Evaluation of the reproducibility and performance characteristics of the Phagomagnetic separation (PhMS)-qPCR assay for rapidly detecting viable Mycobacterium avium subsp. paratuberculosis in bovine milk and faeces

Irene R. Grant^1,2*Iker A. Sevilla³ELENA MOLINA³BEATRIZ ROMERO MARTINEZ^4,5Victor Lorente-Leal^4,6Virginie C Thibault-Poisson^7,8Martina Cechova⁹Heike U. Köhler¹⁰

• ¹Queen's University Belfast, Belfast, United Kingdom
• ²Rapid-Myco Technologies Limited, Belfast, Northern Ireland, UK, United Kingdom
• ³NEIKER-Instituto Vasco de Investigación y Desarrollo Agrario, Derio, Spain
• ⁴Centro de Vigilancia Sanitaria Veterinaria (VISAVET), Universidad Complutense de Madrid, Madrid, Spain
• ⁵Universidad Complutense de Madrid Facultad de Veterinaria, Madrid, Spain
• ⁶Universidad Complutense de Madrid Departamento de Genetica Fisiologia y Microbiologia, Madrid, Spain
• ⁷Anses Laboratoire de Ploufragan-Plouzane, Ploufragan, France
• ⁸GDS France (National Animal Health Farmers’ Organization), Paris, France
• ⁹Veterinary Research Institute, Brno, Czechia
• ¹⁰Friedrich-Loeffler Institut, Jena, Germany

Inter-laboratory trials were carried out to evaluate the reproducibility and estimate test performance characteristics of the Phagomagnetic separation (PhMS)-qPCR assay, a novel phage-based assay recently developed as a rapid alternative to culture for detecting viable Mycobacterium avium subsp. paratuberculosis (MAP) in bovine milk and faeces. Unique reagents and a detailed instruction manual required for the PhMS-qPCR assay were provided to five European veterinary diagnostic laboratories by Rapid-Myco Technologies Limited. Milk and faeces test panels were prepared at NEIKER and distributed to participant laboratories between April-June 2023 and March-May 2024, respectively. Each test panel comprised of MAP-spiked and/or naturally infected bovine milk or faeces samples (18 samples per panel on two separate occasions for each sample matrix). The six participant laboratories (including organizer) performed automated or manual PhMS and used whatever qPCR instrument they had available. All laboratories used the IDEXX RealPCR MAP DNA test for the qPCR part of the assay. Generally, substantial agreement was observed overall between PhMS-qPCR results and reference culture results for spiked milk (Kappa value 0.5982) and naturally MAP-infected faeces (Kappa value 0.7780 using an amended protocol). Preliminary estimates of the detection (analytical) sensitivity (Se), detection specificity (Sp) and trueness (T) of the PhMS-qPCR assay applied to bovine milk and faeces were obtained. The mean Se, Sp and T values across six laboratories were 93.1%, 67.9% and 88.7% when milk was tested and 84.1%, 93.7% and 88.9% when faeces was tested. Overall, the PhMS-qPCR assay performed well in multiple laboratories and test reproducibility was demonstrated (Cohen's Kappa ≥0.6-1.000). The estimates of performance characteristics of the PhMS-qPCR assay are generally acceptable for a potential diagnostic test. Hence the PhMS-qPCR assay shows considerable promise as a rapid test to detect viable MAP in veterinary specimens such as milk and faeces. Further and fuller validation of the assay will continue to assess its diagnostic potential.