Adoptive transfer of antigen-specific T-cells has shown therapeutic successes in the treatment of tumors in patients with metastatic melanoma. Tumor antigen-specific T-lymphocytes, however, occur only at low frequencies in a small proportion of patients. This low T-lymphocyte frequency together with the difficulties associated with in vitro generation of T-lymphocytes specific for cancers other than melanoma hampers adoptive T cell therapy. To make adoptive T-cell therapy more uniformly applicable, strategies were developed at transferring tumor-specificity to primary human T-lymphocytes via antibody (Ig) or T-cell receptor (TCR) molecules. We exploited the selection power of phage display that allows for the testing of tens of billions of individual clones with a high-throughput selection of Fabs with peptide/MHC complex binding capacity. Following in vitro selection, human "TCR-like" Fab fragments have been functionally expressed on human T-lymphocytes, resulting in MHC-restricted, tumor-specific lysis and cytokine production. Currently, we have extended our selections to a panel of class I and II MHC-restricted MAGE and other tumor-specific epitopes, and would like to propose that phage display represents a technology able to expand T-cell therapy to numerous tumor types.