Phase I Study of Low Dose Decitabine in Patients with Acute Myeloid Leukemia (AML): Pharmacokinetics (PK), Pharmacodynamics (PD), and Clinical Activity.

Decitabine, an inhibitor of DNA methyltransferase (DNMT) enzymes, has clinical activity in myeloid malignancies, even at low doses (>1 log below the MTD) which may be optimal for demethylation rather than only cytotoxicity. We designed a two step trial for AML patients (pts) to determine: Step 1) the minimal effective pharmacologic dose (MEPD) of decitabine as a single agent (defined as lowest dose to induce reexpression of genes commonly methylated in AML in 5/6 pts treated at a single dose level) and Step 2) the MTD of the histone deacetylase inhibitor valproic acid given in combination with decitabine at this MEPD. To date, Step 1 is ongoing; decitabine alone has been administered to 14 pts at two dose levels. Pts had relapsed/refractory AML (N=8) or age>60 and ineligible/refused standard induction therapy (N=6). Pts ranged in age from 57–83 years and had received ≥ 2 prior inductions (N=6), 1 prior induction (2), or were untreated (4). 8 pts received decitabine at 15mg/m²/IV over 1 hr daily (d) for 10 consecutive d, and 6 received 20mg/m²/d on the same schedule, every 28 d. Plasma decitabine levels were analyzed by a validated LC-MS/MS method with a sensitivity of 2 ng/ml developed in our laboratory. Plasma drug concentration-time courses on d 1 and 10 achieved a mean Cmax of 93 ng/ml (N=7) and followed a two compartment infusion model. The mean short and long half-lives were 2.7 and 36.9 min, respectively, with a trend of decreasing the longer half-life on d 10. According to published bone marrow (BM) response criteria (Cheson, JCO 2003), 4/10 pts who completed two cycles of therapy responded (2 at each dose level). 3 had complete response with incomplete recovery of counts (CRi), and one achieved CR. Hematologic improvement (HI) or clinical benefit was seen in 3 additional pts: one achieved neutrophils of 2,200/uL, platelets (plts)>100,000/uL, disappearance of circulating blasts, and disappearance of BM blasts except by flow cytometry; one pt achieved plt transfusion independence (plts>100,000/uL); one pt had stabilization of disease for 6 months. Two more pts who remain on study are not yet evaluable for BM response but have already had significant HI with plts of 259,000 and 74,000/uL, respectively, after one cycle of therapy. Two pts were not evaluable for response due to death from sepsis or death related to decitabine-induced differentiation syndrome (first reported by Blum, ASH 2004), respectively. Decitabine was well tolerated, given typically as an outpatient. There were no severe non-hematologic drug related toxicities. Pts with HI or BM response required plt transfusion support when subsequent cycles were administered. At 15mg/m²/d x 10 d: 4/6 pts experienced at least 100% increase in reexpression genes commonly methylated in AML such as p15 or estrogen receptor; immunoblotting demonstrated DNMT1 protein depletion in 4/6 pts lasting 4–42 d. Clinical trial design based on achieving the MEPD is feasible in AML; we demonstrate clinical activity of low dose decitabine associated with changes in levels of the DNMT1 target and reexpression of epigenetically silenced genes. Complete PK and PD studies including gene reexpression by quantitative RT-PCR, DNMT1 protein levels by immunoblotting, and decitabine-induced promoter demethylation by COBRA/bisulfite sequencing will be presented. (NCI U01 CA 76576-05)