Journal of Clinical Virology 30S1 (2004) S1–S5

Review

The ADVIA Centaur® infectious disease assays: a technical review

Robert Dwyer∗

Bayer HealthCare LLC, Diagnostics Division, 511 Benedict Avenue, Tarrytown, NY 10591-5097, USA

Abstract

Bayer Healthcare, Diagnostics Division, has developed several new several new assays for hepatitis B, hepatitis C and HIV 1/O/2 on theADVIA Centaur® Immunoassay system. The panel for detection of markers for hepatitis B infection includes assays for hepatitis B surfaceantigen (HBsAg) and HBsAg. Confirmatory, antibodies to hepatitis B surface antigen (anti-HBs), IgM and IgG antibodies to hepatitis Bcore antigen (anti-HBc Total) and IgM antibodies to hepatitis B core antigen (anti-HBc IgM). The hepatitis B panel is complemented bythird generation assays for anti–HCV and anti-HIV that concomitantly detects HIV-1 (including group O) and HIV-2. The features of theADVIA Centaur and the design of the assays combine to deliver state-of-the-art performance. Features, such as disposable sample pipettetips to prevent sample carryover, clot detection and management to verify sample addition, reagent integrity checks to assure accurate reagentaddition, auto repeat capabilities and the capacity to maintain up to 30 assays onboard at 2–8◦C all contribute to the systems’s utility forinfectious disease testing. Furthermore, the flexibility of the ADVIA Centaur has allowed for selection of formats to provide optimal assayperformance. The anti-HBs, anti-HBc Total and anti-HIV 1/O/2 assays are antigen-bridging assays. The anti-HBc IgM assay is a class-capturetwo step assay configured for detection of IgM anti-HBc antibodies during acute HBV infection. The anti-HCV assay is an indirect assayutilizing streptavidin-coated microparticles preformed with a combination of recombinant and synthetic peptide antigens from the NS2, NS4,NS5 and core regions of the HCV genome. The HBsAg assay is a sandwich assay with an incubation time of 28 minutes and the HBsAgConfirmatory assay is a fully automated neutralization assay. The availability of these assays on the ADVIA Centaur Immunoassay systemenables laboratories to consolidate other immunoassays along with the infectious disease assays for workstation operation and efficiencies.© 2004 Elsevier B.V. All rights reserved.

Keywords:ADVIA Centaur; Chemiluminescence; Immunoanalyser; Hepatitis B; Hepatitis C; HIV

1. Introduction

The need for automation in diagnostic virology is rapidlyincreasing. Serological tests are needed not only for the rapiddetection of an infection but also because they greatly con-tribute to determination of the specific nature of the diseaseand, thus, to the initiation and monitoring of an appropriatetreatment.

2. Development and characteristics of the new assaysfor infectious disease

Bayer HealthCare, Diagnostics Division has developednew assays for hepatitis B (HBV), hepatitis C (HCV) andHIV on the ADVIA Centaur® immunoassay system, thusexpanding its menu of infectious disease assays. The HBV

Abbreviations:IVD, in vitro diagnostics; IVDD, in vitro diagnosticsdirective; EU, European Union; CTS, common technical specifications;CE, certification

∗ Tel.: +1-914-524-2708; fax:+1-914-524-2543.E-mail address:robert.dwyer.b@bayer.com (R. Dwyer).

assay panel includes methods for HBsAg and HBsAg Con-firmatory, anti-HBc Total, anti-HBc IgM and anti-HBs. Theinfectious disease panel is complemented by an assay foranti-HCV and an assay for the concomitant detection ofanti HIV-1 (including group O) and anti HIV-2. Additionalassays currently under development include methods foranti-HAV IgM, anti-HAV Total, HBeAg and anti-HBe.

2.1. The ADVIA Centaur system—a fully automatedplatform for infectious disease testing

The new infectious disease assays have been developedfor the fully automated ADVIA Centaur immunoassay sys-tem. This system has been designed specifically for usein large-volume laboratories with high sample throughputthat require a high degree of reliability and automation.A unique bi-directional incubation and processing ring al-lows the ADVIA Centaur to process antibody capture assaysin a true random access mode with competitive and sand-wich immunoassays. Features, such as disposable pipettetips, cascade reflex testing, clot detection and management,auto-repeat and auto-dilute capabilities, and the capacity to

1386-6532/$ – see front matter © 2004 Elsevier B.V. All rights reserved.doi:10.1016/j.jcv.2004.02.002

S2 R. Dwyer / Journal of Clinical Virology 30S1 (2004) S1–S5

Fig. 1. ADVIA Centaur HBsAg assay format.

maintain up to 30 assays onboard at 2–8◦C all contribute tothe system’s utility for infectious disease testing. Further-more, the ADVIA Centaur system enables laboratories toconsolidate other immunoassays along with the infectiousdisease assays for workstation operation and efficiencies.

2.2. Serological tests for HBV markers

The laboratory diagnosis and monitoring of both acuteand chronic HBV infections requires the use of several testsincluding hepatitis B surface antigen (HBsAg), the e anti-gen (HBeAg), antibody to the core antigen (anti-HBc), anti-body to the e antigen (anti-HBe), and antibody to the surfaceantigen (anti-HBs). Immunity conferred by HBV vaccina-tion can be assessed by measuring levels of anti-HBs (EASLJury, 2003). Panels of tests, as shown below, are particularlyhelpful in differentiating acute hepatitis, chronic hepatitisand recovered patients:

(1) Suspected acute viral hepatitis—HBsAg and anti-HBcIgM.

(2) Chronic hepatitis—HBsAg and anti-HBc Total.(3) Monitoring of chronic HBV infection—HBsAg,

HBeAg, anti-HBe and qualitative or quantitative HBVDNA.

The ADVIA Centaur system has been optimized for thedevelopment of assays for HBV. Its flexibility allows formany different assay formats and protocols. For example, theHBsAg assay is a sandwich assay with a pre-incubation ofsample and antibodies prior to addition of solid phase. TheAnti-HBs and anti-HBc total assays are antigen-bridgingassays, one with a single incubation and wash step and theother with two incubations and wash steps. The HBc IgMassay is a�-capture format with two wash steps and anindirect label.

2.2.1. ADVIA Centaur® HBsAg assay and ADVIACentaur® HBsAg confirmatory assay

The ADVIA Centaur HBsAg assay is a sandwich assayusing direct chemiluminometric technology (Fig. 1). It uti-lizes a biotinylated anti-HBs monoclonal antibody and anacridinium ester conjugated anti-HBs monoclonal antibodywhich are incubated with sample. Streptavidin coated para-magnetic microparticles are added to bind the complex.After washing, the microparticles are flashed by acid/baseaddition. The resulting signal is read against a stored stan-dard curve and the result expressed as an INDEX value(positive cutoff of 1.0; equivalent to 0.066 IU/ml—WHOFirst Reference Preparation 80/549-1).

The European performance evaluation (n = 5745 sam-ples) resulted in a resolved sensitivity of 100% (438/438)and a resolved specificity of 99.94% (5259/5262) when com-pared to the Abbott AxSYM HBsAg method. Within-runand total assay imprecision for serum and plasma (EDTA,Na Heparin and Li Heparin) samples ranged from 1.3% to11.4% and from 2.9% to 14.7%, respectively. Complete as-say characteristics are shown inTable 1.

The ADVIA Centaur HBsAg confirmatory assay is a fullyautomated neutralization assay (Table 1). The method usestwo additional reagents (one containing anti-HBs negativeprocessed plasma and one containing high titer anti-HBshuman processed plasma) that are used in combination withthe HBsAg reagents to measure neutralization of the HB-sAg complex. Results are reported as either ‘not-confirmed’or ‘confirmed’.

2.2.2. ADVIA Centaur® HBc total assayThe ADVIA Centaur HBc total assay is a two-wash

antigen sandwich immunoassay in which antigens (recom-binant HBcAg) are bridged by antibody (anti-HBc) presentin sample. The method uses a mild chaotropic buffer toreduce non-specific binding and a chemiluminometric de-

R. Dwyer / Journal of Clinical Virology 30S1 (2004) S1–S5 S3

Table 1ADVIA Centaur system assay characteristics

Assay Format Samplevolume (�l)

Time to first result(TTFR, min)

Throughputtests (h)

Number oftests (pack)

On-board stability (OBS)/calibrator interval (days)

HBsAg 20 min sandwich 100 29 240 200 41/21HBsAg Conf 20/20 Ab neutralization 2× 100 58 120 100 41/21Anti-HBs 7.5 min Ag bridging 100 18 240 200 41/28Anti-HCV 20/20 indirect 10 58 120 200 41/28Anti-HBc IgM 20/20 µ capture 15 58 120 100 41/28Anti-HBc total 20/20 Ag bridging 50 58 120 200 28/14HIV 1/O/2 20/20 Ag bridging 50 58 120 200 28/14

Table 2ADVIA anti-HBc total clinical sensitivity

ADVIA CentaurHBc Total assay

AxSYM anti-HBc total assay

Positive Negative Total

Positive 364 29 393Equivocal 5 0 5Negative 4 15 19Total 373 44 417

tection system. Results are reported as index values witha positive cutoff of 1.0. Complete assay characteristics areshown inTable 1.

The performance of the ADVIA Centaur HBc total assaywas determined by testing a total of 5579 samples at two Eu-ropean trial sites. The samples included the following pop-ulations: HBV positive samples, normal blood donors, andhospitalized patients. The European performance evaluationresulted in a resolved specificity of 99.94% (5135/5138)when compared to the Abbott AxSYM Core method. Clin-ical sensitivity was determined by testing a populationof 417 HBV patient samples previously determined to bepositive for HBsAg (Table 2). 373 of these HBV patientsamples were found to be positive for anti-HBc Total usingthe reference assay and 393 were positive using the ADVIACentaur HBc Total assay. 15 samples were found to be neg-ative by both assays. An additional 29 samples that werefound to be negative by AxSYM and positive by ADVIACentaur HBc Total were found to be positive upon retestand/or resolution testing. The initial sensitivity was 98.91%(364/368) for ADVIA Centaur HBc Total compared to92.8% (373/402) for the reference assay. The ADVIA Cen-taur HBc total method is precise with within-run and totalimprecision ranging from 3.6% to 5.8% and 6.5% to 8.4%in samples with index values of 0.06–4.5, respectively.

2.2.3. ADVIA Centaur® HBc IgM assayThe ADVIA Centaur HBc IgM assay utilizes a two-step,

two-wash protocol in which IgM in the patient specimenis captured by paramagnetic microparticles in the first step,followed by quantitation of HBc specific IgM by recombi-nant HBc antigen linked to acridinium ester in the secondstep. Results are reported as index values with a positivecutoff of 1.0.

The European performance evaluation (n = 800 samples)resulted in a resolved sensitivity of 98.5% (201/204) and a

resolved specificity of 100% (434/434) when compared tothe Abbott AxSYM Core M method. Within-run and totalassay imprecision for serum and plasma (EDTA, Na Heparinand Li Heparin) samples ranged from 3.0% to 7.6% andfrom 3.9% to 8.6%, respectively.

2.2.4. ADVIA Centaur® anti-HBs assayThe assay utilizes a one-step “reverse sandwich” protocol

using native HBsAg coupled to paramagnetic microparticlesas the solid phase, and native HBsAg linked to acridiniumester as the chemiluminescent tracer. The assay is standard-ized to the WHO First International Reference Preparationfor anti-HBs (1977) and offers linear measurement between1 and 1000 mIU/ml. Results are reported in mIU/ml witha cut-off equivalent to 10 mIU/ml. An equivocal zone of±2.5 mIU/ml is set around the cutoff.

The European performance evaluation (n = 703 sam-ples) resulted in a resolved sensitivity of 99.0% (200/202)and a resolved specificity of 100% (496/496) when com-pared to the Abbott AxSYM Anti-HBs method. The AD-VIA Centaur assay had equivalent performance to the otheranti-HBs assays when 10 seroconversion panels were eval-uated. In a precision study performed over 20 days, thewithin-run and total assay imprecision of all positive sam-ples ranged from 2.6% to 9.7% and from 3.7% to 10.4%,respectively.

2.3. Serological tests for HCV infection

Diagnostic tests for HCV (Richter, 2002) can be dividedinto two categories:

(a) Serological assays (EIA and recombinant immunoblotassay (RIBA)) that detect the antibody to HCV(anti-HCV).

(b) Molecular assays that detect, quantify and/or character-ize HCV RNA genomes in an infected patient.

(c) Genotyping techniques.

Serological assays can detect HCV antibodies to indicatepresent or previous infection, but these assays cannot dis-criminate acute from chronic or resolved infection.

Three generations of anti-HCV tests have been devel-oped with each generation resulting in an improvement inthe sensitivity of detecting anti-HCV (Richter, 2002). Thefirst-generation immunoassays used the c100-3 epitope of

S4 R. Dwyer / Journal of Clinical Virology 30S1 (2004) S1–S5

an NS protein (NS4). The sensitivities of these immunoas-says were low for a high-prevalence population (approxi-mately 80%), and the fraction of positive results that werefalse positive was as high as 70% for a low-prevalencepopulation (blood donors). This led to the development ofmore sensitive and specific second-generation immunoas-says that incorporated additional antigens from NS (c33c)and structural (c22-3) proteins. Third generation assays,which incorporate a fourth antigen, NS5, enable the detec-tion of HCV antibodies on an average of 26 days earlier inindividuals with transfusion-transmitted HCV. Furthermore,sensitivity is also improved (approaching 97%) specificallyin a high-prevalence population.

2.3.1. ADVIA Centaur® HCV assay1

The HCV assay utilizes streptavidin-coated microparti-cles (preformed with a combination of recombinant and syn-thetic peptide antigens from the NS3, NS4, NS5 and the coreregions of the HCV genome) and a monoclonal anti-humanIgG antibody conjugated to acridinium ester. Results are re-ported as index values with a cutoff of 1.0.

The European performance evaluation (n = 5677 sam-ples) resulted in a resolved sensitivity of 100% (449/449)and a resolved specificity of 99.9% (5217/5222) when com-pared to the Abbott AxSYM HCV 3.0 method.

Significant interference with the other viral infections anddisease state samples tested were not detected. Precisionestimates were made with data generated over 20 runs on10 days. Total imprecision was less than 10% over the assayrange with the exception of negative samples.

2.4. Serological tests for HIV infection

When risk assessment and/or clinical presentation indi-cate the possibility of acute HIV infection, laboratory test-ing ensures correct diagnosis. The testing alogrithm for HIVare usually first-line serological screening immunoassays forHIV antibody and/or antigen testing, followed by WesternBlots and/or NAT assays to confirm reactivity (Doerr, 1998).

First generation HIV assays detected only HIV-1. Theseassays had a poor specificity, a suboptimal sensitivity andproduced inconsistent results. The second generation assaysdetected both HIV-1 and HIV-2. These assays were morespecific, showed an improved sensitivity and gave moreconsistent results. Currently, third generation assays detectHIV-1 including group ‘O’ as well as HIV-2. These assaysusually incorporate an antigen-bridging format with recom-binant antigens and/or synthetic peptides for capture anddetection. They have an increased sensitivity, as they detectboth IgG and IgM HIV antibodies.

1 The ADVIA Centaur HCV and HIV assays are developed, manu-factured, and sold by Bayer HealthCare Corporation, LLC for OrthoClinical Diagnostics.

Table 3ADVIA Centaur HIV 1/O/2 assay—seroconversion sensitivity

Panel Days to first reactive result Days to first reactive result

ADVIACentaur

AxSYM Panel ADVIACentaur

AxSYM

221 14 14 AI 7 7241 15 15 AM 9 9251 22 X 18 ANE 103 103291 6 6 AP X 7 11331 20 20 AT 14 14341 21 21 AZ 28 2868106 21 21 AG X 27 34AC 111 111 AU 13 13AD X 18 25 BB 14 14AE 7 7 BE 14 X 12AF 28 28 BG X 23 28AH 21 21 RP-002 X 75 77AJ 43 43 P 30 30AQ 18 18 Z 27 27AS 19 19

2.4.1. ADVIA Centaur® HIV 1/O/2 assay1

The application for determining HIV using the ADVIACentaur HIV 1/O/2 Assay is for the qualitative detection ofantibodies to human immunodeficiency virus type 1 (includ-ing group O) and type 2 in serum or plasma.

The ADVIA Centaur HIV 1/O/2 assay uses recom-binant antigens corresponding to the viral envelope andcore proteins. Recombinant antigens include an HIV-1envelope protein, HIV-1 core protein p24, and an HIV-2envelope-oriented protein, and an HIV-1 group O peptide.The recombinant antigens and peptide are biotinylated foruse as the solid phase with Streptavidin microparticles andlabeled with acridinium ester as detection. Use of theserecombinants and peptides improves detection of HIV-1 an-tibody, including group O, and/or HIV-2 antibody positivespecimens and minimizes nonspecific reactions. The cutofffor positivity is an index value of 1.0.

The European performance evaluation was performed us-ing a total of 5842 samples (419 HIV-1 and 100 HIV-2 in-fected patients, 25 specimens with HIV subtypes A, B, C,D. E, F, G. and O and 25 non-B subtypes as well as 5069blood donors). The resolved sensitivity was 100% (554/554)and the resolved specificity was 99.9% (5281/5288).

The ADVIA Centaur assay was found to be equivalentand, in four cases have better performance, than the predi-cate device when 29 seroconversion panels were evaluated.In two other cases the predicate device detected antibodiesto HIV one bleed earlier (Table 3). Specimens with poten-tially interfering substances, including HAMA, rheumatoidfactor and other infections did not cause false positive re-sults. The within-run and total assay imprecision rangedfrom 2.1% to 7.4%.

3. Conclusion

New immunoassays for HBV, HCV, HIV-1 (includinggroup O) and HIV-2 have been developed for the fully au-

R. Dwyer / Journal of Clinical Virology 30S1 (2004) S1–S5 S5

tomated ADVIA Centaur system. This system and these as-says are specifically designed for use in the large-volumelaboratories with high sample throughput that require au-tomation and a high degree of reliability. An extensive per-formance evaluation was carried out at two evaluation sitesin Europe. The results of this evaluation indicate these as-says to be state-of-the-art for infectious disease testing.

References

Doerr HW, Viral infections. In: Thomas L, editor. Clinical laboratorydiagnostics. Frankfurt: Th-Books; 1998. p. 1218–59.

EAS Jury. Consensus statement. In: EASL International Consensus Con-ference on Hepatitis B, 13–14 September 2002; Geneva, Switzerland.J. Hepatol 2003;38:533–40.

Richter SS. Laboratory assays for diagnosis and management of HepatitisC virus infection. J Clin Microbiol 2002;40:4407–12.

Further reading

Anon, 2002. Commission decision on common technical specificationsfor in vitro-diagnostic medical devices. Off J Eur Commun L131:17–30.

Anon, 1998. Directive 98/79/EC of the European Parliament and of theCouncil of 27 October 1998 on in vitro diagnostic medical devices.Off J Eur Commun L331:1–37.