Evaluation of two methods for quantitation of hepatitis C virus RNA

Background: Accurate quantitation of hepatitis C virus (HCV) RNA in serum may provide a means to predict disease course and response to interferon-alpha therapy. Several quantitative assays are commercially available, but none have been accepted as the gold standard. Methods and Results: The branched DNA quantitative hybridization assay (Quantiplex HCV 1.0, Chiron, Emeryville, CA) and a quantitative reverse transcription polymerase chain reaction (Amplicor HCV MONITOR, Roche Diagnostic Systems, Branchburg, NJ) were compared using a panel of 53 sera from patients with chronic hepatitis C. All sera contained HCV RNA of known genotype. Overall, there was a positive correlation between the results for the 41 sera that gave discrete values in both tests (r = .81, linear regression; P less than or equal to .01, Kendall's rank test); however, the mean number of HCV copies per milliliter was 13.5-fold higher with Quantiplex (P less than or equal to .01). A plot of the difference between methods against their means showed poor agreement between the methods. No correlation between the results of the two tests was observed for sera with MONITOR values greater than 5.0 x 10(5) copies/mL. Discrete MONITOR values were obtained for all 12 sera that were below the lower limit of quantitation of Quantiplex (mean, 1.78 x 10(5)). Parallel testing of serial dilutions of two sera showed that each method gave linear responses over the stated dynamic ranges; however, the proportional systematic error was greater with MONITOR. The mean coefficient of variation for replicate determinations was 23% for Quantiplex and 45% for MONITOR (P = .13). Conclusions: Despite a positive correlation, systematic differences exist between the two methods for quantitation of HCV RNA and they cannot be used interchangeably.