Two Nonmyeloablative HLA-Matched Related Donor Allogeneic Hematopoietic Cell Transplantation Regimens in Patients with Severe Sickle Cell Disease

Abstract

Nonmyeloablative (NMA) conditioning is being used increasingly with success in matched related donor (MRD) and alternative donor allogeneic hematopoietic cell transplantation (allo-HCT) in individuals with sickle cell disease (SCD). Advantages include decrease toxicity and applicability in patients otherwise unable to tolerate conditioning regimens due to end-organ damage or age. We aimed to add to published data outcomes of two similar NMA conditioning protocols, termed Protocol 1 (ClinicalTrials.gov ID NCT00061568) and Protocol 2 (ClinicalTrials.gov ID: NCT02105766)) in mainly adult patients with SCD to evaluate the safety, toxicity, and success of these regimens in individuals at high-risk for poor transplantation-related outcomes. We also evaluated the tolerability and outcomes of Protocol 2, which included preconditioning immunodepletion, in patients at even higher risk of T cell-mediated rejection or plasma/B cell-mediated anti-donor erythrocyte antibody production-the latter due to ABO incompatibility or recipient RBC alloimmunization to a donor antigen. Finally, we evaluated the incidence and trajectory of mixed donor myeloid chimerism over time following allo-HCT. In this retrospective analysis of the 2 prospective phase 2 NMA transplant protocols, 91 individuals with SCD or transfusion-dependent β-thalassemia underwent MRD allo-HCT at the National Heart, Lung, and Blood Institute; regimens contained alemtuzumab, low-dose radiation, and sirolimus for graft-versus-host disease (GVHD) prophylaxis with or without preconditioning immunodepletion with pentostatin and oral cyclophosphamide (Protocol 2). In the total cohort of 91 transplantation recipients, outcomes were favorable with timely neutrophil and platelet engraftment (median, 21 days [range, 7 to 67 days] and 21 days [range, 10 to 112 days], respectively), minimal high-grade acute GVHD and no chronic GVHD, overall survival of 90%, sickle-free survival of 85%, and mixed donor myeloid chimerism in 43% at a median follow up of 7.3 years (range, 0.8 to 20 years). Most patients with mixed myeloid chimerism at 2-years post-HCT remained stable in their values. In analyzing each protocol separately, outcomes were comparable except for higher cytomegalovirus reactivation necessitating treatment in Protocol 2 without an associated increase in graft failure. In the combined cohort, graft failure occurred in 11 patients, and hematologic malignancy or abnormal cytogenetics on bone marrow evaluation developed in 7 patients. In a subanalysis of factors that may implicate transplantation outcomes, the number of RBC units transfused post-HCT was significantly higher in recipients with pre-HCT history of alloimmunization to donor RBC antigens. There was no difference in the number of RBC units transfused, duration of transfusion, or red cell engraftment in those with major ABO incompatibility; preconditioning immunodepletion and pretreatment with rituximab likely were helpful. Both NMA allo-HCT protocols were successful in achieving adequate engraftment and sickle-free survival with minimal toxicity, including in individuals with mixed donor myeloid chimerism. The addition of preconditioning immunodepletion was well-tolerated and reduced the rate of graft failure in high-risk recipients.

Keywords: Alemtuzumab; Allogeneic hematopoietic cell transplantation; Immunodepletion; Mixed myeloid chimerism; Nonmyeloablative; Sickle cell disease; Sirolimus.

Figure 1.. Mean Sorted Peripheral Blood (PB) Chimerism and Survival Outcomes by Protocol.

(A) Mean peripheral blood chimerism. Solid-filled bars represent myeloid (CD14+/15+) lineage, while diagonally-filled bars depict the lymphoid (CD3+) lineage. Protocol 1 is in black and Protocol 2 in purple/blue. P-values indicate significant differences between Protocol 1 and 2. Error bars show standard error of mean (SEM). (B) Overall Survival (OS) using last follow-up (defined as the last encounter with available laboratory results, including chimerism analysis). (C) Disease Free Survival (DFS). (D) GVHD-Free/Relapse-Free Survival (GRFS). Survival analyses were by Kaplan-Meier, P <0.05 are considered significant. Protocol 1 is in black and Protocol 2 in dotted blue.

Figure 2.. Trajectory of Average Myeloid and Lymphoid PB Chimerism by Myeloid Chimerism Category.

Chimerism values were analyzed in the combined NMA allo-HCT cohort (i.e., 91 recipients). Chimerism categories were based on levels of donor myeloid chimerism at 2-years post-HCT. Error bars represent standard error of the mean (SEM); Sem was narrow in full donor myeloid chimerism. Myeloid chimerism is in black and lymphoid chimerism is in blue. (A) Full Donor Myeloid Chimerism (>95%), n=54. (B) Stable Mixed Myeloid Chimerism (5–95%), n=14. (C) Dropping Mixed Myeloid Chimerism (5–95%), n=11. Patients with graft failure by 2-year post-HCT (n=9), patients without follow up beyond 2-years (n=1), and patients with unsorted PB mixed chimerism (n=2) were excluded from the analysis.

Figure 3.. RBC and Platelet Transfusions at 1-Year Post-HCT By ABO Incompatibility and Pre-HCT Recipient RBC Alloimmunization Status.

Horizontal lines in the boxes represent medians and plus signs show means. Bars depicts min and max values. P-values are shared if significant. The dotted vertical line in each window separates general yes/no categories with subcategories within individuals with ABO incompatibility or alloimmunization; the ‘no’ categories are the same in each respective window. There were two patients in the alloimmunized cohort that could not be subcategorized due to unidentified antigen and/or lack of information on antigen presence in the donor (A) Number of RBC units transfused by ABO category, further separated by ABO mismatch subcategories. (B) Number of platelet transfusions by ABO category, further separated by ABO mismatch subcategories. (C) Number of RBC units transfused by alloimmunization status, further analyzed by the presence of the cognate antigen in the donor. (D) Number of platelet transfusions by alloimmunization status.

Figure 4.. Post-HCT Hematologic Malignancy.

Hematologic malignancy is separated by type of disease and whether it meets criteria for malignancy (light blue) versus abnormal cytogenetics without associated peripheral BP changes (light pink). Gender, age at transplant, and protocol are shared on the Y-axis. Patients with mixed chimerism (MC) at malignancy diagnosis are noted at the end of the respective swimmer’s plot. Events (age at malignancy diagnosis, graft failure, or death) relative to HCT are depicted on the Y-axis. Time 0 = allo-HCT day.