No: 1539 Presentation at ESCV 2015: Poster 2 A European Multicentric Study on the analytical performance of DxN VERIS MDx system HBV assay R. Delgado1, M. Drago2, D. Fanti2, H. Fleury3, M.R. Gismondo5, J. Izopet4, A. Lombardi5, M.A. Marcos6, D. Mileto5, K. Saune4, P. Trimoulet3, J. Vila6, D. Whittaker7 1 Microbiology Department, Hospital Universitario 12 de Octubre, Madrid, Spain 2 Clinical Chemistry and Microbiology laboratories, Niguarda Ca’ Granda Hospital, Milan, Italy 3 Virology Department, Hopital Pellegrin, Bordeaux, France 4 Department of Virology, Federative Institute of Biology, CHU Toulouse, France 5 Virology Laboratory, Sacco Hospital, Milan, Italy 6 Department of Clinical Microbiology, Hospital Clinic, School of Medicine, University of Barcelona, Centre for International Health Research (CRESIB), Barcelona, Spain 7 Laboratory Medicine Building, NGH, Sheffield, United Kingdom Background: Beckman Coulter’s VERIS Molecular Diagnostic (MDx) system is a new fully automated system for quantitative and qualitative analysis of molecular targets. Seven European Virology Laboratories assessed theVERIS MDx HBV assay in order to determine its analytical performance. Methods:All sites followed the sameprotocol providedbyBeckman Coulter. Precision was performed on 5 QC controls provided by Beckman Coulter, at different HBV DNA levels (1.4 log IU/ml; 3.3 log IU/ml; 5.9 log IU/ml; 8.8 log IU/ml), tested in duplicate for 20 days. Analytical specificity was assessed on 786 anti-HBc and HBV-DNA negative patient samples. The limit of detection was determined using serial dilutions of HBV WHO standard run in replicate of 12 per day for 3 days. Determination was performed using Probit transformation analysis. Method comparison between VERISMDxHBVassay, and theRocheHBVassayand theAbbottHBV assay was determined at a total of 4 sites on 184 and 274 samples respectively. Results: For precision, the total standard deviation for all sites combined is ≤0.16 log IU/ml. The between site standard deviation and CV% are lower than 0.15 log IU/ml and 8% respectively. The overall analytical specificity is 100% (95% CI: 99.1–100%). Limit of detection for all sites combined is 3.43 IU/ml (3.18–3.67). Linearity is excellent in the range of 1.68–8.82 log IU/ml, with the following linear regression: y=0.996x+0.007 (r2 =1). The correlation between VERIS versus Abbott and Roche is good. Passing-Bablok regression shows VERIS log IU/ml =−0.089+0.93 Abbott log IU/ml with bias of 0.215, 0.271 and 0.386 log IU/ml respectively for 25th, 50th, 75th percentiles. The Pearson correlation shows a good correlation between the two assays (Pearson r=0.96, p<0.001). Bland–Altman analysis onVERIS versus Abbott assay showed an overall bias of −0.381 log IU/ml with a standard deviation of 0.596. Similar analysis versus Roche HBV assay shows VERIS log IU/ml =−0.224+0.96 Roche log IU/ml with Passing-Bablock analysis, bias of −0.297, −0.322 and −0.366 log IU/ml respectively for 25th, 50th and 75th percentiles. The Pearson regression is r=0.93. Bland–Altman shows an overall bias of −0.394 log IU/ml with a standard deviation at 0.557. Conclusion: VERIS MDx HBV assay shows good analytical performances at each of evaluation sites. The consistency of the results among the different sitesmake VERISMDxHBV assay a helpful and reliable tool for HBV DNA viral load determination. http://dx.doi.org/10.1016/j.jcv.2015.07.200