Diagnostic performance of hepatitis C virus core antigen testing for detecting hepatitis C in people living with hepatitis B: a systematic review and meta-analysis

Background: The current diagnostic strategy for hepatitis C virus (HCV) infection involves a two-step approach: antibody HCV screening followed by confirmatory nucleic acid testing. This study aimed to evaluate the diagnostic performance of the Abbott ARCHITECT HCV Ag assay in serum/plasma samples as a potential one-step alternative for diagnosing active HCV infection in people living with hepatitis B virus (PLWHB) through a systematic review and meta-analysis.

Methods: A systematic review and meta-analysis were conducted following PRISMA-DTA guidelines. This protocol was registered on PROSPERO (CRD42023402093). A comprehensive search of electronic databases identified studies published up to 1 November 2024, comparing the ARCHITECT HCV Ag assay to an HCV-RNA reference standard. Sensitivity, specificity, and likelihood ratios were pooled using a random-effects model within the MIDAS module of Stata software. Study quality was assessed using QUADAS-2. Heterogeneity was evaluated using the Q statistic, quantified using the I², and further explored through meta-regression.

Results: Ten studies (n = 494 participants) met inclusion criteria. The Abbott ARCHITECT HCV Ag assay demonstrated high sensitivity [91%, 95% confidence interval (CI): 76-97%] and specificity (99%, 95% CI: 99-100%). The positive likelihood ratio (PLR) was 81.20 (95% CI: 12.34-534.36), and the negative likelihood ratio (NLR) was 0.09 (95% CI: 0.03-0.27). The area under the summary receiver operating characteristic curve (AUC-SROC) was 99% (95% CI 98-100%). In regions with high HCV prevalence (≥ 10%), the test accurately confirmed active HCV infection in over 90% of cases. However, confirmatory testing remains necessary in low-prevalence settings (≤ 5%). The assay demonstrated an excellent ability to identify individuals without HCV infection, with a low false-negative rate (≤ 2%) regardless of HCV prevalence. Heterogeneity analysis revealed moderate to substantial variation in test performance (I² = 72.09% for sensitivity, 35.47% for PLR, and 78.33% for NLR). QUADAS-2 applicability concerns predicted heterogeneity, but differences were likely insignificant due to minimal variations and limited studies.

Conclusions: The Abbott ARCHITECT HCV Ag assay exhibited promising accuracy in detecting active HCV infection among PLWHB. This test might help diagnose active HCV infection in high-prevalence scenarios (≥ 10%) but needs further confirmation in low-prevalence settings (≤ 5%).