Evaluation of Hepatitis B core-related antigen (HBcrAg) as a biomarker in cohorts from the United Kingdom and South Africa.
Downs LO., Delphin M., van Schalkwyk M., Hugo S., Lumley SF., Waddilove E., Wang T., Martin J., de Lara C., Babbs A., Andersson MI., Glashoff RH., Ansari MA., Agarwal K., Dusheiko G., Taljaard J., Preiser W., Barnes E., Kelly G., Carey I., Shimakawa Y., Maponga TG., Matthews PC.
OBJECTIVES: We set out to evaluate Hepatitis B core-related antigen (HBcrAg) as a proxy for hepatitis B (HBV) viral load (VL) and liver disease in two different population settings. METHODS: We undertook a cross-sectional retrospective observational study using samples and data from adults living with chronic HBV infection from the United Kingdom (UK, n=142) and South Africa (SA, n=211). We assessed HBcrAg distribution, relationship with other biomarkers, and risk stratification performance, applying point of care test (POCT) thresholds. RESULTS: SA and UK cohorts differed by ethnicity, HIV coinfection, HBeAg-positivity and proportion with HBV VL >200,000 IU/ml (all p<0.001). HBcrAg positively correlated with alanine aminotransferase (ALT) (in both settings p<0.01), and fibrosis/cirrhosis by APRI score (p=0.03 in UK, p=0.008 in SA), but not with elastography or FIB-4 scores. HBcrAg ≥4.3 log10U/ml (POCT threshold) was 100% sensitive and 92% specific for predicting VL >200,000 IU/ml in the UK cohort, compared to 94% sensitive and 86% specific in the SA population. CONCLUSIONS: HBcrAg correlated with VL, but less so with liver disease. Use of this biomarker needs tailoring for use in diverse populations.