Immune Reconstitution and Clinical Outcome after α/β T-Cell Depleted Allogeneic Stem Cell Transplantation from Matched Related and Unrelated Donors

Introduction: The outcome of allo-SCT in patients with hematological malignancies is still hampered by GVHD and relapse. Specific depletion of αβ T- cells is proposed to result in a decreased incidence of aGVHD, whereas the remaining innate cells such as NK cells and γδT cells may provide control of infected and transformed cells the first months post SCT. This strategy has been pioneered in haploindentical transplantation with promising results. Within this study, we extend αβT- cell depleted allo-SCT to patients with a matched related and unrelated donor. The primary aim is to develop an allogeneic SCT protocol with a low incidence of aGVHD without an increased incidence of infections or relapse to serve as a platform for post-allo interventions such as a pre-emptive DLI or transfer of genetically modified T cells.

Methods: Patients with hematological malignancies (including AML, ALL, MM, NHL) who received an αβT-cell depleted allo-SCT of a HLA matched sibling (MRD) or HLA matched (9 or 10/10) unrelated donor (MUD) were analysed. αβT-cell reduction was performed by negative selection with anti-αβTCR antibodies in combination with magnetic microbeads, using the automated CliniMACS device (Miltenyi Biotec, Bergisch Gladbach, Germany). The maximal contamination with αβT-cells was 5x10⁵/kg. The conditioning regimen consisted of: ATG (Genzyme®) 6 mg/m² + fludarabine 120 mg/m² + busilvex AUC=90. Immune suppression consisted of 28 days of mycophenolic acid. A cohort of 32 patients was retrospectively analyzed for clinical parameters including immune reconstitution, engraftment, infections, GVHD, relapse, NRM and OS and compared to an historical control cohort of recipients of a T cell replete allo-SCT. In addition in a subset of patients NGS of the TCRβ chain was performed using the Illumina/MiSeq sequencing platform after isolation diverse immune subsets within the αβT-cell repertoire.

Results: The combination of ATG/fludarabine/busilvex was well tolerated with hematological recovery within 3 weeks. Primary engraftment (chimerism > 95%) was observed in all patients (n=32). Immune reconstitution primarily consisted of NK cells. In addition, γδT cells were detectable at normal numbers the first half year post SCT, whereas the adaptive immune repertoire showed a delayed reconstitution. As compared to the historical control cohort, the incidence of CMV (54% vs 38%; p = 0,48) and EBV (32% vs 9%' p=0,148) infections did not show a significant increase. The incidence of aGVHD > grade II within 100 days in patients of a αβT-cell depleted allo-SCT was 0%. During this relative short time of follow-up (1-14 months) 2 patients developed a relapse (both > 6 months) and 2 patients deceased (one with mucormycosis, one with GVHD post DLI). With NGS of the TCRβ repertoire, a surprising diversity was observed in defined immune subsets ranging from clonal expansion of regulatory T cells to broad repertoires in effector memory cells.

Conclusion: αβT-cell depletion in MRD/MUD results in a swift reconstitution of innate cells (NK cells and γδT-cells) the first 6 months post transplantation, followed by a subsequent reconstitution of the adaptive immune repertoire. The diversity appears to be different for diverse subsets of the αβT-cell repertoire, which remains to be confirmed in an extended pool of patients. The incidence of severe aGVHD is low, without a significant increase in infections or relapse shortly post allo-SCT. These results will be confirmed during extended follow-up and in a planned prospective multicenter study.