Antiviral Therapy 2014; 19: (doi: /IMP2723)

Transcription

1 Antiviral Therapy 214; 19: (doi: /IMP2723) Original article Performance characteristics of the COBAS Ampliprep/COBAS TaqMan v2. and the Abbott RealTime hepatitis C assays implications for response-guided therapy in genotype 1 infections Ninon Taylor 1, Elisabeth Haschke-Becher 2, Richard Greil 1, Michael Strasser 3, Hannes Oberkofler 2 * 1 Department of Internal Medicine III with Hematology, Medical Oncology, Hemostaseology, Infectious Diseases, Rheumatology, Oncologic Center, Laboratory of Immunological and Molecular Cancer Research, Paracelsus Medical University, Salzburg, Austria 2 Department of Laboratory Medicine, Paracelsus Medical University, Salzburg, Austria 3 Department of Internal Medicine I, Paracelsus Medical University, Salzburg, Austria *Corresponding author h.oberkofler@salk.at These authors contributed equally Background: With the advent of the protease inhibitors boceprevir and telaprevir a novel therapy approach for HCV genotype 1 infected subjects has become standard of care. Quantification of HCV viral load (VL) represents an important predictor of treatment response. Methods: Two different real-time PCR platforms, the COBAS Ampliprep/COBAS TaqMan v2. (CAP-CTM v2.) and the Abbott RealTime (ART) HCV assay are most widely used. We performed a comparative evaluation of both systems focusing on genotype 1 HCV quantification using clinical specimens, the fourth WHO International Standard for HCV and the Paul Ehrlich National Standard, respectively. Results: The HCV VL assays showed an excellent overall agreement in the clinical specimens studied (R 2 =.912). Discrepant results were obtained at the low VL end. Four samples tested negative with CAP-CTM v2. but were detectable with ART and two samples were undetectable with ART but tested positive with CAP-CTM v2.. The coefficient of variation in replicate measurements of both reference materials was higher for CAP-CTM v2. as compared with ART at the clinical decision point for boceprevir ( 1 IU/ml), but was similar for the two assays at the clinical decision point for telaprevir ( 1, IU/ml). The tendency for underestimation of the diluted standards was higher for ART than for CAP-CTM v2.. Conclusions: Although both assays allowed accurate determination of VL levels in clinical samples, careful interpretation of results at the low VL end is essential. Furthermore, discontinuation of therapy based on single HCV RNA measurement should be carefully reconciled, unless the issue of assay variability has been addressed adequately. Introduction HCV infection is a worldwide public health problem and a major cause of chronic liver disease with an estimated 2 million people infected worldwide [1]. Patients suffering from chronic HCV infection are at high risk for progressive hepatic fibrosis and cirrhosis that may accumulate to end-stage liver disease, a leading indication for liver transplantation [2]. In the past decade, standard of care (SOC) therapy was based on a combination of pegylated interferon-a and ribavirin and has been associated with rates of sustained virological response (SVR) below 5% in treatment-naive subjects harbouring chronic genotype 1 infections [3,4]. Fortunately, regulatory approval of direct-acting antiviral (DAA) substances including the NS3/4A serine protease inhibitors boceprevir and telaprevir initiated a new era in the treatment of genotype 1 infected individuals [5,6]. Inclusion of these novel DAAs in pegylated interferon-a/ribavirin combinatory triple therapy has since led to a substantial improvement in SVR rates, and the option of abbreviated regimens resulted in a revision of treatment guidelines [7,8]. Given the high economic burden and severe potential toxicity associated with HCV therapy, early predictors of non-response 214 International Medical Press (print) (online) 449

2 N Taylor et al. were established to identify patients who should cease therapy because of futility. These futility rules allow minimization of drug exposure thereby avoiding adverse side effects or the emergence of viral resistance. Boceprevir-based treatment regimens should be stopped if the HCV RNA level is 1 IU/ml at treatment week 12 or if HCV RNA is detectable at treatment week 24 in treatment-naive patients. For telaprevir, futility rules include stopping of treatment if the HCV RNA level is 1, IU/ml at treatment week 4 or 12 and/or detectable HCV RNA levels are found at week 24 [7,8]. Similar recommendations have been developed for treatment-experienced patients [9,1]. Cutoff levels for the futility rules have been defined in clinical trials using the High Pure System (Roche Molecular Systems, Pleasanton, CA, USA) for RNA extraction and the COBAS TaqMan HCV Test, v2. real-time PCR assay (Roche Molecular Systems) with a lower limit of quantification (LLOQ) of 25 IU/ml [5,6]. In addition, patients without cirrhosis and an undetectable HCV RNA level at week 8 and week 24 for boceprevir and week 4 and 12 for telaprevir, respectively, may be considered for a shortening of treatment. For both DAAs, SVR is defined as HCV RNA less than 25 IU/ml at 24 weeks after triple therapy end [7]. Meanwhile several commercial assays are available that allow accurate measurement of HCV viral load (VL). These assays are based on real-time PCR technology, which is also recommended by international Clinical Practice Guidelines, due to the increased dynamic range and excellent sensitivity [11]. Among them, fully automated and workflow improved systems have been developed including the combination of the COBAS AmpliPrep extraction platform (Roche Molecular Systems) with the COBAS TaqMan 48 Real-Time Analyzer (CAP-CTM) and the Abbott m24sp automated sample preparation system (Abbott Molecular, Des Plaines, IL, USA) in combination with the Abbott m2rt Real-Time PCR instrument (ART). Since accurate quantification of HCV RNA is mandatory for response-guided therapy in boceprevir/telaprevir-containing therapy regimens we evaluated the two platforms focusing on genotype 1 HCV RNA quantification in a university hospital diagnostic laboratory setting. We performed a method comparison and assessed accuracy, reproducibility as well as concordance of both assays using residual clinical samples as well as replicate measurements of internationally recognized reference materials (the fourth WHO HCV International Standard and the Paul Ehrlich National Standard) diluted to nominal concentrations representing the clinical decision points (1 IU/ml and 1, IU/ml for boceprevir and telaprevir, respectively). Methods Undiluted clinical serum samples from 84 individuals infected with HCV genotype 1a (n=27) or 1b (n=57) covering various VL levels were recruited during 212 as part of routine testing from our hepatology outpatient clinic in Salzburg, Austria. Patients provided written informed consent and the study was approved by the local Ethics Committee in accordance with the 1975 Declaration of Helsinki. Blood samples were collected into a Vacutainer tube (Greiner Bio One; Kremsmünster, Austria) and centrifuged at 33 g for 1 min upon receipt in the diagnostic laboratory. Serum samples were immediately frozen at -8 C. After routine testing appropriate residual aliquots were prepared and subsequently stored again at -8 C until parallel testing with the two real-time PCR platforms to guarantee identical assay conditions. The Versant HCV Amplification 2. Kit (Siemens Healthcare Diagnostics Inc., Tarrytown, NY, USA) was used for reverse transcription and amplification of a 24-base-pair fragment of the 5 -untranslated region (5 -UTR) of the HCV genome. PCR products were purified with the QIAquick PCR purification kit (Qiagen, Hilden, Germany) according to the manufacturer s instructions. The TRUGENE HCV 5 NC Genotyping Kit (Siemens Healthcare Diagnostics Inc.) was used for bidirectional DNA sequencing of the 5 -UTR with two oligonucleotides labelled with different fluorescent dyes, followed by electrophoresis and data analysis on the OpenGene DNA sequencing system (Siemens HealthCare Diagnostics Inc.). Each bidirectional sequence was automatically aligned to a panel of reference sequences with the GeneLibrarian module 4.1 (Library: HCV_5NC_92) of the Gene-Objects software (Siemens HealthCare Diagnostics Inc.), allowing genotype assignment based on percent sequence identity. Samples with indeterminate HCV subtype were excluded from the study. For direct comparison of the different platforms, two internationally recognized standards were tested in parallel: the fourth WHO International Standard for Hepatitis C Virus for Nucleic Acid Amplification Techniques (NIBSC code: 6/12; National Institute for Biological Standards and Control [NIBSC]) with an assigned unitage of 26, IU/ml and the PEI Reference Preparation HCV RNA (number 3443/4; Paul- Ehrlich-Institut, Federal Agency for Sera and Vaccines, Langen, Germany) with an assigned unitage of 8, IU/ml. Dilutions of 1, IU/ml and of 1 IU/ml were prepared using HCV-RNA-negative human serum and 15 aliquots each were analysed in a single run using the CAP-CTM v2. and the ART assay, respectively. For the CAP-CTM v2. assays, automated viral RNA extraction was performed starting from 5 ml International Medical Press

3 The CAP-CTM v2. and ART assays in HCV genotype 1 infections patient s serum or diluted reference material using the COBAS Ampliprep system (Roche Molecular Systems, Inc.). PCR amplification was then performed on the COBAS TaqMan 48 system (Roche Diagnostics) using the HCV CTM v2. assay according to the manufacturer s instructions. Data were analysed with the Amplilink software version and HCV RNA levels were expressed in IU/ml. The assay has a dynamic range from 15 to 1,, IU/ml (8. log 1 IU/ ml). For the ART assay the Abbott m24sp automated sample preparation system (Abbott Diagnostics) was used for HCV RNA isolation starting from 5 ml patient s serum or diluted standard material. Extracted samples and controls were then amplified and detected with the Abbott m2rt Real-Time PCR instrument, using the Abbott RealTime HCV assay, according to the manufacturer s instructions. Data analysis was performed with the m2rt software version 6., and the Abbott RT HCV assay has a dynamic range from 12 IU/ml to 1,, IU/ml (8. log 1 IU/ml). The limit of detection (LOD) of both assays is equal to the LLOQ. The Abbott RealTime HCV assay was standardized against the second WHO International Standard for Hepatitis C RNA (NIBSC 96/ 798). Results are expressed as IU/ml or log 1 transformed IU/ml, as appropriate. The correlation coefficient (R value) was obtained using Deming regression analysis and the rates of agreement between RealTime HCV and CAP/CTM were determined from the mean differences in quantification for averaged log 1 values by Bland Altman plot analysis (Abacus Validation Systems, Jena, Germany). Intra-assay variability was expressed as the sd and the coefficient of variation (CV) based on the mean HCV RNA concentrations. The Levene s test was used to test for homogeneity of variance between repeat measurements (IBM SPSS statistics software, version 21; SPSS Inc., Chicago, IL, USA). Results We first compared HCV RNA levels as measured with the ART and the CAP-CTM v2. assay in 84 undiluted clinical serum specimens from genotype 1 infected individuals showing detectable HCV RNA with at least one assay (Figure 1). Three samples that tested positive but below the LLOQ and one sample with a VL of 26 IU/ ml with the ART assay were undetectable with the CAP- CTM v2. assay, whereas two specimens that tested positive but below the LLOQ with the CAP-CTM v2. assay had undetectable HCV RNA levels as measured with the ART assay. Linear regression analysis of logtransformed VLs of the remaining 78 paired samples quantifiable with both assays showed a good correlation with an R 2 value of.912 and a slope of 1.4 (Figure 1). Viral loads were within.5 log 1 IU/ml in 7 samples (89.7%). All samples with an absolute difference of >.5 log 1 IU/ml originated from subtype 1b infected subjects and in all samples the ART assay showed lower VL levels. The absolute difference in the mean HCV VL was.16 log 1 (the Abbott HCV test results being lower) in our clinical samples (range, ). All samples with HCV RNA levels >5 IU/ml measured by the CAP- CTM v2. assay were also found within the quantifiable range with the ART assay. To evaluate the accuracy of the two HCV real-time PCR assays at the novel clinical decision points for DAA-based therapy regimens two reference materials, the fourth WHO International Standard for HCV, NIBSC code 6/12 (WHO) and the Paul Ehrlich Institute regional HCV standard (PE), both HCV genotype 1 based, were measured at the relevant concentrations on both assay platforms. Intra-assay variation was evaluated by paired testing of 15 replicates each of diluted reference standards with nominal concentrations of 1 IU/ml and 1, IU/ml, respectively (Table 1). Using the WHO standard at a concentration of 1 IU/ ml the mean log 1 difference from standard was similar for both assays (.19 versus.18, CAP-CTM v2. versus ART). However, the CV was borderline significantly higher (P=.56) for the CAP-CTM v2. assay (CV=.51) in comparison to the ART assay (CV=.26). Similar results were observed for the PE standard, also demonstrating a higher CV for the CAP-CTM v2. assay. We observed that, using the PE standard at the concentration of 1 IU/ml, the ART assay showed a higher degree of under-quantification with a mean log 1 difference of.29 versus.14 (ART versus CAP-CTM v2.). At the nominal concentration of 1, IU/ml the Figure 1. Scatter plot of HCV viral load values as determined by the CAP-CTM v2. and the ART assay Abbott RealTime HCV, log 1 IU/ml COBAS Ampliprep/COBAS TaqMan v2., log 1 IU/ml COBAS Ampliprep/COBAS TaqMan HCV v2. assay (CAP-CTM) versus Abbott RealTime HCV assay (ART). The regression fitted line is superimposed (solid line). Antiviral Therapy

4 N Taylor et al. Table 1. Quantification of the fourth international WHO standard for HCV RNA and the Paul Ehrlich regional HCV standard using the CAP/CTM v2. and the Abbott HCV assay CAP/CTM v2. Abbott Real-Time HCV Mean Mean (±sd), Mean log 1 Mean D Mean (±sd), Mean log 1 Mean D (log 1 IU/ml) IU/ml CV (range) from standard IU/ml CV (range) from standard WHO standard nominal input (2.) (38.3) ( ) (17.8) ( ) 1, 1, (3.) (2.3) ( ) (96.8) ( ) Paul Ehrlich standard nominal input (2.) (33.4) ( ) (1.85) ( ) 1, (3.) (165.9) ( ) (69.2) ( ) CAP-CTM v2., COBAS Ampliprep/COBAS TaqMan v2.; CV, coefficient of variation; mean D, mean difference. CAP-CTM v2. had an excellent performance with mean differences from standard of.5 log 1 and a CV of.17 and.18 for both, the WHO and the PE reference material, respectively. HCV RNA levels were again slightly underestimated with the ART assay, especially in replicate measurements using the PE standard material showing a mean log 1 difference from standard of.26, but the CV was comparable to the CAP-CTM v2. assay. Bland Altman plots of the HCV RNA quantifications for the WHO and PE standard were used to determine the agreement between the CAP-CTM v2. and the ART assay by plotting the differences between the two measurement procedures against the mean log 1 results (Figure 2). For the PE standard at a nominal concentration of 1 IU/ml assay, the mean difference between the values of the CAP-CTM v2. versus ART assays was -.2 log 1 IU/ml HCV RNA, with limits of agreement (95% CI) of -.6,.3 (Figure 2A). Results for the WHO standard at the 1 IU/ ml yielded a mean difference of -.8 IU/ml (limit of agreement of -.5.5; Figure 2B). The mean difference between the values for the PE standard at a nominal concentration of 1, IU/ml -.2 was log 1 IU/ml, with limits of agreement of -.4 and -.4 log 1 IU/ml (Figure 2C). The mean difference for the WHO standard (concentration 1, IU/ml) was -.2 log 1 IU/ml, with limits of agreement of -.4 and.3 log 1 IU/ml (Figure 2D). Discussion The effective usage of boceprevir- and telaprevir based antiviral therapy regimens critically relies on the availability of accurate HCV RNA VL results at pre-specified time points to make correct decisions concerning duration of therapy as well as stopping of treatment due to futility or the emergence of resistance-associated variants [12]. Shortly after introduction of the first version of the CAP/CTM HCV assay [13], observations accumulated concerning the misquantification of several genotype 4 strains but also of a small percentage of genotype 1 strains [14 18]. Inclusion of an additional reverse primer as well as a dual-probe design for target detection resulted in a novel CAP-CTM v2. assay, with increased genotype inclusivity and minimized impact of possible sequence mismatches [19 21]. The ART assay utilizes a single-probe design with a fluorescent moiety covalently linked to the 5 -end and a quenching moiety covalently linked to the 3 -end of the probe to detect its target, which is also located in the 5 -UTR region of the HCV genome [22]. We compared the improved CAP-CTM v2. and the ART assay using genotype 1 restricted clinical samples and reference materials. There is one recent, initial performance evaluation comparing these assay platforms across all viral genotypes using clinical specimens, albeit measurements were obtained from two separate diagnostic laboratories in the Netherlands [23]. To our knowledge, this is the first direct, head to head comparison within the same diagnostic setting, which allows for the exclusion of pre-analytic and analytic handling issues. Despite an excellent overall agreement of the two assays over a wide dynamic range, it was of interest, that all results with an >.5 log 1 IU/ml difference originated from subtype 1b assigned samples, which indicates a consistent albeit relatively small difference between the assays, which is likely to be related to the 1b subtype. In addition VL results from these discrepant samples as well as mean VLs across all samples were lower as determined with the ART assay compared with results as determined with the CAP-CTM v2. assay. A similar observation was reported by the Dutch group [23] International Medical Press

5 The CAP-CTM v2. and ART assays in HCV genotype 1 infections Figure 2. Bland Altman plot analyses of HCV RNA levels as determined by the CAP-CTM v2. and the ART assay A [log 1 ART] [log 1 ART] C PE 1 IU/ml Average [(log 1 CAP-CTM V2. + log 1 PE 1, IU/ml WHO 1, IU/ml Average [(log 1 Average [(log 1 B [log 1 ART] D [log 1 ART] WHO 1 IU/ml Average [(log 1 CAP-CTM V2. + log 1 Paired testing of 15 replicates of diluted reference standards using the COBAS Ampliprep/COBAS TaqMan HCV v2. assay (CAP-CTM v2.) and the Abbott RealTime HCV assay (ART): (A) Paul Ehrlich Institute regional HCV standard (PE) with nominal concentrations of 1 IU/ml, (B) fourth WHO International Standard for HCV (WHO) with a concentration of 1 IU/ml, (C) PE standard with a concentration of 1, IU/ml and (D) WHO standard with a concentration of 1, IU/ml. In the Bland Altman figures, the difference between the HCV RNA levels obtained by the two assays is plotted as a function of the mean of the two values. The dotted lines represent the 95% CI. New SOC treatment guidelines are being developed for subjects harbouring genotype 1 infections [7] while stopping algorithms for pegylated interferon/ribavirin therapy regimens have been established on the basis of a less than 2-log 1 viral load drop at week 12 or detectable HCV RNA at week 24 [11], futility rules for the novel DAA protease inhibitors are now based on absolute measures of HCV RNA ( 1 IU/ml at week 12 for boceprevir and 1, IU/ml at week 4 or 12 for telaprevir, respectively) [7]. We have therefore evaluated the accuracy of the CAP-CTM v2. and the ART assay using replicate measurements of different genotype 1 based reference materials including the fourth WHO international HCV standard and the PE diluted to the clinically relevant concentrations. The coefficient of variation of the CAP- CTM v2. assay at the clinical decision point of 1 IU/ ml was significantly higher for both reference materials as compared with the ART assay. At the nominal concentration of 1, IU/ml the precision of the assays was similar. Both assays showed a slight underquantification of the reference materials diluted to the lower nominal concentration of 1 IU/ml in our analyses. The CAP- CTM v2. measured HCV RNA concentrations highly accurate at the concentration of 1, IU/ml, whereas the Abbott assay showed a clear trend for underestimation. Bland Altman analyses revealed a good overall agreement between the two assays. Of note, despite the reported high sensitivity of both assays [21,22] discrepant results were observed at the low VL end with four clinical samples not detected with the CAP-CTM v2. assay and two falsely negative clinical samples measured using the ART assay. This Antiviral Therapy

6 N Taylor et al. is consistent with our observation regarding a higher coefficient of variation at the lower end of the linear range. Our observations might therefore have substantial implications for current therapy guidelines suggesting treatment shortening in patients without cirrhosis whose HCV RNA level is undetectable at weeks 8 and 24 (boceprevir) or at weeks 4 and 12 (telaprevir). In these cases diagnostic laboratories and clinicians must clearly differentiate between HCV RNA results <LLOQ where the target RNA could still be detected and undetectable HCV RNA results (target not detected) and should also be aware of the fact that detectability of HCV RNA below the LLOQ critically depends on how close the actual amount of HCV RNA in the patient s sample is to IU/ml [12]. We therefore believe that decisions concerning discontinuation of therapy due to futility or treatment shortening should not be based on single HCV RNA measurements. Additional studies are needed to more robustly assess the analytical performance of these assays. Importantly, a switch of assays during therapy is not recommended due to the observed variation between different assays and clinicians must understand performance characteristics of HCV RNA real-time PCR assays to effectively manage patients treated with the novel DAAs. Acknowledgements We thank Iris Kremser, Helene Schnaitl and Evelin Wölbl for excellent technical assistance. This study was supported by a grant from the PMU Forschungsförderungsfonds (PMU-FFF). Disclosure statement The authors declare no competing interests. References 1. Global surveillance and control of hepatitis C. Report of a WHO Consultation organized in collaboration with the Viral Hepatitis Prevention Board, Antwerp, Belgium. J Viral Hepat 1999; 6: Freeman RB, Steffick DE, Guidinger MK, et al. Liver and intestine transplantation in the United States, Am J Transplant 28; 8: Fried MW, Shiffman ML, Reddy KR, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 22; 347: Manns MP, McHutchison JG, Gordon SC, et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 21; 358: Poordad F, McCone J, Bacon BR, et al. Boceprevir for untreated chronic HCV genotype 1 infection. N Engl J Med 211; 364: Accepted 4 December 213; published online 16 January Sherman KE, Flamm SL, Afdhal NH, et al. Response-guided telaprevir combination treatment for hepatitis C virus infection. N Engl J Med 211; 365: Ghany MG, Nelson DR, Strader DB, et al. An update on treatment of genotype 1 chronic hepatitis C virus infection: 211 practice guideline by the American Association for the Study of Liver Diseases. Hepatology 211; 54: Liang TJ, Ghany MG. Current and future therapies for hepatitis C virus infection. N Engl J Med 213; 368: Bacon BR, Gordon SC, Lawitz E, et al. Boceprevir for previously treated chronic HCV genotype 1 infection. N Engl J Med 211; 364: Zeuzem S, Andreone P, Pol S, et al. Telaprevir for retreatment of HCV infection. N Engl J Med 211; 364: EASL Clinical Practice Guidelines. management of hepatitis C virus infection. J Hepatol 211; 55: Cobb B, Pockros PJ, Vilchez RA, et al. HCV RNA viral load assessments in the era of direct-acting antivirals. Am J Gastroenterol 213; 18: Sarrazin C, Dragan A, Gartner BC, et al. Evaluation of an automated, highly sensitive, real-time PCR-based assay (COBAS Ampliprep/COBAS TaqMan) for quantification of HCV RNA. J Clin Virol 28; 43: Sarrazin C, Gartner BC, Sizmann D, et al. Comparison of conventional PCR with real-time PCR and branched DNAbased assays for hepatitis C virus RNA quantification and clinical significance for genotypes 1 to 5. J Clin Microbiol 26; 44: Chevaliez S, Bouvier-Alias M, Brillet R, et al. Overestimation and underestimation of hepatitis C virus RNA levels in a widely used real-time polymerase chain reaction-based method. Hepatology 27; 46: Vermehren J, Kau A, Gartner BC, et al. Differences between two real-time PCR-based hepatitis C virus (HCV) assays (RealTime HCV and Cobas AmpliPrep/Cobas TaqMan) and one signal amplification assay (Versant HCV RNA 3.) for RNA detection and quantification. J Clin Microbiol 28; 46: Chevaliez S, Bouvier-Alias M, Castera L, et al. The Cobas AmpliPrep-Cobas TaqMan real-time polymerase chain reaction assay fails to detect hepatitis C virus RNA in highly viremic genotype 4 clinical samples. Hepatology 29; 49: Akhavan S, Ronsin C, Laperche S, et al. Genotype 4 hepatitis C virus: beware of false-negative RNA detection. Hepatology 211; 53: Chevaliez S, Bouvier-Alias M, Rodriguez C, et al. The Cobas AmpliPrep/Cobas TaqMan HCV test, version 2., real-time PCR assay accurately quantifies hepatitis C virus genotype 4 RNA. J Clin Microbiol 213; 51: Zitzer H, Heilek G, Truchon K, et al. Second-generation Cobas AmpliPrep/Cobas TaqMan HCV quantitative test for viral load monitoring: a novel dual-probe assay design. J Clin Microbiol 213; 51: Vermehren J, Colucci G, Gohl P, et al. Development of a second version of the Cobas AmpliPrep/Cobas TaqMan hepatitis C virus quantitative test with improved genotype inclusivity. J Clin Microbiol 211; 49: Chevaliez S, Bouvier-Alias M, Pawlotsky JM. Performance of the Abbott real-time PCR assay using m2sp and m2rt for hepatitis C virus RNA quantification. J Clin Microbiol 29; 47: Pas S, Molenkamp R, Schinkel J, et al. Performance evaluation of the new Roche cobas AmpliPrep/cobas TaqMan HCV test, version 2., for detection and quantification of hepatitis C virus RNA. J Clin Microbiol 213; 51: International Medical Press