Validation of sample pooling for high-throughput RT-qPCR subtyping of avian influenza A(H5)

Highly pathogenic avian influenza A(H5N1) is a global health threat due to its high historical virulence and increasing mammalian host range, as demonstrated by the 2024–2025 outbreak among U.S. dairy cattle, with 71 confirmed human infections. While human-to-human transmission has not yet been documented, limited pre-existing human immunity, and uncertainty about evolutionary barriers to human infection make H5N1 a potential pandemic threat. Expanding testing capacity for H5N1 beyond its currently limited scope would be critical for response to potential outbreaks. Thus, we developed and validated a dual-target reverse transcription PCR assay for H5 subtyping of pooled influenza A-positive samples. The assay targets two regions of the hemagglutinin (HA) gene of influenza A(H5). Using contrived samples containing inactivated bovine H5N1 virus, we determined the assay’s 95% lower limit of detection for individual samples (58 copies/mL; 95% confidence interval [CI]: 50 to 66 copies/mL) and eight-sample pools (2,465 copies/mL; 95% CI: 1,969 to 3,756 copies/mL). The assay demonstrated 100% positive percent agreement, detecting H5 in all contrived H5-positive pools (32/32 pools; 95% CI: 89.1% to 100.0%) where the contrived positive samples ranged from 500,000 to 3,906 copies/mL. We observed 100% negative percent agreement in H5-negative, influenza A-positive pools (32/32; 95% CI: 89.1% to 100.0%). Though sample pooling reduced analytical sensitivity, contrived samples with H5 virus present at levels typically found in human respiratory infections were detected in eight-sample pools. This approach has the potential to increase testing capacity for influenza A (H5) while maintaining adequate sensitivity, enhancing preparedness for potential outbreaks.