Randomized multicenter phase II trial of timed-sequential therapy with flavopiridol (alvocidib), cytarabine, and mitoxantrone (FLAM) versus “7+3” for adults with newly diagnosed acute myeloid leukemia (AML).

Abstract

7002 Background: Serial studies have demonstrated that induction therapy with flavopiridol (50 mg/m2 days 1-3), a multi-serine-threonine cyclin-dependent kinase inhibitor, followed by cytarabine (667 mg/m2/days 6-8) and mitoxantrone (40 mg/m2 day 9) yields complete remission (CR) rates of nearly 70% in pts with newly diagnosed, poor-risk AML. This trial compares "FLAM" with 7+3 in newly diagnosed AML pts. Methods: Between May 2011-July 2013, 165 (FLAM, n=109; 7+3, n=56) newly diagnosed AML pts (18-70 years) with non-favorable cytogenetics were randomized 2:1 to receive FLAM or 7+3 (cytarabine 100 mg/m2/day, daunorubicin 90 mg/m2) across 10 institutions. Randomization was stratified by age, secondary AML and leukocyte count. Pts with residual leukemia on day 14 received 5+2 on the 7+3 arm, whereas pts treated with FLAM were not retreated on day 14. The primary endpoint was to compare CR rates between 1 cycle of FLAM and 1 cycle of 7+3. Secondary endpoints were safety, CR rates after 1 cycle of FLAM vs 7+3 + 5+2, overall survival (OS), and disease-free survival (DFS). Results: The majority of pts on both arms had at least 1 poor-risk feature, excluding age (FLAM=77%, 7+3=68%). FLAM resulted in higher CR rates compared to 7+3 (70% vs 46%, p=0.003), though this difference was less when compared to 7+3 + 5+2 (70% vs 57%, p=0.08). FLAM also produced higher CR rates in pts with secondary AML (60% vs 35%, p=0.05), pts with >1 poor-risk feature (61% vs 34%, p=0.01), pts without poor-risk features (100% vs 72%, p=0.009), and pts <60 years (79% vs 52%, p=0.02). Relapse rates as of this analysis were similar (FLAM: 36% vs 7+3: 38%). Toxicities were similar between both arms, including grade >3 toxicities, early treatment-related mortality, and time to count recovery. Conclusions: FLAM induction results in significantly higher CR rates compared with 7+3 without increased toxicity. FLAM appears to be more active than 7+3 in pts <60 years of age and those with poor-risk features. Although follow-up is too early to assess OS and DFS, these results are promising and a phase 3 comparison of FLAM vs 7+3 is being explored. Clinical trial information: NCT01349972.