CD34+ Cell Dose Effects on Clinical Outcomes of Patients with Acute Myeloid Leukemia after T-Replete Haploidentical Allogeneic Hematopoietic Stemcell Transplantation Using Peripheral Blood Stem Cells. a Study from the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation (EBMT)

Abstract

The absolute number of infused CD34+ donor cells represents a critical step towards the achievement of a meaningful marrow engraftment after allogeneic hematopoietic stem cell transplantation (HSCT). Previous observations have reported controversial conclusions regarding the impact of CD34+ donor cell doses and post-transplant clinical end-points. The vast majority of analysis conducted so far have considered myeloablative and reduced-intensity conditioning allogeneic HSCT setting, with HLA-matched sibling and/or unrelated donors. The T-cell replete haploidentical setting has not been extensively studied yet. We conducted a retrospective analysis on 414 adult patients (median age 54 years; range, 18-74) with acute myeloid leukemia (AML) in first (70%) and second (30%) complete remission who had received a T-cell replete allogeneic HSCT from a haploidentical donor, using peripheral blood stem cells (PBSC), between 2006-2018, based on ALWP-EBMT registry data. Median donor age was 37 years (range, 20-71). Time from disease diagnosis to HSCT was 6.3 months (range, 1.3-97.9). Seventy-three (18%) patients had secondary-AML. Eighty-seven patients (21%) had unfavorable cytogenetics. Karnofsky performance-status at the time of HSCT was ≥ 90 in 318 (77%) patients. Graft vs. host disease (GVHD) prophylaxis was post-transplant cyclophosphamide (PT-Cy)-based in 293 (71%) patients and anti-lymphocyte serum (ATG)-based in 121 (29%) patients. Conditioning was myeloablative in 179 (43%) patients and reduced-intensity in 235 (57%) patients. Median number of infused CD34+ cells was 6.58 x 106/kg (range, 2.2-31.2). Thirteen patients (3.2%) experienced graft failure while the remaining 399 (96.8%) engrafted uneventfully. Neutrophils engraftment at day 30 was 91.2% (95% Confidence interval [CI]: 88 - 93.6) while sustained platelets engraftment (>20.000/uL at 6-months) was observed in 83.7% (95% CI: 79.6 - 87). Median time to neutrophils and platelets engraftment were 20 (range, 10-79) and 21 days (range, 1-37) respectively. After a median follow-up of 23.3 months (range, 12.1-41.8) 2-year overall survival (OS) was 64.5 % (95% CI 59.3-69.7) with a leukemia-free survival (LFS) of 57.3 % (95% CI 51.8-62.7) and non-relapse mortality (NRM) of 23.3 % (95% CI: 19.0-27.7). The cumulative incidence of grade II-IV and III-IV acute GVHD at day-100 were 32.3 % (95% CI: 27.8 - 36.9) and 14.6 % (95% CI: 11.3 - 18.2), respectively. Incidence of 2-years chronic GVHD overall and extensive were 36.3 % (95% CI: 30.9 - 41.6) and 14.4 % (95% CI: 10.7 - 18.6), respectively. GVHD-free/relapse-free survival was 43.5 % (95% CI: 38 - 48.9) at 2-years. We found the optimal CD34+/kg threshold defining high-dose (n= 334) versus low-dose (n= 80)to be 4.96 x 106, using the Hothorn and Zeileis method (Hothorn, T. and Zeileis, A. Generalized maximally selected statistics. Biometrics 2008;64(4):1263-1269). Patients in the high-dose group experienced higher rates of engraftment compared to those in the low-dose group both for neutrophils (92.1% vs. 87.3%, p= 0.005) and platelets (86% vs. 73.3%, p= 0.001) while median time to neutrophils engraftment was 19 days (range, 10-79) in the high-dose group and 22 days (range, 14-46) in the low-dose group (p= 0.0001). Recipients of > 4.96 x 106/kg CD34+ cells experienced lower NRM (hazard ratio [HR] 0.48; 95% CI:0.29-0.75) and prolonged LFS (HR 0.62; 95% CI 0.43-0.90) and OS (HR 0.59; 95% CI:0.40-0.87) than recipients of ≤ 4.96 x 106/kg CD34+cell dose. Patients who received ATG as GVHD prophylaxis experienced less grades II-IV acute GVHD respect those in the PT-Cy cohort (HR 0.56; 95% CI: 0.36-0.87) without any differences in grades III-IV acute GVHD incidence at day 100 between the two groups. On Cox analysis, a female donor to male recipient combination was associated with higher extensive chronic GVHD incidence compared to other donor-recipient sex combinations, with CD34+ categorized according to the optimal threshold of 4.96 x 106 (HR 2.21; 95% CI:1.23-3.97; p= 0.008). AML patients who underwent a PBSC T-cell replete HSCT from haploidentical donors receiving ≥ 4.96 x 106CD34+/kg cells experienced a lower NRM and extended survival rates compared to those receiving ≤ 4.96 x 106/kg CD34+. The latter findings may have a significant practical impact and can guide prospective approaches. Disclosures Labopin: Jazz Pharmaceuticals: Honoraria. Blaise:Molmed: Consultancy, Honoraria; Pierre Fabre medicaments: Honoraria; Jazz Pharmaceuticals: Honoraria; Sanofi: Honoraria. Chevallier:Jazz Pharmaceuticals: Honoraria; Daiichi Sankyo: Honoraria; Incyte: Consultancy, Honoraria. Socie:Alexion: Consultancy. Mohty:Jazz Pharmaceuticals: Honoraria, Research Funding.