Usefulness of the Hepatitis C Virus Core Antigen Assay for Screening of a Population Undergoing Routine Medical Checkup
http://www.100md.com
微生物临床杂志 2005年第4期
Unite Virus et Pseudovirus: Morphogenese et Antigenicite LEA3865, Universite Franois Rabelais, Tours
Institut inter Regional pour la Sante, La Riche, France
ABSTRACT
We studied the usefulness of the recently designed Trak-C assay for the detection and quantification of the hepatitis C virus (HCV) core antigen (Ag) for the screening of HCV infection in 4,201 subjects selected from 74,150 consecutive volunteers undergoing routine medical checkups. Subjects were selected for screening because they had risk factors (group II, n = 321) and/or elevated alanine transaminase activity (group I, n = 3847). Initially, the anti-HCV antibody assay and the Trak-C assay were performed on each patient. Subsequently, the Trak-C assay was performed only when the anti-HCV enzyme immune assay (EIA) was positive. Positive samples were further evaluated for anti-HCV antibodies by a third-generation strip immunoblot assay and for HCV RNA. Four samples (1.2%) from group II and 113 (2.9%) from group I were anti-HCV EIA positive. We also tested 33 subjects who previously tested positive for anti-HCV in our medical center. Among the 150 anti-HCV EIA-positive samples, the HCV core Ag result was in accord with the HCV RNA result in 146 cases (97.3%). When the EIA result was positive, the HCV core Ag concentration and the HCV RNA load were correlated (r2 = 0.78; P < 0.001). Four samples with low viral loads were Trak-C negative but HCV RNA positive. Among the 2,395 anti-HCV EIA-negative serum samples collected during the first part of the study, 17 (0.7%) were found to contain very low levels of HCV core Ag (<8.5 pg/ml, the cutoff value being 1.5 pg/ml). All these samples were HCV RNA negative and considered to be false positives. This was confirmed by HCV core Ag neutralization analysis. The HCV core Ag assay is a useful method in the screening strategy of HCV infection and provides a reliable means of distinguishing between current and cleared HCV infections that is well correlated with HCV RNA testing.
INTRODUCTION
The method currently recommended for identifying subjects with hepatitis C virus (HCV) infection is an enzyme immune assay (EIA) for the detection of anti-HCV antibodies (16). However, this assay sometimes generates false-positive or false-negative results. Furthermore, it is not always possible to distinguish between current and past cleared infections. Thus, supplementary tests are required. The strip immunoblot assay, a more specific anti-HCV serological test, is useful to distinguish true-positive from false-positive EIA results. Positive results from nucleic acid testing (NAT) for HCV RNA indicate active HCV infection (1). Assays that detect the HCV core antigen (Ag) have been developed to diagnose current HCV infection. The first such assay developed was a qualitative assay, conceived for screening blood donations. This assay increased viral safety by significantly reducing the length of the window preceding seroconversion (4, 13). A second test was subsequently developed for the detection and quantification of total HCV core antigen (Trak-C; Ortho-Clinical Diagnostics, Raritan, N.J.). This assay, incorporating an immune complex dissociation step, was designed for blood donation screening (3) and also for treatment monitoring (11, 12).
We investigated the usefulness of this new HCV core antigen assay for the screening of HCV infection in subjects undergoing routine medical checkups provided by the French national health insurance system. We looked at whether this assay is a useful initial test for the diagnosis of HCV infection in this kind of population and whether it can efficiently discriminate between previous and current infections. Finally, we asked whether it gives reliable information for an accurate medical follow-up.
MATERIALS AND METHODS
Population and study design. We investigated a series of 74,150 consecutive subjects who underwent routine medical checkups provided by social security medical centers between December 2001 and December 2002. These patients all lived in 10 administrative areas (called departments) in the western part of France. These medical centers are part of the French national health insurance system. Every 5 years, they offer biomedical examination to individuals who spontaneously attend the health center or who are directly invited to attend the medical center. The medical checkup includes a series of biological tests, followed by a clinical examination. The eligible population includes working people and their families as well as individuals in a "precarious" social situation, such as those with no paid employment and recipients of the welfare fund. Precarious populations can have a checkup every year. During the study period, 16,921 (23%) of the 74,150 individuals who underwent routine medical checkups were in precarious situations. The study population is described in Table 1.
Subjects were selected for HCV screening on the basis of two criteria. Group I (n = 3,847) included individuals showing a significant elevation of serum alanine transaminase (ALT), defined as ALT activity 20% above a norm (N) adjusted for age and sex. Results are expressed as ALT value/N. This screening strategy has been described previously (5, 7). Many of the subjects from group I who were found to be positive for anti-HCV antibodies had no record of HCV seropositivity in our center. We contacted their general practitioners to determine whether they were aware of previous positive tests. As recommended by the French Ministry of Health, a second blood sample was taken from subjects who were considered to have newly discovered HCV infection, and these subjects were retested with an anti-HCV assay differing from that used for the initial screening (see "Laboratory methods").
Group II (n = 321) included individuals with HCV risk factors, such as blood transfusion before 1991, past or current intravenous (i.v.) drug abuse, contact with HCV-infected relatives, or multiple operations. This group only included individuals with normal ALT values and no previous positive anti-HCV tests. During the same period, we also identified a group of 33 subjects with normal ALT values who had previously tested positive for anti-HCV antibodies in our medical centers and who returned for medical checkup 2 to 8 years after their initial screening (group III).
A questionnaire was sent to the general practitioner of all positive subjects to explore HCV risk factors and to record the medical follow-up of subjects who were known to be HCV positive. The questionnaire was approved by the Commission Nationale de l’Informatique et des Libertes (no. 26674, version 4).
Laboratory methods. Screening for anti-HCV antibodies was performed with a third-generation EIA (Ortho HCV 3.0 enzyme-linked immunosorbent assay). All EIA-positive samples were further analyzed using a third-generation strip immunoblot assay (Chiron RIBA HCV 3.0). According to the manufacturer's recommendations, the RIBA was scored as positive if an antibody reaction was observed with two or more viral proteins.
The HCV core antigen assay (Trak-C) is an enzyme immunoassay that allows the detection and quantification of total HCV core Ag despite the presence of anti-HCV antibodies. The detection limit of the method, as established by the manufacturer, is 1.5 pg/ml. Tests were performed in duplicate and were scored as positive if a reaction was observed in both cases. The HCV core antigen assay was carried out on every sample during the first part of the study (7 months) and only for anti-HCV antibody-positive samples during the second part of the study (6 months). In total, 41,768 subjects underwent the medical checkup during the first period, and 2,459 (group I, 2,340 subjects; group II, 119 subjects) were selected for screening; 32,382 individuals were seen during the second period, and 1,709 (group I, 1,507 subjects; group II, 202 subjects) were screened.
When the anti-HCV assay, the HCV core Ag assay, or both were positive, samples were further investigated for HCV RNA with a quantitative assay (bDNA HCV RNA 3.0 assay; Bayer Corporation, Tarrytown, N.Y.). A more sensitive qualitative assay (Amplicor HCV detection kit 2.0; Roche Diagnostics Corporation, Branchburg, N.J.) was used when the quantitative assay and/or the HCV core Ag assay was negative. Complementary assays were performed in case of discrepancies between HCV core Ag and HCV RNA assays: HCV core Ag-positive and HCV RNA-negative samples were tested with a neutralization assay (Ortho Trak-C confirmatory assay; Ortho-Clinical Diagnostics); HCV core Ag-negative and HCV RNA-positive samples were genotyped (VERSANT HCV genotype assay [LiPA]; Bayer Corporation Diagnostics Division).
RESULTS
Anti-HCV antibody screening. (i) Group I (elevated ALT value). Of the 3,847 subjects in group I, 113 (2.9%) were found to be anti-HCV EIA positive. Precarious populations, who accounted for 23% of the studied population, accounted for 50% of the anti-HCV-positive results (Table 1).
Sixty-seven of these 113 subjects (59%) had a previous positive anti-HCV test. RIBA confirmed that all were positive. Forty-six individuals were not aware of their HCV seropositivity before entering the study. Forty-one (89%) of these cases were confirmed positive by RIBA. Of the five others (11%), two reacted with only one viral antigen (core and NS5, respectively), two reacted with only human superoxide dismutase antigen, and one was completely nonreactive. Identical results were obtained for the initial and control samples for these five cases; thus, they were considered to be false positives. Thus, of the 108 anti-HCV-positive subjects with elevated ALT, 41 (38%) were not aware of their positivity before entering the study.
(ii) Group II (normal ALT value, risk factors, no previous positive tests). In group II, 4 (1.2%) of 321 subjects were anti-HCV EIA positive. All belonged to the precarious population. None were aware of their HCV seropositivity before their medical checkup, and all were confirmed positive by RIBA.
(iii) Group III (normal ALT value, known positive tests). All 33 subjects in group III were confirmed to be anti-HCV positive by RIBA.
HCV core Ag and HCV RNA in anti-HCV antibody EIA-positive subjects. The results of the HCV core Ag and/or HCV RNA tests are shown in Table 1. The relationships between the results of the HCV core Ag and HCV RNA tests among the 150 anti-HCV EIA-positive subjects (113 from group I, 4 from group II, and 33 from group III) are shown in Table 2. Of these 150 samples, 121 were HCV RNA positive, and 117 were HCV core Ag positive (Table 2). Most (146 of 150; 97.3%) yielded concordant results with the two methods (positive, 117; negative, 29). We found that 115 (95%) of the 121 HCV RNA-positive samples had viral loads above 20,000 IU/ml; all of these samples were also HCV core Ag positive. The five samples that were anti-HCV EIA positive but negative or indeterminate with RIBA (false-positive anti-HCV tests) were negative according to the HCV core Ag assay and HCV RNA assay. The sensitivity of the HCV core Ag assay was 96.7% (117 of 121 results were positive). HCV core Ag levels (logs) and HCV RNA loads (logs) were strongly correlated, with a correlation coefficient of 0.78 and a slope of 0.88. From the relationship between HCV core Ag values and HCV-RNA load, log (HCV RNA) = 0.841 x log (HCV Ag) + 1.097, the detection limit of the HCV RNA assay could be estimated to be 5,000 IU/ml. HCV core Ag levels were not correlated with ALT values (r2 < 0.03; P = 0.08).
Four subjects were HCV RNA positive but HCV core Ag negative. These four subjects (false HCV core Ag negative) had low HCV RNA viral loads (ranging from 1,900 to 15,600 IU/ml). They were all aware of their HCV seropositivity before the medical checkup (three from group I and one from group III). Details of their HCV status and medical information are shown in Table 3.
HCV core Ag and HCV RNA in anti-HCV antibody EIA-negative subjects. Among the 2,395 anti-HCV EIA-negative samples that were tested for HCV core Ag, 17 (0.71%; 15 in group I and 2 in group II) were repeatedly found to be positive. All these serum samples exhibited low levels of HCV core Ag (1.5 to 8.5 pg/ml). They were all negative by a neutralization assay and were all negative for HCV RNA by reverse transcription (RT)-PCR. Thus, the specificity of the HCV core Ag assay was 99.3% and reached 100% after neutralization analysis. Besides, the examination of a consecutive serum sample revealed no case of late seroconversion, indicating an absence of HCV infection in these patients.
Relationship between HCV status and other medical factors. Among the 100 subjects who were aware of their anti-HCV positivity before their medical checkup, 36 had been treated. HCV viremia was significantly more frequent when subjects had not been treated (91 versus 61%; P > 103) and when the ALT level was elevated (95 versus 48%; P > 106). Among the 20 subjects negative for viremia, 14 had been treated, and HCV infection had spontaneously cleared in 6 cases.
DISCUSSION
Several different serological tools are available to identify HCV infection. Positivity for anti-HCV antibodies indicates past or current HCV infection but gives no indication about viral replication. In contrast, molecular detection of HCV RNA indicates active viral replication and calls for thorough clinical evaluation (10, 16). Qualitative RNA assays are currently used to diagnose ongoing HCV infection among anti-HCV antibody-positive individuals. They are also used to detect viremia in subjects during the early phase of infection prior to anti-HCV seroconversion (window period), for instance, in blood banks (8). Quantitative HCV RNA assays are mainly used to monitor anti-HCV treatment (21). Trak-C, a new serological assay that can be used to detect and to quantify circulating HCV core antigens, was initially designed to monitor the efficacy of HCV treatment as an alternative to measurement of HCV RNA viral load (2, 12, 20). This test has also been shown to be appropriate for screening for HCV infection in blood donors (3). The ability of this assay to identify viremic subjects in the general population has never been evaluated prospectively.
NATs are now routinely used by French blood banks to detect HCV viremia in donors during the window period. Retrospective studies carried out with the Trak-C assay showed that the HCV core antigen can detect early HCV viremia during this window period (9) and might represent an alternative to NAT. However, further studies are necessary to evaluate the HCV core antigen assay prospectively before it can be used routinely in blood banks.
In France, the prevalence of blood donors who are susceptible to being tested during such a window period is very low (1 in 106 donations; 95% confidence interval, 1 in 3.1 x 106 to 1 in 0.4 x 106) (14). In the present study, the population included groups at higher risk of HCV infection, such as i.v. drug abusers (6), thus increasing the probability of recruiting individuals in this window period. However, none of the 2,395 anti-HCV-negative subjects was positive for HCV core antigen in our study. Current recommendations, stating that the anti-HCV assay should be used to screen for HCV infection in populations undergoing routine medical checkups, appear largely valid.
Our results show that the HCV core antigen assay is an efficient alternative to NAT to distinguish between past and ongoing HCV infections. Among anti-HCV antibody-positive subjects, we estimated that the sensitivity of the HCV core antigen assay to detect HCV viremia compared to RT-PCR was 96.7% (117 of 121; 95% confidence interval, 92 to 99%) (Table 2). When the HCV core antigen assay and HCV RNA were both positive, a good correlation was observed between HCV core Ag level (logs) and HCV RNA load (logs) (correlation coefficient = 0.78; slope = 0.88). Similar results have been reported previously (2, 20). The four discrepant results concerned subjects with low HCV RNA loads (range, 1,900 to 15,600 IU/ml). These results are in agreement with those of other studies, showing that the HCV core antigen assay can lack sensitivity when the viral load is close to 10,000 IU/ml (2, 11). This lack of sensitivity was not related to a particular HCV genotype (Table 3). In addition, it should be pointed out that a previous study has demonstrated that the HCV core Ag assay exhibits a similar efficiency for quantifying the different HCV genotypes (20). In our study, viral loads below 20,000 UI/ml were found in only 6 of 121 HCV RNA-positive samples. This indicates that a possible lack of sensitivity of the HCV core antigen assay concerns only a minor proportion of the infected patients (5%). Among anti-HCV antibody-positive subjects, the specificity of the HCV core antigen assay was also good because all HCV core Ag-positive subjects were confirmed to be viremic by RT-PCR.
We further explored the specificity of the HCV core Ag assay in 2,395 anti-HCV antibody-negative subjects. This revealed 17 samples that were weakly positive for the HCV core antigen (<8.5 pg/ml) and PCR negative. Neutralization analysis confirmed that all 17 samples were HCV core antigen-negative, leading to a specificity of 100%. Similar results were reported previously (17, 20).
HCV core Ag assay positivity was associated with elevated ALT concentration in most cases, but there was no correlation between ALT elevation and antigen levels. In the four samples that gave discrepant results with the HCV core antigen assay and RT-PCR, ALT levels were above normal, and RIBA reactivity levels were strongly positive, suggesting active HCV infections. Thus, when HCV core negativity is associated with an elevated ALT concentration and/or strongly positive RIBA results, NAT must be performed. Conversely, when HCV core negativity is associated with normal ALT concentrations and/or a weakly positive RIBA pattern, past HCV infection is the most likely diagnosis.
A laboratory algorithm is proposed for HCV screening (Table 4). As previously published (1), the testing for the presence of anti-HCV antibody is still recommended for the initial identification of HCV infection in the general population. In anti-HCV EIA-positive subjects, the HCV core Ag assay and other serological tools, such as RIBA and ALT level testing (19), will contribute to establishing the diagnosis of an ongoing HCV infection.
When the HCV core Ag assay is positive, current HCV infection should be suspected. The RIBA positivity strengthens the diagnosis, and its pattern may help to distinguish between acute and chronic infection (15).
If the HCV core Ag assay is negative, the RIBA negativity indicates a false-positive anti-HCV EIA screening result. A weakly positive RIBA assay asserts the resolution of HCV infection. A strongly positive RIBA assay should be completed by NAT to rule out the possibility that the HCV core Ag assay gave a false-negative result. Following this algorithm in our present study, NAT would have been necessary for only 2 patients (4%) out of the 46 who were not aware of their HCV seropositivity before entering the study. When the PCR is positive, this indicates a low viral load and predicts a good prognosis for a sustained response to the treatment.
Conclusion. The HCV core Ag assay, which is easier to perform than RT-PCR and less expensive, gives reliable information about viral load and might help to ensure an accurate medical follow-up. The HCV core Ag assay is an appropriate serological tool to indicate HCV status. It can efficiently distinguish recovery from current infection.
ACKNOWLEDGMENTS
This study was supported by Ortho Clinical Diagnostic.
REFERENCES
Alter, M. J., W. E. Kuhnert, and L. Finelli. 2003. Guidelines for laboratory testing and result reporting of antibody to hepatitis C virus. Morb. Mortal. Wkly. Rep. 52:1-12.
Bouvier-Alias, M., K. Patel, H. Dahari, S. Beaucourt, P. Larderie, L. Blatt, C. Hezode, G. Picchio, D. Dhumeaux, A. U. Neumann, J. G. McHutchison, and J. M. Pawlotsky. 2002. Clinical utility of total HCV core antigen quantification: a new indirect marker of HCV replication. Hepatology 36:211-218.
Cano, H., M. J. Candela, M. L. Lozano, and V. Vicente. 2003. Application of a new enzyme-linked immunosorbent assay for detection of total hepatitis C virus antigen in blood donors. Transfus. Med. 13:259-266.
Courouce, A. M., N. Le Marrec, F. Bouchardeau, A. Razer, M. Maniez, S. Laperche, and N. Simon. 2000. Efficacy of HCV core antigen detection during the preseroconversion period. Transfusion 40:1198-1202.
Desenclos, J. C., F. Dubois, N. Mariotte, and A. Goudeau. 1997. Analyse des strategies de depistage oriente de l'infection par le virus de l'hepatite C. Gastroenterol. Clin. Biol. 20:S25-S32.
Dubois F., J. C. Desenclos, N. Mariotte, A. Goudeau, and the Collaborative Study Group. 1997. Hepatitis C in a French population-based survey, 1994: seroprevalence, frequency of viremia, genotype distribution, and risk factor. Hepatology 25:1490-1496.
Dubois, F., M. Franois, N. Mariotte, E. Caces, S. Vol, P. Roigeard, F. Barin, A. Goudeau, and J. Tichet. 1994. Serum alanine aminotransferase measurement as a guide to selective testing for hepatitis C during medical checkup. J. Hepatol. 21:837-841.
Kolk, D. P., J. Dockter, J. Linnen, M. Ho-Sing-Loy, K. Gillotte-Taylor, S. H. McDonough, L. Mimms, and C. Giachetti. 2002. Significant closure of the human immunodeficiency virus type 1 and hepatitis C virus preseroconversion detection window with a transcription-mediated-amplification-driven assay. J. Clin. Microbiol. 40:1761-1766.
Laperche, S., N. Le Marrec, N. Simon, F. Bouchardeau, C. Defer, M. M. Maniez-Montreuil, T. Levayer, J. P. Zappitelli, and J. J. Lefrere. 2003. A new HCV core antigen assay based on dissociation of immune complexes: an alternative to molecular biology in the diagnosis of early HCV infection. Transfusion 43:958-962.
Lauer, G. M., and B. D. Walker. 2001. Hepatitis C virus infection. N. Engl. J. Med. 345:41-51.
Lunel, F., P. Veillon, I. Fouchard-Hubert, V. Loustaud-Ratti, A. Abergel, C. Sylvain, H. Rifflet, A. Blanchi, X. Causse, Y. Bacq, C. Payan, and the Fontevraud Study Group. 2003. Antiviral effect of ribavirin in early non-responders to interferon monotherapy assessed by kinetics of hepatitis C virus RNA and hepatitis C virus core antigen. J. Hepatol. 39:826-833.
Maynard, M., P. Pradat, P. Berthillon, G. Picchio, N. Voirin, M. Martinot, P. Marcellin, and C. Trepo. 2003. Antiviral effect of ribavirin in early non-responders to interferon monotherapy assessed by kinetics of hepatitis C virus RNA and hepatitis C virus core antigen. J. Hepatol. 10:318-323.
Peterson, J., G. Green, K. Iida, B. Caldwell, P. Kerrison, S. Bernich, K. Aoyagi, and S. R. Lee. 2000. Detection of hepatitis C core antigen in the antibody negative "window" phase of hepatitis C infection. Vox Sang. 78:80-85.
Pillonel, J., S. Laperche, and Groupe "Agents Transmissibles par Transfusion" de la Societe Franaise de Transfusion Sanguine, l'Etablissement franais du sang, Centre de transfusion sanguine des armees. 2004. Trends in residual risk of transmission (HIV, HCV, HBV) in France between 1992 and 2002 and impact of nucleid acid testing. Transfus. Clin. Biol. 11:81-86.
Proust, B., F. Dubois, Y. Bacq, S. Le Pogam, S. Rogez, R. Levillain, and A. Goudeau. 2000. Two successive hepatitis C virus infections in an intravenous drug user. J. Clin. Microbiol. 38:3125-3127.
Richter, S. S. 2002. Laboratory assays for diagnosis and management of hepatitis C virus infection. J. Clin. Microbiol. 40:4407-4412.
Schuttler, C. G., C. Thomas, T. Discher, G. Friese, J. Lohmeyer, R. Schuster, S. Schaefer, and W. H. Gerlich. 2004. Variable ratio of hepatitis C virus RNA to viral core antigen in patient sera. J. Clin. Microbiol. 42:1977-1981.
Reference deleted.
Valcavi, P., M. C. Medici, F. Casula, M. C. Arcangeletti, F. De Conte, F. Pinardi, A. Calderaro, C. Chezzi, and G. Dettori. 2004. Evaluation of a total hepatitis C virus (HCV) core antigen assay for detection of antigenaemia in anti-HCV positive individuals. J. Med. Virol. 73:397-403.
Veillon, P., C. Payan, G. Picchio, M. Maniez-Montreuil, P. Guntz, and F. Lunel. 2003. Comparative evaluation of the total hepatitis C virus core antigen, branched-DNA, and Amplicor monitor assays in determining viremia for patients with chronic hepatitis C during interferon plus ribavirin combination therapy. J. Clin. Microbiol. 41:3212-3220.
Zeuzem, S., E. Hermann, J. H. Fricke, A. U. Neumann, M. Modi, G. Colucci, and W. K. Roth. 2001. Viral kinetics in patients with chronic hepatitis C treated with standard or peginterferon alpha2a. Gastroenterology 120:1438-1447.(Catherine Gaudy, Catherin)
Institut inter Regional pour la Sante, La Riche, France
ABSTRACT
We studied the usefulness of the recently designed Trak-C assay for the detection and quantification of the hepatitis C virus (HCV) core antigen (Ag) for the screening of HCV infection in 4,201 subjects selected from 74,150 consecutive volunteers undergoing routine medical checkups. Subjects were selected for screening because they had risk factors (group II, n = 321) and/or elevated alanine transaminase activity (group I, n = 3847). Initially, the anti-HCV antibody assay and the Trak-C assay were performed on each patient. Subsequently, the Trak-C assay was performed only when the anti-HCV enzyme immune assay (EIA) was positive. Positive samples were further evaluated for anti-HCV antibodies by a third-generation strip immunoblot assay and for HCV RNA. Four samples (1.2%) from group II and 113 (2.9%) from group I were anti-HCV EIA positive. We also tested 33 subjects who previously tested positive for anti-HCV in our medical center. Among the 150 anti-HCV EIA-positive samples, the HCV core Ag result was in accord with the HCV RNA result in 146 cases (97.3%). When the EIA result was positive, the HCV core Ag concentration and the HCV RNA load were correlated (r2 = 0.78; P < 0.001). Four samples with low viral loads were Trak-C negative but HCV RNA positive. Among the 2,395 anti-HCV EIA-negative serum samples collected during the first part of the study, 17 (0.7%) were found to contain very low levels of HCV core Ag (<8.5 pg/ml, the cutoff value being 1.5 pg/ml). All these samples were HCV RNA negative and considered to be false positives. This was confirmed by HCV core Ag neutralization analysis. The HCV core Ag assay is a useful method in the screening strategy of HCV infection and provides a reliable means of distinguishing between current and cleared HCV infections that is well correlated with HCV RNA testing.
INTRODUCTION
The method currently recommended for identifying subjects with hepatitis C virus (HCV) infection is an enzyme immune assay (EIA) for the detection of anti-HCV antibodies (16). However, this assay sometimes generates false-positive or false-negative results. Furthermore, it is not always possible to distinguish between current and past cleared infections. Thus, supplementary tests are required. The strip immunoblot assay, a more specific anti-HCV serological test, is useful to distinguish true-positive from false-positive EIA results. Positive results from nucleic acid testing (NAT) for HCV RNA indicate active HCV infection (1). Assays that detect the HCV core antigen (Ag) have been developed to diagnose current HCV infection. The first such assay developed was a qualitative assay, conceived for screening blood donations. This assay increased viral safety by significantly reducing the length of the window preceding seroconversion (4, 13). A second test was subsequently developed for the detection and quantification of total HCV core antigen (Trak-C; Ortho-Clinical Diagnostics, Raritan, N.J.). This assay, incorporating an immune complex dissociation step, was designed for blood donation screening (3) and also for treatment monitoring (11, 12).
We investigated the usefulness of this new HCV core antigen assay for the screening of HCV infection in subjects undergoing routine medical checkups provided by the French national health insurance system. We looked at whether this assay is a useful initial test for the diagnosis of HCV infection in this kind of population and whether it can efficiently discriminate between previous and current infections. Finally, we asked whether it gives reliable information for an accurate medical follow-up.
MATERIALS AND METHODS
Population and study design. We investigated a series of 74,150 consecutive subjects who underwent routine medical checkups provided by social security medical centers between December 2001 and December 2002. These patients all lived in 10 administrative areas (called departments) in the western part of France. These medical centers are part of the French national health insurance system. Every 5 years, they offer biomedical examination to individuals who spontaneously attend the health center or who are directly invited to attend the medical center. The medical checkup includes a series of biological tests, followed by a clinical examination. The eligible population includes working people and their families as well as individuals in a "precarious" social situation, such as those with no paid employment and recipients of the welfare fund. Precarious populations can have a checkup every year. During the study period, 16,921 (23%) of the 74,150 individuals who underwent routine medical checkups were in precarious situations. The study population is described in Table 1.
Subjects were selected for HCV screening on the basis of two criteria. Group I (n = 3,847) included individuals showing a significant elevation of serum alanine transaminase (ALT), defined as ALT activity 20% above a norm (N) adjusted for age and sex. Results are expressed as ALT value/N. This screening strategy has been described previously (5, 7). Many of the subjects from group I who were found to be positive for anti-HCV antibodies had no record of HCV seropositivity in our center. We contacted their general practitioners to determine whether they were aware of previous positive tests. As recommended by the French Ministry of Health, a second blood sample was taken from subjects who were considered to have newly discovered HCV infection, and these subjects were retested with an anti-HCV assay differing from that used for the initial screening (see "Laboratory methods").
Group II (n = 321) included individuals with HCV risk factors, such as blood transfusion before 1991, past or current intravenous (i.v.) drug abuse, contact with HCV-infected relatives, or multiple operations. This group only included individuals with normal ALT values and no previous positive anti-HCV tests. During the same period, we also identified a group of 33 subjects with normal ALT values who had previously tested positive for anti-HCV antibodies in our medical centers and who returned for medical checkup 2 to 8 years after their initial screening (group III).
A questionnaire was sent to the general practitioner of all positive subjects to explore HCV risk factors and to record the medical follow-up of subjects who were known to be HCV positive. The questionnaire was approved by the Commission Nationale de l’Informatique et des Libertes (no. 26674, version 4).
Laboratory methods. Screening for anti-HCV antibodies was performed with a third-generation EIA (Ortho HCV 3.0 enzyme-linked immunosorbent assay). All EIA-positive samples were further analyzed using a third-generation strip immunoblot assay (Chiron RIBA HCV 3.0). According to the manufacturer's recommendations, the RIBA was scored as positive if an antibody reaction was observed with two or more viral proteins.
The HCV core antigen assay (Trak-C) is an enzyme immunoassay that allows the detection and quantification of total HCV core Ag despite the presence of anti-HCV antibodies. The detection limit of the method, as established by the manufacturer, is 1.5 pg/ml. Tests were performed in duplicate and were scored as positive if a reaction was observed in both cases. The HCV core antigen assay was carried out on every sample during the first part of the study (7 months) and only for anti-HCV antibody-positive samples during the second part of the study (6 months). In total, 41,768 subjects underwent the medical checkup during the first period, and 2,459 (group I, 2,340 subjects; group II, 119 subjects) were selected for screening; 32,382 individuals were seen during the second period, and 1,709 (group I, 1,507 subjects; group II, 202 subjects) were screened.
When the anti-HCV assay, the HCV core Ag assay, or both were positive, samples were further investigated for HCV RNA with a quantitative assay (bDNA HCV RNA 3.0 assay; Bayer Corporation, Tarrytown, N.Y.). A more sensitive qualitative assay (Amplicor HCV detection kit 2.0; Roche Diagnostics Corporation, Branchburg, N.J.) was used when the quantitative assay and/or the HCV core Ag assay was negative. Complementary assays were performed in case of discrepancies between HCV core Ag and HCV RNA assays: HCV core Ag-positive and HCV RNA-negative samples were tested with a neutralization assay (Ortho Trak-C confirmatory assay; Ortho-Clinical Diagnostics); HCV core Ag-negative and HCV RNA-positive samples were genotyped (VERSANT HCV genotype assay [LiPA]; Bayer Corporation Diagnostics Division).
RESULTS
Anti-HCV antibody screening. (i) Group I (elevated ALT value). Of the 3,847 subjects in group I, 113 (2.9%) were found to be anti-HCV EIA positive. Precarious populations, who accounted for 23% of the studied population, accounted for 50% of the anti-HCV-positive results (Table 1).
Sixty-seven of these 113 subjects (59%) had a previous positive anti-HCV test. RIBA confirmed that all were positive. Forty-six individuals were not aware of their HCV seropositivity before entering the study. Forty-one (89%) of these cases were confirmed positive by RIBA. Of the five others (11%), two reacted with only one viral antigen (core and NS5, respectively), two reacted with only human superoxide dismutase antigen, and one was completely nonreactive. Identical results were obtained for the initial and control samples for these five cases; thus, they were considered to be false positives. Thus, of the 108 anti-HCV-positive subjects with elevated ALT, 41 (38%) were not aware of their positivity before entering the study.
(ii) Group II (normal ALT value, risk factors, no previous positive tests). In group II, 4 (1.2%) of 321 subjects were anti-HCV EIA positive. All belonged to the precarious population. None were aware of their HCV seropositivity before their medical checkup, and all were confirmed positive by RIBA.
(iii) Group III (normal ALT value, known positive tests). All 33 subjects in group III were confirmed to be anti-HCV positive by RIBA.
HCV core Ag and HCV RNA in anti-HCV antibody EIA-positive subjects. The results of the HCV core Ag and/or HCV RNA tests are shown in Table 1. The relationships between the results of the HCV core Ag and HCV RNA tests among the 150 anti-HCV EIA-positive subjects (113 from group I, 4 from group II, and 33 from group III) are shown in Table 2. Of these 150 samples, 121 were HCV RNA positive, and 117 were HCV core Ag positive (Table 2). Most (146 of 150; 97.3%) yielded concordant results with the two methods (positive, 117; negative, 29). We found that 115 (95%) of the 121 HCV RNA-positive samples had viral loads above 20,000 IU/ml; all of these samples were also HCV core Ag positive. The five samples that were anti-HCV EIA positive but negative or indeterminate with RIBA (false-positive anti-HCV tests) were negative according to the HCV core Ag assay and HCV RNA assay. The sensitivity of the HCV core Ag assay was 96.7% (117 of 121 results were positive). HCV core Ag levels (logs) and HCV RNA loads (logs) were strongly correlated, with a correlation coefficient of 0.78 and a slope of 0.88. From the relationship between HCV core Ag values and HCV-RNA load, log (HCV RNA) = 0.841 x log (HCV Ag) + 1.097, the detection limit of the HCV RNA assay could be estimated to be 5,000 IU/ml. HCV core Ag levels were not correlated with ALT values (r2 < 0.03; P = 0.08).
Four subjects were HCV RNA positive but HCV core Ag negative. These four subjects (false HCV core Ag negative) had low HCV RNA viral loads (ranging from 1,900 to 15,600 IU/ml). They were all aware of their HCV seropositivity before the medical checkup (three from group I and one from group III). Details of their HCV status and medical information are shown in Table 3.
HCV core Ag and HCV RNA in anti-HCV antibody EIA-negative subjects. Among the 2,395 anti-HCV EIA-negative samples that were tested for HCV core Ag, 17 (0.71%; 15 in group I and 2 in group II) were repeatedly found to be positive. All these serum samples exhibited low levels of HCV core Ag (1.5 to 8.5 pg/ml). They were all negative by a neutralization assay and were all negative for HCV RNA by reverse transcription (RT)-PCR. Thus, the specificity of the HCV core Ag assay was 99.3% and reached 100% after neutralization analysis. Besides, the examination of a consecutive serum sample revealed no case of late seroconversion, indicating an absence of HCV infection in these patients.
Relationship between HCV status and other medical factors. Among the 100 subjects who were aware of their anti-HCV positivity before their medical checkup, 36 had been treated. HCV viremia was significantly more frequent when subjects had not been treated (91 versus 61%; P > 103) and when the ALT level was elevated (95 versus 48%; P > 106). Among the 20 subjects negative for viremia, 14 had been treated, and HCV infection had spontaneously cleared in 6 cases.
DISCUSSION
Several different serological tools are available to identify HCV infection. Positivity for anti-HCV antibodies indicates past or current HCV infection but gives no indication about viral replication. In contrast, molecular detection of HCV RNA indicates active viral replication and calls for thorough clinical evaluation (10, 16). Qualitative RNA assays are currently used to diagnose ongoing HCV infection among anti-HCV antibody-positive individuals. They are also used to detect viremia in subjects during the early phase of infection prior to anti-HCV seroconversion (window period), for instance, in blood banks (8). Quantitative HCV RNA assays are mainly used to monitor anti-HCV treatment (21). Trak-C, a new serological assay that can be used to detect and to quantify circulating HCV core antigens, was initially designed to monitor the efficacy of HCV treatment as an alternative to measurement of HCV RNA viral load (2, 12, 20). This test has also been shown to be appropriate for screening for HCV infection in blood donors (3). The ability of this assay to identify viremic subjects in the general population has never been evaluated prospectively.
NATs are now routinely used by French blood banks to detect HCV viremia in donors during the window period. Retrospective studies carried out with the Trak-C assay showed that the HCV core antigen can detect early HCV viremia during this window period (9) and might represent an alternative to NAT. However, further studies are necessary to evaluate the HCV core antigen assay prospectively before it can be used routinely in blood banks.
In France, the prevalence of blood donors who are susceptible to being tested during such a window period is very low (1 in 106 donations; 95% confidence interval, 1 in 3.1 x 106 to 1 in 0.4 x 106) (14). In the present study, the population included groups at higher risk of HCV infection, such as i.v. drug abusers (6), thus increasing the probability of recruiting individuals in this window period. However, none of the 2,395 anti-HCV-negative subjects was positive for HCV core antigen in our study. Current recommendations, stating that the anti-HCV assay should be used to screen for HCV infection in populations undergoing routine medical checkups, appear largely valid.
Our results show that the HCV core antigen assay is an efficient alternative to NAT to distinguish between past and ongoing HCV infections. Among anti-HCV antibody-positive subjects, we estimated that the sensitivity of the HCV core antigen assay to detect HCV viremia compared to RT-PCR was 96.7% (117 of 121; 95% confidence interval, 92 to 99%) (Table 2). When the HCV core antigen assay and HCV RNA were both positive, a good correlation was observed between HCV core Ag level (logs) and HCV RNA load (logs) (correlation coefficient = 0.78; slope = 0.88). Similar results have been reported previously (2, 20). The four discrepant results concerned subjects with low HCV RNA loads (range, 1,900 to 15,600 IU/ml). These results are in agreement with those of other studies, showing that the HCV core antigen assay can lack sensitivity when the viral load is close to 10,000 IU/ml (2, 11). This lack of sensitivity was not related to a particular HCV genotype (Table 3). In addition, it should be pointed out that a previous study has demonstrated that the HCV core Ag assay exhibits a similar efficiency for quantifying the different HCV genotypes (20). In our study, viral loads below 20,000 UI/ml were found in only 6 of 121 HCV RNA-positive samples. This indicates that a possible lack of sensitivity of the HCV core antigen assay concerns only a minor proportion of the infected patients (5%). Among anti-HCV antibody-positive subjects, the specificity of the HCV core antigen assay was also good because all HCV core Ag-positive subjects were confirmed to be viremic by RT-PCR.
We further explored the specificity of the HCV core Ag assay in 2,395 anti-HCV antibody-negative subjects. This revealed 17 samples that were weakly positive for the HCV core antigen (<8.5 pg/ml) and PCR negative. Neutralization analysis confirmed that all 17 samples were HCV core antigen-negative, leading to a specificity of 100%. Similar results were reported previously (17, 20).
HCV core Ag assay positivity was associated with elevated ALT concentration in most cases, but there was no correlation between ALT elevation and antigen levels. In the four samples that gave discrepant results with the HCV core antigen assay and RT-PCR, ALT levels were above normal, and RIBA reactivity levels were strongly positive, suggesting active HCV infections. Thus, when HCV core negativity is associated with an elevated ALT concentration and/or strongly positive RIBA results, NAT must be performed. Conversely, when HCV core negativity is associated with normal ALT concentrations and/or a weakly positive RIBA pattern, past HCV infection is the most likely diagnosis.
A laboratory algorithm is proposed for HCV screening (Table 4). As previously published (1), the testing for the presence of anti-HCV antibody is still recommended for the initial identification of HCV infection in the general population. In anti-HCV EIA-positive subjects, the HCV core Ag assay and other serological tools, such as RIBA and ALT level testing (19), will contribute to establishing the diagnosis of an ongoing HCV infection.
When the HCV core Ag assay is positive, current HCV infection should be suspected. The RIBA positivity strengthens the diagnosis, and its pattern may help to distinguish between acute and chronic infection (15).
If the HCV core Ag assay is negative, the RIBA negativity indicates a false-positive anti-HCV EIA screening result. A weakly positive RIBA assay asserts the resolution of HCV infection. A strongly positive RIBA assay should be completed by NAT to rule out the possibility that the HCV core Ag assay gave a false-negative result. Following this algorithm in our present study, NAT would have been necessary for only 2 patients (4%) out of the 46 who were not aware of their HCV seropositivity before entering the study. When the PCR is positive, this indicates a low viral load and predicts a good prognosis for a sustained response to the treatment.
Conclusion. The HCV core Ag assay, which is easier to perform than RT-PCR and less expensive, gives reliable information about viral load and might help to ensure an accurate medical follow-up. The HCV core Ag assay is an appropriate serological tool to indicate HCV status. It can efficiently distinguish recovery from current infection.
ACKNOWLEDGMENTS
This study was supported by Ortho Clinical Diagnostic.
REFERENCES
Alter, M. J., W. E. Kuhnert, and L. Finelli. 2003. Guidelines for laboratory testing and result reporting of antibody to hepatitis C virus. Morb. Mortal. Wkly. Rep. 52:1-12.
Bouvier-Alias, M., K. Patel, H. Dahari, S. Beaucourt, P. Larderie, L. Blatt, C. Hezode, G. Picchio, D. Dhumeaux, A. U. Neumann, J. G. McHutchison, and J. M. Pawlotsky. 2002. Clinical utility of total HCV core antigen quantification: a new indirect marker of HCV replication. Hepatology 36:211-218.
Cano, H., M. J. Candela, M. L. Lozano, and V. Vicente. 2003. Application of a new enzyme-linked immunosorbent assay for detection of total hepatitis C virus antigen in blood donors. Transfus. Med. 13:259-266.
Courouce, A. M., N. Le Marrec, F. Bouchardeau, A. Razer, M. Maniez, S. Laperche, and N. Simon. 2000. Efficacy of HCV core antigen detection during the preseroconversion period. Transfusion 40:1198-1202.
Desenclos, J. C., F. Dubois, N. Mariotte, and A. Goudeau. 1997. Analyse des strategies de depistage oriente de l'infection par le virus de l'hepatite C. Gastroenterol. Clin. Biol. 20:S25-S32.
Dubois F., J. C. Desenclos, N. Mariotte, A. Goudeau, and the Collaborative Study Group. 1997. Hepatitis C in a French population-based survey, 1994: seroprevalence, frequency of viremia, genotype distribution, and risk factor. Hepatology 25:1490-1496.
Dubois, F., M. Franois, N. Mariotte, E. Caces, S. Vol, P. Roigeard, F. Barin, A. Goudeau, and J. Tichet. 1994. Serum alanine aminotransferase measurement as a guide to selective testing for hepatitis C during medical checkup. J. Hepatol. 21:837-841.
Kolk, D. P., J. Dockter, J. Linnen, M. Ho-Sing-Loy, K. Gillotte-Taylor, S. H. McDonough, L. Mimms, and C. Giachetti. 2002. Significant closure of the human immunodeficiency virus type 1 and hepatitis C virus preseroconversion detection window with a transcription-mediated-amplification-driven assay. J. Clin. Microbiol. 40:1761-1766.
Laperche, S., N. Le Marrec, N. Simon, F. Bouchardeau, C. Defer, M. M. Maniez-Montreuil, T. Levayer, J. P. Zappitelli, and J. J. Lefrere. 2003. A new HCV core antigen assay based on dissociation of immune complexes: an alternative to molecular biology in the diagnosis of early HCV infection. Transfusion 43:958-962.
Lauer, G. M., and B. D. Walker. 2001. Hepatitis C virus infection. N. Engl. J. Med. 345:41-51.
Lunel, F., P. Veillon, I. Fouchard-Hubert, V. Loustaud-Ratti, A. Abergel, C. Sylvain, H. Rifflet, A. Blanchi, X. Causse, Y. Bacq, C. Payan, and the Fontevraud Study Group. 2003. Antiviral effect of ribavirin in early non-responders to interferon monotherapy assessed by kinetics of hepatitis C virus RNA and hepatitis C virus core antigen. J. Hepatol. 39:826-833.
Maynard, M., P. Pradat, P. Berthillon, G. Picchio, N. Voirin, M. Martinot, P. Marcellin, and C. Trepo. 2003. Antiviral effect of ribavirin in early non-responders to interferon monotherapy assessed by kinetics of hepatitis C virus RNA and hepatitis C virus core antigen. J. Hepatol. 10:318-323.
Peterson, J., G. Green, K. Iida, B. Caldwell, P. Kerrison, S. Bernich, K. Aoyagi, and S. R. Lee. 2000. Detection of hepatitis C core antigen in the antibody negative "window" phase of hepatitis C infection. Vox Sang. 78:80-85.
Pillonel, J., S. Laperche, and Groupe "Agents Transmissibles par Transfusion" de la Societe Franaise de Transfusion Sanguine, l'Etablissement franais du sang, Centre de transfusion sanguine des armees. 2004. Trends in residual risk of transmission (HIV, HCV, HBV) in France between 1992 and 2002 and impact of nucleid acid testing. Transfus. Clin. Biol. 11:81-86.
Proust, B., F. Dubois, Y. Bacq, S. Le Pogam, S. Rogez, R. Levillain, and A. Goudeau. 2000. Two successive hepatitis C virus infections in an intravenous drug user. J. Clin. Microbiol. 38:3125-3127.
Richter, S. S. 2002. Laboratory assays for diagnosis and management of hepatitis C virus infection. J. Clin. Microbiol. 40:4407-4412.
Schuttler, C. G., C. Thomas, T. Discher, G. Friese, J. Lohmeyer, R. Schuster, S. Schaefer, and W. H. Gerlich. 2004. Variable ratio of hepatitis C virus RNA to viral core antigen in patient sera. J. Clin. Microbiol. 42:1977-1981.
Reference deleted.
Valcavi, P., M. C. Medici, F. Casula, M. C. Arcangeletti, F. De Conte, F. Pinardi, A. Calderaro, C. Chezzi, and G. Dettori. 2004. Evaluation of a total hepatitis C virus (HCV) core antigen assay for detection of antigenaemia in anti-HCV positive individuals. J. Med. Virol. 73:397-403.
Veillon, P., C. Payan, G. Picchio, M. Maniez-Montreuil, P. Guntz, and F. Lunel. 2003. Comparative evaluation of the total hepatitis C virus core antigen, branched-DNA, and Amplicor monitor assays in determining viremia for patients with chronic hepatitis C during interferon plus ribavirin combination therapy. J. Clin. Microbiol. 41:3212-3220.
Zeuzem, S., E. Hermann, J. H. Fricke, A. U. Neumann, M. Modi, G. Colucci, and W. K. Roth. 2001. Viral kinetics in patients with chronic hepatitis C treated with standard or peginterferon alpha2a. Gastroenterology 120:1438-1447.(Catherine Gaudy, Catherin)