A Phase II Randomized Bayesian Study of Very Low Dose Subcutaneous Decitabine Administered Daily or Weekly Times Three in Patients with Lower Risk Myelodysplastic Syndrome (MDS).

Abstract 119

The role of hypomethylating agents is well established in patients with higher risk MDS. The activity and safety of these agents in patients with lower risk (IPSS low or int-1) is not well understood. We hypothesized based on prior in vitro data and cumulative experience with different routes of administration that subcutaneous (SQ) administration of decitabine (5-aza-2'-deoxycitidine) administered either daily for 3 days or weekly times 3 both every 4 weeks will be safe and active in patients with lower risk MDS. To test this hypothesis, we designed a phase II “play the winner” randomized study to compare the daily versus the weekly schedule of decitabine at a daily dose of 20 mg/m2 SQ. Operating characteristics of the design include a 1:1 randomization procedure with the first 40 patients (pts) and after that an adaptive procedure were pts are allocated based on preceding response characteristics. A maximum of 80 pts are plan to be treated on the study. Eligibility criteria include age older than 18 years with low or int-1 MDS by IPSS, adequate performance status and renal and hepatic functions. Pts that have received prior decitabine or 5-azacitidine are excluded. Pts that have received growth factors, lenalidomide or other investigational agents are required a wash out period of 30 days. Baseline transfusion requirements are recorded as baseline and during therapy. At the present time based on the data cut-off by July 10, 2009, 43 pts have been enrolled and treated on study with 22 on the daily times 3 arm and 21 on the weekly x 3 arm. Pt characteristics are: median age 70 years (range 32-87), median time from diagnosis 2.2 months (range 0-64), 36 (84%) pts with de novo disease, 10 (23%) with IPSS low and 32 (74%) with int-1, median WBC 5.6(range 0.6-26), Hgb 9.6 (range 4.9-14.4, platelets 97 (6-442), diploid cytogenetics 30 (70%), 13 pts (30%) with others (-Y, del5, del20, +8, others). Pt characteristics are well balanced between both arms. Of the 43 pts evaluable for response and using IWG 06 response criteria, 4 (9%) pts achieved complete remissions (CR): 3 (14%) in the daily schedule and 1 (5%) in the weekly. Two marrow responses (5%), 1 partial response (2%) and 4 hematological improvements (10%) that were equally distributed between both arms were also documented. The overall response rate is 25%, 32% for the daily arm and 19% for the weekly (p=0.3). The media number of course administered so far is 4+ (range 1-11+). The median time to initial and best response is 2 cycles (ranges 2 and 2-4 respectively). No significant non-hematological toxicity has been observed with minimal grade 3 or 4 toxicity with either arm. Mortality during the first 8 weeks due to treatment complications has been 0%. Three pts have transformed to AML including 2 pts that have died due to disease progression. All in the weekly arm. Based on data cut-off by July 10, 2009, 35 pts (81%) remain on study. With a median follow up of 4.5 months (range 0.9-10), median overall survival has not been reached. Despite the trend to increase response rates and transformation with the weekly arm, at the present time no significant advantage to any of the arms has been documented . As a pharmacodynamic end point, induction of global DNA hypomethlytion using the LINE bisulfite pyrosequencing assay has been measured in 8 consenting patients. Induction of global hypomethylation is observed (50% of pts) pts but the data is limited to establish relationships between arm and/or response. In summary, these initial results indicate that lower dose subcutaneous schedules of decitabine in patients with low or int-1 disease are safe and active with significant less myelotoxicity, and related complications, compared to standard 5-day IV schedules of decitabine used in higher risk disease.