Abstract
BACKGROUND: Marburg virus disease (MVD) is a rare but highly fatal zoonotic illness. Current serological tools are limited by their reliance on glycoprotein (GP)-based antigens or Marburg virus infected cell lysates, which cannot distinguish natural infection induced immune responses from GP-based vaccines. This diagnostic gap complicates the detection of missed cases particularly in areas of high vaccination coverage. METHODS: We developed and validated two complementary assays: a nucleoprotein C-terminal tail (NPct) Mix-and-Read (MR) assay and a recombinant GP1,2ΔTM ELISA. Serum samples used for validation were collected 6-9 months after the 2024 Rwanda MVD outbreak from RT-PCR-confirmed recovered patients (n = 36) and uninfected contacts (n = 46). Assay validation included additional serum controls from pre-outbreak Rwandan samples, Ebola virus disease recovered patients, and Rousettus aegyptiacus bats experimentally infected with MARV. Receiver operating characteristic (ROC) curve analyses were used to assess clinical accuracy. FINDINGS: The MARV anti-NPct MR assay demonstrated near-perfect discrimination between cases and controls with an area under the curve (AUC) = 0.996 (95% CI, 0.990-1.000), and yielding performance of 94.4% clinical sensitivity (95% CI, 81.9-99.0%) and 99.7% specificity (95% CI, 98.1-100.0%), and no cross-reactivity with Ebola convalescent sera. The MARV GP ELISA achieved 94.4% clinical sensitivity (95% CI, 81.9-99.0%) and 100% specificity (95% CI, 92.3-100%) with an AUC 0.973 (95% CI, 0.922-1.000). Both assays produced concordant results for all samples tested. In experimentally infected bats, the MARV anti-NPct MR assay achieved 100% sensitivity and 100% specificity. INTERPRETATION: The MARV anti-NPct MR assay provides a rapid, high-throughput, species-agnostic tool expected to differentiate natural infection from vaccine-induced immunity, addressing a critical gap in the current MVD diagnostic pipeline. Its robust validation in the largest cohort of recovered patients to date and applicability to reservoir hosts highlight its broad utility for outbreak response, follow-up care, and ecological studies. FUNDING: This work was supported by funding from US Centers for Disease Control and prevention, the Coalition for Epidemic Preparedness Innovations (CEPI), and Rwanda Biomedical Centre.