We evaluated the Roche COBAS TaqMan HCV Test For Use With The High Pure System (TaqMan HPS; Roche Diagnostics), for the extraction, detection and quantitation of hepatitis C virus (HCV) RNA in serum or plasma of HCV-infected individuals. The TaqMan HPS is a real-time PCR assay with a reported linear dynamic range of 3.0x10(1) to 2.0x10(8) HCV RNA IU/ml, and a reported lower limit of detection (LLD) of 10 IU/ml. Calculation of the HCV RNA titre is based upon an external standard curve in the presence of an internal control. Intra-assay and inter-assay variation were small in reference panel members with HCV RNA > or =100 IU/ml. Genotype performance and quantitative correlation between the TaqMan HPS and the bDNA (VERSANT HCV 3.0 assay; Bayer Diagnostics), assessed in 59 patient samples, were good for HCV genotype 1 but poor for genotypes 2, 3 and 4. For genotypes 2, 3 and 4, values obtained from the TaqMan HPS were in general 0.5 log lower than those from the bDNA. Sensitivity was poor in low viral titre samples of genotypes 1, 2, 3 and 4. The LLD (95%) was estimated at 41 HCV RNA IU/ml for genotype 4. The TaqMan HPS underestimates HCV RNA at all levels in plasma and serum from HCV-infected individuals, and the LLD should be reconsidered. This is clinically relevant because underestimation of HCV RNA levels during therapy may lead physicians into making incorrect treatment decisions.