Assay for Hepatitis B Core-related Antigen Identify Patients With High Viral Load: Systematic Review and Meta-analysis of Individual Participant Data

Clin Gastroenterol Hepatol. 2020 Apr 29;S1542-3565(20)30590-5. doi: 10.1016/j.cgh.2020.04.045. Online ahead of print.

Abstract

Background & aims: Scale-up of tests and treatments is needed to eliminate hepatitis B virus (HBV) infection from resource-limited countries. However, access to nucleic acid tests that quantify HBV DNA, to determine treatment eligibility, is severely limited. We performed a systematic review and meta-analysis to assess the performance of the hepatitis B core-related antigen (HBcrAg) immunoassay, a low-cost (less than $15/assay) alternative to the nucleic acid test, to identify highly viremic patients, infected with any HBV genotype.

Methods: We searched Medline, Embase, Scopus, and Web of Science through June 27, 2018 for studies that measured HBV DNA and HBcrAg in the same blood samples. We contacted study authors to obtain data from each study participant, and randomly assigned each participant to the derivation or validation cohorts. We applied optimal HBcrAg cut-off values derived from the derivation set to the validation set to estimate sensitivity and specificity of HBcrAg to indicate high viremia.

Results: Of 74 eligible studies found in the systematic review, we obtained individual participant data from 60 studies (81%). We performed a meta-analysis of individual participant data of 5591 HBV-infected patients who did not receive antiviral therapy and 4806 HBV-infected patients who received antiviral agents. In untreated patients, the pooled area under the receiver operating characteristic curve and the optimal cut-off value for the HBcrAg assay were 0.88 (95% CI, 0.83-0.94) and 3.6 log U/ml for identifying patients with a level of HBV DNA ≥2000 IU/ml, and 0.96 (95% CI, 0.94-0.98) and 5.3 log U/ml for identifying patients with a level of HBV DNA ≥200,000 IU/ml, respectively. In the validation set, the HBcrAg assay identified patients with ≥2000 IU/ml HBV DNA with 85.2% sensitivity and 84.7% specificity; the assay identified patients with ≥200,000 IU/ml HBV DNA with 91.8% sensitivity and 90.5% specificity. Performance did not vary among HBV genotypes. In patients receiving anti-HBV therapy, there was no correlation between levels of HBcrAg and HBV DNA.

Conclusions: In a systematic review and meta-analysis of individual participant data, we found that the assay for HBcrAg identifies treatment-naïve patients with high levels of HBV DNA with high sensitivity and specificity, regardless of genotype. The HBcrAg assay is a good, low-cost alternative to the nucleic acid test to identify highly viremic patients infected with different genotypes.

Keywords: developing nation; diagnosis; diagnostic test; real-world.