Hepatitis C antibody testing: problems associated with non-specific binding

Abstract

The prevalence of antibodies to hepatitis C virus (anti-HCV) was measured in a number of groups known to be at increased risk of blood-borne viral infections, using an enzyme-linked immunosorbent assay (EIA) based on a nonstructural peptide generated by recombinant DNA technology. The assay was repeatably reactive in 75.6% of men with haemophilia, 61.9% of intravenous drug users, 34.1 % of homosexual men who were regular attenders at a gay sauna and 30.8% of prisoners. A lower reactivity was detected in sera collected from female prostitutes (10.4%), patients undergoing maintenance haemodialysis (5.9%), or renal transplantation (6.9%) and patients attending a sexually transmitted diseases clinic (6.2%). We also measured reactivity among inmates of a large institution for the mentally handicapped in which hepatitis B is known to be endemic, and in panels of sera which had been stored for 25–35 years. The test was positive in 41.1% of mentally handicapped patients with Down's syndrome and 7% of subjects with other forms of mental retardation. Similarly some 23 % and 20% of sera collected in 1954 and 1964 from patients with a variety of illnesses were found to be reactive. As most diagnostic assays suffer from some degree of non-specificity and confirmatory tests for the anti-HCV assay were not initially available in Australia, we analysed the distribution of optical density (OD) values in the different groups, in an attempt to obtain an insight into the specificity of the results being obtained. Whereas the ODs of sera collected from patients with haemophilia and IVDU had a bimodal pattern, with two well separated sets of results on either side of the cut-off. Sera collected from institutionalized mentally handicapped subjects and sera which had been stored for long periods were distributed symmetrically around the cut-off, strongly suggesting that some factor other than specific antibody is responsible for much of this activity. With the recent availability of limited quantities of specific neutralizing reagents specimens from several groups were retested and their OD values had either a bimodal distribution or were symmetrically distributed around the cut-off. We found 90% of reactive samples from IV drug users confirmed positive for anti-HCV, whereas only 10% of reactive sera from mentally handicapped patients, and none of the old stored sera which reacted with the screening test could be confirmed.