Assessment of characterised run controls and an internal quality control programme to monitor inter-run assay performance for transplant associated viral pathogens

No: 1578 Presentation at ESCV 2015: Poster 2 Evaluation of the H-DiaCMVQ kit for detecting and quantifying CMD-DNA in whole blood samples C. Mengelle, J.-M. Mansuy, K. Saune, L. Houles, J. Izopet Department of Virology, Federative Institute of Biology, CH Toulouse, France Background: The aim of our study was to evaluate the clinical performances of the H-DiaCMVQ kit (Diagenode, Seraing, Belgium) on whole blood samples by comparing the results with those obtained with the method used in our laboratory for monitoring CMV-DNA inwholeblood specimens (referencemethod–Mengelle et al., 2003). Methods: Clinical performances were evaluated from potassium-EDTA samples collected among hospitalized patients. Whole blood sampleswere tested prospectivelywith the reference technique (limit of detection 74 IU/mL i.e 1.87 log10 IU/mL): positive (n=50) and negative (n=100) samples were selected for further analysis. Selected samples were stored at −20 ◦C pending batch analysis then were tested with the H-DiaCMVQ kit. DNA was extracted from whole blood with the MagNA Pure 96 DNA and Viral NA Small Volume kit® on the MagNA Pure 96TM instrument (Roche Molecular Diagnostics, Meylan, France) according to protocol DNA blood SV 2.0 (input volume: 200 l, output volume: 100 l; resulting in a two-fold concentration). 20 L of internal control was automatically dispensed during extraction. Real-time PCR targeting the glycoprotein B was realized on the Light cycler 480 (Roche Molecular Diagnostics). XLSTAT 2014 was used for all statistical analyses. Results: 136/150 samples gave concordant results (37 positive and 99 negative samples) (p<0.0005). Fourteen samples gave discrepant results: one sample was negative with the reference technique and positive with the H-DiaCMVQ kit (401 IU/ml) and 13 samples gave opposite results. For these samples, mean viral load was 221.7 IU/ml (range 89–530 IU/ml). The Deming regression analysis of log10 concentrations of positive (n=37) clinical specimens showed a good correlation between the two methods (slope, 0.84 [CI95%: 0.36–1.32] and y-intercept value, 1.07 [CI95%:0.94–1.21]). TheBland–Altman representation showed that viral loads obtained with the H-DiaCMVQ kit were higher than those obtained with the reference technique: the average deviation was 1 log10 IU/ml (standard deviation=0.44). The mean viral load obtained with the H-DiaCMVQ kit was 4 log10 IU/ml; it was 3 log10 IU/ml with the reference technique. Conclusion: This study showed that CMV-DNA detection and quantification using the H-DiaCMVQ kit gave comparable results to those obtainedwith the reference technique. These resultsmake suitable the use of this technique in themonitoring of CMV-DNA in whole blood. http://dx.doi.org/10.1016/j.jcv.2015.07.218 Abstract No: 1579 Presentation at ESCV 2015: Poster 2 Assessment of characterised run controls and an internal quality control programme to monitor inter-run assay performance for transplant associated viral pathogens A. Ricketts1, C. Hinds1, R. de Boer2, H.G.M. Niesters2 1 Qnostics, UK 2 University Medical Center Groningen, The