Restoring hematopoiesis after low dose decitabine of poor graft function post allogenic hematopoietic stem cell transplantation

Abstract Introduction: Poor graft function (PGF), characterized by pancytopenia, is an intractable complication following allogeneic hematopoietic stem cell transplantation (allo-HSCT). Decitabine, a well-recognized hypomethylating agent, has documented efficacy for the alleviation of cytopenia in recent years. Here, we conducted a prospective, randomized multicenter trial to validate the efficacy and safety of low-dose decitabine in PGF patients (NCT05907499), and further explored the underlying mechanisms. Methods: Patients were eligible if they were hematological malignancies with a confirmed diagnosis of PGF after allo-HSCT. Primary PGF was defined by: (i) failure to ever achieve count recovery in at least two lineages (neutrophils > 1.5x109/L, platelets > 30 x109/L and hemoglobin > 85 g/L in the absence of transfusion) after transplantation; (ii) a hypoplastic bone marrow; (iii) the absence of relapse; and (iv) the presence of donor cells as detected by peripheral blood chimerism studies. Secondary PGF was defined as the recurrence of cytopenia after the initial achievement of engraftment. The eligible patients who agreed to have their samples and information collected were enrolled in the study. Patients were immediately randomly assigned in a 2:1 ratio, to receive the standard supportive care combined with decitabine (6 mg/m2, subcutaneously for consecutive 3 days) (Arm A) or the standard supportive care (Arm B). Bone marrow from 2 cases of Arm A (Case 1 achieved hematological response and case 2 no response) was collected at week 0 and week +4 after intervention, respectively. Single-cell sequence analysis was performed to unveil the immune cellular changes. Result: Among the 62 evaluable patients, response rates were 78.0% and 33.3% for the 2 arms, respectively. Of the 27 patients in Arm A achieving hematological response, 13 patients maintained the response in the subsequent follow-up period and 1 patient died from severe GVHD. There was fluctuation of response of other 13 patients during the follow-up. One-year cumulative incidence of GVHD was 26.8% vs 38.1%, and relapse 7.3% vs 9.5% of 2 arms. The 1.6-year overall survival rate was 85.3% (± 6.3%) and 33.7% (± 13.5%) in 2 arms, respectively. The event-free survival was 54.9% (± 9.3%) in Arm A and 18.8% (± 10.5%) in Arm B. Intervention and relapse were proved to be the independent factors affecting overall survival. The main adverse events were hematologic and related toxicities. Deeper myelosuppression was found in Arm A, and the incidence of infection and bleeding events was comparable between the 2 arms. Single cell sequencing of bone marrow identified a small cluster of NK cells expressing FNDC3B, CHST11, TGFBR3 and RUNX1 (NK5 cells), which actively communicate with HSPCs comparing the 4 samples. NK5-HSPCs interactions analysis showed activation of IL4 / IL4R and IL7 / IL7R, as well as several downstream signaling cascades, including the PI3K-Akt and JAK-STAT pathways of NK5 cells after decitabine intervention, together with the reconstitution of hematopoiesis of case 1. Regulatory network analysis showed transcription factors related with hematopoiesis (ETV6, SP3), NK cell proliferation (KLF12), immune response (ELF1, SREBF2, CREB1) and cell function (ZNF534, PBX3) were markedly upregulated in NK5 cells compared to other NK cell subsets after decitabine intervention. We further found NK5 cells proportions were relatively high in PGF and notably decreased in decitabine-responding patients, but not in non-responding patients. Conclusion: This study demonstrates the efficacy and safety of decitabine for patients with PGF after allo-HSCT, and delineates a distinctive profiling and notably close interaction of NK5 cells with HSPCs. While the exact mechanism of decitabine regulating NK5 cells and affecting the process of hematopoiesis remains unknown and needs further exploration.