Influence of Graft Versus-Host Disease Prophylaxis Regimen On T-Cell Repertoire Diversity Following Reduced-Intensity HLA-Matched Unrelated Donor Allogeneic Hematopoietic Stem Cell Transplantation.

A clearly superior graft-versus-host disease (GVHD) prophylaxis regimen has not been established for patients undergoing reduced intensity allogeneic hematopoetic stem cell transplantation (HSCT) from matched unrelated donors (URD). Encouraging results have been reported with both the combination of alemtuzumab and cyclosporine (AC) and the regimen of tacrolimus, methotrexate, and sirolimus (TMS) in the URD setting. These two regimens work by biologically distinct mechanisms and may have markedly different effects on immune reconstitution. T-cell receptor (TCR) spectratyping analysis, which provides information on antigen receptor diversity, is a valuable method for monitoring post-transplant immune reconstitution. As part of a randomized pilot study, we prospectively assessed the effects of AC vs. TMS on TCR Vb repertoire diversity in patients undergoing reduced intensity HLA-matched unrelated donor transplantation.

Twenty patients (median age 53 yrs; range 24–70 yrs) with hematologic malignancies received reduced intensity conditioning (fludarabine 30 mg/m²/day and cyclophosphamide 1200 mg/m²/day IV Day -6 to -3) followed by a 10/10 HLA-matched unrelated donor T-cell replete mobilized peripheral blood allograft. Patients were randomized to receive either: AC (n=10): alemtuzumab 20 mg/day IV over 8 hours Days -8 to -4 and cyclosporine starting at Day -1 with a 10% per week taper starting at Day +100 or TMS (n=10): tacrolimus and sirolimus starting at Day -3 with a 33% taper at Day +63 and Day +119 and methotrexate 5 mg/m² IV, Days +1, +3, +6, and +11. Blood samples were collected from the donor and patient at baseline and the patient at 1, 3, 6 and 12 months post-transplant for TCR spectratyping analysis. All comparisons are based on an exact Wilcoxon rank sum test; p values < 0.01 were significant because of multiple comparisons.

Patients on the AC arm had significantly fewer T-cells on Day +14 compared with the TMS arm (median CD3⁺ = 1 cells/μl vs 356 cells/μl; CD4⁺ = 0 cells/μl vs 243 cells/μl; CD8⁺ = 0 cells/μl vs. 59 cells/μl; each p<0.0001); there was less disparity at Day +28 (median CD3⁺ = 45 cells/μl vs. 398 cells/μl; CD4⁺ = 36 cells/μl vs. 218 cells/μl; CD8⁺= 5 cells/μl vs 152 cells/μl; each p 0.002). By Day +100, lymphocyte recovery was not appreciably different between the two arms (median CD3⁺ = 242 cells/μl vs. 445 cells/μl (p = 0.095): CD4⁺ = 106 cells/μl vs. 212 cells/μl (p=0.28); CD8⁺ = 72 cells/μl vs. 135 cells/μl (p = 0.03). NK-cell recovery was slightly less in the AC vs. TMS arm at Day +14 (median NK = 27 cells/μl vs. 70 cells/μl; p = 0.01) and at Day +28 (median NK = 29 cells/μl vs. 150 cells/μl; p=0.02). There was no difference by Day +100 (median NK = 124 cells/μl vs. 88 cells/μl; p=0.31). B-cell reconstitution was negligible in both arms through Day +100. Assessment of CD4⁺ TCR Vb repertoire diversity by spectratyping demonstrated significantly lower diversity in patients receiving AC at 1 (p = 0.0003), 3 (p = 0.0003) and 6 (p=0.003) months post transplant compared with patients receiving TMS. CD8⁺ TCR spectratyping similarly revealed significantly reduced diversity in the AC arm at 3 (p = 0.001) and at 6 months (p = 0.003), and a trend toward significance at 12 months (p = 0.07). On each of the 2 arms, 2 of 10 patients developed acute Grade II-IV GVHD. Of the 5 patients on the AC arm who were seropositive for CMV, all 5 reactivated CMV by PCR within the first 60 days and reactivated 2–5 times in the first year. In contrast, only 3 of 5 seropositive patients reactivated CMV on the TMS arm and only one reactivated in the first 60 days.

Two factors may have contributed to the loss of repertoire diversity in the AC arm. First, the alemtuzumab regimen may have severely depleted the infused donor T-cells. Second, stimulation by reactivating virus may have induced expansion of CMV-specific memory and effector T-cells, resulting in a skewed and oligoclonal T-cell repertoire. Especially in CD8⁺ T-cells, CMV has been shown to produce significant oligoclonal expansion (including CD4⁺: CD8⁺ ratio inversion). The loss of T-cell numbers and repertoire may in turn have contributed to the prevalence of early CMV reactivation. Thus, despite the similarities in frequency of acute GVHD in this small sample, it appears that these two commonly used GVHD prophylaxis regimens have very different effects on post-transplant immune reconstitution in the first 6 months after allogeneic HSCT.

No relevant conflicts of interest to declare.