Search

The current approach for preventing hemolysis of red blood cells (RBCs) in major ABO-incompatible bone marrow (BM) grafts after infusion is to deplete RBCs from BM products before transplantation. Traditionally, manual density separation (MDS) using Ficoll-Hypaque (Cytiva Sweden AB, Uppsala, Sweden has been used to accomplish RBC depletion. This process yields good CD34+ cell recovery, but it requires open manipulation and is labor-intensive and time-consuming. We hypothesized that an alternative automated method using Haemonetics Cell Saver 5+ (Haemonetics Corporation, Boston, MA, USA) would offer equivalent RBC depletion and CD34+ cell recovery. Small marrow volumes from pediatric donors can be processed using Cell Saver (CS) without adding the third-party RBCs necessary for other automated methods. This retrospective analysis comprised data from 58 allogeneic BM grafts. RBC depletion and CD34+ cell recovery from BM using MDS (35 grafts) were compared with CS (14 grafts). Nine products underwent RBC depletion using CS with Ficoll (CS-F) when RBC volume was less than 125 mL. Linear regression analysis of log transformation of CD34+ cell recovery adjusted for log transformation of both baseline CD34+ cell content and baseline total volume showed no significant difference between MDS and CS (estimated coefficient, –0.121, P = 0.096). All products contained an RBC volume of less than 0.25 mL/kg post-processing. CD34+ cell recovery with CS-F was comparable to MDS and CS and suitable for pediatric recipients of allogeneic hematopoietic cell transplantation. We provide evidence that an automated method using Haemonetics Cell Saver 5+ achieves RBC depletion and CD34+ cell recovery comparable to MDS when adjusting for baseline factors. [ABSTRACT FROM AUTHOR]

Disease relapse remains a major cause of treatment failure in pediatric patients receiving allogeneic hematopoietic cell transplantation (HCT) for high-risk acute leukemias (AL) or myelodysplastic syndrome (MDS). Comprehensive data on prognostic factors and outcomes after post-transplant relapse in pediatric patients is lacking. The goal of this study was to identify factors associated with survival following post-transplant relapse. We performed a retrospective study of pediatric patients who experienced disease relapse following their 1st HCT for AL or MDS between 1990 – 2018 at our institution. Patient, disease and transplant-related characteristics were extracted from the medical record. Overall survival (OS) was estimated by the Kaplan-Meier (KM) method and compared using the log-rank test. Cox and logistic regression were performed to identify factors statistically associated with outcomes after post-transplant relapse. During the study period, 703 patients received 1st HCT, of which 221 (31%) relapsed at a median of 3.9 months after HCT. Of these relapsed patients, 60% (n = 132) received some chemotherapy or supportive care after post-transplant relapse, 13% (n = 28) received donor lymphocyte infusion only and 28% (n = 61) were able to proceed to a 2nd HCT. Patients who received a 2nd HCT were more likely to have been in first remission at the time of 1st HCT (P = 0.02), experienced relapse more than 6 months after 1st HCT (P < 0.001), experienced acute or chronic GVHD (P = 0.05 and 0.01, respectively) and relapsed in the current decade (2011 – 2018) (P = 0.001). Of the 61 patients who received a 2nd HCT, 18 (30%) were alive at last follow-up, compared to 7 (4%) of 160 patients who had received other therapies. In a multivariable model, longer interval from HCT to relapse (P < 0.001), being able to receive a 2nd HCT (P < 0.001) and receiving a HCT in the current decade (P < 0.001) were found to be significantly associated with OS (Fig 1 & 2). Factors that are typically associated with poor outcomes after HCT, such as disease status or type of HCT, donor and graft source were not predictive of survival following post-transplant relapse. Duration of remission before post-transplant relapse was identified as the strongest independent predictor of a patient's ability to receive a 2nd HCT as well as survival after relapse. The 3-year OS for patients who received a 2nd HCT was 6 times higher than that for patients who did not receive a 2nd HCT. Although the outcomes after 2nd HCT are not commendable, they are improving with time. Our data supports the argument that, when feasible, pediatric patients experiencing post-transplant relapse should be considered for 2nd HCT. [ABSTRACT FROM AUTHOR]

Donor T-cell reconstitution following hematopoietic cell transplantation (HCT) is critical for infection and leukemic control. Donor lymphocyte infusions (DLI) provide additional donor T cells, but may cause graft-versus-host-disease (GVHD) from alloreactive T-cells. As GVHD risk increases with increasing HLA-mismatch, conventional haploidentical DLI doses are often limited to 1 × 105 T cells/kg. Since the majority of alloreactive T cells reside in the naïve (CD45RA-positive) subset, CD45RA-depleted DLI may provide a diverse lymphocyte repertoire to aid in infection and leukemic disease control, with decreased GVHD risk. Herein, we report on feasibility, safety, and early efficacy of escalating doses of haploidentical memory DLI. Donors underwent leukapheresis without mobilization; 2 total blood volumes were collected. Depletion of CD45RA-positive cells was performed using the CliniMACS device (Miltenyi Biotec). DLI dosing typically started at 1 × 105 T cells/kg, with dose escalation every 4 weeks based on clinical need. All donors collected successfully in one attempt. Following CD45RA-depletion, median residual T cell dose was 45 × 106/kg and CD45RA-positive T cell dose was <0.001 × 106/kg (>3.9 log depletion), allowing multiple doses to be stored for future use (Figure 1). Twelve patients received a total of 29 DLIs (median infusions=2, range 1–5; Figure 2). Indications for DLI included detectible leukemia (6), mixed chimerism (6), and lymphopenia/viral infection (3); 3 recipients had multiple indications. First DLI was infused a median of 101.5 days post-HCT (range 41-356). Twenty-three doses had no detectible CD45RA-positive T cells (<0.001 × 106/kg) and 6 contained 0.001-0.002 × 106 CD45RA-positive T cells/kg. No infusion related acute adverse events occurred. One patient developed acute GVHD (clinical grade IV, stage 3 skin/stage 4 gut), 21 days post-DLI (dose=1 × 105 T cells/kg). After DLI, the majority of patients showed some clinical response (chimerism improvement/stability, decreased viral load, and/or rise in peripheral CD3 count). Eradication of detectable leukemia however was not successful, with all 6 patients eventually having disease progression (1 rapid progression; 5 frank relapse 2–6 months after first DLI). With median length of follow-up after last DLI of 243 days (range 20-924), no patients have developed chronic GVHD. CD45RA-depletion is feasible and effective in removing naïve donor T-cells. Haploidentical CD45RA-depleted DLIs were well tolerated and associated with low incidence of acute GVHD, even with T-cell doses as high as 10 × 106/kg. The majority of patients showed clinical benefit from DLI, with the exception of leukemia disease control. We are currently further investigating the efficacy and safety of CD45RA-depleted DLI as part of a prospective clinical haploidentical HCT trial. [ABSTRACT FROM AUTHOR]