This review is a comprehensive account of the current status of tumor immunology, beginning with the identification of human tumor antigens and ending with a summary of clinical applications of what we know about the immunology of tumors.

The immune system has the greatest potential for the specific destruction of tumours with no toxicity to normal tissue and for long-term memory that can prevent cancer recurrence. The last 30 years of immuno-oncology research have provided solid evidence that tumours are recognised by the immune system and their development can be stopped or controlled long term through a process known as immunosurveillance. Tumour specificity of the immune response resides in the recognition of tumour antigens. Viral proteins in tumours caused by viruses and mutated proteins from oncogenes or other genes, as well as nonmutated but abnormally expressed self proteins found on all tumours, have been shown to be good antigens and good targets for immunosurveillance. In many cancers, however, malignant progression is accompanied by profound immune suppression that interferes with an effective antitumour response and tumour elimination. Initially, most of the escape from immunosurveillance was ascribed to changes in the tumour cells themselves (loss of tumour antigens, loss of human leukocyte antigen molecules, loss of sensitivity to complement, or T cell or natural killer (NK) cell lysis), making them a poor target of an immune attack. However, it has become clear that the suppression comes from the ability of tumours to subvert normal immune regulation to their advantage. The tumour microenvironment can prevent the expansion of tumour antigen-specific helper and cytotoxic T cells and instead promote the production of proinflammatory cytokines and other factors, leading to the accumulation of suppressive cell populations that inhibit instead of promote immunity. The best described are regulatory T cells and myeloid-derived suppressor cells. Great conceptual and technical advances in the field of immuno-oncology over the past 30 years have provided us with the knowledge and techniques to develop novel immunotherapeutic approaches for the treatment of cancer. These include methods that enhance tumour immunity by blocking inhibitory pathways and inhibitory cells in the tumour microenvironment (e.g. antibodies against cytotoxic T-lymphocyte-associated antigen-4, programmed death 1 or its ligand programmed death ligand 1, or low-dose chemotherapy). Of equal importance, they include methods that can enhance the specificity of antitumour immunity by inducing the expansion of T cells and antibodies directed to well-defined tumour antigens (e.g. cancer vaccines, potent adjuvants, immunostimulatory cytokines). Even as monotherapies, these approaches are having a substantial impact on the treatment of some patients with advanced, previously untreatable, malignancies. Most exciting of all, these successes provide a rationale to expect that used in various combinations or earlier in disease, current and future immunotherapies may transform cancer treatment, improving a prognosis for many patients.

Immunity to tumor Ags in patients is typically weak and not therapeutic. We have identified a new mechanism by which potentially immunogenic glycoprotein tumor Ags, such as MUC1, fail to stimulate strong immune responses. MUC1 is a heavily glycosylated membrane protein that is also present in soluble form in sera and ascites of cancer patients. We show that this soluble protein is readily taken up by dendritic cells (DC), but is not transported to late endosomes or MHC class II compartments for processing and binding to class II MHC. MUC1 uptake is mediated by the mannose receptor, and the protein is then retained long term in early endosomes without degradation. Long-term retention of MUC1 does not interfere with the ability of DC to process and present other Ags. We also demonstrate inhibited processing of another important glycoprotein tumor Ag, HER-2/neu. This may, therefore, be a frequent obstacle to presentation of tumor Ags and an important consideration in the design of cancer vaccines. It should be possible to overcome this obstacle by providing DC with a form of tumor Ag that can be better processed. For MUC1 we show that a 140-aa-long synthetic peptide is very efficiently processed by DC.

The field of tumor immunology has grown around the idea that one of the important roles of the immune system is to eliminate cancer. This idea was difficult to reconcile with the accepted notion that the immune system evolved to distinguish self from nonself and therefore tumors derived from self-tissues would not be recognized. Lack of appropriate animal models prevented experimental testing of cancer immunosurveillance. This changed with the realization that the immune system evolved to recognize danger and with the advent of mouse models deficient in one or more immune function, which showed predicted increases in susceptibility to cancer. Simultaneously, technical advances that enabled the study of the human immune system provided data for the existence of tumor-specific T cells and Abs and led to molecular identification of tumor Ags, fully validating the cancer immunosurveillance hypothesis. Immunotherapy designed to strengthen cancer immunosurveillance has achieved unprecedented clinical successes.

In contrast to protein antigens, processing of glycoproteins by dendritic cells (DCs) for presentation to T cells has not been well studied. We developed mouse T cell hybridomas to study processing and presentation of the tumor antigen MUC1 as a model glycoprotein. MUC1 is expressed on the surface as well as secreted by human adenocarcinomas. Circulating soluble MUC1 is available for uptake, processing, and presentation by DCs in vivo and better understanding of how that process functions in the case of glycosylated antigens may shed light on antitumor immune responses that could be initiated against this glycoprotein. We show that DCs endocytose MUC1 glycopeptides, transport them to acidic compartments, process them into smaller peptides, and present them on major histocompatability complex (MHC) class II molecules without removing the carbohydrates. Glycopeptides that are presented on DCs are recognized by T cells. This suggests that a much broader repertoire of T cells could be elicited against MUC1 and other glycoproteins than expected based only on their peptide sequences.