Pentostatin Is Advantageous Relative to Fludarabine for in Vivo Murine T Cell Depletion, Suppression of T Cell Cytokine Secretion, and Inhibition of HVGR

Although fludarabine and pentostatin are variably utilized for conditioning prior to clinical allogeneic transplantation, limited data exists with respect to their relative efficacy in terms of host immune T cell depletion and T cell suppression. To directly compare these agents in vivo in a murine model, we compared a regimen of fludarabine plus cyclophosphamide (FC) similar to one that we previously developed (Petrus et al, BBMT, 2000) to a new regimen of pentostatin plus cyclophosphamide (PC). Cohorts of mice (n=5–10) received a three-day regimen consisting of P alone (1 mg/kg/d), F alone (100 mg/kg/d), C alone (50 mg/kg/d), or combination PC or FC. Similar to our previous data, administration of P, F, or C alone yielded minimal host T cell depletion (as measured by enumeration of splenic CD4+ and CD8+ T cells) and minimal T cell suppression (as determined by CD3, CD28 co-stimulation of a constant number of remaining splenic T cells and measuring resultant cytokine secretion by multi-analyte assay). The PC and FC regimens were similar in terms of myeloid suppression (p=.2). However, the PC regimen was more potent in terms of depleting host CD4+ T cells (remaining host CD4 number [× 10^6/spleen], 2.1±0.3 [PC] vs. 4.4±0.6 [FC], p<0.01) and CD8+ T cells (remaining host CD8 number, 1.7±0.2 [PC] vs. 2.4±0.5 [FC], p<0.01). Moreover, the PC regimen yielded greater T cell immune suppression than the FC regimen (cytokine values are pg/ml/0.5×10^6 cells/ml; all comparisons p<0.05) with respect to capacity to secrete IFN-γ (13±5 [PC] vs. 48±12 [FC]), IL-2 (59±44 [PC] vs. 258±32 [FC]), IL-4 (34±10 [PC] vs. 104±12 [FC]), and IL-10 (15±3 [PC] vs. 34±5 [FC]). In light of this differential in both immune T cell depletion and suppression of T cell effector function, we hypothesized that T cells from PC-treated recipients would have reduced capacity to mediate a host-versus-graft rejection response (HVGR) relative to FC-treated recipients. To directly test this hypothesis, we utilized a host T cell add-back model of rejection whereby BALB/c hosts were lethally irradiated (1050 cGy; day -2), reconstituted with host-type T cells from PC- or FC-treated recipients (day -1; 0.1 × 10^6 T cells transferred), and finally challenged with fully MHC-disparate transplantation (B6 donor bone marrow cells, 10 × 10^6 cells; day 0). In vivo HVGR was quantified by the following method at day 7 post-BMT: harvest of splenic T cells, stimulation with host- or donor-type dendritic cells, and use of six-color flow cytometry to detect host T cells, CD4 and CD8 subsets, and cytokine secretion by capture method. Consistent with our hypothesis, PC-treated cells acquired greatly reduced alloreactivity in vivo relative to FC-treated cells: the percentage of host CD4+ T cells secreting IFN-γ in an allospecific manner was 2.3±0.8% in recipients of PC-treated T cells and 62.7±13.4% in recipients of FC-treated cells (p<0.001). Similarly, the percentage of host CD8+ T cells secreting IFN-γ in an allospecific manner was 8.6±2.8% in recipients of PC-treated T cells and 92.7±4.1% in recipients of FC-treated T cells (p<0.001). We therefore conclude that at similar levels of myeloid suppression, the PC regimen is superior to the FC regimen in terms of murine T cell depletion, suppression of global T cell cytokine secretion, and inhibition of in vivo capacity to acquire allospecificity in response to fully genetically disparate marrow allografts. These data provide a rationale to develop PC regimens as an alternative to currently utilized FC regimens.