Human tumour-associated and tumour-specific antigens: some concepts in relation to clinical oncology.

Abstract

The concept of tumour-specific antigens is constantly undergoing reappraisal with the development of more sensitive methods for their detection. This has resulted in the finding that the many 'new' antigens produced by human tumours or materials immunologically closely related to them are also present in non-neoplastic tissues, albeit in small amounts. However, other antigens still appear to exist almost entirely in or on tumour cells so that the antigens of human tumours may be subdivided into either tumour-associated macromolecules or tumour-specific antigens. The elucidation of the chemical nature of the tumour-specific antigens may result in important advances in cancer diagnosis and therapy. As many are organ specific, it should be possible to evolve test systems which will enable tumours to be diagnosed and located before they become apparent clinically. On the other hand the tumour-associated macromolecules, of which the oncofetal antigens are the principal examples, are found in elevated amounts in some non-neoplastic disorders. It is now clear that serial estimation of the levels of these macromolecules is of considerably more diagnostic value than single random measurements. Current work is establishing their value in the detection of recurrent and metastatic tumours before they become apparent by other methods, which is probably their most important role, and also their value as aids to monitor therapeutic efficacy. The future use of both types of antigen may unfold a new era in cancer detection and therapy but many basic chemical and immunological studies are needed before their clinical use can be fully defined.