T-cell receptor (TCR) transfer is an attractive strategy to increase the number of cancer-specific T cells in adoptive cell therapy. However, recent clinical and pre-clinical findings indicate that careful consideration of the target antigen is required to limit the risk of off-target toxicity. Directing T cells against mutated proteins such as frequently occurring frameshift mutations may thus be a safer alternative to tumor-associated self-antigens. Furthermore, such frameshift mutations result in novel polypeptides allowing selection of TCRs from the non-tolerant T-cell repertoire circumventing the problem of low affinity TCRs due to central tolerance. The transforming growth factor beta Receptor II frameshift mutation (TGFbetaRIImut) is found in Lynch syndrome cancer patients and in approximately 15% of sporadic colorectal and gastric cancers displaying microsatellite instability (MSI). The -1A mutation within a stretch of 10 adenine bases (nucleotides 709-718) of the TGFbetaRII gene gives rise to immunogenic peptides previously used for vaccination of MSI+ colorectal cancer patients in a Phase I clinical trial. From a clinically responding patient, we isolated a cytotoxic T lymphocyte (CTL) clone showing a restriction for HLA-A2 in complex with TGFbetaRIImut peptide. Its TCR was identified and shown to redirect T cells against colon carcinoma cell lines harboring the frameshift mutation. Finally, T cells transduced with the HLA-A2-restricted TGFbetaRIImut-specific TCR were demonstrated to significantly reduce the growth of colorectal cancer and enhance survival in a NOD/SCID xenograft mouse model.