Phase II Study of Decitabine in Combination with Imatinib Mesylate in Patients with Accelerated (AP) or Blastic Phase (BP) of Chronic Myeloid Leukemia (CML).

Abstract

Decitabine, a hypomethylating agent, has clinical activity in imatinib refractory chronic myeloid leukemia (Issa et al. JCO 2005). The objective of this phase II study is to investigate the activity of decitabine in combination with imatinib in patients with Ph-positive CML-AP or non-lymphoid CML-BP. To be eligible for this study, patients who had previously been treated with imatinib must have clinical evidence of imatinib failure. Treatment schedule was imatinib orally at 600mg daily and decitabine intravenously at 15mg /m2 x 5 days/week for two consecutive weeks every 6 weeks. At least two courses were planned for all patients. Global DNA methylation of peripheral blood mononuclear cells was measured using a LINE1 bisulfite/pyrosequencing assay. From January 2003 to July 2005, 27 patients (8 BP, 19 AP) were enrolled into this study. Clonal evolution was observed in 21 patients (77%). Twenty-four patients (88%) had failed prior therapy with imatinib, and 3 (12%) were newly diagnosed patients with CML (2 AP and 1 BP). All patients received at least 1 cycle of the treatment (median 3, range 1–12). Toxicity was evaluated in all patients; grade 3/4 toxicity included infection (n=9), CNS bleed (n=2), GI bleed (n=2), dyspnea (n=1), diarrhea (n=1), and edema (n=1). Seven patients (26%) received only 1 cycle of treatment, owing to early death from acute disease progression (n=5), recurrent subdural hematoma (n=1), and a patient's decision due to grade 3 diarrhea (n=1). Twenty patients (74%) received 2 or more treatment cycles and were evaluated for response. Nine patients (33%) achieved complete hematologic response (CHR: BM blast <5%, normalization of peripheral blood and disappearance of signs and symptoms of the disease), and 3 patients (11%) achieved partial hematologic response (PHR: same criteria except for persistence of immature cells, splenomegaly or thrombocytosis but <50% of pretreatment). These included 7 patients (26%) that achieved cytogenetic response (3 complete cytogenetic response (Ph 0%), 1 partial cytogenetic response (Ph<35%), and 3 minor cytogenetic response (Ph<90%)). Median response duration was 13 weeks (range 4–107+). Four patients (15%) who achieved CHR underwent allogeneic stem cell transplantation. Median overall survival of patients with AP and BP were similar in this study (49 and 53 weeks, respectively). Median overall survival of patients with cytogenetic response (n=7), hematologic response only (n=5), and non-responders including those who received only 1 cycle (n=15) were 86, 63, and 18 weeks, respectively (p<0.01). LINE1 methylation decreased from 71.6%±0.9% (mean±standard error of mean) to 60.4%±2.0% on day 5, 60.5%±1.8% on day 12, and returned to 68.8%±1.4% at recovery of peripheral blood counts. Absolute decrease in LINE1 methylation trended greater in non-responders than responders on day 5 (15.4%±3.9% vs 7.6%±2.1%, p=0.14) or day 12 (14.4%±4.5% vs 9.3%±1.8%, p=0.34), though the difference was not statistically significant. In summary, decitabine in combination with imatinib is an active regimen for CML-AP and BP, including patients previously treated with imatinib. Global hypomethylation was observed after decitabine therapy, both in responders and non-responders.