Background: A goal of AML maintenance therapy is to decrease the risk of relapse by suppressing growth of residual leukemic cells post-induction. CC-486 is an oral hypomethylating agent that allows for extended dosing schedules (>7 days [d]/28d cycle) to sustain therapeutic activity. In the QUAZAR AML-001 trial (NCT01757535), CC-486 maintenance treatment (Tx) significantly prolonged overall (OS) and relapse-free survival vs. placebo (PBO) in pts with AML in first remission following induction chemotherapy (IC), who were not candidates for hematopoietic stem cell transplant (HSCT). Pts initially received CC-486 or PBO for 14d/cycle, but pts who relapsed with 5–15% blasts could receive escalated 21d/cycle dosing. We review outcomes of pts who received 21d dosing in QUAZAR AML-001. Methods: Pts were aged ≥ 55 years, with intermediate- or poor-risk cytogenetics and ECOG PS ≤ 3, and had achieved first CR/CRi after IC ± consolidation. Within 4 mo of CR/CRi, pts were randomized 1:1 to CC-486 300 mg or PBO QD on d 1–14 of 28d Tx cycles. CR/CRi status was assessed every 3 cycles. Pts relapsing with 5%–15% blasts in blood or bone marrow could receive study drug for 21d/cycle at the investigator’s discretion. Tx could continue until > 15% blasts, unacceptable toxicity, or HSCT. Results: 91 patients (CC-486, 51/238 [21%]; PBO, 40/234 [17%]) were assigned to ≥ 1 21d/cycle dosing schedule. Median time to dose escalation was 9.2 mo (range 1.0-52.7) for CC-486 and 6.0 mo (0.5-19.3) for PBO. Median number of 21d dosing cycles was 2.0 (range 1-45) in the CC-486 arm and 2.0 (1-16) in the PBO arm; 43% and 18% of pts, respectively, received > 3 cycles of 21d dosing. Among 78 evaluable pts, 10/43 (23%) CC-486 pts and 4/35 (11%) PBO pts regained CR/CRi (central review) during dose escalation. Median OS from randomization was 22.8 mo vs. 14.6 mo with CC-486 vs. PBO, respectively (HR 0.66 [95%CI 0.42, 1.0]; P = 0.073), and 1-year survival rates were 80.4% vs. 59.5% (+20.9% [95%CI 2.1%, 39.7%]). The most common AEs with first onset during 21d dosing were febrile neutropenia (CC-486 24%, PBO 3%), thrombocytopenia (22%, 23%), anemia (22%, 20%), and neutropenia (20%, 10%). Conclusions: Escalated 21d CC-486 dosing was well tolerated and resulted in prolongation of OS and restoration of remission in approximately one-fourth of pts. Hematologic AEs first reported during escalated dosing in both Tx arms may be due in part to disease relapse. A 21d dosing schedule should be considered for pts receiving CC-486 who experience relapse with 5–15% blasts. Clinical trial information: NCT01757535.