INTRODUCTION: Allogeneic hematopoietic stem cell transplantation (HSCT) remains the treatment of choice for pediatric patients with high risk or refractory leukemia. In the absence of related or unrelated HLA-matched donors, alternative approaches such as HLA-haploidentical HSCT have been implemented. Our group has developed one such approach, αβ T-cell/CD19 B-cell depletion (αβhaplo-HSCT), and demonstrated its clinical efficacy (Bertaina A, Blood 2018). In αβhaplo-HSCT, anti-thymocyte globulin (ATG) is used for preventing graft rejection and graft-versus-host disease (GvHD). However, the optimal dose still needs to be elucidated. Here, we present the first analysis comparing 2 different Thymoglobulin® ATG doses: one fixed at 3.75 mg/Kg, as established in previous European studies with the closely related ATG Grafalon®, and one based on a newly developed algorithm that also integrates absolute lymphocyte count (ALC).

METHODS: Between March 2017 and April 2020, 27 pediatric patients (median age 12 years) with hematological malignancies were transplanted at Stanford University's Lucile Packard Children's Hospital. Importantly, 60% of these patients were in CR2 or greater and 2/27 had active disease at the time of HSCT. All children received a fully myeloablative conditioning regimen. On days -9/-8/-7 before αβhaplo-HSCT, patients were given a regimen of Thymoglobulin® ATG. No patient received post-HSCT pharmacological GvHD prophylaxis. The fixed dosage was given to 14 patients in cohort 1 (ATG: 3.75 mg/Kg) and the novel ALC-based dosage to 13 in cohort 2 (ATG ranging between 3-6 mg/Kg). All patients enrolled in this study (BMT Protocol 179 and 351 approved from our IRB) had a minimum follow-up of 100 days, evaluated at Day 30 and 90. Following Admiraal R (Lancet Hematology 2015), we defined CD4 immune reconstitution (IR) as more than 50 CD4 T-cells/ul twice within the first 100 days after HSCT. PBMC were enriched by Ficoll-Hypaque (Sigma Aldrich) density gradient centrifugation. Flow analyses were performed on fresh cells resuspended in PBS 2% FBS on Cytek DxP 10 flow-cytometer. At least 5x104 events of total cells were acquired and analyzed using FloJo software.

RESULTS: With a median follow-up of 555 and 124 days for cohort 1 and 2 respectively, 12 patients (85.7%) in cohort 1 and 7 patients (54%) in cohort 2 achieved CD4 immune reconstitution. In cohort 2, CD3 αβ T cells were significantly lower at both Day 30 and Day 90 (P=0.0003, Figure 1A). At Day 90, both the CD4 and CD8 subpopulations were significantly depressed (P=0.01 and P=0.056, respectively, Figure 1C). In both subpopulations, the memory compartment was the most reduced (Figure 1D). The absolute numbers of CD3 γδ T cells did not differ between the cohorts at either Day 30 or 90 (Figure 1B). Viral reactivations were higher in cohort 1 (10/14, 71%) than in cohort 2 (6/13, 46%, P=NS). Half of the reactivations were CMV reactivations, but no patient developed organ disease. There was no statistically significant difference in overall survival and the incidence of relapse in the two cohorts. Three patients developed grade III-IV aGvHD: 2 in cohort 1 (14%) and 1 in cohort 2 (8%, P=NS). Remarkably, the only patient, who developed grade IV aGvHD in cohort 2, did not experience symptoms until Adenovirus reactivation 138 days after HSCT.

CONCLUSION: Our analysis confirms that the optimal dose of ATG Thymoglobulin® before αβhaplo-HSCT remains elusive. There were no significant clinical differences between the 2 ATG regimens. However, the ALC-based regimen resulted in the more pronounced reduction of donor-derived memory T cells. Our analysis suggests two intriguing explanations for the observed pattern of results. First, the selective depletion of the memory compartment in both CD4 and CD8 T cells may well be due to a priming effect of ATG Thymoglobulin® on the few αβ T cells left over in the graft. Second, the equivalent reconstitution of naive T cells in the 2 cohorts is likely because the ATG has no impact on the thymus-dependent IR. Remarkably, in our overall cohort, 70% of the patients achieved CD4 IR by 90 days after αβhaplo-HSCT. This result is superior to the best results from other ex vivo T-cell depleted approaches 54%, recently reported by Van Roessel (Cytotherapy 2020). In vivo studies of the pharmacokinetics of ATG in αβhaplo-HSCT recipients and a comparison with the use of Grafalon® are required to shed more light on this crucial topic.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.