A Phase 1 Study of Dose-Dense 5-Aza-2′-Deoxycitidine (decitabine) in Relapse Refractory Acute Lymphocytic Leukemia (ALL).

Abstract 2030

Poster Board II-7

Aberrant DNA methylation of multiple promoter associated CpG islands is frequent in primary ALL and predicts for poor prognosis in adult and pediatric disease. Treatment of ALL cell lines with the hypomethylating agent decitabine results in induction of global and gene specific hypomethylation, reactivation of epigenetically silenced genes and induction of apoptosis at low concentrations and prolonged exposures (Leuk Res 2005;29:739-48). Prior studies of decitabine in myeloid leukemias indicated that induction of global and gene specific hypomethylation using a 5-day schedule of decitabine is transient peaking 7 to 10 days after initiation of therapy. In view of the proliferative nature of ALL and in vitro modeling results, we designed a phase 1 study of decitabine administered daily x 5 every other week in patients with relapsed or refractory ALL. The main objective of the study is to determine the safety, activity, pharmacodynamic effects and optimal dose (based on clinical activity, toxicity profile and hypomethylating properties) of decitabine in relapsed refractory ALL. The study design follows a standard “3+3” rule with an expansion cohort of N=10 patients at the optimal dose. Patients of any age with relapsed refractory ALL are elegible regardless of performance status or organ function. Initial dose level of decitabine was 10 mg/m2 IV infused over 1 hour daily x 5 days every other week with courses of therapy repeated every 28 days. Use of steroids was allowed during the first course of therapy at the discretion of treating physician. 23 patients have been treated in 7 dose levels (10, 20, 40, 60, 80, 100 and 120 mg/m2 IV QD x 5 every other week). Cumulative doses per course ranged from 100 to 1200 mg/m2. Patient characteristics are: median age 36 years (range 8-67), median WBC 5.3 (range 0.2 to 97), median % peripheral blasts 23% (0-97), cytogenetics diploid in 4 (17%), Ph + 2 (8%), complex 17 (73%), phenotype preB/B in 15 (65%). Median number of prior therapies was 3 (range 1 to 7). No severe drug related grade 3 or 4 toxicity was observed at any dose level. Frequent toxicities included diarrhea, fatigue and liver function abnormalities that were limited and probably related to disease. Overall response rate was 23% (6 pts) including 1 CRp (complete remission with incomplete platelet recovery) and 5 complete marrow responses (blasts less than 5%). All responses lasted at least 4 weeks. Responses were observed at multiple dose leves (#1, 2, 4, 5, 7). Global and gene specific methylation was analyzed on days 0, 2,5, 14,16,19 and 28 of cycle 1. Samples were collected from 18 consenting patients. Global methylation was analyzed using the LINE bisulfite pyrosequencing assay. Median day 0 methylation was 63%, declined to 55% (p=0.01) on day 14 and increased to 61% on day 28. The most effective dose in inducing global hypomethylation was 60 mg/m2 : 61% baseline to 21% on day 28. The following genes were analyzed for gene specific methylation: p73, p15, p15, HES5, Notch3 and Jag1. Induction of hypomethylation was observed in informative patients, a process associated with gene expression reactivation. The analysis is not powered to detect association between response and hypomethylating effect. Finally, depletion of DNMT1 was measured using a Western blot assay. Depletion was only observed in 1 patient treated at 60 mg/m2 that had achieved a response. In summary, single agent decitabine is safe at higher doses than used in myeloid leukemias with clinical activity in patients with advanced refractory relapsed ALL. Of importance, hypomethylating effect is observed at cumulative doses of up to 1200 mg/m2 with a maximal effect at 600 mg/m2, doses that are considered cytotoxic in myeloid leukemias. The study continues at the expansion cohort of 60 mg/m2 IV QD x 5 every other week, the dose considered to be optimal based on toxicity, response and hypomethylating effects. Two patients have been treated but are early for assessment. A parallel study of decitabine combined with hyperCVAD is ongoing in patients that do not respond or progress after single agent decitabine. The activity of decitabine should be tested in patients in first relapse ALL.