Combined Methyltransferase/Histone Deacetylase Inhibition with 5-Azacitidine and MS-275 in Patients with MDS, CMMoL and AML: Clinical Response, Histone Acetylation and DNA Damage.

Optimal re-expression of genes silenced through promoter methylation requires treatment with a DNA methyltransferase inhibitor (DNMTi) followed by the addition of a histone deacetylase inhibitor (HDACi). MS-275 is an orally bioavailable HDACi with a 50 hour half-life. A phase I study combining 5-azacitidine (5AC; 30, 40, or 50 mg/m²/dose) and MS-275 (2, 4, 6, 8 mg/m²/dose) was conducted in patients with MDS, CMMoL, and AML. Subcutaneous 5AC was self-administered daily for 10 days as per the most successful dose schedule of a prior study with 5AC plus sodium phenylbutyrate. MS-275 was administered on days 3 and 10 of each 28 day treatment cycle, following the in vitro paradigm requiring initial exposure to the DNMTi alone for optimal reversal of epigenetic transcriptional silencing. 31 patients were treated: 13 MDS, 4 CMMoL (8 IPSS Int-1, 7 Int-2, 2 proliferative CMMoL), 14 AML (7 AML with trilineage dysplasia [AML-TLD], 4 relapsed, 3 primary-refractory). Median age was 63 years (35 – 84). 5 patients received < 4 cycles and are not considered evaluable for response (declining performance status- 3, sepsis and death - 1, recurrent dose limiting toxicity [DLT]-1). DLT occurred in four patients treated with MS-275 8 mg/m²/dose (5AC 40 and 50 mg/m²/dose): laryngeal edema (2), delayed neutrophil recovery > 21 days (1), asthenia (1). No DLT occurred at any combination with lower doses of MS-275; however, grade 2 fatigue was common at MS-275 6 mg/m²/dose and was considered excessive for chronic administration. 12/27 evaluable patients responded, including 2 CR, 4 PR, 6 bilineage hematologic improvement (2000 IWG criteria). Responses occurred at all dose combinations, in patients with MDS (7), CMMoL (1), AML-TLD (3), and relapsed AML (1). Combining MDS and AML-TLD, responses developed in 10/20. The median number of cycles administered to date to responding patients is 11 (6 – 22). Median time to first objective hematologic response was 2 cycles (1 – 5); median time to best hematologic response was 4 cycles (2 – 9). 20 patients had clonal cytogenetic abnormalities including 6 clinical responders: all 6 had a minimum decrease in abnormal metaphases of 50%, with 4 cytogenetic CR. Median time to best cytogenetic response was four cycles (4 – 6). Following 3 days of 5AC alone, H3 or H4 acetylation (PBMC) was increased in 10/30 patients. Overall, H3 acetylation increased in 25/29 patients (median 2.5 fold, range 1.5 - >1000-fold); H4 acetylation increased in 28/29 patients (median 4-fold, range 1.5 −154-fold). The DNA damage-associated H2AXγ was induced in 7/29 patients after 3 days of 5AC (PBMC) and 20/29 patients after the administration of MS-275. Median fold-increase in H2AXγ-expression was 5.3 (range 1.5 - >1000). The 5AC/MS-275 combination is clinically tolerable and leads to substantive cytogenetic remissions. The relative contribution of the HDACi is under examination in a current Intergroup study randomizing patients to the 5AC/MS-275 combination versus the comparable dose schedule of 5AC alone.