A Two Step Myeloablative Approach to Allogeneic Hematopoietic Stem Cell Transplantation (HSCT) for Hematological Malignancies From HLA Partially-Matched (Haploidentical) Related Donors.

Abstract 2292

Poster Board II-269

Lack of a matched donor remains a major obstacle to the application of allogeneic HSCT in all appropriate patients. We developed a haploidentical 2 step myeloablative transplant regimen that used Cyclophosphamide (CY) tolerization to avoid severe GVHD while still allowing rapid post-HSCT immune reconstitution. We refer to this as a two step approach because the patients receive the lymphoid and stem cell portions of the graft at different times during the transplant regimen rather than as a single transplant inoculum. Between 2006 and 2009 twenty-seven patients, median age of 52 years (range 19-67), with high risk hematological malignancies were transplanted from haploidentical donors that were mismatched for HLA-A, B, C, and DR in the GVHD direction at 4 antigens (13), 3 antigens (11), and 2 antigens (2). One patient had no mismatches in the GVHD direction but was a 3 antigen mismatch in the rejection direction. The patients were given a conditioning regimen consisting of eight fractions (total dose 12 Gy) of total body irradiation (TBI) followed immediately thereafter by a donor lymphocyte infusion (DLI) of 2 ×10e8 CD3+ donor T cells, the first step of the transplant process. Within 24 hours, this dose of T cells consistently produced an in-vivo allogeneic reaction characterized by fever (median temperature 103.8f) and in many cases rash and diarrhea. Skin biopsies performed on 2 of the patients with rash were consistent with GVHD. After a median time of 63 hours (range 60-66 hours) following the DLI, CY (60 mg/kg/day) was given for 2 days to eliminate alloactivated donor and host T cells. The symptoms caused by the DLI typically disappeared 24 hours after completion of CY. The second step of the transplant occurred when a CD34 selected HSC product from the same donor was infused 24 hours after the end of the infusion of CY. GM-CSF was used post HSCT to accelerate white cell recovery and to promote a TH1 type immune response. Tacrolimus and mycophenylate mofetil were used for GVHD prophylaxis. Donor apheresis was performed over 2 days for lymphocyte collection followed by administration of G-CSF for 5 days and an additional 2 aphereses for HPC collection on days 4 and 5 of G-CSF administration. Fifteen of 27 patients (56%) are alive without evidence of their disease 3 to 32 months past HSCT. One additional patient is alive but with relapsed disease. Kaplan-Meier estimates of survival are 70% in patients without disease at the time of HSCT and 43% in patients not in remission at HSCT. Eleven patients (41%) died, 5 from relapsed disease, 3 from toxicity, and 3 from infection. Of the 6 relapses, 5 occurred in mothers who received transplants from their children. There were no deaths attributable to GVHD and only 1 brief occurrence of severe (grade III gut) GVHD. Fifteen patients (55%) developed grade 1-2 GVHD, mostly of the skin and easily controlled with steroids in 11 or with steroids plus photopheresis in 4 patients. Three patients developed limited chronic GVHD, one of whom is off immunosuppressive therapy. Immune reconstitution was brisk. Patients typically reached CD3+CD4+ counts of >100 cells/ul within a few months of discontinuation of immunosuppressive therapy. Two multiparous females with pre-existing anti-donor antibodies experienced humorally mediated rejections. Both died of toxicities related to a second HSCT. The other toxic death occurred in the patient with no mismatches in the GVHD direction who experienced a flare of his Crohn's disease shortly before conditioning. Three infectious deaths occurred, one due to progression of preexisting aspergillus lung disease and bacteremia, 1 due to RSV pneumonia, and 1 due to brain abscess.

This 2 step technique; 1) allows for the administration of a prescribed amount of tolerized lymphocytes in an effort to promote consistent immunologic outcomes post transplant, 2) prevents exposure of the donor HSC to CY, 3) prevents exposure of the donor lymphocytes to G-CSF thus avoiding skewing to a TH2 T cell phenotype, and 4) allows for a greater separation between TBI and CY in the conditioning regimen which may help decrease regimen related toxicity. The high overall survival rate in patients in remission at the time of transplant illustrates the importance of early identification of patients for haploidentical HSCT who are without well matched donors. These encouraging clinical outcomes suggest that this novel 2 step approach to HSCT should be further explored.