Amplification Testing in Clinical Samples for Cytomegalovirus Nucleic Acid Performance of the MagNA Pure 96 System

diagnostic medical devices (1). Accordingly,CE marking on the MagNA Pure 96 system declares the capabilityoftheproducttoextractpurenucleicacidswithhighrecoveryandreproducibility, which ensures reliable clinical diagnostics. Evalu-ation of MagNA Pure 96 nucleic acid extraction using clinicalrespiratory specimens for identifying viruses by qualitative real-time PCR was recently reported (2). Rapid, sensitive, accurate,and predictive molecular cytomegalovirus (CMV) assays in theroutinediagnosticlaboratoryareofincreasinginterestforthepre-vention and management of CMV diseases in immunocompro-mised patients, such as transplant recipients (3, 4). In addition,CMV screening is of particular importance concerning preg-nancy, because CMV is one of the most common causes of con-genital infections (5, 6).The present study evaluated the performance of the MagNAPure96systemforCMVDNAisolationfrom200lofcellculture(standard material) and of clinical plasma specimens. The extrac-tion on the MagNA Pure 96 system was carried out with the “ViralNA Plasma Protocol” and/or with the “Pathogen Universal Pro-tocol” (elution volume, 50 l). In comparison, the “Cell-free Pro-tocol” on the QIAsymphony SP platform (Qiagen, Hilden, Ger-many) was used for sample processing (elution volume, 85 l) inour in-house CMV workflow. Data of studies that have evaluatedthe QIAsymphony SP extraction platform in diagnostic laborato-ries were previously published (7–9).IsolatedCMVDNAwasquantifiedusingareal-timePCRassayaccording to Mori et al. (10) by applying a two-step protocol on aRoche LightCycler 2.0 instrument. All samples were processedtogether with 100 copies of a plasmid-based internal control oflambda DNA flanked by CMV primer sequences to exclude false-negative results.At the beginning of the study, the viral load of our CMV stan-dard material was verified on a Cobas Amplicor Analyzer usingthe Cobas Amplicor CMV Monitor test (Roche Diagnostics).Then, dilutions of the standard material were subjected to multi-ple processing in one run and in runs on several days according tothe recommendations of the European Pharmacopoeia to deter-mine performance parameters. The obtained low intra- and inter-run variations demonstrated a good recovery of the viral loadstested(coefficientofvariation,2%;standarddeviation,lessthan0.5 copies/ml). The limit of detection (LOD), the point at which95% of the replicates of a given viral load are detected, was deter-mined by a statistical approach (Probit analysis). The LOD of theCMV workflow that included MagNA Pure 96 extraction with theViral NA Plasma Protocol as well as with the Pathogen UniversalProtocol was found to be 330 copies/ml. This result represents ananalytical sensitivity slightly higher than that determined for ourroutine CMV workflow with the QIAsymphony SP (LOD 750copies/ml). Excellent linearity of quantitative CMV testing wasobserved at all times (Pearson’s correlation coefficient