In Oral Maintenance Therapy in AML, Do All Roads Lead to the Epigenome?

Michael R. Savona, MD
Professor of Medicine and Cancer Biology, Vanderbilt University Medical Center, Nashville, Tennessee
David T. Bowen, MD
Professor of Haematology and Consultant Hematologist, Leeds Teaching Hospital, U.K.

Maintenance therapy to reduce relapse of acute myeloid leukemia (AML) and prolong overall survival (OS) has been explored since the 1960s.1 Chemotherapeutic and immunologic approaches have all failed to improve OS. Only one intervention has been approved for this indication (in 2008, and in Europe only): the rarely used combination of histamine dichloride and interleukin-2.

The DNA methyltransferase inhibitors (DNMTi) azacitidine and decitabine – also called hypomethylating agents (HMA) – have demonstrated clinically meaningful activity as frontline therapy for myelodysplastic syndromes (MDS), chronic myelomonocytic leukemia (CMML), and AML. Subcutaneous azacitidine was first approved by the FDA for this indication in 2004, and intravenous decitabine was approved for MDS and CMML in 2006. However, the parenteral route of administration of these drugs in oncology clinics – administration on 5 to 7 days of every month, with therapy continued until toxicity or disease progression – is inconvenient, to say the least.

In 2020, two oral DNMTi analogues became available in the U.S.: a fixed-dose combination of oral decitabine and cedazuridine (ASTX727) and oral azacitidine (CC-486). One of the challenges of developing oral DNMTi was first-pass metabolism by cytidine deaminase in the gut and liver, which rendered far less active drug in circulation compared with the equivalent parenteral dosing and with major differences in drug levels between patients.

ASTX727 is a fixed-dose combination of the DNMTi decitabine 35 mg and of cedazuridine 100 mg, a specific inhibitor of cytidine deaminase. In a series of bioequivalence studies, the fixed-dose combination led to blood levels of decitabine equivalent to those achieved with intravenous decitabine.2,3 These data supported the FDA’s decision to approve Inqovi on July 7 for the first-line treatment of MDS and CMML. A combination of cedazuridine with azacitidine, ASTX030, is currently in clinical trials.

On September 1, the FDA approved CC-486, not as first-line therapy but for “continued treatment of patients with AML who achieved first complete remission (CR) or complete remission with incomplete blood count recovery (CRi) following intensive induction chemotherapy and are not able to complete intensive curative therapy,” also known as maintenance therapy. Early dose-finding efforts attempted to dose escalate CC-486 to 600 mg to overwhelm cytidine deaminase and approximate the area under the curve of parenteral azacitidine, but pharmacokinetic analysis did not show bioequivalence.4 Administering CC-486 over 14 or 21 days at various dose levels revealed meaningful drug activity; as expected, though, the pharmacokinetic profile and pharmacodynamic impact were considerably different from parenteral azacitidine, leading to the reasonable conclusion that functionally, CC-486 is not purely an oral formulation of parenteral azacitidine, and oral azacitidine should not be used as a simple substitute for injectable azacitidine for first-line therapy in any setting.5,6

CC-486’s efficacy was evaluated in the phase III, randomized, double-blind, placebo-controlled QUAZAR trial, which compared CC-486 maintenance therapy (300 mg/day for 14 days of 28-day treatment cycles) versus placebo in patients older than 55 whose AML was in CR or CRi after intensive induction. Of note, participants were required to be within 90 days of initial CR/CRi. Later in the trial, this was amended to 120 days due to poor enrollment. Eligible patients also may have received 0 (20%), 1 (31%), 2 (45%) or an unknown number of consolidation courses (4%).

With a median follow-up of 41 months, median overall survival (OS) was 24.7 months for patients treated with CC-486 and 14.8 months for those in the placebo arm, for a 31% reduction in death in favor of maintenance therapy with CC-486 (hazard ratio [HR] = 0.69; figurE). Relapse-free survival (RFS) also was prolonged (HR=0.65). The survival benefit with CC-486 was independent of relevant prognostic factors, including cytogenetic risk (intermediate vs. poor) and number of consolidation courses. While the improvement in OS is noteworthy, putting this in context and describing the applicable patient population vis-à-vis the FDA approval label is necessary.

First, while the optimal number of cycles of consolidation chemotherapy in the treatment of AML is debatable, most experts would agree it is more than one, although tolerability of intensive consolidation in older patients may limit this. Thus, the question that was answered by QUAZAR was: Is CC-486 superior to placebo for patients who receive induction plus a median of one cycle of consolidation? The FDA label could be construed to reflect an incomplete intensive therapy schedule, but no strict definition of the term “not able to complete intensive curative therapy” is actually provided in the prescribing information. So, the question of whether CC-486 is superior to placebo after conclusion of ”full” consolidation (with likely greater curative potential) remains unanswered.

Younger patients with AML who received parenteral azacitidine maintenance in CALGB 10503 had no benefit over placebo.7 By contrast, HOVON97 was an attempt to use parenteral azacitidine in older patients who received at least 2 cycles of consolidation chemotherapy. This led to a promising improvement in disease-free survival, but not OS.8 Whether CC-486 would have fared better in these patients is intriguing, as is an argument that parenteral azacitidine might have yielded similar results to QUAZAR if more limited consolidation were offered in HOVON97.

Second, CC-486 did improve median OS and median relapse-free survival compared with placebo, but in the end, “All roads lead to Rome.” The survival curves appear to converge several months beyond the median follow-up, with the caveat of censoring and small numbers to be clarified when further data are published. As such, it is plausible that the intervention did not increase cure rates, but rather delayed relapse and also death. CC-486 was well tolerated with only 5% of patients discontinuing therapy for treatment-related adverse events (predominantly gastrointestinal) and an impressive median number of 12 cycles administered (compared with 6 for placebo).

Finally, QUAZAR was designed and recruited prior to the approval of targeted induction therapies including gemtuzumab ozogamicin, novel delivery technology for chemotherapy (such as CPX-351), and the use of venetoclax in AML. In addition, a high (and growing) proportion of patients with AML proceed to transplant rather than consolidation and CC-486 was not studied as a bridge to transplant. Thus, the QUAZAR trial data clearly support a benefit for CC-486 maintenance in a selected cohort of AML patients, but its role in the era of recent drug approvals for AML needs further evaluation.


  1. Rashidi A, Walter RB, Tallman MS, et al. Maintenance therapy in acute myeloid leukemia: an evidence-based review of randomized trials. Blood. 2016;128:763-773.
  2. Savona MR, Odenike O, Amrein P, et al. A phase 1 dose-escalation study of ASTX727, a combination of the oral cytidine deaminase inhibitor (CDAi) E7727 with oral decitabine, in subjects with myelodysplastic syndromes (MDS). Lancet Haematol. 2019;6:194-203.
  3. Garcia-Manero G, Griffiths E, Steensma D, et al. Oral cedazuridine/decitabine: a phase 2, multicenter pharmacokinetic/pharmacodynamic randomized, crossover study in MDS and CMML. Blood. 2020;136:674-683.
  4. Garcia-Manero G, Gore SD, Cogle C, et al. Phase I study of oral azacitidine in myelodysplastic syndromes, chronic myelomonocytic leukemia, and acute myeloid leukemia. J Clin Oncol. 2011;29:2521-2527.
  5. Laille E, Savona MR, Scott BL, et al. Pharmacokinetics of different formulations of oral azacitidine (CC-486) and the effect of food and modified gastric pH on pharmacokinetics in subjects with hematologic malignancies. J Clin Pharmacol. 2014;54:630-639.
  6. Garcia-Manero G, Gore SD, Kambhampati S, et al. Efficacy and safety of extended schedules of CC-486 (oral azacitidine) in patients with lower risk myelodysplastic syndromes. Leukemia. 2016; 30:889-896.
  7. Blum W, Sanford BL, Klisovic R, et al. Maintenance therapy with decitabine in younger patients with acute myeloid leukemia in first remission: a phase II Cancer and Leukemia Group B study (CALB 10503). Leukemia. 2017;31:34-39.
  8. Huls G, Chitu D, Havelange V, et al. Azacitidine maintenance after intensive chemotherapy improves DFS in older AML patients. Blood. 2019;133:1457-1464.