Although highly active antiretroviral therapy (HAART) can provide sustained control of human immunodeficiency virus 1 (HIV-1) replication, it allows for only a partial immune reconstitution. Moreover, the use of HAART is clearly limited by the lack of elimination of virus reservoirs, the adverse effects of therapy and the potential for development of viral resistance. Following HAART, the number of both memory and naive T cells in the periphery may increase, but a significant immune reconstitution is not observed in all treated patients.1 In addition, immunological measures generally do not normalize to levels seen in persons without HIV infection, and virus-specific T-cell responses frequently remain impaired.1,2 Approaches aimed at correcting the immune deficiencies of HIV-1 infection, and stimulating virus-specific responses, have mostly targeted the restoration of defective elements of adaptive immunity, and have been the subject of several recent reviews.3–5 These approaches include the use of cytokines and growth factors, virus-specific therapeutic vaccines, auto-vaccination through supervised structured treatment interruptions (STI), and gene therapy. Variable and limited degrees of success have been achieved with some of these immune-based strategies; nevertheless, a clearcut clinical benefit still awaits demonstration in large, controlled, randomized multicentre trials.