T cells engineered to express chimeric antigen receptors (CARs) to CD19 produced high rates of complete responses in patients with B-cell malignancies. However, CAR T cell immunotherapy of solid tumors remains largely ineffective. Besides the differences in target antigens and tumor biology, the type of effector cells could play a major role in determining the efficacy of CAR-redirected immunotherapy. My presentation will elucidate the potential advantages of Vα24-invariant NKT cells (iNKTs) compared with conventional T cells as a carrier of CARs or recombinant TCRs for immunotherapy of solid tumors. First, human iNKTs and CAR iNKTs better localize to the tumor site compared with T and CAR T cells from the same PBMC preparations in a xenogeneic model of neuroblastoma in NSG mice. Second, in addition to direct killing of tumor cells via a tumor-specific CAR, iNKTs can kill or functionally reprogram tumor-supportive M2-like macrophages in a CD1d-dependent manner. In contrast, polyclonal T cells are devoid of CAR-independent anti-tumor activities. Third, unlike T cells, iNKTs from allogeneic donors are not expected to induce graft-versus-host disease in HLA-mismatched recipients. Moreover, we recently engineered CAR constructs that co-express shRNA targeting expression of HLA molecules on the cell surface of iNKTs without making them susceptible to NK-cell cytotoxicity. The resultant CAR iNKTs can be universally tolerated and produced in bulk from selected healthy volunteers for a cost-effective “off-the-shelf” cancer immunotherapy.