Harnessing self-reactivity in cancer immunotherapy.

Recent years have witnessed a significant advance in our understanding of the immune response to cancer through the identification of human tumor antigens at the molecular level. The growing list of normal self proteins recognized on tumors by T cells of cancer patients is consistent with the existence in humans of a self-reactive T-cell repertoire analogous to that described in murine models of autoimmune disease. The presence of these cells in normal individuals reveals the benign self-reactivity which lies dormant in the T-cell repertoire of humans. These autoreactive T cells, although pathogenic when activated in autoimmune disease, could be directed towards a beneficial effect in tumor immunotherapy. Potential targets for such a limited anti-self/anti-tumor immune response include overexpressed proteins, tissue-specific differentiation antigens and developmental proteins expressed aberrantly in tumor cells. Depending on the determinant chosen, recruitment of these self-directed T cells may require overcoming some degree of non-deletional peripheral tolerance. By specifically targeting immune responses to subdominant and cryptic determinants from these proteins, it may be possible to maximize the anti-tumor effect resulting from self-directed immunity while minimizing unwanted damage to normal tissue.