"Here we seek to identify the genetic program at the basis of T memory stem cell persistence in vivo and use their properties to improve the adoptive T cell transfer for the treatment of human cancer. Adoptive cell transfer with tumour-specific T lymphocytes is currently under test in the clinic for the treatment of solid tumours such as melanoma. However, the poor persistence in vivo of adoptively-transferred cells is the major caveat to successful immunotherapy. Recent evidence suggests that CD8+ T cells derived from lymphocytes at early stages of differentiation better mediate tumour rejection due to their increased persistence, survival and proliferation in vivo. We recently identified the least differentiated human memory T cell subset, the T stem cell memory (TSCM), capable to self-renew and simultaneously generate more differentiated progeny. These TSCM are also endowed with superior immune reconstitution capacity and anti-tumor immunity (Gattinoni L, Lugli E et al. Nat Med, 2011;17:1290-97). However, the genetic program at the basis of persistence of these cells, especially under chronic stress, still needs to be defined. By studying immune reconstitution in haploidentical, T-repleted bone marrow transplantation, we will address this question and subsequently exploit this information to improve anti-tumor immunotherapy. Specifically, we will:1. Determine whether TSCM cells are major contributors to the process of immune reconstitution as suggested by non-human primate and humanized mouse models.2. Identify the genetic program responsible for the persistence of TSCM cells in vivo under conditions of continuous stress, i.e. the lymphopenia induced by the conditioning regimen3. Genetically-modify tumor-specific T cells with candidate genes to improve persistence in humanized mouse tumor models.Our data will impact not only the field of adoptive immunotherapy but also of vaccination and transplantation."