Follow Up On CD8-Depleted Donor-Lymphocyte Infusions After T-Cell Depleted Allogeneic Hemopoietic Stem Cell Transplantation

We applied prophylactic CD8-depleted (CD8^depl) donor lymphocyte infusions (DLI) in the setting of T-cell depleted reduced-intensity allogeneic hematopoietic stem cell transplantation (HSCT) in a phase I/II trial. T-cell depletion was carried out by the use of high-dose Alemtuzumab (100 mg or 60 mg for unrelated or sibling donor transplantation, respectively). We have demonstrated the feasibility of this approach after having treated 23 patients in this protocol (Meyer et al. Blood 2007). From 2004 to 2011, 134 patients with different hematologic diseases were included and followed for a median observation time of 1.5 years after transplantation (range, 1.5-9 years). Median age was 56 years (range, 19-71). Stem cell source was peripheral blood from either matched siblings (n=23), matched unrelated (n=62), or donors with single HLA mismatches (unrelated n=48, related n=1). Tapering of cyclosporine A was started in the 6^th week after transplantation. Subsequently, CD8^depl DLI were administered prophylactically in escalating doses starting with 1x10⁶ CD4 T cells/kg bodyweight. 53 patients received at least one dose of DLI. Among 81 patients who did not qualify for DLI, 60 patients had primary mild GVHD. Following DLI, acute GVHD was the major reason for withholding subsequent DLI-doses (66%), mainly because of acute GVHD ≤ 2°. Extensive chronic GVHD was diagnosed in 9.4% of the patients. Overall survival after 1 and 3 years was 60% and 43%, respectively. Survival significantly differed between the DLI and non DLI group after 3 years (63.8% vs. 30.1%, p=0.002). The inferior outcome of the non DLI group was confirmed when only those patients who did not receive DLI for other reasons than GVHD were considered (28.6%). The largest patient cohort in our trial were patients with AML and MDS (n=21 with and n=33 without DLI). In these patients, the survival benefit for the DLI group after 3 years was significant (75.8% vs 28.9%, p=0.0012). Interestingly, the relapse rate did not differ between DLI and non-DLI patients but the non-relapse mortality. The presence of GVHD at any time was associated with a reduced relapse rate (57.9% vs. 31.4%, p=0.0015), independent of DLI-application. However, GVHD had no impact on overall survival. Decreasing donor T-cell chimerism (TCC) was found in 34 patients who subsequently received DLI and 13 who did not. Following CD8^depl DLI, 31 patients (88%) converted to full donor. In contrast, only 2 of the patients with decreasing TCC in the non-DLI group (15.3%) converted spontaneously.

In summary, we observed that the application of prophylactic CD8^depl DLI was associated with a survival benefit and the re-establishment of full donor T-cell chimerism. In AML/MDS patients the improved survival was associated with lower treatment-related mortality rather than a reduced relapse-rate. Our data strongly suggest randomized trials comparing prophylactic application of CD8depl DLI vs. no DLI as well as CD8depl vs. non-manipulated DLI in a preemptive setting.