Comparison of different HCV viral load and genotyping assays

Background: We report an interlaboratory comparison of methods for the determination of hepatitis C virus (HCV) serum load and genotype between a recently, established molecular laboratory at the Alaska Native Medical Center (ANMC) and two independent laboratories using different assays. At ANMC, a Real-time quantitative RT-PCR amplification methodology (QPCR) has been developed in which HCV viral loads are determined by interpolation of QPCR results to those of standards calibrated to the World Health Organization (WHO) First International Standard for HCV. HCV genotype is subsequently determined by direct sequencing of the DNA fragment generated from the QPCR assay. Objectives and Study Design: The above methods were statistically compared to results obtained for the same patient sera by two independent laboratories using different commercially available viral load assays; Quantiplex(TM) HCV RNA (Bayer Diagnostics) and Amplicor(TM) HCV Monitor(TM) (v 2.0) (Roche Molecular Systems), as well as two different genotyping assays; restriction fragment length polymorphism, (RFLP) and INNO-LiPA HCV 11 (Innogenetics). Results: ANMC's Real-time QPCR HCV viral load results compared moderately well with those obtained by the Quantiplex(TM) HCV RNA method (R-2 = 0.3813), and compared quite well with recent lot numbers of Amplicor(TM) HCV Monitor(TM) in which viral loads are derived in IU/ml (R-2 = 0.6408), but compared poorly with earlier lot numbers of Amplicor(TM) HCV Monitor(TM) in which viral loads were derived in copies/ml (R-2 = 0.0913). The ANMC direct sequencing method for genotype determination compared moderately to very well with both the RFLP (84-86%) and INNO-LiPA (85-97.5%) methods. Conclusions: These viral load comparisons highlight the discrepancies that may occur when patient HCV viral loads are monitored using different types of assays. Comparison of HCV genotype by different methods is more reliable statistically and important clinically for predicting probability of response to antiviral therapy. However, viral loads are important for monitoring response once therapy has begun. (C) 2002 Elsevier Science B.V. All rights reserved.