High-throughput multiplex immunoassay for the detection of mpox and MVA-BN vaccination up to 2 years after exposure in Belgium: a retrospective diagnostic accuracy study.

BACKGROUND: Mpox, caused by monkeypox virus (MPXV), has gained global attention following several international outbreaks. Accurate and high-throughput detection of MPXV-specific antibodies is essential for surveillance programmes, diagnosis, and vaccine trials. We therefore evaluated the long-term accuracy of an mpox multiplex immunoassay that measures IgG responses to 11 orthopoxvirus antigens. METHODS: The assay was validated in a diagnostic accuracy study involving a Belgian cohort of clade IIb MPXV-infected individuals (n=211) and modified vaccinia virus Ankara-Bavarian Nordic (MVA-BN) vaccine recipients (n=202), who were followed up for 2 years after exposure. We used receiver operating characteristic analysis to assess diagnostic performance at multiple timepoints after infection or vaccination compared with negative controls (n=287). To explore the assay's broader utility, we also investigated correlations between antigen-specific IgG responses and neutralising antibody titres. FINDINGS: In individuals without previous smallpox vaccination (n=149), the assay accurately detected mpox infection up to 2 years after infection, achieving 92% sensitivity (95% CI 89-97) at 95% specificity (91-98) when multiple antigens were combined. By contrast, diagnostic performance was low in previously vaccinated individuals (n=62), with sensitivities of most antigens dropping below 60% (38-77) after 6 months. Most antigen-specific IgG responses correlated strongly with neutralising antibody titres. INTERPRETATION: The assay shows high sensitivity and specificity for detecting previous mpox infection in unvaccinated individuals, with reliable performance up to 2 years after infection. Strong concordance with neutralising antibody titres supports the assay's potential for monitoring long-term infection and vaccine-induced immunity. However, previous vaccination status should be considered when interpreting results. FUNDING: Research Foundation-Flanders (FWO), Department of Work, Economy, Science, Innovation and Social Economy of the Flemish Government (WEWIS), and Netherlands Organisation for Health Research and Development (ZonMw).