Several studies have shown a relationship between pretreatment hepatitis C virus (HCV) viral load and the response to interferon (IFN) therapy, creating a need for quantitative HCV-RNA assays. Here, we compared three commercial methods: nucleic acid sequence-based amplification NASBA (Organon), branched DNA 2.0 (bDNA) (Chiron), and Monitor (Roche), with reverse-transcription polymerase chain reaction (RT-PCR) as the reference. We assessed sensitivity and reproducibility on a well-characterized panel of sera (EUROHEP), a Chimp Rodney plasma pool, and samples from IFN-treated and -untreated patients with chronic hepatitis C caused by different HCV genotypes. The reproducibility of the NASBA and bDNA methods was slightly better than that of Monitor, especially for genotypes 2 and 4. NASBA had the highest sensitivity (99% vs. 94% and 88% with Monitor and bDNA, respectively), especially for the follow-up of patients on IFN. NASBA gave the highest HCV-RNA concentrations, which were approximately 10-fold more than with the bDNA assay and 100-fold more than with the Monitor kit. The linearity, tested on the chimp Rodney plasma pool, was better with bDNA for high viral load than with NASBA and Monitor, although for low concentration of HCV RNA, bDNA was negative. Pretreatment viral load was lower in patients who had a sustained virological response to IFN, although the bDNA method was not sensitive enough to quantify all pretreatment samples. This study indicates that gene amplification methods (NASBA or Monitor) have better sensitivity than bDNA assays for quantification of HCV RNA in patients with chronic HCV infection, although the bDNA and NASBA methods are more likely to quantify all genotypes. Prospective studies are needed to demonstrate the usefulness of quantitative assays for the follow-up of patients with chronic hepatitis C.