Response to upfront azacitidine in juvenile myelomonocytic leukemia in the AZA-JMML-001 trial

Abstract

Allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative therapy for most children with juvenile myelomonocytic leukemia (JMML). Novel therapies controlling the disorder prior to HSCT are needed. We conducted a phase 2, multicenter, open-label study to evaluate the safety and antileukemic activity of azacitidine monotherapy prior to HSCT in newly diagnosed JMML patients. Eighteen patients enrolled from September 2015 to November 2017 were treated with azacitidine (75 mg/m2) administered IV once daily on days 1 to 7 of a 28-day cycle. The primary end point was the number of patients with clinical complete remission (cCR) or clinical partial remission (cPR) after 3 cycles of therapy. Pharmacokinetics, genome-wide DNA-methylation levels, and variant allele frequencies of leukemia-specific index mutations were also analyzed. Sixteen patients completed 3 cycles and 5 patients completed 6 cycles. After 3 cycles, 11 patients (61%) were in cPR and 7 (39%) had progressive disease. Six of 16 patients (38%) who needed platelet transfusions were transfusion-free after 3 cycles. All 7 patients with intermediate- or low-methylation signatures in genome-wide DNA-methylation studies achieved cPR. Seventeen patients received HSCT; 14 (82%) were leukemia-free at a median follow-up of 23.8 months (range, 7.0-39.3 months) after HSCT. Azacitidine was well tolerated and plasma concentration--time profiles were similar to observed profiles in adults. In conclusion, azacitidine monotherapy is a suitable option for children with newly diagnosed JMML. Although long-term safety and efficacy remain to be fully elucidated in this population, these data demonstrate that azacitidine provides valuable clinical benefit to JMML patients prior to HSCT. This trial was registered at www.clinicaltrials.gov as #NCT02447666.

Graphical abstract

Figure 1.

DNA-methylation levels in JMML BM granulocytes during treatment with azacitidine. (A) GDMS for all CpG probes with β ≥0.75 at predose on C1D1. (B) JDMS derived from differentially methylated probes (JMML DMPs) across all time points. Data illustrated as absolute values or relative change. Statistics performed via Kruskal-Wallis one-way analysis of variance/Dunn test. Significance vs C1D1: *P < .05; **P < .01; ***P < .001; ****P < .0001. C, cycle; D, day; NR, nonresponder; R, responder.