Calibrating Hepatitis E Virus Serological Assays Using Asymptomatic Specimens Obtained in Japan.

ABSTRACT

This study aimed to calibrate hepatitis E virus (HEV) serological assays. We optimized the previously developed in-house HEV antibody enzyme-linked immunosorbent assay (ELISA) by setting the cutoff with an in-house serological performance panel consisting of broad HEV antibody titers and subtracting nonspecific background values for anti-HEV IgM, IgA, and IgG. We also compared the assay's performance with that of commercial serological assay kits (four kits for IgM, one for IgA, and two for IgG). Although all serological assays readily detected HEV antibodies at high titers in the symptomatic hepatitis E population, considerable variations between assays were observed in the asymptomatic population. The in-house ELISA showed a higher sensitivity for HEV IgM, IgA, and IgG than the commercial kits and detected the seroconversion of HEV IgM and IgG earlier when testing a commercially available HEV seroconversion panel. The low sensitivity of the commercial kits was due to the high setting of the original cutoff, which was demonstrated by receiver operating characteristic analysis. However, the corrected cutoff value reduced assay specificity. Background subtraction is essential to achieve high specificity because the in-house ELISA without background subtraction reduced its specificity. These results indicate that asymptomatic specimens and background subtraction contribute to the optimization of HEV serological assays. IMPORTANCE Accurate diagnosis of hepatitis E virus (HEV) infection is essential for public health surveillance and for preventing HEV-contaminated blood transfusion. Anti-HEV IgM or IgA is used as a reliable marker of recent HEV infection. However, considerable variability in the sensitivity and specificity of HEV antibody detection is observed among several commercially available assay kits. In addition, none of the HEV antibody detection methods have been approved by the U.S. Food and Drug Administration (FDA). Here, we show that the in-house enzyme-linked immunosorbent assay (ELISA) could detect HEV IgM and IgA more sensitively than commercial kits in the asymptomatic population. We also suggest that the assay performance of commercial kits might be improved by optimizing the cutoff and reducing nonspecific background noise. A sensitive serological (IgM or IgA) assay in addition to HEV RNA testing will contribute to accurate diagnosis of acute HEV infection because HEV RNA-positive duration is relatively short.