Clinical utility of total HCV core antigen quanti

Clinical Utility of Total HCV Core AntigenQuantification:A New Indirect Markerof HCV ReplicationMagali Bouvier-Alias,1Keyur Patel,2Harel Dahari,3St´e phanie Beaucourt,1Patrick Larderie,1Lawrence Blatt,2Christophe Hezode,1,4Gaston Picchio,5Daniel Dhumeaux,1,4Avidan U.Neumann,3John G.McHutchison,2and Jean-Michel Pawlotsky1Hepatitis C virus(HCV)RNA detection,viral load quantification,and HCV genotyping arewidely used in clinical practice.Recently,the availability of an anticore antigen(Ag)mono-clonal antibody allowed development of an enzyme-linked immunosorbent assay(ELISA)detecting and quantifying total HCV core Ag in peripheral blood of HCV-infected patients.The aims of the present study were to investigate the biologic significance of this new markerin HCV infection,to establish the intrinsic performance of the current assay,and to deter-mine its potential utility in the management of HCV-infected patients.A panel of infectedsera calibrated to the World Health Organization International Standard and657serumsamples from infected patients receiving antiviral treatment were studied.We showed thattotal HCV core Ag quantification is an accurate,precise,and specific indirect marker of HCVreplication.We estimated that1pg/mL of total HCV core Ag is equivalent to approximately8,000HCV RNA international units(IU)/mL,although minor between-patient differencesmay exist.In conclusion,total HCV core Ag quantification can be used in the variousindications of viral load monitoring,including the evaluation of baseline viral load beforetherapy,the assessment of the virologic response to antiviral treatment,and the study of earlyviral kinetics during therapy.Nevertheless,the total HCV core Ag assay cannot be used as amarker of viral replication for HCV RNA values below20,000IU/mL,limiting its use in themonitoring of late events during and after antiviral treatment.(H EPATOLOGY2002;36:211-218.) H epatitis C virus(HCV)is a single-strand RNAvirus belonging to the Flaviviridae family.Itsgenome is contained in an icosahedral capsid(or core),itself contained within the viral envelope.Thecapsid is formed by polymerization of the HCV core pro-tein,a structural viral protein encoded by the5Јend of theHCV open reading frame.1After translation,host signalpeptidases cleave the HCV core protein from the precur-sor polyprotein and remove the signal peptide located atits C-terminus.1The mature HCV core protein is a21-kdphosphoprotein made of the191first amino acids of thepolyprotein.In the cytoplasm of infected cells,it is locatedin close vicinity to the perinuclear membranes and theendoplasmic reticulum,where it polymerizes in the pres-ence of genomic RNA to form viral capsids.1The HCVcore protein is antigenic,interacts with numerous cellu-lar proteins,induces specific cellular and humoral re-sponses,2,3and through various pathways could play animportant role in the pathogenesis of HCV infection.4Virologic diagnosis and monitoring of HCV infectionare based on the use of serologic assays detecting specificanti-HCV antibodies(including anticore antibodies),and assays that can detect,quantify or characterize thecomponents of HCV viral particles.5In past years,clinicalhepatitis C studies have for the most part used molecularbiology–based HCV RNA techniques.Indeed,HCVRNA detection,viral load quantification,and HCV geno-typing are widely used in clinical practice.6Recently,theavailability of an anticore antigen(Ag)monoclonal anti-Abbreviations:HCV,hepatitis C virus;ELISA,enzyme-linked immunosorbentassay;OD,optical density;bDNA,branched DNA.From the1Departments of Virology(EA3489)and4Hepatology and Gastroen-terology,Hoˆpital Henri Mondor,Universite´Paris XII,Cre´teil,France;2the Divi-sion of Gastroenterology and Hepatology and the5Department of Immunology,Scripps Clinic,La Jolla,CA;and3Faculty of Life Sciences,Bar-Ilan University,Ramat-Gan,Israel.Received December19,2001;accepted April12,2002.Address reprint requests to:Jean-Michel Pawlotsky,M.D.,Ph.D.,Service deBacte´riologie-Virologie,Hoˆpital Henri Mondor,51avenue du Mare´chal de Lattrede Tassigny,94010Creteil,France.E-mail:jean-michel.pawlotsky@hmn.ap-hop-paris.fr;fax:(33)14-981-2839.Copyright©2002by the American Association for the Study of Liver Diseases.0270-9139/02/3601-0027$35.00/0doi:10.1053/jhep.2002.34130211body allowed development of an enzyme-linked immu-nosorbent assay(ELISA)that showed the presence of HCV core Ag in peripheral blood of HCV-infected pa-tients.7,8Results obtained in Japan with afirst-generation assay showed a significant relationship with the presence of HCV RNA and suggested that HCV core Ag detection, and quantification might be used to follow patients and predict the response to interferon alfa(IFN-␣)therapy.9 A standardized commercial assay using the same monoclonal antibody has been developed for the qualita-tive detection of HCV core Ag(Ortho Antibody to Hep-atitis C Core Antigen ELISA Test System;Ortho-Clinical Diagnostics,Raritan,NJ).This assay,conceived for the screening of blood donations,increases viral safety by sig-nificantly reducing the serologic window occurring before seroconversion during acute infection.Several studies have shown that core Ag can be detected,on average,1to 2days after HCV RNA during the preseroconversion period.10-13However,this screening assay exhibits de-creased sensitivity when analyzing HCV antibody–posi-tive samples.This is probably caused by the absence of a preliminary immune-complex dissociating step in the as-say’s procedure.In this regard,a new assay with potential application in the clinical setting,incorporating an im-mune-complex dissociating step and using the same monoclonal antibody,has been recently developed for detection and quantification of total HCV core Ag(Total HCV core Ag assay;Ortho-Clinical Diagnostics).The aims of the present study were(1)to investigate the biologic significance of this new marker in HCV in-fection,(2)to establish the intrinsic performance of the current assay,and(3)to determine its potential utility in the management of HCV-infected patients. Materials and MethodsMaterials.The following materials were used in this study:(1)A panel of infected sera calibrated to the World Health Organization International Standard,14the Nu-cleic Acid Panel(NAP)-HCV RNA(AcroMetrix Inc., San Francisco,CA),including the following panel mem-bers:50HCV RNA international units(IU)/mL,500 IU/mL,5,000IU/mL,50,000IU/mL,200,000IU/mL, 500,000IU/mL,and2,000,000IU/mL.15(2)A total of 392serum samples from44patients from Henri Mondor University Hospital,Cre´teil,France.These patients,in-fected with HCV genotype1b,received various schedules of IFN-␣with or without ribavirin.Blood samples were taken before treatment in all patients and frequently thereafter during thefirst weeks of treatment in20cases (i.e.,14days and2hours before the start of therapy,at the start of therapy,every4hours for24hours,every6hours for36hours,every12hours for48hours,at day14,and at day28of therapy).HCV RNA loads were measured with both the third-generation signal amplification-based branched DNA(bDNA3.0)assay Versant HCV RNA Quantitative Assay(Bayer Diagnostics,Emeryville,CA) and the reverse transcriptase-competitive polymerase chain reaction-based SuperQuant assay(National Genet-ics Institute,Los Angeles,CA).This collection of serum samples covered a wide range of viral loads representative of those routinely encountered in clinical practice.(3) Sixty-five serum samples from patients from Henri Mon-dor University Hospital infected with HCV genotype3a, taken before or at month6of IFN-␣treatment.HCV RNA loads were measured with bDNA3.0.(4)Two hun-dred samples from40patients infected with various HCV genotypes and receiving various schedules of IFN-␣or pegylated IFN-␣and ribavirin treatment at Scripps Clinic,La Jolla,CA.Blood samples were collected at the start of therapy,at weeks4and12of therapy,at the end of treatment,and at the end of follow-up6months later. HCV RNA loads were measured with the SuperQuant assay in35cases and with another reverse transcriptase polymerase chain reaction–based assay,the Amplicor HCV Monitor assay version2.0(Roche Molecular Sys-tems,Pleasanton,CA)in the remaining5cases.(5)A total of124sera from immunocompetent patients referred to Henri Mondor hospital for various medical conditions that tested negative for anti-HCV antibodies(ORTHO HCV3.0ELISA Test System with Enhanced SAVe;Or-tho-Clinical Diagnostics)and for HCV RNA in a quali-tative,nonquantitative polymerase chain reaction assay with a lower detection cutoff of50HCV RNA IU/mL (Amplicor HCV2.0;Roche Molecular Systems).(6)Se-rum samples from3HCV-infected patients containing different amounts of HCV RNA.(7)Serial one-third di-lutions of sera taken from7patients infected with differ-ent HCV genotypes,namely1a,1b,2b,3a,4a,5a,and6a, into a seronegative plasma.HCV Core Ag Testing.The NAP-HCV RNA panel was tested9times in5different runs to assess the dynamic range and linearity of quantification of the assay,as well as its precision and repeatability.To eliminate any interfer-ence,both the concentrated solution and the diluent used for preparing the panel also were tested.The392blood samples from Henri Mondor hospital were tested for total HCV core Ag and the results were compared with HCV RNA testing to assess the value of total HCV core Ag as a marker of viral replication,to determine the sensitivity of the core Ag assay relative to molecular biology techniques, to establish a relationship between total HCV core Ag and HCV RNA units,and to study the potential utility of total HCV core Ag quantification to monitor early HCV212BOUVIER-ALIAS ET AL.HEPATOLOGY,July2002replication kinetics during antiviral therapy.The200 samples from the Scripps Clinic were tested to assess the value of total HCV core Ag quantification as a routine marker of the virologic response to antiviral therapy.The sera from124anti-HCV antibody-negative,HCV RNA–negative individuals were tested to determine the specific-ity of the total core Ag assay.To determine the precision and reproducibility of HCV core Ag quantification,the2 positive controls provided in the kits were tested22times in22distinct runs and the sera from3patients containing different HCV RNA loads were tested4times in3dis-tinct runs.Finally the one-third dilutions of sera from7 patients infected with different genotypes were tested to verify that quantification was linear whatever the HCV genotype.HCV Core Ag Assay.Blood samples were assayed according to the manufacturer’s instructions.The kit lot number was DX-00001.One hundred␮L of samples and controls were mixed with50␮L of a pretreatment buffer containing detergents and pretreated for30minutes at56°C to dissociate HCV core antigen-antibody com-plexes.For the ELISA reaction,100␮L of pretreated samples and controls were incubated for60minutes at 25°C in the monoclonal antibody-coated wells of a mi-crotiter plate.The plates were washed and incubated for 30minutes at25°C with200␮L of conjugate,washed again,and incubated for30minutes at25°C with200␮L of substrate.The optical densities(ODs)were read in a spectrophotometer at490nm using a620nm reference. The samples and controls were tested in duplicate and the mean OD of each duplicate testing was used.The samples that exhibited more than25%variation between the two ODs were considered invalid and retested.As recom-mended by the manufacturer,the lower detection cutoff was established for each run and corresponded to the mean OD of the2negative controls plus0.041.A sample was considered positive only when the mean OD was higher than the cutoff OD of the corresponding run.The amount of total HCV core Ag in pg/mL was calculated by means of a standard curve established in each run by test-ing serial dilutions of a standard containing400HCV core Ag pg/mL.HCV RNA Detection and Quantification.The3 HCV RNA assays were performed according to the man-ufacturer’s instructions.The results were expressed in HCV RNA IU/mL.ResultsTotal HCV Core Ag Is a Marker of HCV Replica-tion.Figure1shows the relationship between total HCV core Ag and HCV RNA in392HCV genotype1b clinical samples covering a wide range of HCV RNA loads.HCV RNA was quantified in IU/mL with bDNA3.0,an accu-rate HCV RNA quantification assay.16In the293samples that were positive in both markers,total HCV core Ag and HCV RNA were significantly related(rϭ0.92;PϽ.0001).The relationship between HCV RNA in IU/mL and total HCV core Ag in pg/mL was given by the fol-lowing equation:HCV core Ag(log10pg/mL)ϭ0.9714ϫHCV RNA(log10IU/mL)Ϫ3.7461 Based on this equation,it was possible to calculate that 1pg/mL of total HCV core Ag is equivalent to7,688 IU/mL.Thus,total HCV core Ag is a reliable marker of HCV replication,and1core pg/mL is equivalent to ap-proximately8,000HCV RNA IU/mL in clinical samples from HCV-infected patients.The correlation between HCV RNA IU/mL and HCV core Ag pg/mL varied around this average ratio when individual samples were considered,with the majority of the ratios lying between 5,000and12,000HCV RNA IU/mL per core Ag pg/mL (Fig.2A).Because we studied several samples taken at different time points in the same patients,we were able to assess whether these differences were sample-or patient-related.As shown in Fig.2B,the correspondence between HCV RNA IU/mL and HCV core Ag pg/mL did not vary substantially within individual patients,but larger(al-though not dramatic)differences were seen between dif-ferent patients.To assess if the total HCV core Ag-HCV RNA rela-tionship was genotype-independent,we tested65geno-Fig.1.Relationship between HCV RNA(in log10IU/mL)and total HCV core Ag(in log10pg/mL)in293samples out of392tested that were positive for both markers.The2markers were significantly related(rϭ0.92;PϽ.0001).HEPATOLOGY,Vol.36,No.1,2002BOUVIER-ALIAS ET AL.213type 3a samples for both markers.Again,total HCV core Ag and HCV RNA were signi ficantly related (r ϭ0.90;P Ͻ.0001;data not shown).The relationship between HCV RNA and total HCV core Ag was given by the following equation:HCV core Ag (log 10pg/mL)ϭ0.9028ϫHCV RNA (log 10IU/mL)Ϫ3.2604One pg/mL of total HCV core Ag was equivalent to approximately 6,711IU/mL,a conversion factor of the same order as in genotype 1b.Performance of the Total HCV Core Ag Assay as a Marker of HCV Replication.Based on 22runs,the lower detection cutoff of the total HCV core Ag assay was on average 2.26Ϯ0.48pg/mL.Thus,based on the rela-tionship with HCV RNA mentioned above,the total core Ag assay did not detect HCV replication below 16,631Ϯ3,374HCV RNA IU/mL (i.e.,below approximately 20,000IU/mL).The clinical sensitivity of the assay was con firmed when testing the NAP –HCV RNA panel (Fig.3).Indeed,3panel members (200,000HCV RNA IU/mL,500,000IU/mL,and 2,000,000IU/mL,respec-tively)were always quanti fied in the core Ag assay.In contrast,the panel member containing 50,000HCV RNA IU/mL was found positive only once out of 9tests,with a value of 2.17pg/mL,just above the cutoff of the corresponding run.As shown in Fig.3,quanti fication was linear within the tested range of viral loads.Overall,the dynamic range of quanti fication of the total HCV core Ag assay was between its lower detection cutoff (approxi-mately 2pg/mL)to at least 250pg/mL.In addition,serial one-third dilutions of clinical samples from patients in-fected with various HCV genotypes (including 1a,1b,2b,3a,4a,5a,and 6a)showed that total HCV core Ag quan-ti fication was linear within its dynamic range of quanti fi-Fig.3.Total HCV core Ag detection and quantification in the NAP–HCV RNA panel,a panel calibrated to the HCV RNA IU/mL (AcroMetrix Inc.).Total HCV core Ag was constantly found in 3panel members (i.e.,200,000HCV RNA IU/mL,500,000IU/mL,and 2,000,000IU/mL,respectively),and its amount was significantly related to HCV RNA load in the corresponding range of values.The panel member containing 50,000HCV RNA IU/mL was found HCV core Ag-positive once out of 9experiments,with a core Ag value higher than the cutoff of the corre-sponding run,but lower than the mean of those in all of theexperiments.Fig. 2.Relationship between HCV RNA IU/mL and HCV core Ag pg/mL.On average,in the 293tested samples that were positive in both assays,the ratio was approximately 8,000HCV RNA IU/mL per HCV core Ag pg/mL.(A)HCV RNA IU/mL to HCV core Ag pg/mL ratio in 293individual samples testing positive in both assays.(B)Mean Ϯstandard deviation of the HCV RNA IU/mL to HCV core Ag pg/mL ratio of different samples taken at different dates in 20patients.214BOUVIER-ALIAS ET AL.HEPATOLOGY,July 2002cation irrespective of the HCV genotype(data not shown).One sample was found positive for total HCV core Ag among124samples from anti-HCV–negative,HCV RNA–negative patients,but was negative on repeat test-ing.The specificity of the assay was thus99.2%(95% confidence interval:94.3%-100.0%).The assay precision and reproducibility were assessed by testing the2positive controls provided in the kits22times in22distinct runs, and by testing sera from3patients containing different HCV RNA loads4times in4distinct runs.Overall,the within-run coefficients of variation ranged from7.5%to 33.2%,and the between-run coefficients of variation ranged from16.3%to26.7%.As recommended by the manufacturer,each sample was tested in duplicate,and the mean of the2determinations was used as thefinal result.Only8samples out of392(2%)had duplicate testing differing by more than25%and were retested in our study.The between-laboratory reproducibility was not evaluated.Clinical Use of Total HCV Core Ag Quantification to Monitor the Response to Therapy.Forty treated patients were tested before and during treatment,at the end of treatment,and at the end of follow-up for both HCV RNA and HCV core Ag,as detailed in the Materials and Methods section.They were classified into3catego-ries based on HCV RNA testing,including sustained vi-rologic responders(negative HCV RNA at the end of treatment and at the end of follow-up6months later); responder-relapsers(negative HCV RNA at the end of treatment,but positive HCV RNA detection at the end-of follow-up);nonresponders(positive HCV RNA detection throughout the entire follow-up period). Total HCV core Ag quantification allowed the same classification as the use of molecular biology testing in35 patients out of40(87%).The remaining5patients (13%)were misclassified with the core Ag assay.One patient with a baseline HCV RNA of265,000IU/mL was found to be HCV core Ag–negative at this time point,and throughout the entire follow-up period;he achieved sus-tained virologic responders after therapy.It is not yet clear whether this patient had no circulating HCV core Ag or the assay was not able to detect it.A nonresponder patient was HCV core Ag-negative at the end of treatment(viral load:10,300IU/mL)and6months later(viral load:150 IU/mL).Another nonresponder was found to be HCV core Ag–negative at day0(despite of a315,000IU/mL viral load)and was then classified as a responder-relapser with the core Ag assay(undetectable core Ag versus 10,000HCV RNA IU/mL at the end of treatment,59 core Ag pg/mL versus more than420,000HCV RNA IU/mL[undiluted sample]at the end of follow-up).Fi-nally,2responder-relapsers were HCV core Ag–negative at the end of follow-up(in spite of viral loads of1,900 IU/mL and2,450IU/mL,respectively).Overall,it ap-peared that most of the response misclassifications were related to the lower sensitivity of the core Ag assay relative to the HCV RNA assay.Clinical Use of Total HCV Core Ag Quantification to Study HCV Kinetics During Therapy.HCV repli-cation kinetics have been recently shown to be character-ized by a2-phase decrease during IFN-␣–based therapy.17,18Thefirst-phase decline occurs at day1,is rapid,and is likely related to direct inhibition of viral production or release.The second-phase decline starts at day2or later,is slower,and could be related to infected cell death in the context of efficient virus production blocking.17,18Our study of44patients receiving various IFN-␣–ribavirin schedules with frequent sampling showed that the changes from1time point to the next in HCV core Ag and in HCV RNA strongly correlated for all samples above20,000HCV RNA IU/mL(rϭ0.80, PϽ.001).The nonlinearfit of thefirst-phase of viral kinetics hada good agreement between HCV RNA and core Ag(Fig.4).This was attributable to the fact the former are deter-mined by early samples with HCV RNA titers generally higher than20,000IU/mL.Indeed,the mean effective-ness of3million units of IFN-␣to block HCV virion production was calculated to be89%Ϯ10%and91%Ϯ10%with core Ag and HCV RNA kinetics,respectively. The mean half-life of free virions was calculated to be 1.7Ϯ0.5and1.9Ϯ0.6hours with core Ag and HCV RNA kinetics,respectively.In contrast,the second-phase kinetics differed between the2assays.This was because of the fact that this phase is more dependent on latter samples that can fall below 20,000HCV RNA IU/mL(Fig.5).Indeed,the calcu-lated half-life of infected cells determining the second-phase slope differed on average between the2assays: 6.9Ϯ3.5and4.2Ϯ2.0days with the core Ag and HCV RNA kinetics,respectively.Overall,7out of10patients with a rapid viral response according to HCV RNA mea-surement(defined by a decrease of more than0.3logs per week during thefirst4weeks)also were identified as such with HCV core Ag measurement.In the remaining3 rapid responders,HCV core Ag kinetics apparently were slower because core Ag gave aflat second-phase slope, being below the limit of detection.DiscussionWe show in this study that total HCV core Ag is an accurate indirect marker of HCV replication that can be used in clinical practice.Indeed,HCV core Ag is signifi-HEPATOLOGY,Vol.36,No.1,2002BOUVIER-ALIAS ET AL.215cantly related to viral load measured by molecular biol-ogy –based techniques in clinical samples.However,in contrast with HCV RNA,HCV core Ag is not a product and direct marker of viral replication.We have established that 1HCV core Ag pg/mL is equivalent,on average,to approximately 8,000HCV RNA IU/mL,the standard-ized HCV RNA unit currently used in commercially available HCV RNA assays.Nevertheless,the core Ag to HCV RNA ratio can vary slightly from one HCV-in-fected patient to another,suggesting minor differences in the amount of core protein per HCV RNA molecule in the peripheral blood of different infected individuals.Be-cause the number of capsomers per virion is constant and each virion supposedly contains only 1genome molecule,our results are in keeping with the presence of core pro-teins not associated with viral genomes in peripheral blood,or vice versa.These discrepancies cannot be ex-plained by the recent report of nonenveloped viral nucleo-capsids in the serum of infected patients.19These various forms of the core protein in peripheral blood and their respective relationships to HCV replication remain to be determined.Despite these differences,our findings suggest that to-tal HCV core Ag quanti fication can be used as a marker of HCV replication in various clinical settings.Nevertheless,with the current version of the assay,assessment of viral replication below approximately 20,000HCV RNA IU/mL is not possible,reducing its utility in situations in which low viral loads are expected.In our hands,the performance of the total HCV core Ag assay was satisfactory.A qualitative assay using the same monoclonal antibody but lacking the pretreatment step failed to detect HCV core Ag in a signi ficant propor-tion of chronically infected patients,whereas it easily de-tected HCV core Ag at the preseroconversion phase,when no antibodies are present.10-13In contrast,the pre-treatment step aimed at dissociating HCV core Ag-anti-body complexes 7,8allowed the total HCV core Ag assay to detect and quantify HCV core Ag in all of the patients with suf ficiently high levels of replication.We found the assay to be speci fic and precise,reproducible,and accu-rate.This good reproducibility led the manufacturertoFig. 5.Two examples (A and B)showing discrepancies between second-phase viral kinetics estimated by HCV RNA testing (●)and total HCV core Ag testing (ᮀ),respectively,in patients receiving IFN-␣and ribavirin therapy.Despite similar results at the beginning of therapy,the curves diverged later on during therapy because of the lower sensitivity of total HCV core Ag assay at low replicationlevels.Fig. 4.Two examples (A and B)showing the close relationship between first-phase (days 1and 2)viral kinetics estimated by HCV RNA testing (●)and total HCV core Ag testing (ᮀ),respectively,in patients receiving IFN-␣and ribavirin therapy.216BOUVIER-ALIAS ET AL.HEPATOLOGY,July 2002recommend single determinations instead of duplicate ones in the currently available version.In addition,total HCV core Ag quantification appeared to be independent of the HCV genotype.We used total HCV core Ag quantification in various clinical settings where viral load assessment is indicated and compared the results with HCV RNA quantification by means of molecular biology assays.Our results show that total HCV core Ag quantification can be used in2 settings,namely before treatment and to monitor treat-ment efficacy.Before treatment,HCV core Ag quantifi-cation may help tailoring treatment dose and/or duration and serves as a reference for subsequent monitoring.Clin-ical trials are now needed to establish the HCV RNA and HCV core Ag thresholds that could be useful to tailor chronic hepatitis C treatment with the combination of pegylated IFN-␣and ribavirin(unpublished data).20To-tal HCV core Ag quantification also may be used during and after treatment to assess the response to therapy with the limitation of its lower detection cutoff,equivalent to approximately20,000HCV RNA IU/mL.The study of HCV RNA kinetics during treatment was recently shown to aid in understanding the mechanisms of action of an-tiviral drugs.17,18,21In addition,early viral kinetics have a good predictive value for the lack of sustained virological response(i.e.,ultimate treatment failure),and can be used to make early clinical decisions aimed at improving the overall results of therapy while reducing side effects(un-published observations).20,22,23HCV RNA kinetics dur-ing IFN-␣–ribavirin-based therapy have been shown to include2successive phases of viral decrease.Thefirst rapid phase occurs at day1and is likely caused by IFN-␣blocking of HCV production or release,whereas the sec-ond phase starts at day2or later,is slower,and could be related to infected cell death in the context of efficient blocking of viral production.17,18The latter leads to HCV RNA disappearance in the absence of breakthrough dur-ing therapy and to sustained virologic response in the absence of posttreatment relapse.A sustained response corresponds to a cure of the disease in the vast majority of cases,as a result of definitive infected cell clearance.24,25 Our study also shows that total HCV core Ag quantifica-tion can be reliably used to studyfirst-phase kinetics.This could prove to be particularly useful in research studies and in clinical practice to evaluate the antiviral effect of IFN-␣and make early predictions of response and appro-priate clinical decisions.Nevertheless,one should be more cautious when using total HCV core Ag quantification for monitoring the long-term response to therapy.Indeed, the assay does not correlate with HCV viral load below 20,000HCV RNA IU/mL,so that the assessment of sec-ond-phase kinetics,end-of-treatment response and sus-tained virologic response may differ between the2assays. In such indications,HCV RNA detection with a sensitive assay(lower detection cutoff of the order of50IU/mL) remains the standard.6In summary,total HCV core Ag quantification is an accurate and precise indirect marker of replication in HCV-infected patients.One total HCV core Ag pg/mL is equivalent to approximately8,000HCV RNA IU/mL. Total HCV core Ag quantification can be used in vari-ous settings in which there is a need to monitor viral load,although clinically relevant decision thresholds need to be established in prospective clinical trials.Nev-ertheless,the assay cannot be used as a marker of vi-ral replication for HCV RNA values below20,000IU/ mL.A new assay with better sensitivity is currently under development.Acknowledgment:The total HCV core Ag kits used in this study kindly were provided by Ortho-Clinical Diag-nostics.The authors thank David Hendricks and Fran-c¸oise Huisse(Bayer Diagnostics)for providing the bDNA 3.0kits,Paul Neuwald(AcroMetrix Inc.)for providing the NAP-HCV RNA panel,and Charles Tackney for helpful discussion.References1.Reed KE,Rice CM.Overview of hepatitis C virus genome structure,polyprotein processing,and protein properties.Curr Top Microbiol Im-munol2000;242:55-84.2.Nelson DR,Marousis CG,Davis GL,Rice CM,Wong J,Houghton M,Lau JY.The role of hepatitis C virus-specific cytotoxic T lymphocytes in chronic hepatitis C.J Immunol1997;158:1473-1481.3.Satoi J,Murata K,Lechmann M,Manickan E,Zhang Z,Wedemeyer H,etal.Genetic immunization of wild-type and hepatitis C virus transgenic mice reveals a hierarchy of cellular immune response and tolerance induc-tion against hepatitis C virus structural proteins.J Virol2001;75:12121-12127.i MMC,Ware CF.Hepatitis C Virus Core Protein:Possible Roles inViral Pathogenesis.In:Hagedorn CH,Rice CM,eds.The hepatitis C viruses.Berlin:Springer-Verlag;2000;117-134.5.Pawlotsky JM.Diagnostic tests for hepatitis C.J Hepatol1999;31(Suppl1):71-79.6.Pawlotsky JM.Molecular diagnosis of viral hepatitis.Gastroenterology2002(in press).7.Orito E,Mizokami M,Tanaka T,Lau JY,Suzuki K,Yamauchi M,et al.Quantification of serum hepatitis C virus core protein level in patients chronically infected with different hepatitis C virus genotypes.Gut1996;39:876-880.8.Aoyagi K,Ohue C,Iida K,Kimura T,Tanaka E,Kiyosawa K,Yagi S.Development of a simple and highly sensitive enzyme immunoassay for hepatitis C virus core antigen.J Clin Microbiol1999;37:1802-1808.9.Tanaka E,Ohue C,Aoyagi K,Yamaguchi K,Yagi S,Kiyosawa K,Alter HJ.Evaluation of a new enzyme immunoassay for hepatitis C virus(HCV) core antigen with clinical sensitivity approximating that of genomic am-plification of HCV RNA.H EPATOLOGY2000;32:388-393.10.Courouce AM,Le Marrec N,Bouchardeau F,Razer A,Maniez M,La-perche S,Simon N.Efficacy of HCV core antigen detection during the preseroconversion period.Transfusion2000;40:1198-1202.HEPATOLOGY,Vol.36,No.1,2002BOUVIER-ALIAS ET AL.217。