AbstractAbstract 3616We recently reported that in addition to down-regulating DMNT1 mRNA and inducing DNA hypomethylation, AZA has the ability to reduce the mRNA and protein expression of M2, a subunit of Ribonucleotide Reductase (RNR) in AML cells [Blood 119(22): 5229–38, 2012]. RNR is an enzyme critical to DNA synthesis and repair that converts endogenous ribonucleotides to deoxynucleotides. Thus, we hypothesize that AZA-mediated downregulation of M2 results in decreased levels of endogenous triphosphate deoxynucleotides (dNTP). Thus, when AZA is combined with cytarabine (AraC), the backbone for several primary and salvage regimens in AML, the AraCTP/dNTP ratio would be expected to increase in favor of AraCTP DNA incorporation, thereby overcoming AraC resistance. Based on this rationale, we conducted a phase I trial of escalating doses of AZA (q24 h) on days 1–8 combined with unchanged mitoxantrone (6 mg/m2 q24h), etoposide (80 mg/m2 q24 h), and AraC (1 g/m2 q24 h) (MEC) salvage chemotherapy on days 3–8 in patients (pts) with relapsed or refractory AML. AZA doses were escalated (30, 50, or 75 mg/m2 q24h) with each dose level (DL) at which 3–6 pts were treated. Thirteen pts (39% male) with median age 52 (range 36–64) were enrolled. Five pts had primary refractory disease and 8 had relapsed AML (4 with 1st complete remission duration <12 months). Four pts had prior allogeneic stem cell transplant, and 4 pts had secondary AML (2 treatment related, 2 prior MDS). According to European LeukemiaNet AML Risk Groups, 2, 4, 4, and 3 pts had favorable, intermediate-I, intermediate-II, and adverse disease, respectively. One pt died on day 18 of sepsis, was not evaluable for response, and was replaced. Three pts are currently being treated on the last planned DL. In addition to anticipated myelosuppression, grade 3/4 non-hematologic adverse events at least possibly related to AZA included infections (n=10), diarrhea (n=5), increased bilirubin (n=1), increased AST/ALT (n=1), and vomiting (n=1). One pt (relapsed within 100 days of allogeneic transplant) did not recover platelets until day 58 after achieving morphologic complete remission (CR). Maximally tolerated dose has not been reached. CR and CR with incomplete count recovery (CRi) were achieved in 6/9 (67%; 5 CR and 1 CRi) pts who completed treatment. In responding pts, median time to ANC>500 was 35 days (range 23–45) and median time to platelets>50K was 35 days (range 22–45). Pharmacokinetics (PK) of AZA showed post-infusion biexponential decays with fast and slow plasma half-lives in the range of 10 minutes and 2 hours, respectively. Significant pharmacodynamic reduction in M2 mRNA and/or protein levels was observed in 3 pts analyzed, thus far. In conclusion, the combination of AZA with MEC salvage chemotherapy in advanced AML is well tolerated, not significantly different than MEC alone, and without prolongation in anticipated cytopenias. Early response rate and evidence of reduction in M2 as a proposed target of AZA are encouraging considering that most of the pts had high risk clinical and cytogenetic features. Updated clinical, PK and pharmacodynamic data will be presented. Supported by the National Cancer Institute of the National Institutes of Health Grant Number U01CA076576. Disclosures:Off Label Use: Azacitidine in AML.