GVHD remains a major source of morbidity and mortality after non-T cell-depleted BMT. The use of donor T lymphocytes expressing a suicide gene could lead to specific immunomodulation after BMT. We are currently evaluating such an approach in a phase I clinical study. A 12-day ex vivo expansion is required to generate gene-modified donor T lymphocytes. CsA is commonly used for GVHD prophylaxis. We analyzed, in a murine GVHD model, the effects of CsA administration on the alloreactivity of fresh or ex vivo-expanded T cells. Variable amounts of fresh or ex vivo-expanded T cells were administered in conjunction with a marrow graft to lethally irradiated allogeneic mice. As expected, a protective effect of CsA with a delayed GVHD-related mortality (P < 0.01 vs saline treatment) was observed in mice receiving fresh splenocytes. However, CsA treatment had no effect (P = NS) in mice experiencing lethal GVHD induced by ex vivo-expanded T cells whether or not the T cells had been 'rested' in low-dose IL-2 prior to in vivo administration. In agreement with the in vivo findings, CsA also inhibited the in vitro proliferation of alloreactive fresh T cells while having no significant inhibitory effect on the alloreactivity of ex vivo-expanded T lymphocytes. Overall, we demonstrate that the alloreactivity of ex vivo-expanded T lymphocytes is not sensitive to CsA and that this differential effect of CsA is not related to the alloreactive potential of the infused T cells. These findings could be highly relevant when considering allogeneic T cell therapy approaches.