Abstract
Murine models of transplantation established that nonmyeloablative conditioning using repeated low doses of irradiation targeted to lymphoid tissues (TLI) and depletive anti-T cell antibodies protects against GVHD by skewing residual host T cell subsets to favor regulatory natural killer (NK) T cells that suppress GVHD by polarizing donor T cells toward secretion of non-inflammatory cytokines such as IL-4. We recently translated the murine protocol to a clinical study using non-myeloablative TLI and ATG host conditioning with HLA matched related and unrelated donors, and showed a marked reduction in the incidence of acute GVHD while retaining graft anti-tumor activity (Lowsky et al., in Press NEJM). Engrafted donor CD4+ T cells showed a marked increase in IL-4 production as compared to CD4+ T cells from controls. We now adapted the TLI and ATG nonmyeloablative host conditioning regimen to a clinical study of allogeneic HCT using haploidentical matched (3/6 HLA matched) related donors to determine if it will result in donor hematopoietic cell engraftment and also protect against acute GVHD. Blood derived hematopoietic progenitor cells were collected by apheresis from donors mobilized with G-CSF and the product was T cell depleted using CD34+ selection. CD3+ T cells were added back to the donor inoculum according to a dose escalation schedule. The initial T cell dose was 1 x105 CD3+ cell/kg with designated increments based on clinical outcomes of up to a maximum of 1 x107 CD3+ cells/kg. The desired CD34+ cell dose was >5 x 106 CD34+ cells/kg for all patients. Seven patients were transplanted; the median age was 53 years (range 27 to 61 years). Five patients had acute myelogenous leukemia, two with disease in remission and three not in remission at the start of TLI and ATG, one with myelodysplastic syndrome, and one with progressive peripheral T cell lymphoma. The median follow-up for all patients is 265 days with three of seven patients alive and free of disease at the last observation period. Sustained donor hematopoietic cell engraftment was achieved in three of three patients only after the T cell dose was increased to 1 x107 CD3+ cells/kg. No patient developed acute GVHD. None of the three patients receiving the highest dose of T cells had any invasive fungal or viral infections. Monitoring of sorted host T cell subsets before TLI and ATG, and immediately after but before the infusion of donor cells, revealed in five of five patients a highly significant skewing of residual host T cells favoring invariant NK T (CD3+ CD161hi Va24 +Vb11 +) cells. The mean absolute number of host CD3+, and CD4+ and CD8+ T cells decreased by 99, 163 and 121 fold, respectively, immediately after conditioning compared to the absolute numbers before the start of TLI and ATG, whereas the mean absolute number of invariant NK T cells decreased by only 11%. In conclusion, we have determined the conditions for successful hematopoietic cell engraftment using a non-myeloablative regimen of TLI and ATG that appears associated with a reduced aGVHD risk yet retained graft anti-tumor activity. As in the pre-clinical model, we show direct evidence that the low incidence of aGVHD is associated with a significant alteration in residual host T cell subsets markedly favoring invariant NK T cells.
Author notes
Corresponding author