to detect an effect difference of 0.45 between the two endpoints for a variant with minor allele frequency of 5% in the SUGAR-MGH cohort. For a variant with a 20% minor allele frequency, there would be 85% power to detect an effect difference of 0.26. Given the large number of potential questions that can be tested using SUGAR-MGH, appropriate statistical thresholds, correcting for the number of hypotheses tested, must be used to limit false positive findings. In summary, the SUGAR-MGH cohort illustrates a paradigm for the construction of a pharmacogenetic resource in humans, which is free of the uncontrolled nature of retrospective clinical datasets, achievable at a local site with an investigator-initiated award, and simple enough to enable the recruitment of a large cohort with excellent retention rates and short duration. The receptor tyrosine kinase fms-like tyrosine kinase 3 is expressed at high levels on malignant blasts in 70% to 100% of cases of acute myeloid leukemia and is mutated, most commonly by internal tandem duplication, in 20 to 30 percent of AML cases in different series. FLT3-ITD mutations result in constitutive FLT3 signaling and, clinically, are associated with short disease-free survival following chemotherapy. FLT3 signaling may also be activated in AML cells by autocrine stimulation by FLT3 ligand. Diverse kinase inhibitors inhibit signaling by both FLT3-ITD and CNQX disodium salt wild-type FLT3. However first-generation inhibitors, including lestaurtinib, midostaurin, tandutinib sorafenib and sunitinib, lack optimal potency, CP 99994 dihydrochloride selectivity and pharmacokinetic properties, resulting in limited activity and/or problematic toxicities, and have produced limited single-agent therapeutic benefit, mainly consisting of transient decreases in blasts. The single randomized trial of a first-generation FLT3 inhibitor, lestaurtinib, in conjunction with chemotherapy reported to date did not demonstrate clinical benefit. The second-generation bis-aryl urea FLT3 inhibitor quizartinib has excellent kinase selectivity and pharmacokinetic properties inhibits FLT3-ITD and wild-type FLT3 at 0.1 and 0.5 mM, respectively, in vivo and has shown favorable tolerability and single-agent activity in phase I and II trials. Of note, the dose-limiting toxicity of quizartinib is prolongation of the QT interval, which occurred in 38% and 6% of patients receiving continuous daily doses of 300 and 200 mg, respectively. Following completion of early-phase clinical testing, quizartinib will be tested in combination with chemotherapy. The ATP-binding cassette proteins ABCB1 and ABCG2 are drug efflux proteins that are frequently expressed on AML cells.