First attempts of cancer immunotherapy were made approximately 100 years ago on the assumption that tumor cells are recognized as ‘foreign’ by the immune system. Later on, a whole series of experimental animal tumor models were developed. They included the use of syngeneic tumors, spontaneously arising tumors and human tumor xenografts in immunodeficient mice. The experimental data contributed to our understanding of the interaction between immunocompetent cells and their products on the one hand and tumor cells on the other. On this basis, various immunotherapeutic protocols have been devised which included the use of ‘nonspecific’ components such as bacterial adjuvants, cytokines, NK cells and macrophages and attempts were made to raise specific T and B cell responses against tumor cells. Many human tumor-associated antigens have been characterized, and various ways of increasing the immunogenicity of human tumor cells have been described. Moreover, more insight has been achieved in defining ‘high risk’ populations on the basis of genetic background, the role of environmental factors and the characterization of ‘precancerous’ cells. Some cancer vaccines have been used in clinical trials which have resulted in partially beneficial therapeutic effects but have not provided a full solution for a rational use of immunotherapy against human neoplasia.