Efficacy, Safety, and Pharmacokinetics (PK) of Azacitidine (AZA) in Children and Young Adults with Acute Myeloid Leukemia (AML) in the Phase 2 AZA-AML-004 Trial

Introduction: Pediatric AML accounts for approximately 20% of pediatric leukemias and represents a clinical challenge due to the poor prognosis for some patients (pts). AZA is a hypomethylating agent shown to slow leukemia progression; however, there are few data on the efficacy of AZA in children. Here we report results from the multicenter, open-label, phase 2 AZA-AML-004 study (NCT02450877) to determine the efficacy, safety, and PK of AZA in children and young adults with AML in molecular relapse after their first complete remission (CR1).

Methods: Eligible pts were aged 3 months to < 18 years with a diagnosis of AML according to WHO classification and ≥ 1 of the following molecular aberrations: t(8;21), RUNX1-RUNX1T1; inv(16), CBFb/MYH11; t(9;11), MLL-AF9; NPM1, or FLT3-ITD mutations. Following CR1, pts had confirmed molecular relapse using RT-qPCR, defined as ≥ 1 log greater than previous sensitivity to a minimal residual disease (MRD) level of ≥ 5 × 10⁻⁴. MRD assessment was carried out centrally. This study was designed to consist of 2 parts: a safety run-in part and a 2-arm randomized part. The primary endpoint of the safety run-in part was identification of a safe and tolerable dose of AZA for the randomized part of the study. The secondary endpoints of the safety run-in part included AZA plasma PK parameters and molecular response at Day 84 of treatment. The randomized part of the study was not conducted; results are reported for the safety run-in part.

Pts were treated with daily intravenous administration of AZA 100 mg/m² on Days 1-7 of a 28-day cycle for up to 3 cycles; if the 100 mg/m² dose was deemed unsafe or intolerable, an additional cohort of pts were to be enrolled and treated with AZA 75 mg/m² on days 1-7 of a 28-day cycle.

Results: As of October 8, 2019, a total of 7 pts experiencing molecular relapse after CR1 had received AZA 100 mg/m². Median age was 6.7 years (range 2.0-12.0); 5 of 7 pts were male. Median time since initial AML diagnosis to enrollment was 10.3 months (range 9.0-13.1). Four pts had a inv(16), CBFb/MYH11, and 3 had t(8;21), RUNX1-RUNX1T1, 1 of which also had an associated FLT3-ITD mutation. Median treatment duration was 12.0 weeks (range 8.0-13.9). Five pts completed 3 cycles of treatment and 2 pts discontinued AZA prior to cycle 3, 1 due to disease progression and 1 due to death. Median treatment cycles received was 3 (range 2-3).

AZA-related treatment-emergent adverse events (TEAEs) occurred in all 7 pts. Five pts experienced AZA-related grade 3 or 4 TEAEs; the most common was neutropenia (n = 4). No dose-limiting toxicities were observed. One pt experienced serious AZA-related TEAEs of febrile neutropenia and pneumonia. Treatment in 2 pts was interrupted due to febrile neutropenia and neutropenia, respectively. No pts experienced TEAEs leading to discontinuation or dose reductions.

Six pts were included in the PK evaluation. Following multiple doses of AZA 100 mg/m², the geometric means for C_max and AUC_0-∞ on Cycle 1 Day 7 were 1,556 ng/mL and 787.6 ng∙h/mL, respectively, with high inter-pt variability observed (CV% 201.6% and 88.8%, respectively). AZA plasma concentrations declined with a geometric mean t_1/2 of 0.38 h; the geometric means for clearance and volume of distribution were 127.2 L/h and 70.2 L, respectively.

Five pts had MRD assessment at Day 84 with 4 pts achieving either molecular stabilization (n = 3) or molecular improvement (n = 1); 1 pt had clinical relapse. Six pts treated with AZA underwent hematopoietic stem cell transplantation (HSCT). The median time from initiation of AZA treatment to HSCT was 118 days (range 100-167). At median follow-up time of 14.5 months (range 14.0-36.0) from enrollment, 5 pts were alive. Two pts died > 28 days after AZA discontinuation, 1 due to cardiac failure (30 days post last dose), and 1 due to multiorgan failure (276 days post last dose).

Conclusions: In the safety run-in part of this phase 2 study evaluating AZA in children and young adults with AML, 100 mg/m² on Days 1-7 of a 28-day cycle was established as the safe and tolerable dose. The observed TEAEs were consistent with the known safety profile of AZA. PK data were consistent with historical AZA data, and exposures were similar compared with adult patients treated with AZA. These data suggest that AZA may potentially reduce MRD level prior to HSCT. The study is continuing as an investigator-sponsored study to further explore efficacy of AZA in AML with molecular relapse in children and young adults.