Post Hematopoietic Cell Transplantation Maintenance Therapy with Low-Dose Azacitidine in a Pediatric Population with High-Risk Myeloid Neoplasms

Patients with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) undergoing allogeneic hematopoietic cell transplantation (HCT) remain at risk for relapse. Children with relapsed AML and high-risk features including prior HCT, early relapse, and high-risk cytomolecular features experience survival of ≈30%. Given the even more dismal survival and limited therapy options for post-HCT relapse, azacitidine (aza) has been explored as post-HCT maintenance therapy based on potential disease targeting activity and enhancement of graft versus leukemia. Studies to date suggest that this approach is safe with signals of favorable survival. A phase 3 trial in adults did not show a survival benefit, although results were confounded by a high rate of withdrawal and a minority of patients received all intended cycles.

We performed a descriptive analysis of all patients with high-risk myeloid neoplasms who underwent allogeneic HCT at Seattle Children's Hospital (SCH) from November 2017 to January 2023 and received post-HCT maintenance aza. Patients were considered high-risk if they had undergone prior HCT, measurable residual disease at HCT, ≥CR2, or high-risk cytomolecular features. Patients received myeloablative conditioning according to standard protocols. All patients were in remission prior to aza initiation. Aza 36-50 mg/m ² was administered as IV or subcutaneous infusions daily for 5 consecutive days on a 28-day cycle with intention to continue therapy with cycles beginning through 1-year post-HCT.

Nineteen patients began aza therapy at a median of 109 days (range 66-263) post-HCT. Demographics and disease information plus the primary high-risk feature for consideration of post-HCT aza are reported in Table 1. Following aza initiation, 9 (47.4%) patients continued subsequent aza cycles at their referring institution with concurrent remote management by SCH as needed. Median post-HCT follow-up was 30.4 months (range 7-66). The median number of completed aza cycles was 8 (range 3-12) and 14 (73.7%) patients received all intended cycles through 1 year following HCT. In 3 patients, aza cycles were delayed >4 weeks due to non-neutropenic fever or infection; de-escalation to 36 mg/m ² for subsequent courses in the 2 patients who were receiving 50 mg/m ² led to no further delays. There were no delays due to marrow suppression, no increased transfusion needs, and no hospitalizations attributed to aza.

Fourteen (73.7%) patients had grade II acute graft versus host disease (GVHD) prior to initiation of aza. Two (10.5%) of these patients subsequently developed chronic oral GVHD, 1 diagnosed incidentally during their last aza cycle and 1 diagnosed after completion of aza. Two (10.5%) patients experienced ≤ grade II acute GVHD flares while tapering systemic immunosuppressive treatment (IST) and receiving aza, prompting escalation of same systemic IST (n=1) or addition of topical IST (n=1). Seventeen of 18 patients who were in remission at 1-year post-HCT were tapering or off IST. Five (26.3%) patients discontinued aza before 1 year post-HCT: 2 due to relapse, 2 per family preference, and 1 for inflammatory process with persistent dual chimerism from double cord HCT. Three-year estimates for overall and relapse free survival were 92% (95% CI 54-99) and 76% (95% CI 46-91), respectively, and cumulative incidence of relapse was 24% (95% CI 7-48) (Figure 1). The 6 patients with history of prior HCT are all are alive with ongoing remission at ≥3 years following second HCT with aza maintenance.

Our experience demonstrates that post-HCT maintenance therapy with low-dose aza for high-risk pediatric AML/ MDS is tolerable with limited toxicity and without significant impact on marrow suppression or GVHD. The ability to successfully implement a post-HCT aza regimen in partnership with local institutions allowed families to return home, potentially reducing emotional hardship and financial toxicity due to travel/ housing and supporting feasibility. Outcomes in this cohort were improved compared to historical controls; this was especially notable in patients undergoing 2 ^nd HCT and in patients ≥CR2. Our findings suggest a potential role for post-HCT low-dose aza in children with high-risk AML/ MDS but further prospective pediatric studies are needed to define optimal dosing and evaluate for benefit.