Accelerated qualification of diagnostic tests during an ongoing outbreak of Bundibugyo virus disease: a collaborative analytical validation study.

BACKGROUND

Recognition of the 2026 Bundibugyo virus (BDBV) disease outbreak was delayed by adverse political circumstances and logistical constraints. The deployed test was intended for Ebola virus rather than BDBV, a distinct viral species. A virus isolate representing the outbreak strain is unavailable, and regulatory demands on diagnostics have increased. We aimed to provide reference virus material and validate diagnostic tests to aid in regulatory clearance.

METHODS

We generated an inactivated, patient-derived BDBV genome reference standard (GRS; 2·4 × 106 copies per mL) from a throat swab of a patient admitted to Charité-Universitätsmedizin Berlin, Berlin, Germany. We distributed the GRS internationally; tested its stability over 90 h at -80°C, 4°C, 20°C, and 37°C; and analytically validated four pilot real-time RT-PCR diagnostic tests. We assessed the sensitivity and run-to-run consistency of the diagnostic tests at the coordinating laboratory and specificity in a network of nine laboratories, using cultured pathogen materials (52-62 samples per test, representing 30 pathogens) and clinical leftover samples (120-172 samples per assay, representing 22 differential diagnostic aetiologies). Detection outcomes from all four tests were pooled in a probit model with assay class as a covariate to estimate the ratio of 95% limits of detection between BDBV-specific and broad-range assays.

FINDINGS: Relative to storage at -80°C, GRS viral load after 90 h was reduced 1·15-fold (95% CI 0·96-1·37, p=0·12) at 4°C and at 20°C, and 2·67-fold (2·24-3·19, p<0·0001) at 37°C. 95% limits of detection were 8·4 copies per μL (95% CI 3·2-87·2) for the Altona Diagnostics RealStar Filovirus Screen RT-PCR Kit 1.0, 2·4 copies per μL (1·0-20·6) for the Altona Diagnostics RealStar Bundibugyo RT-PCR Kit 1.0, 11·9 copies per μL (4·7-139·4) for the Roche LightMix Modular Ebola Virus Test Kit, and 2·5 copies per μL (1·2-14·4) for the Roche LightMix Modular Bundibugyo Virus Test Kit. In the pooled probit model, the 95% limits of detection of BDBV-specific assays were 0·24 times (95% CI 0·13-0·47) that of broad-range assays-ie, approximately four times lower. Decentralised evaluation of specificity in nine laboratories was completed within 15 days, and no cross-reactivity occurred in cultured or clinical panels.

INTERPRETATION: Decentralised evaluation of diagnostic test kits can facilitate emergency-use decisions, national authorisation, or regulatory review in ongoing outbreaks. This approach defines a course of action for laboratory networks to accelerate the development and assessment of diagnostic medical countermeasures. It bridges the interval between first outbreak recognition and availability of fully certified commercial diagnostic tests, especially for rare, high-consequence pathogens for which virus isolates or clinical materials are scarce.

FUNDING: None.