- Email: [email protected]
2 Confirmatory Assay and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay
Journal of Virological Methods 224 (2015) 91–94
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
Evaluation of two HIV antibody confirmatory assays: GeeniusTM HIV1/2 Confirmatory Assay and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay I. Friedrichs ∗ , C. Buus, A. Berger, O.T. Keppler, H.F. Rabenau Institute of Medical Virology, University Hospital Frankfurt/Main, Goethe University, National Reference Laboratory for Retroviruses, Paul-Ehrlich-Str. 40, 60596 Frankfurt/Main, Germany
a b s t r a c t Article history: Received 30 July 2014 Received in revised form 19 August 2015 Accepted 21 August 2015 Available online 24 August 2015 Keywords: Serological HIV diagnosis Confirmatory assay Line immunoassay Immunochromatographic assay
The laboratory diagnosis of an HIV infection mainly depends on the detection of HIV-specific antibodies/HIV p24 antigen whereby different algorithms for the confirmation of reactive screening assays exist. The objective of the present study was to compare the performance of two supplemental HIV antibody confirmatory assays: the GeeniusTM HIV1/2 Confirmatory Assay and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay. Therefore 279 serum samples previously analyzed for HIV during routine diagnostics at the Institute for Medical Virology, National Reference Center for Retroviruses, University Hospital Frankfurt, were analyzed retrospectively. 96.8% samples had concordant results in both HIV confirmatory assays, whereby the Geenius Assay showed a discrimination rate of 100% while two HIV-1 samples were not typeable with the recomLine Assay. Overall assay sensitivity was 100% in both assays and specificity was 99.0% (recomLine Assay) and 93.4% (Geenius Assay), respectively. The -values for both assays indicated high agreement. Overall nine samples had discordant results from which four were from acutely EBV/CMV-infected patients and one from a patient with primary HIV-1 infection during seroconversion. In conclusion, both assays are well suited for the detection, confirmation and discrimination of HIV-1and -2-specific antibodies. © 2015 Elsevier B.V. All rights reserved.
1. Introductions HIV infection remains a major public health concern around the globe. According to UNAIDS an estimated number of 35 million adults and children worldwide were infected with HIV in 2013, of which around 2.1 million were newly infected and 1.5 million deaths were reported (UNAIDS, 2014). The highest rate of HIV transmission occurs during acute HIV infection (Brenner et al., 2007) which is typically characterized by high viral loads and low antibody titres. A reliable laboratory diagnosis might contribute to decrease transmission rates at this stage and can reduce morbidity and mortality and improve quality of life through early initiation of antiretroviral therapy (Chou et al., 2012; Cohen et al., 2011; Severe et al., 2010). The laboratory diagnosis of an HIV infection mainly depends on the detection of HIV-specific antibodies and/or HIV p24
∗ Corresponding author. Tel.: +49 69 6301 83062; fax: +49 69 6301 83061. E-mail address: [email protected] (I. Friedrichs). http://dx.doi.org/10.1016/j.jviromet.2015.08.015 0166-0934/© 2015 Elsevier B.V. All rights reserved.
antigen. According to the European Guideline on HIV testing and the recommendations of the Centers for Disease Control and Prevention (CDC) and the Association of Public Health Laboratories (APHL) fourth generation screening assays for simultaneous detection of HIV-1/2 antibodies and HIV p24 antigen should be used. In different countries different algorithms exist for the confirmation of reactive screening assays. Supplemental antibody testing is recommended to confirm repeatedly reactive fourth generation assay results whereby such assays should distinguish between HIV-1 and HIV-2 antibodies. If the confirmatory assay is positive for either HIV-1 or HIV-2 antibodies the person is considered to be HIV-infected. In case of a negative confirmatory assay, nucleic acid testing (NAT) for HIV RNA is recommended and, if positive, an acute HIV infection has to be assumed (Malloch et al., 2013; Pandori and Branson, 2010; Poljak et al., 2009). The objective of the present study was to compare the performance of two supplemental HIV confirmatory assays for the detection and differentiation of HIV-1- and HIV-2-specific antibodies.
92
I. Friedrichs et al. / Journal of Virological Methods 224 (2015) 91–94
Table 1 Assay characteristics of the two HIV confirmatory assays. Geenius Assay
recomLine Assay Line-immunoassay (recombinant proteins) 20 l (serum/plasma)
HIV-1 and HIV-2 type differentiation Controls
Immunchromatographic assay (recombinant proteins and synthetic peptides) 5 l (serum/plasma) 15 l (venous blood) Manual 30 min Computer-assisted (GeeniusTM Reader, Notebook, GeeniusTM Software)* Yes (detection within one cassette/cartridge) (Function) control band on each cartridge
CE-marked Manufacturer
Yes Bio-Rad
Test principle Sample volume (specimen type) Processing Processing time Evaluation
* **
(Semi)automatic (Dynablot-Plus Strip Processor) 330 min Computer-assisted (Scanner, PC with analysis software recomScan)** Yes (detection with one strip) (Function) control band, conjugate control band and a cut-off control band are included on each strip Yes Mikrogen
Software considers band intensity in its decision algorithm. Assessment of band intensity in relation to the cut-off band.
2. Materials and methods 2.1. Sample and patients’ characteristics 279 serum samples from patients admitted to the University Hospital Frankfurt/Main (Germany) between 2011 and 2013 were analysed retrospectively with two HIV confirmatory assays: the GeeniusTM HIV1/2 Confirmatory Assay (Bio-Rad, Munich, Germany) and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay (Mikrogen, Neuried, Germany). All serum samples had been previously tested for HIV during routine diagnostics at the Institute of Medical Virology, National Reference Centre for Retroviruses, University Hospital Frankfurt/Main with the HIV Ag/Ab Combo Screening Assay (Abbott, Delkenheim, Germany; Architect System; REF 4J27) working with the Architect system detecting HIV-1/2 antibodies and HIV-1 p24 antigen. Samples with reactive screening assay results (positive, indeterminate) had been further evaluated using the western blot New Bio-Rad LAV-Blot I and LAV-Blot II (Bio-Rad, Munich, Germany). Before testing all serum samples were stored at −20 ◦ C. The mean patient age was 38 years (range 3–83 years), 166 were male and 101 female (unknown gender: n = 12). According to results from the HIV Ag/Ab Combo Screening Assay, New Bio-Rad LAV-Blot I and LAV-Blot II (hereinafter named “reference confirmatory assay” [RCA]) 173 samples were HIV-positive [HIV-1 (n = 171) and HIV-2 (n = 2)]. 106 samples were from HIVnegative patients including theoretically “tricky” samples with potentially interfering factors leading to cross-reactivity and/or false-reactive HIV-test results: 27 samples from pregnant women, 14 from patients with acute EBV infection (EBV-VCA-IgM pos. and -IgG pos./indeterminate, EBV-EBNA1-IgG neg.), five from patients with CMV-IgM positivity, two from HTLV-1- (HTLV-1-IgG pos.) and two from HTLV-2-positive patients (HTLV-2-IgG pos.). 2.2. Evaluated assays The GeeniusTM HIV1/2 Confirmatory Assay (Bio-Rad) hereinafter named “Geenius Assay” is a single use immunochromatographic test using the following HIV proteins (HIV-1: p31 [POL], gp160 [ENV], p24 [GAG], gp41 [ENV] and HIV-2: gp36 [ENV], gp140 [ENV]). The gp160 and the p24 antigen are recombinant proteins but all the others are synthetic HIV peptides. This cartridge assay allows rapid evaluation within 30 min. The assay was performed using the Geenius Reader S/N DP2C000309 and Mini PC Lenovo for test interpretation. Each cartridge includes a control band, which indicates that sample and reagents have been applied. Independent of its intensity all visible bands were considered “reactive” and were included to interpret reactivity. The Geenius Assay was
interpreted as positive for HIV-1 if two bands of the HIV-1 test lines with at least one ENV [gp160 or gp41] were visible. Positivity for HIV-2 was assumed if both HIV-2 test lines [gp36 and gp140] were present. The recomLine HIV-1 & HIV-2 IgG Line Immunoassay (Mikrogen) hereinafter named “recomLine Assay” is a line immunoassay using exclusively recombinant HIV proteins. The assay works with the HIV-1-specific proteins gp120 [ENV], gp41 [ENV]; p24 and p17 [GAG], p51 and p31 [POL], and the HIV-2 specific proteins gp105 [ENV] and gp36 [ENV] whereby the specific proteins gp41 and gp36 allow the differentiation between HIV-1- and HIV2-specific antibody reactivity. The assay was performed using Dynalab-Blot-Plus Strip Processor and recomScan for automated analysis and computer-assisted evaluation (processing and evaluation time: 4–5 h). Each test strip includes a reactivity control band, a conjugate control band and a cut-off control band. The band intensity was assessed in relation to the cut-off band whereby only bands with intensity equivalent or higher to the cut-off band were included in evaluation. Positivity is given when two ENV bands of the same HIV type (HIV-1: gp120 and gp41/HIV-2: gp105 and gp36) or one ENV band (HIV-1: gp41/HIV-2: gp36) and at least one GAG band (p17, p24) or POL band (p31, p51) are present. Both are CE-marked assays able to detect HIV-1- and -2-specific antibodies in serum or plasma samples and in case of the Geenius Assay, also in whole blood obtained by fingerstick. Typically negative screening samples were not further investigated by a confirmatory assay during routine diagnostic. Consequently testing for both assays was done outside of manufacturer’s instructions because also negative screening samples had to be evaluated. The outcome of both assays was interpreted in a computer-assisted fashion. Further details about the assays are found in Table 1. 2.3. Statistics Sensitivity and specificity were determined, whereby all reactive results (positive and indeterminate) were included in the calculation. Statistical analysis was done using the 95% confidence interval [95% CI] with a significance level of p < 0.05. Furthermore, Cohen’s kappa (Ä) was calculated (using the program BiAS for Windows 9.04, Epsilon Verlag, Hochheim Darmstadt 2009). A Ä-value >0.7 indicates that two ratings agree well; a value of 1 indicates full concordance (Kwiecien et al., 2011). 3. Results In the present study the sensitivity for the reference tests and both evaluated confirmatory assays (Geenius and recomLine Assay) was 100% [95%-CI: 97.9; 100] irrespective of whether
I. Friedrichs et al. / Journal of Virological Methods 224 (2015) 91–94
93
Table 2 Comparison of the Geenius Assay and the recomLine Assay results with reference assay (RCA) results (computer-assisted evaluation).
Positive Indeterminate Negative Positive Indeterminate Negative
Geenius assay
recomLine assay
* ** *** a
RCA positive (n = 173)
RCA negative** (n = 106)
Sensitivity [95% CI]
Specificitya [95% CI]
Cohens Kappa [95% CI]
173 0 0 172*** 1 0
2 5 99 1 0 105
100% [97.9; 100]
93.4% [86.9; 93.3]
0.96* [0.84; 1.00]
100% [97.9; 100]
99.0% [94.9; 100]
0.99* [0.87; 1.00]
According to Landis and Koch this indicates “high correlation”. In case of negative screening assay, the confirmatory assay was performed only in exceptional cases. 2 not typeable. Excluding indeterminate test results from calculation specificity for Geenius Assay increased to 98.0% [93.0, 99.8] while specificity for recomLine did not change.
Table 3 Overview of samples with discordant test results. Sample no.
Screening (S/CO)
RCA
Geenius assay
RecomLine assay
Comment
89
5.94
HIV-1-positive (gp160 and gp41)
Indeterminate (gp41)
187
Negative
HIV-1-positive (p52, p55, p68, gp160) Not done
Indeterminate (p31)
Negative
194
Negative Negative
234
Negative
Negative
Positive (p31, gp160) Indeterminate (p31) Negative
Negative
198
Indeterminate (p18) Negative
HIV-1 RNA: 982.000 copies/ml (plasma) → primary infection (seroconversion phase) EBV-VCA-IgM/IgG and Paul-Bunnel test positive, EBNA negative; HIV-negative follow-up sample (16 month later) CMV-positive (IgM > IgG), CMV-DNA: 2.238 copies/ml CMV-positive (IgM > IgG), no NAT
243
Negative
Negative
266
Negative
Negative
271
Negative
Negative
278
Negative
Indeterminate (p18)
Indeterminate (p31) Indeterminate (gp140) Indeterminate (p31) Indeterminate (gp140)
indeterminate test results were included or excluded from the calculation. Specificity for the recomLine Assay (99.0%) was higher than for the Geenius Assay (93.4%) when considering indeterminate results as positive (Table 2). If indeterminate samples were excluded from calculation specificity for the Geenius Assay increased to 98.0%, while the specificity for recomLine Assay did not change. The concordance of both assays in respect to the reference RCA was very high for both the Geenius (Ä = 0.94) and the recomLine Assay (Ä = 0.98). The Geenius Assay showed a discrimination rate of 100% [95%CI: 98.7; 100.0] and the recomLine Assay of 99.3% [95%-CI: 97.4; 99.9]. Two serum samples were HIV-positive in the recomLine Assay, which were, however, not typeable using the computerassisted evaluation. Visually, both of these samples could be clearly interpreted as HIV-1-positive. Both assays identified the two HIV-2 antibody-positive samples correctly. Nine samples were discordant when comparing results from either Geenius or recomLine Assay with the reference tests. From these four samples with previously negative screening assay and RCA were so called “tricky” sera, containing potentially interfering factors. Two samples were from acute EBV-infected patients (nos. 187, 278) and two from CMV-IgM-positive individuals (nos. 194, 198), respectively. Further information about the discordant samples are given in Table 3. 4. Discussion Persons who are unaware of their HIV status are of major concern considering the high rate of HIV transmission during the acute phase (Brenner et al., 2007). The current CDC
Negative Positive (gp41, p51) Negative
no NAT, no follow-up sample
Negative
HIV-negative follow-up sample (6 month later) HIV-1 RNA negative
Negative
HIV-1 RNA negative
Negative
EBV-VCA-IgM and Paul Bunnell assay positive, EBNA and VCA-IgG negative
recommendations for HIV testing and the European Guidelines on HIV testing aim to reduce the time between infection and diagnosis by initially using fourth-generation Ag/Ab combination immunoassays for screening. Sera which are repeatedly reactive in screening assays should be further tested using a confirmatory immunoassay, which differentiates between HIV-1 and HIV-2 antibodies and by NAT (Malloch et al., 2013; Pandori and Branson, 2010; Poljak et al., 2009; Wesolowski et al., 2013). Nevertheless in different countries slightly different algorithms are used for the laboratory diagnosis and confirmation of an HIV infection. In the present study specificity and sensitivity of two HIV antibody confirmatory assays, the Geenius and recomLine Assay were evaluated relative to RCA. According to the manufacturers’ data sensitivity in patients confirmed as HIV-1-/HIV-2-infected was 100% in both assays. Furthermore manufactures’ data suggest specificity of 100% for Geenius and 98.8% for recomLine Assay in samples drawn from hospitalized patients. It should be pointed out, that indeterminate samples had been excluded in the calculations of sensitivity and specificity by manufacturers for the Geenius Assay irrespective of whether they were false reactive or not. In contrast, for the recomLine Assay all reactive results (positive and indeterminate) had been included according to the manufacturer to calculate sensitivity and specificity. Present study data showed a sensitivity of 100% for both assays. Assay specificity (including indeterminate results) was 93.4% in the Geenius Assay and 99.0% in the recomLine Assay, respectively. Based on these results, the recomLine Assay complies with WHO recommendations for HIV testing which require sensitivity and specificity of at least 99% and 95%, respectively (van Dyck et al., 1999). It has to be pointed out that if indeterminate test results
94
I. Friedrichs et al. / Journal of Virological Methods 224 (2015) 91–94
were excluded from the analysis also the Geenius Assay complies with WHO criteria with increased specificity of 98%. Furthermore, the Geenius Assay had a HIV-1/HIV-2 discrimination rate of 100% [95%-CI: 98.7; 100.0], while the recomLine Assay was able to discriminate 99.3% [95%-CI: 97.4; 99.9] of the HIV1 samples and both HIV-2 samples. It should be noted that the two HIV-positive, but untypeable samples (performing computerassisted evaluation in recomLine Assay) could visually be easily interpreted as being HIV-1-positive. Positivity for HIV-1 in both assays is defined by the presence of certain band constellations, whereby the viral envelope glycoprotein precursor gp160, which is processed by a host cell protease to form the cleavage products gp120 and gp41, plays an important role. Positivity in the Geenius Assay is given if two bands are present from which at least one should be an ENV band (gp41 or gp160). In contrast, the recomLine Assay interprets samples positive if from two visible bands at least one band is the ENV band gp41, although recomLine includes another ENV band (gp 120). Thus the presence of gp160 and gp41 in the Geenius Assay are of equal importance if only gp41 is present in the recomLine Assay. 4.1. Potential cross reactivity/false reactivity False-reactive HIV test results can be caused by various factors. Interferences might occur after vaccinations, during pregnancy, due to HTLV infection or when the test kit is near expiration date (Araujo et al., 2009; Bronze et al., 1998; Facente et al., 2009; Wesolowski et al., 2011; Willman et al., 2001). Furthermore, active infection with e.g. CMV or EBV might result in false-positive reactivity. In the current study the Geenius Assay scored false-reactive in four samples (one positive, three indeterminate) from active CMV- and EBV-infected patients. According to the manufacturers’ evaluation data the Geenius Assay and the recomLine Assay showed reactivity in 2.4% and 3.6% of samples with potentially interfering factors, respectively. However, such false-reactive results are less dramatic because HIV-NAT-testing should be performed additionally to give further information (Malloch et al., 2013; Pandori and Branson, 2010) and follow-up sera would be required in such a case. 4.2. Processing/handling The Geenius Assay has the clear advantage of rapid evaluation (within 30 min) and easy handling, so that only limited lab experience and no further lab equipment are necessary. Thus, this rapid assay might also be used in settings like emergency units or in resource-limited countries, were surveillance and diagnosis of HIV infections are major health challenges (Louie et al., 2008; Sato et al., 1994). In addition, the Geenius Assay is the only confirmatory system from the evaluated assays that can be used on whole blood samples. In contrast, recomLine Assay takes at least 4 h and specific lab experience and equipment are needed. The major strength of this assay is its higher specificity (98.1%) and the fact that visible bands are interpreted in relation to a cut-off band. Although immunoassays are often visually assessed the manufacture recommends computer-assisted evaluation because the human eye is limited to distinguish shades of grey and might be influenced by other factors increasing subjectivity. In addition, test results might vary from one operator to the next. Nevertheless two HIV-1-positive samples were evaluated positive but untypeable performing computerassisted evaluation in the recomLine, but experienced laboratory workers could have interpreted these samples visually as HIV-1reactive.
5. Conclusion In conclusion, both evaluated assays are well suited for the detection, confirmation and discrimination of HIV-1- and HIV-2specific antibodies. Acknowledgements The authors would like to thank Mikrogen (Neuried, Germany) and Bio-Rad (Munich, Germany) for the supply of free assay kits. References Araujo, P.R., Albertoni, G., Arnoni, C., Almeida, K., Ribeiro, J., Rizzo, S.R., Carvalho, F.O., Baretto, J.A., Mangueira, C., 2009. Rubella vaccination and transitory false-positive test results for human immunodeficiency virus Type 1 in blood donors. Transfusion 49, 2516–2517. Brenner, B.G., Roger, M., Routy, J.P., Moisi, D., Ntemgwa, M., Matte, C., Baril, J.G., Thomas, R., Rouleau, D., Bruneau, J., Leblanc, R., Legault, M., Tremblay, C., Charest, H., Wainberg, M.A., 2007. High rates of forward transmission events after acute/early HIV-1 infection. J. Infect. Dis. 195, 951–959, Quebec Primary H.I.V.I.S.G. Bronze, M.S., Warr, A.G., Spigel, D., Smith, V.D., Smalley, D., 1998. False-positive enzyme immunoassay for human immunodeficiency virus due to acute cytomegalovirus infection. Clin. Infect. Dis. 27, 221–222, An official publication of the Infectious Diseases Society of America. Chou, R., Selph, S., Dana, T., Bougatsos, C., Zakher, B., Blazina, I., Korthuis, P.T., 2012. Screening for HIV: systematic review to update the 2005 U.S. Preventive Services Task Force recommendation. Ann. Intern. Med. 157, 706–718. Cohen, M.S., Chen, Y.Q., McCauley, M., Gamble, T., Hosseinipour, M.C., Kumarasamy, N., Hakim, J.G., Kumwenda, J., Grinsztejn, B., Pilotto, J.H., Godbole, S.V., Mehendale, S., Chariyalertsak, S., Santos, B.R., Mayer, K.H., Hoffman, I.F., Eshleman, S.H., Piwowar-Manning, E., Wang, L., Makhema, J., Mills, L.A., de Bruyn, G., Sanne, I., Eron, J., Gallant, J., Havlir, D., Swindells, S., Ribaudo, H., Elharrar, V., Burns, D., Taha, T.E., Nielsen-Saines, K., Celentano, D., Essex, M., Fleming, T.R., Team, H.S., 2011. Prevention of HIV-1 infection with early antiretroviral therapy. N. Engl. J. Med. 365, 493–505. Facente, S.N., Dowling, T., Vittinghoff, E., Sykes, D.L., Colfax, G.N., 2009. False positive rate of rapid oral fluid HIV tests increases as kits near expiration date. PLoS ONE 4, e8217. Kwiecien, R., Kopp-Schneider, A., Blettner, M., 2011. Concordance analysis: part 16 of a series on evaluation of scientific publications. Deutsches Arzteblatt Int. 108, 515–521. Louie, B., Wong, E., Klausner, J.D., Liska, S., Hecht, F., Dowling, T., Obeso, M., Phillips, S.S., Pandori, M.W., 2008. Assessment of rapid tests for detection of human immunodeficiency virus-specific antibodies in recently infected individuals. J. Clin. Microbiol. 46, 1494–1497. Malloch, L., Kadivar, K., Putz, J., Levett, P.N., Tang, J., Hatchette, T.F., Kadkhoda, K., Ng, D., Ho, J., Kim, J., 2013. Comparative evaluation of the Bio-Rad Geenius HIV-1/2 Confirmatory Assay and the Bio-Rad Multispot HIV-1/2 Rapid Test as an alternative differentiation assay for CLSI M53 algorithm-I. J. Clin. Virol. 58 (Suppl. 1), e85–e91, The official publication of the Pan American Society for Clinical Virology. Pandori, M.W., Branson, B.M., 2010. Expert review of anti-infective therapy. In: 2010 HIV Diagnostics Conference, pp. 631–633, vol. 8. Poljak, M., Smit, E., Ross, J., 2009. 2008 European guideline on HIV testing. Int. J. STD AIDS 20, 77–83. Sato, P.A., Maskill, W.J., Tamashiro, H., Heymann, D.L., 1994. Strategies for laboratory HIV testing: an examination of alternative approaches not requiring Western blot. Bull. World Health Organ. 72, 129–134. Severe, P., Juste, M.A., Ambroise, A., Eliacin, L., Marchand, C., Apollon, S., Edwards, A., Bang, H., Nicotera, J., Godfrey, C., Gulick, R.M., Johnson Jr., W.D., Pape, J.W., Fitzgerald, D.W., 2010. Early versus standard antiretroviral therapy for HIV-infected adults in Haiti. N. Engl. J. Med. 363, 257–265. UNAIDS, July 2014. The Gap Report. http://www.unaids.org/en/media/unaids/ contentassets/documents/unaidspublication/2014/UNAIDS Gap report en.pdf van Dyck, E., Meheus, A., Piot, P., 1999. Laboratory Diagnosis of Sexually Transmitted Diseases. World Health Organization (WHO), Geneva. Wesolowski, L.G., Delaney, K.P., Lampe, M.A., Nesheim, S.R., 2011. False-positive human immunodeficiency virus enzyme immunoassay results in pregnant women. PLoS ONE 6, e16538. Wesolowski, L.G., Delaney, K.P., Meyer 3rd, W.A., Blatt, A.J., Bennett, B., Chavez, P., Granade, T.C., Owen, M., 2013. Use of rapid HIV assays as supplemental tests in specimens with repeatedly reactive screening immunoassay results not confirmed by HIV-1 Western blot. J. Clin. Virol. 58, 240–244, The official publication of the Pan American Society for Clinical Virology. Willman, J.H., Hill, H.R., Martins, T.B., Jaskowski, T.D., Ashwood, E.R., Litwin, C.M., 2001. Multiplex analysis of heterophil antibodies in patients with indeterminate HIV immunoassay results. Am. J. Clin. Pathol. 115, 764–769.
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
Evaluation of two HIV antibody confirmatory assays: GeeniusTM HIV1/2 Confirmatory Assay and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay I. Friedrichs ∗ , C. Buus, A. Berger, O.T. Keppler, H.F. Rabenau Institute of Medical Virology, University Hospital Frankfurt/Main, Goethe University, National Reference Laboratory for Retroviruses, Paul-Ehrlich-Str. 40, 60596 Frankfurt/Main, Germany
a b s t r a c t Article history: Received 30 July 2014 Received in revised form 19 August 2015 Accepted 21 August 2015 Available online 24 August 2015 Keywords: Serological HIV diagnosis Confirmatory assay Line immunoassay Immunochromatographic assay
The laboratory diagnosis of an HIV infection mainly depends on the detection of HIV-specific antibodies/HIV p24 antigen whereby different algorithms for the confirmation of reactive screening assays exist. The objective of the present study was to compare the performance of two supplemental HIV antibody confirmatory assays: the GeeniusTM HIV1/2 Confirmatory Assay and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay. Therefore 279 serum samples previously analyzed for HIV during routine diagnostics at the Institute for Medical Virology, National Reference Center for Retroviruses, University Hospital Frankfurt, were analyzed retrospectively. 96.8% samples had concordant results in both HIV confirmatory assays, whereby the Geenius Assay showed a discrimination rate of 100% while two HIV-1 samples were not typeable with the recomLine Assay. Overall assay sensitivity was 100% in both assays and specificity was 99.0% (recomLine Assay) and 93.4% (Geenius Assay), respectively. The -values for both assays indicated high agreement. Overall nine samples had discordant results from which four were from acutely EBV/CMV-infected patients and one from a patient with primary HIV-1 infection during seroconversion. In conclusion, both assays are well suited for the detection, confirmation and discrimination of HIV-1and -2-specific antibodies. © 2015 Elsevier B.V. All rights reserved.
1. Introductions HIV infection remains a major public health concern around the globe. According to UNAIDS an estimated number of 35 million adults and children worldwide were infected with HIV in 2013, of which around 2.1 million were newly infected and 1.5 million deaths were reported (UNAIDS, 2014). The highest rate of HIV transmission occurs during acute HIV infection (Brenner et al., 2007) which is typically characterized by high viral loads and low antibody titres. A reliable laboratory diagnosis might contribute to decrease transmission rates at this stage and can reduce morbidity and mortality and improve quality of life through early initiation of antiretroviral therapy (Chou et al., 2012; Cohen et al., 2011; Severe et al., 2010). The laboratory diagnosis of an HIV infection mainly depends on the detection of HIV-specific antibodies and/or HIV p24
∗ Corresponding author. Tel.: +49 69 6301 83062; fax: +49 69 6301 83061. E-mail address: [email protected] (I. Friedrichs). http://dx.doi.org/10.1016/j.jviromet.2015.08.015 0166-0934/© 2015 Elsevier B.V. All rights reserved.
antigen. According to the European Guideline on HIV testing and the recommendations of the Centers for Disease Control and Prevention (CDC) and the Association of Public Health Laboratories (APHL) fourth generation screening assays for simultaneous detection of HIV-1/2 antibodies and HIV p24 antigen should be used. In different countries different algorithms exist for the confirmation of reactive screening assays. Supplemental antibody testing is recommended to confirm repeatedly reactive fourth generation assay results whereby such assays should distinguish between HIV-1 and HIV-2 antibodies. If the confirmatory assay is positive for either HIV-1 or HIV-2 antibodies the person is considered to be HIV-infected. In case of a negative confirmatory assay, nucleic acid testing (NAT) for HIV RNA is recommended and, if positive, an acute HIV infection has to be assumed (Malloch et al., 2013; Pandori and Branson, 2010; Poljak et al., 2009). The objective of the present study was to compare the performance of two supplemental HIV confirmatory assays for the detection and differentiation of HIV-1- and HIV-2-specific antibodies.
92
I. Friedrichs et al. / Journal of Virological Methods 224 (2015) 91–94
Table 1 Assay characteristics of the two HIV confirmatory assays. Geenius Assay
recomLine Assay Line-immunoassay (recombinant proteins) 20 l (serum/plasma)
HIV-1 and HIV-2 type differentiation Controls
Immunchromatographic assay (recombinant proteins and synthetic peptides) 5 l (serum/plasma) 15 l (venous blood) Manual 30 min Computer-assisted (GeeniusTM Reader, Notebook, GeeniusTM Software)* Yes (detection within one cassette/cartridge) (Function) control band on each cartridge
CE-marked Manufacturer
Yes Bio-Rad
Test principle Sample volume (specimen type) Processing Processing time Evaluation
* **
(Semi)automatic (Dynablot-Plus Strip Processor) 330 min Computer-assisted (Scanner, PC with analysis software recomScan)** Yes (detection with one strip) (Function) control band, conjugate control band and a cut-off control band are included on each strip Yes Mikrogen
Software considers band intensity in its decision algorithm. Assessment of band intensity in relation to the cut-off band.
2. Materials and methods 2.1. Sample and patients’ characteristics 279 serum samples from patients admitted to the University Hospital Frankfurt/Main (Germany) between 2011 and 2013 were analysed retrospectively with two HIV confirmatory assays: the GeeniusTM HIV1/2 Confirmatory Assay (Bio-Rad, Munich, Germany) and the recomLine HIV-1 & HIV-2 IgG Line Immunoassay (Mikrogen, Neuried, Germany). All serum samples had been previously tested for HIV during routine diagnostics at the Institute of Medical Virology, National Reference Centre for Retroviruses, University Hospital Frankfurt/Main with the HIV Ag/Ab Combo Screening Assay (Abbott, Delkenheim, Germany; Architect System; REF 4J27) working with the Architect system detecting HIV-1/2 antibodies and HIV-1 p24 antigen. Samples with reactive screening assay results (positive, indeterminate) had been further evaluated using the western blot New Bio-Rad LAV-Blot I and LAV-Blot II (Bio-Rad, Munich, Germany). Before testing all serum samples were stored at −20 ◦ C. The mean patient age was 38 years (range 3–83 years), 166 were male and 101 female (unknown gender: n = 12). According to results from the HIV Ag/Ab Combo Screening Assay, New Bio-Rad LAV-Blot I and LAV-Blot II (hereinafter named “reference confirmatory assay” [RCA]) 173 samples were HIV-positive [HIV-1 (n = 171) and HIV-2 (n = 2)]. 106 samples were from HIVnegative patients including theoretically “tricky” samples with potentially interfering factors leading to cross-reactivity and/or false-reactive HIV-test results: 27 samples from pregnant women, 14 from patients with acute EBV infection (EBV-VCA-IgM pos. and -IgG pos./indeterminate, EBV-EBNA1-IgG neg.), five from patients with CMV-IgM positivity, two from HTLV-1- (HTLV-1-IgG pos.) and two from HTLV-2-positive patients (HTLV-2-IgG pos.). 2.2. Evaluated assays The GeeniusTM HIV1/2 Confirmatory Assay (Bio-Rad) hereinafter named “Geenius Assay” is a single use immunochromatographic test using the following HIV proteins (HIV-1: p31 [POL], gp160 [ENV], p24 [GAG], gp41 [ENV] and HIV-2: gp36 [ENV], gp140 [ENV]). The gp160 and the p24 antigen are recombinant proteins but all the others are synthetic HIV peptides. This cartridge assay allows rapid evaluation within 30 min. The assay was performed using the Geenius Reader S/N DP2C000309 and Mini PC Lenovo for test interpretation. Each cartridge includes a control band, which indicates that sample and reagents have been applied. Independent of its intensity all visible bands were considered “reactive” and were included to interpret reactivity. The Geenius Assay was
interpreted as positive for HIV-1 if two bands of the HIV-1 test lines with at least one ENV [gp160 or gp41] were visible. Positivity for HIV-2 was assumed if both HIV-2 test lines [gp36 and gp140] were present. The recomLine HIV-1 & HIV-2 IgG Line Immunoassay (Mikrogen) hereinafter named “recomLine Assay” is a line immunoassay using exclusively recombinant HIV proteins. The assay works with the HIV-1-specific proteins gp120 [ENV], gp41 [ENV]; p24 and p17 [GAG], p51 and p31 [POL], and the HIV-2 specific proteins gp105 [ENV] and gp36 [ENV] whereby the specific proteins gp41 and gp36 allow the differentiation between HIV-1- and HIV2-specific antibody reactivity. The assay was performed using Dynalab-Blot-Plus Strip Processor and recomScan for automated analysis and computer-assisted evaluation (processing and evaluation time: 4–5 h). Each test strip includes a reactivity control band, a conjugate control band and a cut-off control band. The band intensity was assessed in relation to the cut-off band whereby only bands with intensity equivalent or higher to the cut-off band were included in evaluation. Positivity is given when two ENV bands of the same HIV type (HIV-1: gp120 and gp41/HIV-2: gp105 and gp36) or one ENV band (HIV-1: gp41/HIV-2: gp36) and at least one GAG band (p17, p24) or POL band (p31, p51) are present. Both are CE-marked assays able to detect HIV-1- and -2-specific antibodies in serum or plasma samples and in case of the Geenius Assay, also in whole blood obtained by fingerstick. Typically negative screening samples were not further investigated by a confirmatory assay during routine diagnostic. Consequently testing for both assays was done outside of manufacturer’s instructions because also negative screening samples had to be evaluated. The outcome of both assays was interpreted in a computer-assisted fashion. Further details about the assays are found in Table 1. 2.3. Statistics Sensitivity and specificity were determined, whereby all reactive results (positive and indeterminate) were included in the calculation. Statistical analysis was done using the 95% confidence interval [95% CI] with a significance level of p < 0.05. Furthermore, Cohen’s kappa (Ä) was calculated (using the program BiAS for Windows 9.04, Epsilon Verlag, Hochheim Darmstadt 2009). A Ä-value >0.7 indicates that two ratings agree well; a value of 1 indicates full concordance (Kwiecien et al., 2011). 3. Results In the present study the sensitivity for the reference tests and both evaluated confirmatory assays (Geenius and recomLine Assay) was 100% [95%-CI: 97.9; 100] irrespective of whether
I. Friedrichs et al. / Journal of Virological Methods 224 (2015) 91–94
93
Table 2 Comparison of the Geenius Assay and the recomLine Assay results with reference assay (RCA) results (computer-assisted evaluation).
Positive Indeterminate Negative Positive Indeterminate Negative
Geenius assay
recomLine assay
* ** *** a
RCA positive (n = 173)
RCA negative** (n = 106)
Sensitivity [95% CI]
Specificitya [95% CI]
Cohens Kappa [95% CI]
173 0 0 172*** 1 0
2 5 99 1 0 105
100% [97.9; 100]
93.4% [86.9; 93.3]
0.96* [0.84; 1.00]
100% [97.9; 100]
99.0% [94.9; 100]
0.99* [0.87; 1.00]
According to Landis and Koch this indicates “high correlation”. In case of negative screening assay, the confirmatory assay was performed only in exceptional cases. 2 not typeable. Excluding indeterminate test results from calculation specificity for Geenius Assay increased to 98.0% [93.0, 99.8] while specificity for recomLine did not change.
Table 3 Overview of samples with discordant test results. Sample no.
Screening (S/CO)
RCA
Geenius assay
RecomLine assay
Comment
89
5.94
HIV-1-positive (gp160 and gp41)
Indeterminate (gp41)
187
Negative
HIV-1-positive (p52, p55, p68, gp160) Not done
Indeterminate (p31)
Negative
194
Negative Negative
234
Negative
Negative
Positive (p31, gp160) Indeterminate (p31) Negative
Negative
198
Indeterminate (p18) Negative
HIV-1 RNA: 982.000 copies/ml (plasma) → primary infection (seroconversion phase) EBV-VCA-IgM/IgG and Paul-Bunnel test positive, EBNA negative; HIV-negative follow-up sample (16 month later) CMV-positive (IgM > IgG), CMV-DNA: 2.238 copies/ml CMV-positive (IgM > IgG), no NAT
243
Negative
Negative
266
Negative
Negative
271
Negative
Negative
278
Negative
Indeterminate (p18)
Indeterminate (p31) Indeterminate (gp140) Indeterminate (p31) Indeterminate (gp140)
indeterminate test results were included or excluded from the calculation. Specificity for the recomLine Assay (99.0%) was higher than for the Geenius Assay (93.4%) when considering indeterminate results as positive (Table 2). If indeterminate samples were excluded from calculation specificity for the Geenius Assay increased to 98.0%, while the specificity for recomLine Assay did not change. The concordance of both assays in respect to the reference RCA was very high for both the Geenius (Ä = 0.94) and the recomLine Assay (Ä = 0.98). The Geenius Assay showed a discrimination rate of 100% [95%CI: 98.7; 100.0] and the recomLine Assay of 99.3% [95%-CI: 97.4; 99.9]. Two serum samples were HIV-positive in the recomLine Assay, which were, however, not typeable using the computerassisted evaluation. Visually, both of these samples could be clearly interpreted as HIV-1-positive. Both assays identified the two HIV-2 antibody-positive samples correctly. Nine samples were discordant when comparing results from either Geenius or recomLine Assay with the reference tests. From these four samples with previously negative screening assay and RCA were so called “tricky” sera, containing potentially interfering factors. Two samples were from acute EBV-infected patients (nos. 187, 278) and two from CMV-IgM-positive individuals (nos. 194, 198), respectively. Further information about the discordant samples are given in Table 3. 4. Discussion Persons who are unaware of their HIV status are of major concern considering the high rate of HIV transmission during the acute phase (Brenner et al., 2007). The current CDC
Negative Positive (gp41, p51) Negative
no NAT, no follow-up sample
Negative
HIV-negative follow-up sample (6 month later) HIV-1 RNA negative
Negative
HIV-1 RNA negative
Negative
EBV-VCA-IgM and Paul Bunnell assay positive, EBNA and VCA-IgG negative
recommendations for HIV testing and the European Guidelines on HIV testing aim to reduce the time between infection and diagnosis by initially using fourth-generation Ag/Ab combination immunoassays for screening. Sera which are repeatedly reactive in screening assays should be further tested using a confirmatory immunoassay, which differentiates between HIV-1 and HIV-2 antibodies and by NAT (Malloch et al., 2013; Pandori and Branson, 2010; Poljak et al., 2009; Wesolowski et al., 2013). Nevertheless in different countries slightly different algorithms are used for the laboratory diagnosis and confirmation of an HIV infection. In the present study specificity and sensitivity of two HIV antibody confirmatory assays, the Geenius and recomLine Assay were evaluated relative to RCA. According to the manufacturers’ data sensitivity in patients confirmed as HIV-1-/HIV-2-infected was 100% in both assays. Furthermore manufactures’ data suggest specificity of 100% for Geenius and 98.8% for recomLine Assay in samples drawn from hospitalized patients. It should be pointed out, that indeterminate samples had been excluded in the calculations of sensitivity and specificity by manufacturers for the Geenius Assay irrespective of whether they were false reactive or not. In contrast, for the recomLine Assay all reactive results (positive and indeterminate) had been included according to the manufacturer to calculate sensitivity and specificity. Present study data showed a sensitivity of 100% for both assays. Assay specificity (including indeterminate results) was 93.4% in the Geenius Assay and 99.0% in the recomLine Assay, respectively. Based on these results, the recomLine Assay complies with WHO recommendations for HIV testing which require sensitivity and specificity of at least 99% and 95%, respectively (van Dyck et al., 1999). It has to be pointed out that if indeterminate test results
94
I. Friedrichs et al. / Journal of Virological Methods 224 (2015) 91–94
were excluded from the analysis also the Geenius Assay complies with WHO criteria with increased specificity of 98%. Furthermore, the Geenius Assay had a HIV-1/HIV-2 discrimination rate of 100% [95%-CI: 98.7; 100.0], while the recomLine Assay was able to discriminate 99.3% [95%-CI: 97.4; 99.9] of the HIV1 samples and both HIV-2 samples. It should be noted that the two HIV-positive, but untypeable samples (performing computerassisted evaluation in recomLine Assay) could visually be easily interpreted as being HIV-1-positive. Positivity for HIV-1 in both assays is defined by the presence of certain band constellations, whereby the viral envelope glycoprotein precursor gp160, which is processed by a host cell protease to form the cleavage products gp120 and gp41, plays an important role. Positivity in the Geenius Assay is given if two bands are present from which at least one should be an ENV band (gp41 or gp160). In contrast, the recomLine Assay interprets samples positive if from two visible bands at least one band is the ENV band gp41, although recomLine includes another ENV band (gp 120). Thus the presence of gp160 and gp41 in the Geenius Assay are of equal importance if only gp41 is present in the recomLine Assay. 4.1. Potential cross reactivity/false reactivity False-reactive HIV test results can be caused by various factors. Interferences might occur after vaccinations, during pregnancy, due to HTLV infection or when the test kit is near expiration date (Araujo et al., 2009; Bronze et al., 1998; Facente et al., 2009; Wesolowski et al., 2011; Willman et al., 2001). Furthermore, active infection with e.g. CMV or EBV might result in false-positive reactivity. In the current study the Geenius Assay scored false-reactive in four samples (one positive, three indeterminate) from active CMV- and EBV-infected patients. According to the manufacturers’ evaluation data the Geenius Assay and the recomLine Assay showed reactivity in 2.4% and 3.6% of samples with potentially interfering factors, respectively. However, such false-reactive results are less dramatic because HIV-NAT-testing should be performed additionally to give further information (Malloch et al., 2013; Pandori and Branson, 2010) and follow-up sera would be required in such a case. 4.2. Processing/handling The Geenius Assay has the clear advantage of rapid evaluation (within 30 min) and easy handling, so that only limited lab experience and no further lab equipment are necessary. Thus, this rapid assay might also be used in settings like emergency units or in resource-limited countries, were surveillance and diagnosis of HIV infections are major health challenges (Louie et al., 2008; Sato et al., 1994). In addition, the Geenius Assay is the only confirmatory system from the evaluated assays that can be used on whole blood samples. In contrast, recomLine Assay takes at least 4 h and specific lab experience and equipment are needed. The major strength of this assay is its higher specificity (98.1%) and the fact that visible bands are interpreted in relation to a cut-off band. Although immunoassays are often visually assessed the manufacture recommends computer-assisted evaluation because the human eye is limited to distinguish shades of grey and might be influenced by other factors increasing subjectivity. In addition, test results might vary from one operator to the next. Nevertheless two HIV-1-positive samples were evaluated positive but untypeable performing computerassisted evaluation in the recomLine, but experienced laboratory workers could have interpreted these samples visually as HIV-1reactive.
5. Conclusion In conclusion, both evaluated assays are well suited for the detection, confirmation and discrimination of HIV-1- and HIV-2specific antibodies. Acknowledgements The authors would like to thank Mikrogen (Neuried, Germany) and Bio-Rad (Munich, Germany) for the supply of free assay kits. References Araujo, P.R., Albertoni, G., Arnoni, C., Almeida, K., Ribeiro, J., Rizzo, S.R., Carvalho, F.O., Baretto, J.A., Mangueira, C., 2009. Rubella vaccination and transitory false-positive test results for human immunodeficiency virus Type 1 in blood donors. Transfusion 49, 2516–2517. Brenner, B.G., Roger, M., Routy, J.P., Moisi, D., Ntemgwa, M., Matte, C., Baril, J.G., Thomas, R., Rouleau, D., Bruneau, J., Leblanc, R., Legault, M., Tremblay, C., Charest, H., Wainberg, M.A., 2007. High rates of forward transmission events after acute/early HIV-1 infection. J. Infect. Dis. 195, 951–959, Quebec Primary H.I.V.I.S.G. Bronze, M.S., Warr, A.G., Spigel, D., Smith, V.D., Smalley, D., 1998. False-positive enzyme immunoassay for human immunodeficiency virus due to acute cytomegalovirus infection. Clin. Infect. Dis. 27, 221–222, An official publication of the Infectious Diseases Society of America. Chou, R., Selph, S., Dana, T., Bougatsos, C., Zakher, B., Blazina, I., Korthuis, P.T., 2012. Screening for HIV: systematic review to update the 2005 U.S. Preventive Services Task Force recommendation. Ann. Intern. Med. 157, 706–718. Cohen, M.S., Chen, Y.Q., McCauley, M., Gamble, T., Hosseinipour, M.C., Kumarasamy, N., Hakim, J.G., Kumwenda, J., Grinsztejn, B., Pilotto, J.H., Godbole, S.V., Mehendale, S., Chariyalertsak, S., Santos, B.R., Mayer, K.H., Hoffman, I.F., Eshleman, S.H., Piwowar-Manning, E., Wang, L., Makhema, J., Mills, L.A., de Bruyn, G., Sanne, I., Eron, J., Gallant, J., Havlir, D., Swindells, S., Ribaudo, H., Elharrar, V., Burns, D., Taha, T.E., Nielsen-Saines, K., Celentano, D., Essex, M., Fleming, T.R., Team, H.S., 2011. Prevention of HIV-1 infection with early antiretroviral therapy. N. Engl. J. Med. 365, 493–505. Facente, S.N., Dowling, T., Vittinghoff, E., Sykes, D.L., Colfax, G.N., 2009. False positive rate of rapid oral fluid HIV tests increases as kits near expiration date. PLoS ONE 4, e8217. Kwiecien, R., Kopp-Schneider, A., Blettner, M., 2011. Concordance analysis: part 16 of a series on evaluation of scientific publications. Deutsches Arzteblatt Int. 108, 515–521. Louie, B., Wong, E., Klausner, J.D., Liska, S., Hecht, F., Dowling, T., Obeso, M., Phillips, S.S., Pandori, M.W., 2008. Assessment of rapid tests for detection of human immunodeficiency virus-specific antibodies in recently infected individuals. J. Clin. Microbiol. 46, 1494–1497. Malloch, L., Kadivar, K., Putz, J., Levett, P.N., Tang, J., Hatchette, T.F., Kadkhoda, K., Ng, D., Ho, J., Kim, J., 2013. Comparative evaluation of the Bio-Rad Geenius HIV-1/2 Confirmatory Assay and the Bio-Rad Multispot HIV-1/2 Rapid Test as an alternative differentiation assay for CLSI M53 algorithm-I. J. Clin. Virol. 58 (Suppl. 1), e85–e91, The official publication of the Pan American Society for Clinical Virology. Pandori, M.W., Branson, B.M., 2010. Expert review of anti-infective therapy. In: 2010 HIV Diagnostics Conference, pp. 631–633, vol. 8. Poljak, M., Smit, E., Ross, J., 2009. 2008 European guideline on HIV testing. Int. J. STD AIDS 20, 77–83. Sato, P.A., Maskill, W.J., Tamashiro, H., Heymann, D.L., 1994. Strategies for laboratory HIV testing: an examination of alternative approaches not requiring Western blot. Bull. World Health Organ. 72, 129–134. Severe, P., Juste, M.A., Ambroise, A., Eliacin, L., Marchand, C., Apollon, S., Edwards, A., Bang, H., Nicotera, J., Godfrey, C., Gulick, R.M., Johnson Jr., W.D., Pape, J.W., Fitzgerald, D.W., 2010. Early versus standard antiretroviral therapy for HIV-infected adults in Haiti. N. Engl. J. Med. 363, 257–265. UNAIDS, July 2014. The Gap Report. http://www.unaids.org/en/media/unaids/ contentassets/documents/unaidspublication/2014/UNAIDS Gap report en.pdf van Dyck, E., Meheus, A., Piot, P., 1999. Laboratory Diagnosis of Sexually Transmitted Diseases. World Health Organization (WHO), Geneva. Wesolowski, L.G., Delaney, K.P., Lampe, M.A., Nesheim, S.R., 2011. False-positive human immunodeficiency virus enzyme immunoassay results in pregnant women. PLoS ONE 6, e16538. Wesolowski, L.G., Delaney, K.P., Meyer 3rd, W.A., Blatt, A.J., Bennett, B., Chavez, P., Granade, T.C., Owen, M., 2013. Use of rapid HIV assays as supplemental tests in specimens with repeatedly reactive screening immunoassay results not confirmed by HIV-1 Western blot. J. Clin. Virol. 58, 240–244, The official publication of the Pan American Society for Clinical Virology. Willman, J.H., Hill, H.R., Martins, T.B., Jaskowski, T.D., Ashwood, E.R., Litwin, C.M., 2001. Multiplex analysis of heterophil antibodies in patients with indeterminate HIV immunoassay results. Am. J. Clin. Pathol. 115, 764–769.