Purpose: To evaluate T cell recovery and donor chimerism following haploidentical hematopoietic cell transplantation (HCT) with a non-myeloablative conditioning approach that includes T cell depletion of host and donor and delayed DLI.

Methods: Eighteen patients, 3 cohorts of 4 patients each and 1 cohort of 6 evaluable patients/10 transplanted, with chemorefractory hematologic malignancies, received related HLA 1–3 of 6, A, B, or DR antigen mismatched donor HCT after non-myeloablative conditioning with Medi-507 (anti-CD2 humanized mAb; Biotransplant, Inc.), cyclophosphamide, thymic irradiation and peritransplant cyclosporine. The patients in Protocols A received a MEDI-507 test dose of 0.1 mg/kg on Day -2 followed by 0.6 mg/kg on Days −1, 0 and +1 and transplantation of unmanipulated bone marrow. In Protocol B, the timing and dose of Medi-507 was modified. The patients in Protocol C and D received the latter Medi-507 protocol, but were given Isolex ®-selected CD34+ cells from G-CSF mobilized PBSC. Protocol D differs from Protocol C with the addition of fludarabine to more reliably achieve sustained chimerism. Donor leukocyte infusions were administered in an effort to convert mixed to full donor chimerism and to achieve a graft-versus-tumor effect. Chimerism was measured by peripheral blood microsatellite markers or by flow cytometry using HLA-specific mAbs. T cell recovery and phenotype were followed by flow cytometry. Because a high percentage of CD4 T cells post- transplant were CD25high, we performed quantitative RTPCR for Foxp3 and CTLA-4 on sorted PBMC populations.

Results: T cell depletion early post-HCT was detected in all patients. There was a marked difference in the percentage of graft acceptance/loss, GVHD prevalence, and T cell phenotype related to each protocol modification. The majority (>90%) of CD4 T cells appearing in the first 100 days post-SCT were CD45RO+/CD45RA- “memory” cells and CD8 T cells were CD45RO+CD45RA-/CD62L-. In addition, a remarkably high percentage (19.5–75%, mean 38.1%) of CD4 T cells expressed high levels of CD25 in recipients of Protocols A, B, and C early post-HCT. CD25 expression decreased as T cell counts increased. By quantitative RTPCR, we found that sorted CD25highCD4 T cells expressed Foxp3 and CTLA-4, consistent with a regulatory phenotype. The addition of fludarabine in Protocol D resulted in an overall reduction in the percentage of peripheral CD4CD25high T cells compared to Protocol C at 4 weeks post-HCT (C 23.43% +/−4.7% versus D 2.1% +/− 0.9%, p<0.00005). The addition of fludarabine improved sustained engraftment from 2/4 in Protocol C to 5/6 in Protocol D. In Protocol C, T cell chimerism conversion occurred following DLI in 2/4 patients with no or grade II skin-limited GVHD. In Protocol D, full or near full donor chimerism was achieved in 5/6 patients, 3 spontaneously and 2 following DLI, and grade I–IV GVHD developed in 5/6 patients.

Conclusions: These data suggest that the addition of fludarabine may increase the incidence of sustained and full donor chimerism. Fludarabine efficiently depleted host CD4+CD25high Foxp3+ cells, which may have a regulatory role, preventing spontaneous chimerism conversion and associated GVHD.

Author notes

Corresponding author