Targeted immunotherapy using T cells bearing chimeric antigen receptors (CAR T cells) is a promising treatment of end-stage leukemia patients. 1–4 Currently, CAR-T must be produced from autologous cells, complicating their production and delaying their administration to the patients, thereby preventing their application to fast-progressing diseases such as acute leukemias. Production of allogeneic universal donor T cells, as an ‘off-theshelf’reagent, could significantly facilitate CAR T-cell therapy and greatly increase the number of benefiting patients. 5 To this end, one must overcome host-mediated rejection and potential graft versus host disease (GvHD) that are mediated by allogeneic cells. While the risk of GvHD may be adequately addressed by gene editing of the endogenous T-cell receptor (TCR), suicide gene transduction or selection of non-GvHD inducing cells, 6–9 graftrejection represents a more difficult barrier. Here, we present a method for preventing rejection of HLA mismatched CAR-T cells, while preserving their selective anti-tumor activity, by using donor-derived mouse anti-3rd-party central memory CD8 veto cells (Tcm), previously shown to be devoid of GvH and capable of inducing tolerance to allografts. 10, 11 The immune-regulatory properties of anti-3rd-party veto Tcm were established in fully mismatched T-cell-depleted bone marrow transplantation (TDBMT), under reduced intensity conditioning (RIC). 10, 11 However, whether Tcm can act as a tolerizing agent independently of allogeneic TDBMT, without support of the CD34+ veto cell compartment, remains unknown. We addressed this question by initially determining whether adoptively transferred mismatched Tcm can withstand hostmediated rejection. Thus, allogenic F1-Tcm (H2db) cells were infused in conjunction with syngeneic TDBMT into lethally (8 Gy TBI) or sublethally (5.5 Gy TBI) irradiated Balb/c mice (see Supplementary Materials and Methods). F1-Tcm survived in the blood, persisting> 130 days post transfer. Average survival was similar under both lethal (0.72%±0.28/Lymphocytes, 8.8%±2.2/CD8+-compartment) and sublethal (0.69%±0.25/Lymphocytes, 6.2%±3.1/CD8+-compartment) regimens, indicating that RIC did not hinder survival of F1-Tcm (Supplementary Figure 1). Comparable results were obtained in C57BL/6 recipients (data not shown). Next, we tested whether Tcm could survive in the absence of TDBMT under RIC. Such a simple and safe protocol for immune tolerance induction, could potentially enable application of ‘offthe-shelf’allogeneic CAR T-cells, or other genetically modified T cells, from a single donor. To better simulate the clinical setting we tested both fully allogeneic Tcm (H2b) and F1-Tcm (H2db) cells, which were able to successfully engraft and sustain their presence under RIC (5.5–5 Gy TBI), with an average of 0.8%±0.42 F1-Tcm and 0.48%±0.67 allo-Tcm in peripheral blood on day 47 post transplant (Figure 1a; Supplementary Figure 2A). In line with our previous findings, 11 adoptive transfer of allo-Tcm was not associated with GvHD symptoms (ruffled fur, hunched back, or reduced body weight)(Supplementary Figure 2B). Notably, a gradual decrease of Tcm percentage in the peripheral blood occurred over time (Supplementary Figure 2C), potentially indicating gradual rejection of the Tcm cells. Nevertheless, Tcm populations remained detectable in the blood of the mice at> 8 months post injection, indicating their long-term persistence.