Virtual quantification of influenza A virus load by real-time RT-PCR

Abstract

BackgroundThe pan-influenza A real-time RT-PCR detection assay developed by the Centers for Disease Control and Prevention (CDC) during the 2009 pandemic is widely utilized. A quantitative version of the assay may be useful to monitor influenza A infection and response to treatment. ObjectivesTo prove in principle the possibility that a virtual quantification tool (VQT) would allow conversion of CDC real-time RT-PCR cycle threshold (Ct) values in virus RNA copy number. Study designA plasmid carrying the CDC real-time RT-PCR target region of the influenza A Matrix (M) gene was generated. In a multicenter study, a set of 5 ten-fold dilutions (equivalent to 1×102 to 1×106copies/reaction) were prepared and distributed to the 4 participating virology laboratories and then amplified to generate a virtual quantification standard curve. Clinical samples (n=120) were quantified in parallel by interpolation with locally generated standard curves and using the VQT. ResultsA total of 40 standard curves were obtained by the participating centers (ten from each center). The intra- and inter-laboratory variability showed a coefficient of variation (CV) ≤5%. Influenza A virus quantification in 120 respiratory samples showed a significant correlation between interpolation with locally generated standard curves and the VQT (R2=0.9655). Bland Altman analysis showed that the majority (no. 111, 92.5%) of clinical samples had <0.5log10 variation. ConclusionsVQT proofs the concept that qualitative results from real-time RT-PCR assays can be converted into quantitative determination of virus load in clinical samples without running standard curves in parallel.