A competitive reverse transcription (RT)-nested polymerase chain reaction (PCR) assay for HCV RNA was compared with the Roche Amplicor HCV Monitor assay, based on non-competitive, single step RT-PCR. A total of 83 serum samples were tested in parallel by both assays. All samples could be quantified by competitive RT-PCR (cPCR), whereas seven were negative by the non-competitive assay (ncPCR). The HCV RNA titre of the discordant samples assessed by cPCR was significantly lower than that of the remaining 76 ( P < 0.001). Absolute HCV RNA titres were higher by cPCR than by ncPCR ( P < 0.001), even if the results of the two methods were statistically correlated ( P < 0.001). HCV RNA titre tested by cPCR was not significantly different between samples infected with genotype 1 or 2. However, values obtained by ncPCR were higher in samples with genotype 1 ( P < 0.001). Furthermore, all seven discordant samples were infected with genotype 2. When both methods were used to measure serial dilutions of standard HCV RNA, we observed a bias to lower measurements with the ncPCR kit. This study shows a good correlation between the results of two PCR-based methods for the quantification HCV RNA. However, the degree of sensitivity and the absolute HCV RNA titre measured may vary according to the assay used.