Diagnostic Assay Recommended by the World Health Organization for Swine Origin Influenza A (H1N1) Virus Cross-Reacts with H5N1 Influenza Virus

Abstract

Pandemic influenza poses a significant risk to global human health. The unpredictable nature of the influenza virus necessitates accurate, reliable, and rapid diagnostic assays for the detection of, and characterization of the current pandemic caused by, swine origin influenza virus (SOIV) in order to better understand the nature of the pandemic and implement appropriate health interventions. SOIV emerged in mid-April 2009 (1), leaving laboratories around the world scrambling to establish a diagnostic test to detect this novel influenza virus. The World Health Organization (WHO) responded with remarkable speed by releasing guidelines and protocols for a real-time reverse transcription (RT)-PCR assay 15 days after the reported identification of SOIV (2). This provided a rapid and sensitive assay for the detection of SOIV that was critical for the determination of the spread and extent of SOIV infections. These guidelines recommend three primer-and-probe sets: InfA, amplifying a conserved region of the matrix gene from all influenza A viruses; SW H1, designed to specifically detect the hemagglutinin gene segment (subtype H1) from SOIV; and SW InfA, designed to specifically detect the nucleoprotein (NP) gene segment from all swine influenza viruses. As part of the WHO Global Influenza Surveillance Network, the New Zealand National Influenza Centre regularly participates in WHO external quality assurance panels of 10 RNA samples that are H1N1, H3N2, or H5N1 influenza A virus or, recently, SOIV. In following the WHO recommended guidelines exactly, it was found that the SW InfA primer-and-probe set amplified all six H5N1 RNA samples. This is clinically relevant, as the proficiency panel RNA returned threshold cycle (CT) values (InfA assay, 23.44 to 28.67) well within the range of what we regularly observe with clinical samples. In addition, the SW InfA primer-and-probe set detected two additional H5N1 viruses not provided in the proficiency panel, the human H5N1 strain A/Vietnam/3028/2004 and an avian H5N1 strain, A/Mallard/New Zealand/272-73/2008. The efficiency of this reaction was 79% (three replicates) compared to 94% (five replicates) for the amplification of SOIV RNA, suggesting that mismatches within the primer target region reduced assay efficiency. To prove that the H5N1 NP gene sequence was amplified by the SOIV SW InfA assay, all real-time RT-PCR amplicons were sequenced. The two SOIV RNA samples from the proficiency panel and SOIV-positive control showed 100% identity to the consensus sequence of 394 NP full-length gene segments available as of 29 July 2009. In contrast, all H5N1 RNA samples from the proficiency panel contained at least 31 nucleotide differences from the SOIV consensus sequence but showed between 93.3% and 96.8% identity to A/Vietnam/1203/2004. Alignment of all SOIV and human H5N1 NP gene segments available in GenBank showed that 391 SOIV viruses have one mismatch in the SW InfA forward primer (Table ​(Table1).1). The remaining three SOIV sequences have a further mismatch in either the forward primer or the probe. The majority (74.6%, or 91 sequences) of available human H5N1 sequences are identical within the SW InfA primer-and-probe regions, having two mismatches in the forward primer, two in the probe, and four in the reverse primer region. None of the mismatches occurred within 9 nucleotides of the 3′ end, an important determinant for primer specificity. TABLE 1. Comparison of SOIV and human H5N1 NP gene sequences with SW infA primer and probe sequences The speed at which these assays were developed and deployed to the WHO Global Influenza Surveillance Network was remarkable, particularly with the limited SOIV sequence information available at the time of development. Other factors which constrained the development of RT-PCR primer targets include the origin and genetic rate of change for each gene segment; only five of the eight gene segments are of swine origin (3), while the rapidly evolving HA and NA genes have a propensity to drift at a much higher rate than conserved segments such as NP (5). This report demonstrates that the SW InfA assay is not specific to SOIV and is able to detect both human and avian (H5N1) influenza A viruses and so there is the potential for misidentification. Following WHO guidelines, the utilization of all primer-and-probe sets for the detection of SOIV should provide an accurate diagnosis. However, we have observed that for the same SOIV RNA sample, the SW InfA assay consistently returns lower CT values than the SW H1 assay and so clinical samples with higher CT values can yield an SW H1-negative, SW InfA-positive result. In these cases, we suggest sequencing of real-time SW InfA PCR amplicons to ensure the amplification of SOIV (4). Alternatively, and especially for countries in which H5N1 is endemic, an H5N1-specific assay could be performed to exclude this diagnosis. However, a negative H5N1 result would not exclude other influenza virus subtypes that may cross-react with the SOIV assay.