Background Reactivation of latent CMV following allogeneic hematopoietic stem cell transplantation remains a major cause of morbidity despite improvements in surveillance protocols and antiviral drug therapies. Selective restoration of anti-CMV cellular immunity is an attractive alternative approach if it can be achieved in a non-toxic manner that can be widely utilized. The application of CMV-specific adoptive cellular therapies following the initial proof of principle provided almost a decade ago has been limited be a number of factors including the practical difficulties of exporting technically demanding and labor-intensive methodology to smaller transplant centers. Methods We review the lessons learnt from studies in the setting of EBV-associated post transplant lymphoproliferative disease and the advances in both understanding and technology that have allowed the development of a multitude of new approaches for the generation of CMV-specific T cells suitable for adoptive transfer. These include the use of monoclytes, dendritic cells and B-lymphoblastoid cell lines as the presenting cells of antigen delivered by pulsing with exogenous proteins or peptides, or of antigen processed endogenously following transduction with one of a variety of viral vectors. We also discuss some of the issues surrounding the planning and implementation of trial protocols incorporating these products, and the techniques necessary for monitoring the fate of the infused cells and their efficacy. Finally, we review the preliminary data that is available on the newer generation of clinical trials that are ongoing in this field. Discussion Extensive characterization of many different systems used to culture CMV-specific cells has now been performed. These suggest that it is now possible to generate T cells with specificity for a number of different CMV-specific target antigens (most commonly the lower matrix protein pp65). Clinical trials using these products within the current regulatory guidelines ares till in their infancy. However, preliminary results are beginning to suggest that the newer generation of CMV-specific T cell products can be administered with a relatively low risk of graft-versus-host disease, and that antiviral activity can be demonstrated following infusion.