Development of a monoclonal antibody-based immunoradiometric assay for the measurement of the free alpha-subunit of human chorionic gonadotrophin.

Abstract

Almost a century has elapsed since the antigen-antibody interaction was first recognised as the basis of an immune response (Ehrlich, 1897). However, it was only in the 1930s, with the development of improved technologies that this concept was better understood, and led to the discovery of the amazing diversity and specificity of antibody molecules (Landsteiner, 1933). Theoretically, it is possible to make antibodies to a variety of biological substances and other chemicals, and therefore they are ideally suited as specific recognition elements to be used for analytical, cytological, functional, therapeutic and biochemical purposes. The development of the radioimmunoassay (RIA) thirty five years ago, revolutionised research in many areas of clinical and scientific investigation. This technique evolved rapidly from the discovery made by Berson et al. in 1956 that antibodies to insulin could be detected in patients treated with this hormone, by measuring the binding of radiolabeled insulin to these antibodies. Although in the past RIAs have been the most important assay system employing antibody and labelled tracer, the limitation was that reliance had to be placed on the chance development of a good polyclonal antibody. These shortcomings stimulated the search for monospecific antibodies of reproducible quality and sufficient quantity. The development and introduction of monoclonal antibody technology brought about a revolution in immune serology (Kohler and Milstein, 1975). Establishment of immortal cell lines which contained the genetic elements of antibody-producing cells was achieved by fusion between a myeloma cell line and spleen cells from an immunised donor. The resulting hybrids had the essential properties of both parents, namely, permanent growth and a high capacity for the synthesis and secretion of immunoglobulins, normally characteristics of plasmacytomas, together with the genetic elements defining a specific antibody. Gestational trophoblastic disease (GTD) is a neoplastic condition of the trophoblast and occurs as molar pregnancy in a benign or invasive form, or as choriocarcinoma in a malignant form. Effective therapy has been developed for the treatment of both choriocarcinoma and molar pregnancy, but the key to successful management of these patients lies in their prompt diagnosis and careful monitoring of response to treatment (Green-Thompson, 1986). Fortuitously, these tumours elaborate the human chorionic gonadotrophin hormone (hCG) and its free alpha (a) and beta (B) subunits and hence a ready marker for the tumour exists. Human chorionic gonadotrophin is one of a group of glycoprotein hormones, which includes luteinising hormone (LH), follicle stimulating hormone (FSH) and thyroid stimulating hormone (TSH). These hormones are composed of two dissimilar subunits designated a and B, which are bound non-covalently in the intact molecule. The B-subunit of each glycoprotein hormone is unique and is responsible for the respective biological and immunological properties of the glycoproteins. In contrast, all four hormones possess an identical a-subunit which is coded for by a single gene (Fiddes and Goodman, 1979). The measurement of hCG and its free B-subunit, as so-called BhCG, for the diagnosis and monitoring of therapy in patients with GTD is now routinely practised throughout the world (Vaitukaitis et al., 1972). However it has been demonstrated by Bagshawe (1975) that when serum BhCG can no longer be measured by current RIA methods, up to 10" tumour cells may remain undetected. In addition, there have been isolated reports of two patients with choriocarcinoma in whom BhCG was undetectable in the serum but who appeared to be secreting only the a-subunit (Dawood et al, 1977). Furthermore, it has been suggested that measurement of free a-subunit rather than intact hCG or the free B-subunit is a more effective means of detecting persistent trophoblastic disease as well as tumour recurrence following treatment (Quigley et al, 1980a and b). Radioimmunoassays which measure the free a-subunit of hCG have been developed, but in general lack the specificity and sensitivity required (Gaspard et al, 1980; Kohorn et al, 1981). These assays employ polyclonal antisera which also detect epitopes common to the pituitary gonadotrophins. Thus there is a need to produce monoclonal antibodies which recognise regions of the free a-subunit which are hidden in the intact gonadotrophins. Such antibodies would provide the required specificity for use in RIAs but are limited in their use by their inherent lack of high affinity for the antigen. Fortunately, this drawback may be overcome by using monoclonal antibodies as labelled reagents in an alternative assay system, the immunoradiometric assay (IRMA), described by Miles and Hales (1968). The IRMA, particularly the two-site sandwich version of the assay, has been shown to provide greater sensitivity in addition to allowing enhanced specificity. This is a consequence of the use of two antibodies in excess to detect the analyte, each directed at a different epitope on the target molecule. The first antibody, referred to as the capture antibody, is usually linked to a solid-phase to facilitate easy separation and is added in excess relative to the target hormone to enhance antibody-antigen interaction, thereby allowing increased sensitivity in the measurement of analyte. The second antibody, referred to as the detection antibody, is labelled with a radioactive isotope or an enzyme to detect antigen already bound to the capture antibody. The application of monoclonal antibodies specific for the free a-subunit to a highly sensitive IRMA format is an obvious need. Hence this study was undertaken firstly, to raise and characterise monoclonal antibodies to the free a-subunit, secondly to develop an IRMA using these antibodies and finally to establish whether measurement of free a-subunit has any clinical advantage.