Abstract
BACKGROUND: Talaromycosis, caused by the fungus Talaromyces marneffei, is a leading cause of HIV-associated death in Southeast Asia. Current culture-based diagnosis only identifies late-stage infection, limiting understanding of disease burden and disease spectrum. We evaluated the clinical performance of anti-Mp1p IgM and IgG enzyme immunoassays (EIA) for talaromycosis diagnosis. METHODS: This diagnostic study included 423 adults with advanced HIV disease and culture-confirmed talaromycosis as cases, and 206 non-talaromycosis individuals with and without HIV who have never traveled to Asia as controls. Anti-Mp1p IgM and IgG antibodies were measured using conventional double-sandwich EIA. Diagnostic performance was assessed using the healthy control group and the HIV control group separately. Assay cut-offs were based on both the Youden index generated from the receiver operating characteristic curves, and separately using a pre-defined specificity of 95%. RESULTS: At the pre-defined 95% specificity, IgM had low to moderate accuracy of 62.3% and 87.9%, and a low sensitivity of 8.3% and 21.3%, when evaluated with healthy and HIV controls, respectively. IgG had similarly low accuracy of 52.2% and 78.4%, and a low sensitivity of 21.5% and 30.5%, when evaluated using healthy and HIV controls, respectively. Both IgM and IgG assays performed better in HIV controls than healthy controls. CONCLUSIONS: The anti-Mp1p IgM and IgG EIAs show low utility for the diagnosis of acute talaromycosis. However, at the high specificity cut-off of 95%, the assays have utility in the detection of T. marneffei exposure at both individual and population levels, and. provides a foundation for future sero-epidemiological studies.