LAUSR

Introduction Despite major advances in hematopoietic cell transplant (HCT), 10-50% of patients relapse following allogeneic HCT. Donor lymphocyte infusion (DLI) is one treatment strategy for relapse that potentially provides a graft versus malignancy therapeutic effect. A major complication of DLI is graft versus host disease (GVHD), with incidences ranging from 40-60%. Risk of developing GVHD is related to multiple factors with DLI cell dose speculated to have the strongest correlation. Conversely, too small of a dose can allow disease progression. The literature is quite heterogeneous regarding dosing and consensus data are lacking. The University of North Carolina Bone Marrow Transplant and Cellular Therapy Program (BMT CTP) has a standard operating policy surrounding DLI dosing and timing. As part of our quality audit, we evaluated policy compliance and outcomes to determine if policy changes were needed. Methods We evaluated HCT recipients with Acute Myeloid Leukemia (AML) and myelodysplastic syndrome (MDS) who received at least one DLI between 2001 and 2016. During these 15 years, 282 patients received an allogeneic HCT for AML and MDS. 42 patients with AML (15%) received at least 1 DLI for a total of 59 DLIs and 14 patients with MDS (5%) received at least 1 DLI for a total of 21 DLIs. 4 patients with AML and 1 with MDS received a DLI for reasons not included in the policy and were excluded. The final population included 38 AML patients receiving 54 DLIs and 13 MDS patients receiving 15 DLIs for a total of 51 patients receiving 69 DLIs. The BMT CTP DLI policy outlines the recommended cell dose to be administered (Figure 1). Data collected included disease, donor type, recipient age, graft, reason for DLI, cell dose infused, survival, cause of death, GVHD occurrence after DLI, and if DLI algorithm was followed. Results Figure 2 shows patient characteristics and outcomes following DLI. Time from HCT to first DLI for AML patients was a median of 176.5 days (range 71 days – 10 years) and for MDS, 137 days (range 78 days - > 3 years). DLI dosing ranged from 1 × 106 to 5 × 107 CD3 cells/kg for AML and 1 × 106 to 1.2 × 107 CD3 cells/kg for MDS. For those with follow-up data, 50% (25/50) developed GVHD following DLI and 28.6% (14/49) were alive at a median of 6.2 years after last DLI. For both cohorts, dosing algorithm was followed about 50% of the time. Conclusion While our GVHD occurrence following DLI is within published data range, AML relapse remains high. Heterogeneity in dose and disease precludes precise identification of prognostic factors for successful outcomes, yet these data support the role of DLI in contributing to the survival of a subset of patients following HCT. However, improved uniformity in dosing may aid in understanding appropriate dosing. We have recently made modifications to our current algorithm with a goal that increased consistency will allow systematic evaluation of outcomes and improved care.