Adoptive T cell therapy with patient-derived cells engineered ex vivo to express chimeric antigen receptors (CARs) is a promising therapeutic strategy in cancer, but producing these cells for a large number of patients is both technically challenging and cost prohibitive. Smith et al. describe a method of programming T cells in situ using DNA-carrying nanoparticles. Nanoparticles were targeted to T cells and loaded with DNA encoding the leukaemia-specific CD19−4-1BB−CD3ζ (194-1BBz) CAR; this DNA sequence was flanked by piggyBac transposons, and DNA encoding the piggyBac transposase was also included, to allow chromosomal integration and persistent CAR expression. These nanoparticles induced CAR expression in T cells in vitro and in immunocompetent mice. Furthermore, the nanoparticles induced tumour regression and increased survival of mice bearing B cell acute lymphoblastic leukaemias, with a therapeutic efficacy similar to that of ex vivo engineered T cells expressing the 194-1BBz CAR. As the manufacture and storage of these nanoparticles is fairly easy, this strategy could provide a viable alternative to ex vivo T cell engineering approaches.