Rationale and goal: Gene silencing that occurs as a result of histone deacetylation and subsequent block in cell differentiation is considered an important leukemogenic process. Reinstatement of gene expression through modulation of histone acetylation may be an important therapeutic approach for leukemia. Phase 1 and 2 studies with vorinostat, a potent histone deacetylase (HDAC) inhibitor, have suggested that vorinostat can induce histone acetylation in vivo , and anti-tumor activities against advanced solid tumors and hematologic malignancies have been reported. Our goal was to design a phase 1 trial of vorinostat combined with drugs active in acute myeloid leukemia (AML).
Study design: Based on our preclinical work showing that cytarabine (ara-C) and vorinostat are antagonistic when given concomitantly but that ara-C and etoposide are synergistic with vorinostat when given after vorinostat exposure at proper dose ranges (Proc AACR 2006, 47: abstract 2122), we designed a clinical study with escalating doses of vorinostat given orally on days 1–7, followed by ara-C (1gm/m2 q12h IV over 3h if age ≥65, 2 gm/m2 q12h IV over 3h if age <65) and etoposide (100 mg/m2 IV over 1h once daily) on days 11–14 to pts with advanced leukemias. Three dose levels (DL) of vorinostat are planned: 200 mg PO BID (DL1), 200 mg PO TID (DL2) and 300 mg PO BID (DL3). Since October 2006 we have enrolled 10 pts with advanced leukemia on this study, of whom 9 are evaluable for toxicity, and 7 for efficacy.
Patient characteristics: Median age 60 years (range 20–77); 6 were male and 4 female. Seven pts had AML in first relapse (2 had CR duration <2 months, 1 < 6 months, 2 <12 mos, and 1 with CR of 28 months); one had primary refractory AML. One pt had newly diagnosed secondary AML, and one high-risk MDS. A standard phase 1 study dose escalation scheme was used with 3 pts per DL per age stratum.
Toxicity: To date no DLTs have been observed, and we are enrolling at DL2. As expected for this patient population, the most frequent toxicities were infectious such as febrile neutropenia, bacteremia, candidemia, and pneumonia. One pt treated at DL2 died on day 40 of a disseminated fungal infection. Other toxicities that were related to vorinostat were mostly grade 1–2 GI toxicities such as anorexia, nausea, vomiting, diarrhea, and constitutional symptoms such as fatigue that occurred in majority of pts. Electrolyte abnormalities (including one gr 3 hypophosphatemia) and elevated LFTs (gr 1–2) were documented in 2 pts, while DVT (gr 3), renal insufficiency (gr 1), and rash (gr 2) occurred in 1 pt each. Of the 9 pts given vorinostat, 2 required hydroxyurea to control blast counts before chemotherapy administration on day 11, while one pt with a rising count decided to proceed with therapy off study.
Response: Of the 7 pts evaluable for efficacy, 3 (43%) achieved CR; of these, 2 pts (1 of whom one had FLT-3 ITD primary refractory leukemia) went on to receive allogeneic BMT and are in continuous CR of 11 and 6 months duration. The third CR pt did not receive any further therapy and relapsed after a CR of 10 months.
Pharmacokinetic (PK) data: The clinical PK profiles for vorinostat are typical for this drug given as a single agent, with Cmax of 0.75 μM, t1/2 1.53 hours at the DL1.
Pharmacodynamic studies in progress: These include measurement of cell cycle kinetics, TRAIL death receptors, and key apoptosis mediators in bone marrow blast cells and buccal mucosal cells pre- and during vorinostat therapy.
Conclusions: So far vorinostat appears to be well tolerated when given in combination with ara-C and etoposide. Although patient numbers are too small to make firm conclusions, the relatively high remission rate observed in this highly refractory population of patients is encouraging.