Adaptation and validation of an influenza A subtyping panel for detection of H1pdm09, H3 and H5 on a high-throughput RT-qPCR system

Facing the emergence of highly pathogenic avian influenza virus (HPAIV) H5N1 clade 2.3.4.4b in birds and its circulation in dairy cattle, rapid and reliable assays to detect HPAIV infection in humans are needed in diagnostic laboratories worldwide. We adapted and evaluated the performance of a molecular influenza A subtyping assay for the detection of A(H1N1)pdm09, A(H3N2), A(H5) and a pan-influenza A target on a high-throughput, fully automated platform. Previously published target primers and probes (Panning et al., Terrier et al., and the World Health Organization) were modified for target inclusivity/exclusivity, and adapted to compatibility with the Roche cobas5800/6800/8800 system as multiplex reaction. To evaluate the analytical performance of the new subtyping assay external quality assessment (EQA) samples were used as standards and quantified by digital-PCR. Limits of detection determined by 95% probit analysis were 754.0 digital copies (dcp)/ml for A(H1N1)pdm09, 148.0 dcp/ml for A(H3N2), 156 dcp/ml for A(H5) and 45.2 dcp/ml for the influenza A pan-target. Linearity was assessed for each subtyping target over at least four log-steps (r2: 0.9969–0.9998). The assay showed 100% agreement with EQA samples, eluates from the Friedrich Loeffler Institut (Germany) tested positive for different influenza A subtypes, and CE-IVD manual tests using 132 clinical samples. No false positives were detected in the exclusivity set. Our new subtyping assay is fully automated, easily scalable and can be used in surveillance and routine clinical settings enabling the detection of HPAIV infections in humans and may contribute to limit potential transmission chains at an early stage.