The only proposed T-cell epitope derived from the TEL-AML1 translocation is not naturally processed

Adoptive therapy with T-cell receptor (TCR)-engineered T cells is a promising approach in cancer treatment. While usage of T cells specific for tumor-associated antigens (TAAs) can lead to serious side effects because of autoimmunity, targeting true tumor-specific mutations, such as the products of translocations in leukemias, should reduce such a risk. A potentially ideal target might be the chimeric protein TEL-AML1, which results from the chromosomal translocation 12;21 and represents the most common fusion gene in childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL). Within the fusion region of TEL-AML1, a single epitope has been described by reverse immunology as immunogenic in HLA-A*0201 restriction settings. As a potential source of TCRs specific for this TEL-AML1 epitope, we have used mice expressing a human TCR-αβ repertoire and human MHC class I. Surprisingly, we have found that, although a specific functional CD8(+) T-cell response against this peptide could be evoked, the described epitope was in fact not endogenously processed. Analyses done with a potent antigen-presenting cell line, as well as with purified human proteasomes, support the conclusion that this peptide cannot be proposed as a potential target in immunotherapy of ALL in HLA-A*0201-restricted fashion.