Hepatitis B Core-Related Antigen to Indicate High Viral Load: Systematic Review and Meta-Analysis of 10,397 Individual Participants
Kyoko Yoshida,Alice Desbiolles,Sarah Feldman,Sang Hoon Ahn,Enagnon Kazali Alidjinou,Masanori Atsukawa,Laurence Bocket,M.R. Brunetto,María Buti,Ivana Carey,Gian Paolo Caviglia,En Qiang Chen,Markus Cornberg,Masaru Enomoto,Masao Honda,Christoph Höner zu Siederdissen,Masatoshi Ishigami,Harry L.A. Janssen,Benjamin Maasoumy,Takeshi Matsui,Akihiro Matsumoto,Shuhei Nishiguchi,Mar Riveiro‐Barciela,Akinobu Takaki,Pisit Tangkijvanich,Hidenori Toyoda,Margo J.H. van Campenhout,Bo Wang,Lai Wei,Hwai I. Yang,Yoshihiko Yano,Hiroshi Yatsuhashi,Man Fung Yuen,Eiji Tanaka,Maud Lemoine,Yasuhito Tanaka,Yusuke Shimakawa
DOI: https://doi.org/10.1016/j.cgh.2020.04.045
IF: 13.576
2021-01-01
Clinical Gastroenterology and Hepatology
Abstract:Background & Aims To eliminate hepatitis B virus (HBV) infection, scale-up of testing and treatment in resource-limited countries is crucial. However, access to nucleic acid testing to quantify HBV DNA, an essential test to examine treatment eligibility, remains severely limited. We assessed the performance of a novel immunoassay, HBV core-related antigen (HBcrAg), as a low-cost (less than US $15/assay) alternative to nucleic acid testing to indicate clinically important high viremia in chronic HBV patients infected with different genotypes. Methods We searched Medline, Embase, Scopus, and Web of Science databases through June 27, 2018. Three reviewers independently selected studies measuring HBV DNA and HBcrAg in the same blood samples. We contacted authors to provide individual participant data (IPD). We randomly allocated each IPD to a derivation or validation cohort. We applied optimal HBcrAg cut-off values derived from the derivation set to the validation set to estimate sensitivity/specificity. Results Of 74 eligible studies, IPD were obtained successfully for 60 studies (81%). Meta-analysis included 5591 IPD without antiviral therapy and 4806 treated with antivirals. In untreated patients, the pooled area under the receiver operating characteristic curve and optimal cut-off values were as follows: 0.88 (95% CI, 0.83–0.94) and 3.6 log U/mL to diagnose HBV DNA level of 2000 IU/mL or greater; and 0.96 (95% CI, 0.94–0.98) and 5.3 log U/mL for 200,000 IU/mL or greater, respectively. In the validation set, the sensitivity and specificity were 85.2% and 84.7% to diagnose HBV DNA level of 2000 IU/mL or greater, and 91.8% and 90.5% for 200,000 IU/mL or greater, respectively. The performance did not vary by HBV genotypes. In patients treated with anti-HBV therapy the correlation between HBcrAg and HBV DNA was poor. Conclusions HBcrAg might be a useful serologic marker to indicate clinically important high viremia in treatment-naïve, HBV-infected patients. To eliminate hepatitis B virus (HBV) infection, scale-up of testing and treatment in resource-limited countries is crucial. However, access to nucleic acid testing to quantify HBV DNA, an essential test to examine treatment eligibility, remains severely limited. We assessed the performance of a novel immunoassay, HBV core-related antigen (HBcrAg), as a low-cost (less than US $15/assay) alternative to nucleic acid testing to indicate clinically important high viremia in chronic HBV patients infected with different genotypes. We searched Medline, Embase, Scopus, and Web of Science databases through June 27, 2018. Three reviewers independently selected studies measuring HBV DNA and HBcrAg in the same blood samples. We contacted authors to provide individual participant data (IPD). We randomly allocated each IPD to a derivation or validation cohort. We applied optimal HBcrAg cut-off values derived from the derivation set to the validation set to estimate sensitivity/specificity. Of 74 eligible studies, IPD were obtained successfully for 60 studies (81%). Meta-analysis included 5591 IPD without antiviral therapy and 4806 treated with antivirals. In untreated patients, the pooled area under the receiver operating characteristic curve and optimal cut-off values were as follows: 0.88 (95% CI, 0.83–0.94) and 3.6 log U/mL to diagnose HBV DNA level of 2000 IU/mL or greater; and 0.96 (95% CI, 0.94–0.98) and 5.3 log U/mL for 200,000 IU/mL or greater, respectively. In the validation set, the sensitivity and specificity were 85.2% and 84.7% to diagnose HBV DNA level of 2000 IU/mL or greater, and 91.8% and 90.5% for 200,000 IU/mL or greater, respectively. The performance did not vary by HBV genotypes. In patients treated with anti-HBV therapy the correlation between HBcrAg and HBV DNA was poor. HBcrAg might be a useful serologic marker to indicate clinically important high viremia in treatment-naïve, HBV-infected patients.