Aims/hypothesis Anti-CD3 monoclonal antibodies remain the most promising immune therapy for reversing recentonset type 1 diabetes. However, current clinical trials have revealed their major drawback, namely the narrow therapeutic window in which low doses are ineffective and higher doses that preserve functional beta cell mass cause side effects. Strategies that sidestep these limitations while preserving or improving anti-CD3's therapeutic efficiency are essential. We hypothesised that combining a potent vitamin D 3 analogue (TX527), ciclosporin A (CsA) and anti-CD3 would act to lower the dose while maintaining or even boosting therapeutic efficacy to counteract autoimmune destruction of transplanted islets. Methods This study involved the use of syngeneic islet transplantation, immunofluorescence microscopy, immune phenotyping by flow cytometry, RT-PCR analysis, and in vitro and in vivo suppression assays. Results Combination therapy with TX527, CsA and anti-CD3 was well tolerated on the basis of weight, bone and calcium variables. Remarkably, combining all three agents at sub-therapeutic doses greatly reduced recurrent autoimmune responses to a grafted islet mass (mean±SEM: 79.5±18.6 days; p<0.01), by far exceeding the therapeutic efficacy of monotherapy (24.8±7.3 days for anti-CD3) and dual therapy (25.5±12.4 days for anti-CD3+CsA). Combination therapy surpassed anti-CD3 monotherapy in reducing islet infiltration by effector/memory phenotype CD8 + T cells, as well as by reducing proinflammatory cytokine responses and increasing the frequency of T regulatory cells that were functional in vitro and in vivo, and acted in a cytotoxic T lymphocyte antigen 4-dependent manner. Conclusions/interpretation Combining the immunomodulatory actions of anti-CD3 mAb with CsA and the vitamin D 3 analogue, TX527, delivers therapeutic efficacy in an islettransplanted NOD mouse model of diabetes.