Approaches for T-cell-based immunotherapy that have shown substantial effects in clinical trials are generally based on the adoptive transfer of high numbers of antigen-specific cells, and the success of these approaches is thought to rely on the high magnitude of the tumor-specific T-cell responses that are induced. In this study, we aimed to develop strategies that also yield a T-cell repertoire that is highly skewed toward tumor recognition but do not rely on ex vivo generation of tumor-specific T cells. To this end, the tumor-specific T-cell repertoire was first expanded by DNA vaccination and then infused into irradiated recipients. Subsequent vaccination of the recipient mice with the same antigen resulted in peak CD8(+) T-cell responses of approximately 50%. These high T-cell responses required the presence of antigen-experienced tumor-specific T cells within the graft because only mice that received cells of previously vaccinated donor mice developed effective responses. Tumor-bearing mice treated with this combined therapy showed a significant delay in tumor outgrowth, compared with mice treated by irradiation or vaccination alone. Furthermore, this antitumor effect was accompanied by an increased accumulation of activated and antigen-specific T cells within the tumor. In summary, the combination of DNA vaccination with host conditioning and adoptive transfer generates a marked, but transient, skewing of the T-cell repertoire toward tumor recognition. This strategy does not require ex vivo expansion of cells to generate effective antitumor immunity and may therefore easily be translated to clinical application.