INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

The broad potential diagnostic and therapeutic monitoring applications of PCR have prompted a significant effort toward improving the performance and reliability of hepatitis C virus (HCV) quantitation assays. The occurrence of false-positive results due to amplicon carryover contamination, false-negative results due to enzymatic inhibition, and poor reproducibility of PCR, shown previously by several proficiency studies, has always been a concern (2, 4, 12, 29). However, technical advances marked by the introduction of uracil-N-glycosylase and internal controls have successfully addressed most of these issues and have allowed an increased use of PCR in clinical laboratories (5, 17, 19, 23). Standardization and accuracy of quantitation have been fundamental in the case of HCV, a pathogen that cannot be directly detected by conventional techniques and whose viremia levels have important clinical implications (13, 21, 26). Indeed, the inclusion of HCV RNA levels in the diagnostic algorithm for chronic hepatitis C has been made easier by the partial automation of the procedure achieved with the development of dedicated laboratory instrumentation (3, 9, 28). The recent development of an automated PCR system for the quantitation of HCV RNA may further support the clinical use of viral load monitoring in the management of chronically infected individuals (1, 15).

In this respect, we have evaluated the analytical and clinical performance of a new automated test, the COBAS AMPLICOR HCV MONITOR test, version 2.0 (v2.0), compared to those of the first-generation manual version (AMPLICOR HCV MONITOR) and nested PCR. The COBAS AMPLICOR HCV MONITOR test, v2.0, incorporates changes in the formulation of the amplification solution that are likely to improve its ability to efficiently quantify all major viral subtypes, whereas the first-generation AMPLICOR HCV MONITOR test was reported to underestimate the quantities of genotypes other than genotype 1.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

The COBAS AMPLICOR HCV MONITOR test, v2.0, was evaluated in two phases. The first phase focused on intra-assay, interassay, and interoperator reproducibility, and the second phase focused on the efficiency of quantification of different HCV genotypes and a comparison to nested PCR.

Analytical evaluation. To assess the precision of the automated test, coded samples, obtained by diluting high-titer serum specimens previously quantified by the COBAS AMPLICOR HCV MONITOR test, v2.0, were provided to each of the participating laboratories by Roche Molecular Systems. Two specimens each infected with genotypes 1, 2, and 3 were diluted to approximately 10³, 10⁴, 10⁵, and 10⁶ copies/ml and were tested in six replicates per run for intra-assay reproducibility, two replicates per run for 6 subsequent days for interassay reproducibility, and two replicates per run performed by two operators for interoperator reproducibility.

Clinical evaluation. A total of 211 serum samples, which contained genotype 1a, 1b, 2a, 2b, 3a, 3b, 4, or 5, were obtained from 87 patients during interferon (IFN) therapy for type C chronic hepatitis. The diagnosis was based on serological, biochemical, and histological parameters, and the presence of HCV RNA was determined by an in-house nested PCR or a single-step reverse transcription-PCR (AMPLICOR HCV test; Roche Diagnostic Systems, Branchburg, N.J.) (28). HCV RNA levels had also been previously quantified in 101 specimens by the AMPLICOR HCV MONITOR test (Roche Diagnostic Systems) (3).

Automated HCV RNA quantitation. The concentrations of HCV RNA in both the coded and the clinical samples were determined by the COBAS AMPLICOR HCV MONITOR test, v2.0, by following the manufacturer's instructions. In brief, 100 µl of serum was subjected to chaotropic lysis in the presence of known amounts of an internal quantitation standard (QS). The QS is a synthetic RNA transcript with primer binding regions identical to those of the HCV target sequence, a randomized internal sequence of similar length and base composition as the HCV target sequence, and a unique probe binding region that differentiates the QS from the target amplicon. After isopropanol precipitation and an ethanol wash, the target viral RNA and the QS were resuspended in the Specimen Diluent (Roche), and this mixture was mixed with an equal volume of the amplification ready solution (Master Mix; Roche) containing the primers KY78 (biotinylated) and KY80 (nonbiotinylated), deoxynucleoside triphosphates, AmpErase, and rTth DNA polymerase.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

The performance of the COBAS AMPLICOR HCV MONITOR test, v2.0, for the quantitation of HCV RNA was assessed by the two participating laboratories, laboratories A and B, with coded, mock-infected specimen panels and then with clinical specimens.

As shown in Tables 1 and 2, the interassay, intra-assay, and interoperator reproducibilities of the assay were analyzed by testing several replicates of samples that each contained genotype 1b, 2b, or 3a at expected HCV RNA concentrations of approximately 10³, 10⁴, 10⁵, and 10⁶ copies/ml. The variations between the mean and the observed concentrations for the intra-assay, interassay, and interoperator analyses were 25.7, 34.5, and 25%, respectively. Additionally, the mean CVs for the individual genotypes were similar. Several dilutions of a high-titer, coded specimen containing genotype 1b were also used to analyze the dynamic range of the assay. In these experiments, performed in triplicate, the COBAS AMPLICOR HCV MONITOR test, v2.0, showed a linear range that spanned 3 logarithms, from 10³ to 10⁶ copies/ml (data not shown). Similar results were also obtained with cloned, synthetic transcripts of genotypes 1 to 5 quantified spectrophotometrically at 260 nm (Fig. 1).

View larger version (17K): [in this window] [in a new window]   FIG. 1.   Quantitation of different HCV genotype transcripts by the COBAS AMPLICOR HCV MONITOR test, v2.0 (CA HCM). Serial dilutions of synthetic, genotype-specific HCV transcripts of known concentration were analyzed by the COBAS AMPLICOR HCV MONITOR test, v2.0.

The efficiency of the automated system in quantifying HCV RNA independent of genotype was further evaluated by testing specimens obtained from 72 patients during IFN therapy for chronic type C hepatitis. Compared with the first-generation AMPLICOR HCV MONITOR test, the COBAS AMPLICOR HCV MONITOR test, v2.0, produced a one- to fivefold increase in HCV titers with all genotypes tested. With synthetic transcripts this observation was most marked for genotypes 2, 3, and 4; however, the two tests showed a significant correlation (Fig. 2). The differences in the absolute copy number determined by each test method showed the same profile and did not affect the ability to monitor the viral load changes over the time course of viremia observed in patients under antiviral treatment (Fig. 3).

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Comparison of COBAS AMPLICOR HCV MONITOR test, v2.0, and AMPLICOR HCV MONITOR test in the quantitation of different HCV genotypes. Specimens containing different concentrations of genotype-specific HCV RNA were analyzed by the two tests.

View larger version (12K): [in this window] [in a new window]   FIG. 3.   Viral load profile of a patient infected with genotype 1b, determined by using the COBAS AMPLICOR HCV MONITOR test, v2.0 (), and the first-generation AMPLICOR HCV MONITOR test ().

The viral loads were measured by the COBAS AMPLICOR HCV MONITOR test, v2.0, in a subset of clinical samples taken from 24 patients during IFN therapy and were compared to the nested PCR results. The nested PCR test was able to detect virus in 45 of 64 (70%) specimens infected with HCV genotype 1, 2, or 3. The COBAS AMPLICOR HCV MONITOR test, v2.0, was able to detect and quantify virus in 42 of 64 (66%) of the same set of specimens. The three specimens in which virus was not detected by the COBAS AMPLICOR HCV MONITOR test, v2.0, were infected with genotypes 1a, 1b, and 2. These three specimens were sequential samples with low titers of virus taken from patients who had shown primary biochemical and virological responses to IFN therapy.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The low level of efficiency of IFN therapy in the management of chronic hepatitis C has prompted the search for predictive markers that refine the timing of treatment and improve treatment regimens (3, 6). Viral load appears to be one of the most informative markers both before and during therapy, but its application in clinical practice is still controversial (7, 8, 10, 11, 16, 18, 22, 24, 25, 27). One of the factors that has prevented its widespread use is the less than optimal reproducibility, the variable dynamic range, and the HCV genotype-dependent sensitivities of the available assays (14, 20, 21, 30).

In the present study we have evaluated the performance of an automated, quantitative PCR assay, the COBAS AMPLICOR HCV MONITOR test, v2.0, that has recently been developed to overcome some of these issues.

In a first phase of the evaluation we analyzed the interassay, intra-assay, and interoperator precisions of the COBAS AMPLICOR HCV MONITOR test, v2.0, using standard specimens diluted to several concentrations. Overall, the reproducibility was characterized by mean CVs of 26 to 35%. This is an improvement over the 20 to 60% CVs shown by the first-generation manual version of the test (3). Automation of amplicon dilution and detection postamplification is probably the step that has contributed the most to the increased precision of the COBAS AMPLICOR HCV MONITOR test, v2.0. Although the CVs for some concentrations are quite large, the associated SDs of less than or equal to 0.33 log are not clinically significant.

The assay showed a dynamic range that spanned 3 logarithms, between 10³ and 10⁶ copies/ml. This range appears to be suitable for the monitoring of viremia changes during antiviral therapy because it can discriminate patients with high versus low viremia levels and has adequate sensitivity to assess primary virological responses (8, 10, 11, 16, 25, 27). The extent of viral clearance during the first 4 weeks of treatment is considered a valid marker of the response to treatment by IFN monotherapy. The dynamic range may be useful for determination of the timely modification of treatment for the patients who do not show any significant viral load changes compared to those at the baseline (8, 16, 23, 30). The clinical performance of the COBAS AMPLICOR HCV MONITOR test, v2.0, was also evaluated with samples that have recently been assessed by an in-house, nested PCR assay, previously reported to have a detection limit of 10² copies/ml (22). Only three specimens with viruses that were detectable by nested PCR had levels of viremia below the limit of detection of the COBAS AMPLICOR HCV MONITOR test, v2.0. These samples, obtained from patients who had shown a primary response to IFN therapy, probably contained very low levels of HCV RNA, as often seen in sustained responders prior to viral clearance. Another aspect of test performance investigated was the influence of genotype on quantitation, recently pointed out as one of the drawbacks of the first-generation manual assay (14, 20). Several changes in the formulations of the test reagents have been introduced to improve performance, including a buffer change, the addition of a cosolvent to the amplification solution, and the packaging of Mn²⁺ solution as a separate reagent. It is probable that these modifications have resulted in a more efficient hybridization of the primers to the stem-loop structures of the 5' noncoding region target sequence and in an increase in overall reagent stability. These changes have produced equivalent efficiencies of amplification of genotypes 1 through 5, demonstrated with synthetic transcripts and clinical specimens. Although these improvements have resulted in a difference between the results generated by the two versions of the test, there is no impact on the monitoring of treatment responses, as the kinetics of viremia follow the same pattern. Additionally, the COBAS AMPLICOR HCV MONITOR test, v2.0, is being calibrated to the World Health Organization International Standard for HCV. This will provide a mechanism for easier comparison of results between future versions and other technologies.

Even though the relative change in copy number between serial samples rather than absolute copy number is most important when one is monitoring therapies, the improved genotype-specific performance of this assay will make the data obtained by different tests more easily interpretable and comparable. In addition, the workload reduction and increased accuracy granted by automation may expand the use of viral load monitoring in the management of chronic type C hepatitis and help in better defining its benefits and limitations.