Evaluation of branched DNA signal amplification for the detection of hepatitis C virus RNA


National Institutes of Health, CC, Department of Transfusion Medicine, Bldg 10, Rm 1C-711, Bethesda, MD 20892, USA.


Summary. There is an increasing need for a practical assay to measure HCV RNA to assess the viral burden in chronic hepatitis C virus (HCV) infection as viral load relates to transmission and therapeutic response. This study evaluates branched DNA (bDNA) signal amplification, a technique that avoids many of the pitfalls of polymerase chain reaction (PCR). The bDNA assay uses a microtitre well format and a series of capture, target and amplification probes that bind RNA to the well and then successively bind oligonucleotides to the RNA and branched DNA molecules to the oligonucleotides. Enzyme-labelled probes are bound to the arms of the bDNA and light output from a chemiluminescent substrate is directly proportional to the amount of starting HCV RNA. Appropriate standards provide direct quantitation. Whereas PCR amplifies the HCV genome, bDNA amplifies the hybridization signal.

In testing a standardized, coded panel, bDNA showed 100% specificity and detected five of six sera proven to transmit hepatitis C to the chimpanzee; PCR detected all six infectious sera. Serial samples were measured in two acute and five chronic cases of transfusion-associated hepatitis and in three commercial seroconversion panels. In acute cases, 107–108 molecular equivalents per ml (eq per ml) of HCV RNA were detected prior to peak alanine aminotransferase (ALT) activity and then rapidly declined to non-detectable levels. Similar levels of HCV RNA were observed early in the course of two patients who progressed to chronic hepatitis; the chronic course was characterized by diminished, fluctuating and sometimes non-detectable levels of HCV RNA. In two chronic cases, HCV RNA was not detected, or only transiently detected by bDNA, but was present when assayed by PCR. In one chronic case, the periodicity of HCV RNA levels closely paralleled the fluctuations of ALT suggesting a relationship between viral replication and subsequent hepatocellular injury. In testing 50 blood donors whose anti-HCV reactivity was confirmed by a recombinant immunoblot assay (RIBA), HCV RNA was detected by bDNA in 41 (81%), while PCR was positive in 45 (90%); the overall concordance between bDNA and PCR in 100 anti-HCV enzyme immunoassays (EIA) reactive donor samples was 96%. Lastly, bDNA showed the loss of HCV RNA in six out of six evaluable patients who had complete biochemical responses to interferon; five out of six non-responders also showed appreciable declines in HCV RNA level, but in only two did HCV RNA drop below the detection limit; these two cases remained PCR positive. Seventeen placebo-treated patients did not lose HCV RNA by either bDNA or PCR. Hence the bDNA assay is a practical means to measure HCV RNA in a variety of clinical settings. Although it is not as sensitive as PCR, it has greater specificity, is directly quantitative, and can be used in any routine laboratory that can perform microwell EIAs. This simplified quantitation may be of particular benefit in evaluating the probability of HCV transmission and the response to anti-viral therapy.