• J Morgan Freiman, Trang M Tran, Samuel G Schumacher, Laura F White, Stefano Ongarello, Jennifer Cohn, Philippa J Easterbrook, Benjamin P Linas, and Claudia M Denkinger.
• Ann. Intern. Med. 2016 Sep 6; 165 (5): 345-55.

BackgroundDiagnosis of chronic hepatitis C virus (HCV) infection requires both a positive HCV antibody screen and confirmatory nucleic acid testing (NAT). Testing for hepatitis C virus core antigen (HCVcAg) is a potential alternative to NAT.PurposeTo evaluate the accuracy of diagnosis of active HCV infection among adults and children for 5 HCVcAg tests compared with NAT.Data SourcesEMBASE, PubMed, Web of Science, Scopus, and Cochrane Database of Systematic Reviews from 1990 through 31 March 2016.Study SelectionCase-control, cross-sectional, cohort, or randomized trials that compared any of 5 HCVcAg tests with an NAT reference standard.Data Extraction2 independent reviewers extracted data and assessed quality using an adapted QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) tool.Data Synthesis44 studies evaluated 5 index tests. Studies for the Abbott ARCHITECT HCV Ag assay had the highest quality, whereas those for the Ortho HCV Ag enzyme-linked immunosorbent assay (ELISA) had the lowest quality. From bivariate analyses, the sensitivity and specificity of the assays were as follows: Abbott ARCHITECT, 93.4% (95% CI, 90.1% to 96.4%) and 98.8% (CI, 97.4% to 99.5%); Ortho ELISA, 93.2% (CI, 81.6% to 97.7%) and 99.2% (CI, 87.9% to 100%); and Hunan Jynda Bioengineering Group HCV Ag ELISA, 59.5% (CI, 46.0% to 71.7%) and 82.9% (CI, 58.6% to 94.3%). Insufficient data were available for a meta-analysis about the Fujirebio Lumipulse Ortho HCV Ag and Eiken Lumispot HCV Ag assays. In 3 quantitative studies using Abbott ARCHITECT, HCVcAg correlated closely with HCV RNA levels greater than 3000 IU/mL.LimitationsInsufficient data were available on covariates, such as HIV or hepatitis B virus status, for subgroup analyses. Few studies reported genotypes of isolates, and data for genotypes 4, 5, and 6 were scant. Most studies were conducted in high-resource settings and reference laboratories.ConclusionThe HCVcAg assays with signal amplification have high sensitivity, high specificity, and good correlation with HCV RNA levels greater than 3000 IU/mL and have the potential to replace NAT in settings with high HCV prevalence.Primary Funding SourceNational Institutes of Health.

16 keywords...

hide…