Repeat Peripheral Blood Stem Cell Transplantation for Patients With Sickle Cell Disease and Falling Donor Myeloid Chimerism Levels

Nonmyeloablative allogeneic peripheral blood stem cell (PBSC) transplants are currently being investigated in phase I/II trials assessing engraftment, efficacy, and toxicity at a number of transplant centers. Our ongoing protocol for patients with severe congenital anemias, particularly sickle cell disease (SCD), and an HLA-matched sibling donor has had excellent preliminary results. None of the patients who engrafted had sickle-related events or any evidence of graft versus host disease (GVHD). There was no significant toxicity associated with the conditioning regimen. An additional protocol is ongoing for patients with high risk of graft rejection which employs pentostatin and oral cyclophosphamide (PC) pre-transplant to further deplete recipient lymphocytes in an attempt to decrease the rate of graft rejection, and to date the engraftment rate has substantially improved. Our first haploidentical transplant protocol showed improving engraftment and success rates with the addition of post-transplant cyclophosphamide, but with a median follow-up of 5 years, the disease-free survival was at best 50%. Our new haploidentical transplant protocol has had some encouraging early results in the first 5 patients transplanted.

Based on 67 patients undergoing HLA-matched sibling or haploidentical PBSC transplants at the NIH, we sought to determine what donor chimerism level is necessary to reverse SCD. Three of the patients had falling donor myeloid chimerism (DMC) levels, and when the DMC level fell below 20%, all 3 patients had return of their SCD. Our mathematical model showed that only 20% DMC (which tracks with donor erythroid chimerism) is necessary due to vast differences in donor and recipient red blood cell (RBC) survival. As all 3 patients had persistent but insufficient donor chimerism levels, we performed a PBSC boost using the same donor and busulfan/alemtuzumab conditioning. Haploidentical patients received CD34-selected PBSCs and HLA-matched sibling patients received unmanipulated PBSCs. All 3 patients now remain free of SCD with DMC levels of 100% at 1, 2.5, and 3.5 years post-transplant. Because the same donor was used and no patient experienced graft rejection, the failure of the original protocols likely was due to insufficient myelosuppression and not insufficient immunosuppression. Therefore, in this protocol, we propose repeat PBSC transplants using increased myelosupression and CD34-selected PBSCs in patients with a haploidentical donor and unmanipulated PBSCs in patients with an HLA-matched sibling donor from the same donor in a population of patients who have falling DMC and return of SCD.