Profound T Cell Depletion Does Not Eradicate CMV Specific CD8 T Cells Responsible for Protective Immunity to CMV after T Cell Depleted Allogeneic Stem Cell Transplantation.

Abstract

CMV infection is inversely correlated with T cell immunity and frequently observed after T cell depleted allogeneic stem cell transplantation (TCD alloSCT). In case of profound T cell depletion of the graft, it may be anticipated that donor immunity is not of influence on immunity to CMV after TCD alloSCT, in contrast to conventional alloSCT. To address this hypothesis, we prospectively analyzed CMV load and T cell reconstitution in 20 consecutive CMV+ patients transplanted with a CMV+ or CMV− donor. The grafts were T cell depleted using Campath 20 mg “in the bag” and infused without further manipulation. No post transplantation GVHD prophylaxis had to be given, illustrating the intensity of T cell depletion. CMV DNA load was weekly monitored. As expected, by day 120 no difference in incidence of CMV reactivation was present between CMV+ patients with CMV+ or CMV− donors (8/9 vs 11/11). However, the area under the curve (AUC) representing total CMV DNA load was significantly higher in patients with a CMV− donor (AUC log 6.4 vs 5.5, p=0.03). 2 patients with a CMV− donor developed fatal CMV pneumonitis, whereas no CMV disease was present in patients with a CMV+ donor. To analyze whether this difference in the course of CMV infection was due to a difference in CMV specific T cell reconstitution, we characterized T cells populations after transplantation. At day 90, total T cell numbers in CMV+ patients were significantly increased in patients with a CMV+ donor compared to patients with a CMV− donor (1,2 vs 0.5*10E9/L, p=0.025). This difference was caused by an increase in CD8 T cells (1.0 vs 0.3*10E9/L, p=0.009) but not CD4 T cells (0.2 vs 0.2*10E9/L). To investigate whether this difference was due to expansion of CMV specific T cells, we analyzed 16 patients for whom HLA restricted CMV specific tetramers were available. All patients with a CMV+ donor developed a substantial tetramer positive (tet+) CD8 T cell response (>5*10E6/L) compared to only 33% of patients with a CMV− donor (7/7 vs 3/9, p=0.006). 13% (range 2 – 26%) of the CD8 T cells were tet+ in patients with a CMV+ donor at the peak of the response. This percentage does not include CD8 T cell responses directed against CMV epitopes with other HLA restriction. When extrapolated, this suggests that the increase of CD8 population was largely due to CMV specific T cells. In patients with a CMV− donor the CMV specific CD8 response could have developed from naive donor cells or from residual patient memory cells. We determined the origin of tet+ T cells in 3 patients with a CMV− donor. In all patients, the tet+ cells were of patient origin, indicating absence of a de novo primary donor response. In summary, although the incidence of CMV reactivation was identical in TCD alloSCT with a CMV+ or a CMV− donor, CMV load was lower in patients with a CMV+ donor. In CMV+ patients with a CMV+ donor a CMV specific CD8 response was always present, in contrast to most CMV+ patients with a CMV− donor. Furthermore, when tet+ cells developed early after SCT in CMV+ patients with a CMV− donor, they were of patient origin. We conclude that no primary CMV specific CD8 T cell response of donor origin develops shortly after TCD alloSCT, and that despite profound T cell depletion CMV specific memory CD8 T cells from donor or patient survive the conditioning regimen and provide protective immunity to CMV.