Enhancing diagnostic preparedness for H5N1: a validation study of H5 single-plex assay and detection across multiple platforms

Abstract

The increasing transmission of highly pathogenic avian influenza (HPAI) H5N1 from animals to humans underscores the urgent need for enhanced diagnostic capabilities in clinical microbiology laboratories. Although in silico analysis suggests that commercial multiplex respiratory panels can detect H5N1, these assays lack subtyping H5N1 capabilities, and their real-world performance remains largely unverified. In this study, we evaluated the limit of detection (LoD) for H5N1 using three commercial molecular diagnostic assays routinely employed at our institution: bioMérieux BioFire Respiratory 2.1 Panel, Cepheid Xpert Xpress CoV-2/Flu/RSV Plus, and Hologic Panther Fusion SARS-CoV-2/Flu A/B/RSV assays. All three reliably detected H5N1 were at low viral concentrations. To enable H5N1 detection, we also developed a real-time RT-PCR H5 single-plex assay on the Hologic Panther Fusion Open Access platform for reflex testing of influenza A-positive specimens. In conclusion, although current commercial assays lack influenza A H5 subtype differentiation, our validation data provide critical performance information. When integrated with a targeted H5 assay, these tools can enhance clinical decision-making and public health surveillance by reducing the risk of missed H5N1 infection cases.IMPORTANCEThis study addresses a growing public health concern: the spread of bird flu (H5N1) from animals to humans. Most hospital laboratories use commercial tests to detect respiratory viruses like the flu, but these tests cannot tell if someone has the specific and more dangerous H5N1 strain. To help solve this, we tested three commonly used diagnostic tools and found that they can detect H5N1 even at low levels. However, since they cannot identify the specific H5 subtype, we also developed and validated a follow-up test that runs on one of the existing laboratory high-throughput equipment. This test can confirm whether a patient infected with the flu has the H5N1 strain. By combining these tools, hospital laboratories can improve early detection of H5N1, support better patient care, and help public health officials respond more effectively to outbreaks.

Keywords: Clade 2.3.4.4b; H5N1; LDT; LoD.

Fig 1

Probit analysis of LoD for Influenza A H5N1 on three commercial molecular diagnostic assays and the H5 LDT assay. Probit regression was used to determine the LoD of H5N1 using A/cattle/USA/Texas/56283/2024 for each assay. The x-axis represents viral RNA copies/mL, and the y-axis indicates the probability of detection. The LoD is defined as the viral RNA concentration corresponding to a 95% detection probability. The red circles are data points (dilutions tested) that represent the observed detection rates for each dilution, and the fitted probit curves represent the statistical modelling of detection probabilities. Shaded areas show the 95% CI around the probit curves.