Temozolomide sensitivity is determined by methylation of the O6-methylguanine-DNA methyltransferase (MGMT) promoter. This study assessed whether the temozolomide dose can be tailored byMGMTpromoter status and whether protracted, low-dose temozolomide can “prime” blasts in patients with unmethylated MGMT (unMGMT). Elderly patients with high-risk AML were stratified byMGMTmethylation. Patients with methylatedMGMT(mMGMT) received temozolomide 200 mg/m2 orally for 7 days every 4 weeks, while patients with unMGMT received temozolomide 100 mg/m2 orally for 14 days followed by 200 mg/m2 orally for 7 days every 6weeks. Of 36 patients (median age, 75 years), 31 (86%) had an unMGMT promoter. Overall response rate for the entire cohort was 36%. Patients with mMGMT and unMGMT had similar response rates (40% vs. 29%). Median duration of response and overall survival (OS) among responders were 29 and 35 weeks, respectively. Induction deaths (ID) occurred in 25% of patients, mostly caused by disease progression. Hematological toxicities were the most common adverse event. Toxicities were similar between patients on conventional versus protracted schedules. High HCT-CI scores were predictive of lower CR rate, higher ID, and shorter OS, while bone marrow blast count <50% at screening predicted for improved responses. Temozolomide, dosed according toMGMTmethylation status, demonstrated modest clinical activity in elderly patients with AML, especially in those presenting with fewer comorbidities and low disease burden. The trial was registered onwww.Clinical-Trials.govas #NCT00611247.