This Is a Title in Chidamide, a Novel HDAC Inhibitor, Is Effective for Treating Leukemia Relapse Post Allogeneic Stem Cell Transplantation and Results in Th1 Polarization in AML Patients with AML1-ETO

Background: Although acute myeloid leukemia (AML) with t(8; 21) carries a favorable prognosis, relapse still occursin up to 40% of patients who require allogeneic stem cell transplantation (alloSCT). However, 10-20% of this patients suffer post-transplant relapse. AML1-ETO fusion protein can cause leukemogenesis via epigenetic modulation of AML1 target genes. In addition, emerging evidence support a role of this fusion protein in the downregulation of immune modulators on leukemia and subsequent immune evasion. Both process are histone deacetylase (HDAC)-dependent. Therefore we hypothesize that inhibition of HDAC may represent an effective treatment for AML with AML1-ETO. Chidamide is a novel HDAC inhibitor that has been CFDA approved for PTCL. In this study, we aim to test the safety and efficacy of Chidamide in post-transplant relapse AML with AML1-ETO.

Method: This is a single arm, open label phase I study administrating Chidamide in AML patients with AML1-ETO who has molecular relapse post alloSCT. Chidamide is given 0.5mg/kg PO twice a week for children or 30mg PO twice a week for adults. All patients are treated for 3 months before analysis of molecular study. Treatment is continued for patients with persistently positive AML1-ETO, while discontinued in patients who achieved complete molecular remission (CMR). For patients not able to achieve CMR, DLI is added in addition to Chidamide. The primary objective is to determine the safety of Chidamide in AML post-transplant. The secondary objective is to evaluate overall response. Blood and bone marrow samples were collected prior to the treatment, monthly during the treatment, and 3 months after discontinuation of the treatment. Correlative studies for immune markers were performed by flow cytometry.

Results: Seven patients have been enrolled in this study. No severe adverse event was observed. The most common adverse event was CTCAE grade 2 thrombocytopenia (7/7). Other adverse events that were possibly or probably related to Chidamide included grade 1 constitutional symptoms (7/7), reversible grade 2 transaminitis (3/7), grade 1 renal dysfunction (1/7), grade 1 digestive tract symptoms (4/7), grade 2 neutropenia (2/7), and grade 1 skin rash (4/7). No GVHD was observed in patients received Chidamide alone, whereas 2 patients received Chidamide and DLI developed grade 3 GVHD. Strikingly all 7 patients responded to Chidamide with decreased AML1-ETO level (ORR 100%). Among them, 6 patients achieved complete response (CR 85.7%), 1 patient had partial response (PR 14.3%) and eventually deteriorated.

We also performed correlative studies and made several important findings. First, we assessed the distribution of each immune component from these 7 patients prior to Chidamide treatment, at which point all patients had molecular relapse. Samples from 4 patients who remain CMR post-transplant were used as controls. We observed a significantly decreased absolute counts of CD3+CD4-CD8+T cells in the patients with relapse, compared with that of patients remaining remission (272.0±63.7 vs. 2087.5±536.5, P=0.042), whereas NK cells, B cells, and CD3+CD4+CD8- T cells showed no statistical difference. Importantly there was a significant increase in percentages of Th2 and Th17 cells, while no difference of Th1 and FOXP3+Treg cells was observed. We next examined the level of Th1, Th2, and Th17 from the 7 relapse patients prior to treatment, compared with that from the same patient 1, 2, or 3 months post Chidamide treatment. We found that both percentages and absolute counts of Th1 cells increased gradually post treatment (26.0±12.8 vs. 190.8±43.4, P=0.027). In contrast, Th2 and Th17 cells decreased. Interestingly, at the time of 3 months after Chidamide discontinuation, the increased absolute counts of Th1 cells and decreased percentages of Th2 and Th17 remained. Furthermore, the Th1/Th2 ratio was low in patients with relapse disease. It increased gradually during Chidamide treatment, correlated with the down trend of AML1-ETO titer.

Conclusion: Chidamide is safe and tolerable in AML patients with AML1-ETO who has post-transplant relapse. Although limited by sample size, our study showed promising clinical efficacy of Chidamide in AML. Importantly our study demonstrated a polarization for Th1 and the significant elevated ratio of Th1/Th2 in patients received Chidamide, which may contribute to the treatment efficacy of Chidamide for AML patients with AML1-ETO.