Drs. Wilber and Urdea and the editorial of Tedeschi and Seeff [1] raise several useful points. Although we agree with most of the comments, we need to clarify several issues.

With regard to the letter by Wilber and Urdea, we did not conclude that a prozone effect had occurred. Rather, we documented an unexpectedly high kit-to-kit variation in a subset of high-titer specimens, which showed a greater than 10-fold discordance between our reference assay (quantitative PCR) and the Chiron assay (bDNA). Because the linear range of quantitative PCR exceeds that of bDNA by several orders of magnitude, we were able to identify such discordant specimens, which other investigators have been unable to do. To date, the assay variation has only been observed in a relatively small subset of cirrhotic patients and liver transplant recipients, and the cause of the variation remains unknown.

Regarding the absolute comparison between bDNA and quantitative PCR-based assays, we reiterate that we did all the comparisons in parallel and from similar aliquots; whether the "standards" are RNA equivalents as defined by the bDNA assay or the number of RNA molecules as measured by our methods, the two assays gave similar results. Drs. Wilber and Urdea acknowledge the limited linear range of the bDNA assay compared with the quantitative PCR test. We address this limitation in bDNA sensitivity to help educate clinicians and researchers about the utility and limitations of the assay for diagnostic and therapeutic monitoring. We emphasize that persons with a negative test result may still have persistent infection, whether before or after therapy. Moreover, a negative result after therapy did not predict cure. The fact that the assay has been modified to increase its sensitivity indicates Drs. Wilber and Urdea's concern with the lower sensitivity and reflects their assumption that circulating titers of HCV RNA are important prognostic markers.

Drs. Tedeschi and Seeff raise many important issues in their editorial [1]. We would like to comment on one part of their recommendation. We feel that HCV RNA levels should be measured routinely before, during, and after interferon therapy, both to assess response to therapy and to define whether dose escalation or prolonged therapy is required. The information from this assay is more informative than that obtained from measurement of aminotransferase levels, and the method is less costly than liver biopsy. Quantitative PCR testing and bDNA testing can be accurately done in regional reference laboratories. Shipping of serum samples to such laboratories for testing and storage of these samples for repeat sampling are highly feasible. Through the use of current technology, such regional reference laboratories can be established using appropriate quality controls. We believe the use of such laboratories will greatly improve the clinical care of persons with chronic hepatitis C, will reduce the morbidity from prolonged ineffective therapy, and will ultimately reduce the costs of treating chronic HCV infection.