Evaluation of performance characteristics of hepatitis B e antigen serologic assays

Hepatitis B e antigen (HBeAg) is considered an indicator of high hepatitis B virus (HBV) replication. Performance characteristics of commercially available HBeAg assays have not been determined, thus it is unknown whether lack of HBeAg detection is because of test sensitivity or HBV basal core promoter and precore mutations.

We studied the correlation between HBeAg reactivity with HBV DNA levels in three commercially available HBeAg assays using 335 HBsAg and HBV DNA positive serum/plasma samples.

Diagnostic sensitivity was determined by serial dilutions of a WHO HBeAg standard. The limit of HBeAg detection estimated through regression was 1 IU/mL (Centaur), 97 IU/mL (DiaSorin) and 129 IU/mL (Vitros). Of these 335 samples, enough sample volume remained in 253 samples for head-to-head comparison of the assays.

81 (32%), 41 (16%) and 36 (14%) of the samples were HBeAg positive by the Centaur, DiaSorin and Vitros assays, respectively. Compared to the FDA-approved Centaur assay the specificity of the other two assays was 98%, while sensitivity was 47% for the DiaSorin assay and 41% for the Vitros assay. Significant association was found between HBeAg positive samples and HBV DNA levels >20,000 IU/mL; 31% of HBeAg negative samples (Centaur) had HBV DNA levels >20,000 IU/mL, 26% of HBeAg positive samples had HBV DNA levels <20,000 IU/mL and 5 HBeAg positive samples had HBV DNA levels <2000 IU/mL.

Discordance was seen between these HBeAg assays, indicating reliance on HBeAg alone as a marker of high HBV replication can be misleading. Detection and quantification of HBV DNA remains the accurate and reliable marker of HBV replication.