Primary immunodeficiency diseases (PID) are attractive candidates for a gene therapy approach because many of these disorders convey a poor prognosis while a number of the genes mutated in these conditions have been identified. Gene transfer into hematopoietic stem cells (HSC) should, in theory, lead to a cure of the disease. There are, however, a number of limitations mostly related to the failure of clinically available vectors to enable transgene integration into HSC. Nevertheless PID due to a gene defect leading to failure of cell development could be amenable to gene therapy given the selective advantage conferred to transgene expression in progenitor cells. Terminally differentiated cells are, however, long lived, as is the case for T lymphocytes. This concept led to the first gene therapy trials for adenosine deaminase (ADA) deficiency several years ago. Results were in part disappointing mostly because of the concomitant substitutive treatment by polyethylene glycol-ADA. However, recent application to X-linked severe combined immunodeficiency (gamma (c) deficiency) turned out to be efficient at least on a relatively short term basis (i.e. one year so far). These results demonstrate that this concept is valid and can be the basis for the treatment of other forms of severe T-cell immunodeficiencies. Obviously, development of vectors (lentiviruses) able to efficiently target HSC could in the future considerably enlarge the field of PID treatable by gene transfer.