Conventional therapies active in primary tumors fail in metastatic diseases and colorectal cancer (CRC) metastatic spread to the liver is the major cause of mortality in patients. Adoptive cell therapy (ACT) with tumor redirected T cells holds promise for cancer treatment; however, the difficulty of T-cell trafficking into solid tumor tissues and the suppressive cues generated in the tumor microenvironment (TME) of hepatic metastases can hinder its efficacy.CD1d-restricted invariant natural killer T (iNKT) cells actively participate in tumor immunosurveillance by restricting cancer-supporting myeloid populations in the TME, enforcing immunostimulatory conditions. Moreover, we recently showed that retargeting iNKT cells against cancer cells, by transducing tumor-specific TCR genes, generates an enhanced effector population able to concurrently kill malignant cells and modulate detrimental myeloid cells in TME.The possibility to strengthen iNKT cell anti-tumor functions, together with the monomorphic nature of CD1d, make iNKT cells an attractive potential universal platform for adoptive cell therapy in cancer patients. Here we are investigating the possibility to harness iNKT cells against CRC liver metastases by engineering them with a Chimeric Antigen Receptor (CAR) specific for the human carcinoembryonic antigen (hCEA) and comparing CAR-iNKT anti-tumor efficacy to conventional CAR-T cells both in subcutaneous and metastatic CRC model. Moreover, we also aim to sustain iNKT cell functions by engineering with cytokines (IL15, IL-12, IFNa) that support long term the activity of the transferred cells at the tumor site, improving their resistance to the local immunosuppressive cues. Data collected from this study will provide a proof-of-concept of the efficacy and safety of tumor retargeted iNKT cells for liver metastases treatment, paving the way for future ACT approach in cancer patients.