Epigenetic Priming with 5-Azacytidine Prior to Allogeneic Stem Cell Transplantation for Myeloid Malignancies with In Vivo T Cell Depletion: Results of a Phase II Trial

Background: Relapse is the main cause of death in patients who undergo allogeneic hematopoietic stem cell transplantation (HCT) for myeloid malignancies. Recent improvements in survival have been mainly due to reduction in non-relapse mortality (NRM). Acquired abnormalities of DNA methylation are frequent in myeloid malignancies and are particularly observed among patients with adverse risk features. Epigenetic priming with decitabine to induce DNA hypomethylation was investigated in a phase I trial for acute myeloid leukemia (AML) and has proven safe and feasible (Scandura et al Blood 2011). Applying the principle of chemotherapy-sensitization through hypomethylation in myeloid malignancy, we conducted a phase II trial (NCT02497404) of the epigenetic modulator -5-azacytidine given prior to a reduced intensity conditioning (RIC) regimen. Here we report the results of this trial.

Methods: We designed an open label phase II study to determine the feasibility and efficacy of epigenetic priming with 5-Azacytidine prior to RIC for HCT with in-vivo T-cell depletion. Inclusion criteria were AML, myelodysplastic syndrome (MDS) or myeloproliferative neoplasm (MPN) with an intermediate or poor risk profile by ELN and IPSS-R, respectively, and adequate organ function. Stem cell grafts were GCSF- mobilized peripheral blood stem cells from either a matched related donor (MRD) or matched unrelated donor (MUD), matched for HLA at loci A, B, C, DR, and DQ. The epigenetic priming regimen was 5-Azacytidine 75 mg/m2/daily X 5, given between days -11 to -7 (Dose reduction to 50 mg/m2/daily after 14 patients due to toxicity). The conditioning regimen consisted of Fludarabine 40 mg/m2/daily, days -6 to -3, and Melphalan 140 mg/m2, on day -3. Graft versus host disease (GVHD) - prophylaxis consisted of Alemtuzumab (total dose 60 mg for MUD and 30 mg for MRD), and Tacrolimus for 3-6 months post-transplant. The Primary objectives were overall survival (OS) and progression free survival (PFS) at 1 year. Secondary objectives were incidence of GVHD and NRM.

Results: 39 adult patients with AML, MDS, and chronic neutrophilic leukemia (CNL) were enrolled prospectively between 2/2015 and 6/2018. The median age was 60 years (range, 26-74). The majority of patients had AML (33, 85%), or MDS (5, 13%). Nineteen AML patients (58%) had a poor risk karyotype. Nine cases (23%) had mutated TP53. The donor was MUD in 22 (44%), MRD in 17 (56%) of patients. Other patient characteristics are listed in Table 1. The median follow up for survivors is 6.7 years (5.8 – not reached). Thirty-eight patients engrafted neutrophils at a median of 13 days (9-16) and platelets at a median of 16 days (12-102). One patient was not evaluable due to early death on day 9. OS at 1 year was 64% (95% CI 51-81%) and 36% (95% CI 24-55%) at 4 years, (Figure 1). PFS at 1 year was 54% (95% CI 40-72%) and 33% (95% CI 21-52%) at 4 years. At 1 year, relapse was the cause of death in 21% (95% CI 18-22%) and NRM was 15% (95% CI 13-17%). At 4 years, relapse was the cause of death in 41% (95% CI 38-43%), while NRM reached 23% (95% CI 20-25%). The most common cause of NRM was infection in 5 patients (46%), followed by thrombotic microangiopathy and GVHD. The incidence of acute GVHD by day 100 was 29% (95% CI 13-43%) with grade III-IV acute GVHD found in 8% (95% CI 6-9%). The incidence of chronic GVHD at 1 year was estimated at 22% (95% CI 7.4-35%). No new cases of chronic GVHD were seen thereafter. The most common grade 3/4 toxicities were infections in 8 patients (21%) and renal failure (23%). We found 6/14 patients to have grade 3/4 renal toxicity with 2 requiring hemodialysis. After reducing the 5-azacytidine dose to 50 mg/m2/daily, we observed only 3/25 cases of grade 3/4 renal dysfunction.

Conclusion: Preconditioning treatment with 5-azacytidine as epigenetic priming is tolerable with a toxicity profile overall similar to conditioning with fludarabine/melphalan. But the addition of azacytidine at 75 mg/m2/daily (total 5 doses) causes significant renal toxicity, which abates with dose reduction. The PFS in this very high-risk population with historically high relapse rates is promising. A larger number of patients is needed to further evaluate the efficacy of this approach.