Phase I Study of Vorinostat in Combination with Decitabine in Patients with Relapsed or Newly Diagnosed Acute Myelogenous Leukemia or Myelodysplastic Syndrome

Epigenetic therapies, eg decitabine (dec) a DNA methyltransferase inhibitor (MTI), have added treatment options for myelodysplastic syndrome (MDS) and acute myelogenous leukemia (AML). However, response rates remain relatively low. Preclinical and clinical data suggest that broadening epigenetic targeting by adding HDAC inhibitors to MTIs may improve responses. Preclinical data suggest that outcomes may differ according to the sequence in which epigenetic agents are combined. We present preliminary data from a phase I, open-label, multicenter, dose-escalating study, aiming to find the maximum-tolerated dose and recommended phase II dose of vorinostat combined concurrently or sequentially with dec in patients (pts) with MDS or AML. Other endpoints included tolerability and exploratory assessments of activity. Pts diagnosed with intermediate-high risk MDS, refractory or relapsed AML (≥18 years), or untreated AML(≥60 years; unsuitable for standard chemotherapy), with an ECOG performance status of ≤2 were eligible. See table for vorinostat dosing schedules. Dec 20 mg/m² IV was administered over 1 h, daily on days 1–5. Therapy continued up to 24 months or until progressive disease (PD). 63 pts were randomized to treatment: median age (range, years) 68 (18–85); 65% males; 9 pts with MDS; 27 pts with untreated AML; 24 pts with relapsed/refractory AML. 35 pts have discontinued due to PD/lack of efficacy (n=19), withdrawal of consent (n=8), adverse events (AEs) (n=6), physician decision (n=1), and protocol deviation (n=1). AEs were reported by 50 pts (79.4%), mainly mild to moderate and commonly included nausea (n=19), fatigue (n=18), constipation (n=16), leukopenia (n=16), diarrhea (n=15), and vomiting (n=12). 37 AEs were treatment related. 42 pts (66.7%) had serious AEs, including febrile neutropenia (n=22), grade 3/4 neutropenia (n=7), and pneumonia (n=9). 12 deaths occurred during the study. In 60 pts evaluable for response, the median (range) number of cycles received were 2 (1–4), 6 (4–7), 3 (1–5), 2 (2–5), 2 (1–2), and 1 (1–7) for dose levels 1, 2, 3, 1a, 2a, and 3a, respectively. Dose levels 3 and 3a were expanded. Complete response (CR) was achieved by 22% pts with MDS, 26% with untreated AML, and 8% with relapsed/refractory AML. Hematologic improvement (HI) was reported in 4% and 22% of pts with untreated AML and MDS, respectively. A similar proportion of pts achieved stable disease (SD) in all disease groups (range 30–46%). Overall, best responses recorded in evaluable pts were: CR, n=11 (3 cytogenic CR); partial response (PR), n=1; HI, n=3; SD, n=25. PD was reported in 6 pts. Best response for 14 pts not yet reported. A best response of CR, PR, HI, and SD was achieved in 26%, 3%, 6%, and 32% of pts treated with concurrent therapy, and 10%, 3%, and 45% of pts receiving sequential regimens reported a best response of CR, HI, and SD, respectively. In pts with untreated AML receiving concurrent therapy, CR was achieved in 8 pts, PR in 1 pt, HI in 2 pts, and SD in 10 pts. In those receiving sequential therapy, CR was achieved in 3 pts, HI in 1 pt, and SD in 13 pts. Preliminary data indicate that the combination of vorinostat with dec, either concurrently or sequentially, is possible without significant toxicity. In addition, the combination shows promising activity in MDS and untreated AML.