Diagnostic Microbiology and Infectious Disease 83 (2015) 248–251 Contents lists available at ScienceDirect Diagnostic Microbiology and Infectious Disease journal homepage: www.elsevier.com/locate/diagmicrobio Virology Identifying patients infected with hepatitis B virus in sub-Saharan Africa: potential for misclassification Anders Boyd a,⁎, Sarah Maylin b,d, Raoul Moh e, Delphine Gabillard f,g, Hervé Menan e,h, Nadia Mahjoub b,d, Christine Danel e,f,g, Xavier Anglaret e,f,g, Serge Paul Eholié e,i,j, Pierre-Marie Girard k,l, Fabien Zoulim m, Constance Delaugerre b,c,d, Karine Lacombe k,l, for the ANRS 12240 VarBVA study a INSERM, UMR_S1136, Institut Pierre Louis d’Epidémiologie et de Santé Publique, Paris, France Laboratoire de Virologie, Hôpital Saint-Louis, AP-HP, Paris, France c INSERM U941, Paris, France d Université Paris-Diderot, Paris, France e Programme PAC-CI, ANRS Research Site, Treichville University Hospital, Abidjan, Côte d’Ivoire f INSERM, U897, Epidémiologie-Biostatistique, Bordeaux, France g University of Bordeaux, ISPED, Bordeaux, France h Centre de Diagnostic et de Recherche sur le SIDA et les Maladies Opportunistes (CeDReS), Centre Hospitalier Universitaires (CHU) de Treichville, Abidjan, Côte d’Ivoire i Department of Infectious and Tropical Diseases, Treichville University Teaching Hospital, Abidjan, Côte d’Ivoire j Medical School, University Felix Houphouet Boigny, Abidjan, Côte d'Ivoire k Department of Infectious and Tropical Diseases, Saint-Antoine Hospital, AP-HP, Paris, France l Sorbonne Universités, UPMC Univ Paris 06, UMR_S 1136, Institut Pierre Louis d’Epidémiologie et de Santé Publique, Paris, France m Centre de Recherche sur le Cancer de Lyon, Equipes 15 et 16, INSERM, Unité 1052, CNRS, UMR 5286, Lyon, France b a r t i c l e i n f o Article history: Received 14 March 2015 Received in revised form 7 July 2015 Accepted 16 July 2015 Available online 18 July 2015 a b s t r a c t Most research in sub-Saharan Africa establishes hepatitis B infection via one-time hepatitis B surface antigen (HBsAg) testing. Of 237 HIV-infected patients from two clinical trials testing HBsAg positive (MiniVidas®), 206 (86.9%) had validated serological tests using another assay (Architect). Discrepancies could be due to inactive infection, highlighting the importance of assessing hepatitis B virus infection phase. © 2015 Elsevier Inc. All rights reserved. Keywords: Hepatitis B virus Isolated hepatitis B core antigen Hepatitis surface antigen Inactive hepatitis B infection Bias Chronic infection with hepatitis B virus (HBV) affects roughly 8% of the population in sub-Saharan Africa (SSA) (Ott et al., 2012). Viral hepatitis is the major cause of hepatocellular carcinoma, which is, respectively, the number 1 and 3 cancer among men and women living on the continent (Ferlay et al., 2010). Despite the major health burden of this disease, there is still a poor understanding of the natural history and clinical progression of HBV infection in this setting (Lemoine et al., 2015). Several research projects have been initiated to address these questions; nevertheless, the basis of these studies relies on sufficient means of identifying HBV-infected individuals. We recently conducted a study examining HBV replication among patients also infected with the human immunodeficiency virus (HIV) treated with antiretroviral therapy (ART) (Boyd et al., 2015). Briefly, the VarBVA study is a nested cohort including patients from two prospective, randomized-control trials in Abidjan, Côte d’Ivoire: Trivacan ⁎ Corresponding author. Tel.: +33-1-71-97-05-17; fax: +33-1-49-28-21-49. E-mail address: anders.boyd@upmc.fr (A. Boyd). http://dx.doi.org/10.1016/j.diagmicrobio.2015.07.009 0732-8893/© 2015 Elsevier Inc. All rights reserved. ANRS 1269 (Danel et al., 2006) and Temprano ANRS 12136 (Jean et al., 2014). Study procedures have been detailed elsewhere (Boyd et al., 2015). We report herein our experience with testing hepatitis B surface antigen (HBsAg) at the inclusion of these studies. During the inclusion study visit, all patients were tested for HBsAg using the MiniVidas® assay [bioMérieux, Marcy l’Etoile, France; from package insert and (Weber, 2006): detection limit = 0.05 IU/mL, sensitivity = 97.87%, specificity = 100.00%] performed at the Centre de Diagnostic et de Recherche sur le SIDA et les Maladies Opportunistes, Centre Hospitalier Universitaires de Treichville in Abidjan, Côte d’Ivoire. This test is a purely qualitative, automated, enzyme-linked fluorescent assay using the Vidas instrument (short protocol). An index is calculated comparing the intensity from a reagent–sample mixture to a standard control, with values b0.13 resulting in a negative test. Of the 2465 patients enrolled, 259 (10.5%) were HBsAg positive. As a means of validating these results for another substudy, we decided to retest HBsAg using a different assay [HBsAg Qual II Architect; Abbott Laboratories, Rungis, France; from package insert and (Lou et al., A. Boyd et al. / Diagnostic Microbiology and Infectious Disease 83 (2015) 248–251 2011): detection limit = 0.03 IU/mL, sensitivity = 99.80%, specificity = 99.96%] at the Laboratoire de Virologie, Hôpital Saint-Louis in Paris, France. The assay used here is a qualitative, automated, chemiluminescent microparticle immunoassay. A measure of relative light units from a reagent–sample mixture is compared to a control, with ratios b1.0 resulting in a negative test. Samples were from the same blood draw, stored at −80 °C for a median 6.9 years (range = 4.4–12.4), and had been thawed only once for previous laboratory work. Twenty-two samples were not validated because patients only had 1 study visit (and were hence excluded from the substudy, n = 16) or did not have an available sample (n = 6). Of the 237 included patients, 31 (13.1%) had HBsAg-negative results using this test. As shown in Table 1, patients with discordant results were all hepatitis B “e” antigen (HBeAg) negative, while having significantly lower HBV-DNA levels (P = 0.0001) and median CD4+ cell counts (P = 0.0006) when compared to those with concordant results. In patients with discordant results, one-third (11/31) had detectable HBV-DNA, in whom the median level was 3.03 log10 copies/mL (interquartile range [IQR] = 2.83–3.73). We determined if patients with discordant results had exposure to HBV by testing samples for anti–hepatitis B core antibodies (anti-HBcAb) using the semiquantitative Anti-HBc II test (Architect, positive result N1.0 S/CO). Among those with available samples, 19/27 (70.4%) had anti-HBcAb–positive serology with a median result at 7.6 S/CO (IQR = 4.6–8.7). In this report, we took a validation approach to test for HBV infection, in which HBsAg status was determined at an initial screening visit and a confirmatory HBsAg test was performed on stored samples of the same 249 study visit. We observed that 13.1% of the population did not have confirmed infection, making HBV infection status difficult to establish. Of those with discrepant results, 1/3 had detectable HBV-DNA, and 70% had evidence of HBV exposure. Both of these proportions are much higher than what is expected in HIV-positive populations from the same area (N’Dri-Yoman et al., 2010), suggesting some form of current or past HBV infection. Six of the 31 discordant results had absolutely no other markers of HBV infection, which could indicate a false-positive test with the MiniVidas®. In either case, it is difficult to confirm the results of the MiniVidas® considering we did not have enough serum to perform neutralizing assays. One operational reason for these findings could be that HBsAg levels diminished after prolonged periods of storage, during transportation, or during multiple rounds of freezing and thawing. HBsAg levels from the same blood draw could have passed from detectable with the MiniVidas® to undetectable when determined later with the Architect. This would have assumedly occurred more often in patients with low HBsAg titers, which was mildly prevalent in the study population (3.8% had HBsAg titers of 0.05–10 IU/mL as quantified by the Architect assay). Nevertheless, a previous evaluation of HBsAg quantities in samples stored at −20 °C demonstrated barely any change over a 12-month period (Fung et al., 2011), suggesting that this hypothesis is unlikely. Genetic characteristics of HBV could have also influenced test discordance. HBV genotype E, which was by far the most frequent genotype (Boyd et al., 2015), has been observed to cause reduced sensitivity for some HBsAg tests yet does not appear to affect the MiniVidas® or Architect assays (Scheiblauer et al., 2010). Genetic mutations of the “a” Table 1 Characteristics of patients according to results of the MiniVidas® with validation by the Architect HBsAg assay. HBsAg test result Groups Concordant Discordant Subjects CF124 CF147 CN031 CN097 UC030 UC062 CN002 CN004 CN015 CN040 CN105 NS057 RB001 SM067 SM077 SM151 UC094 US032 US040 PF057 RB134 US114 CF260 CN036 LP022 RB002 SM147 SM148 CN032 PF055 RB162 Age, y Sex HBeAg statusb HBV-DNA, log10 copies/mLb anti-HBc Ab titerb HIV-RNA, log10 copies/mLa CD4+ cell count/mm3a MiniVidas®a Architectb Pos Pos Pos Neg 34 (30–40) 36 (30–38) 68.5% F 67.7% F 19.9% Posc 0% Posc 3.37 (b0.30–6.30)d b0.30 (b0.30–2.90)d Not performed 5.18 (0.38–8.50) 4.93 (4.36–5.37) 4.93 (4.38–5.48) 356 (251–474)e 261 (171–348)e Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Pos Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg 30 31 25 29 57 36 35 38 36 42 25 37 34 39 33 38 46 29 32 23 34 41 26 36 19 36 42 32 36 38 46 F F F F M M F M M M F F F F F M M F F F F F F M F F M F M F F Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg Neg b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 b0.30 2.35 2.51 3.69 3.03 3.03 2.90 3.94 2.81 3.76 2.85 7.10 0.15 0.12 0.12 0.10 0.85 0.13 6.42 5.18 8.52 4.63 8.27 3.22 5.28 9.8 7.62 8.69 3.9 8.97 9.58 Missing 0.06 0.38 9.69 4.33 5.47 8.5 4.39 7.62 Missing Missing Missing 5.86 4.38 5.03 4.52 5.00 4.50 4.75 6.68 5.45 6.72 4.87 4.32 Missing 5.10 4.56 5.37 5.55 3.19 5.95 4.52 5.26 5.48 7.15 3.67 4.15 3.39 3.14 4.82 4.33 6.39 5.01 285 (15.2) 644 (22.9) 205 (13.0) 193 (15.8) 353 (14.8) 269 (23.1) 171 (14.1) 163 (19.8) 57 (14.3) 352 (25.1) 195 (20.5) 528 (28.1) 267 (13.0) 171 (12.2) 511 (20.3) 151 (7.7) 180 (9.1) 348 (22.5) 146 (15.3) 279 (14.9) 175 (13.4) 160 (8.9) 333 (17.2) 334 (18.2) 400 (21.1) 265 (10.9) 146 (15.3) 416 (18.1) 261 (11.9) 188 (10.5) 188 (10.0) All values in the groups are expressed as median (IQR). CD4+ T-cell counts are given in per mm3 and percentages in parenthesis. Samples were taken from the same blood draw and quantified either aduring study inclusion or bduring the retrospective HBsAg validation test. c P = 0.004 between groups using 2-sided Fisher's exact test. d P = 0.0001 and e P = 0.0006 between groups using a Kruskal–Wallis rank test. No other significant differences were observed. Abbreviations: anti-HBc Ab = anti–hepatitis B core antibodies; F = female; HBeAg = hepatitis B “e” antigen; HBsAg = hepatitis B surface antigen; HBV = hepatitis B virus; HIV = human immunodeficiency virus; M = male; Neg = negative; Pos = positive. 250 A. Boyd et al. / Diagnostic Microbiology and Infectious Disease 83 (2015) 248–251 determinant of the S gene have shown to substantially reduce test sensitivity (Chudy et al., 2013; Servant-Delmas et al., 2012), which could have resulted in false-negative tests for one assay but not the other. These types of mutations were fairly common in our study among patients with higher HBV-DNA replication. Unfortunately, HBV-DNA viral loads among those with discrepant tests were too low for HBV sequencing, making it difficult to confirm the role of genetic variability. CD4+ cell counts were also significantly lower in patients with discordant results, suggesting that HIV-induced immunosuppression could play a part in test discrepancy. Previous studies have shown poorer sensitivity and specificity of point-of-care HBsAg tests in HIVinfected individuals from SSA (Geretti et al., 2010), yet the mechanism for this finding remains unclear, as discussed by Njai et al. (2015). Most importantly, patients with discordant results were more likely to be HBeAg negative with much lower levels of HBV-DNA. These characteristics describe more inactive HBV infection, where HBsAg levels are generally low (Jaroszewicz et al., 2010). Considering that discordance is known to occur more frequently at lower HBsAg levels, even between different assay methods with similar detection thresholds (Liu et al., 2014), perhaps patients with more inactive HBV-infection were susceptible of producing discrepant test results. Assuming that inactive carriers were the underlying cause of test discrepancies, it does question the relevance of misclassification. From a clinical perspective, HBV-monoinfected patients with inactive infection would not require treatment, while their risk of developing serious liver-related morbidity and mortality would be deemed low (European Association For The Study Of The Liver, 2012). For these patients, a falsenegative test would result in absent HBV-specific clinical care, which might not be overly detrimental, and false-positive tests would likely require further laboratory evaluation leading to diagnosis of inactive disease, which would again lead to infrequent follow-up. However, inactive carrier status does not make these patients privy to low risk of serious disease or reactivation (Chen et al., 2010). In addition, certain attention should be given during HIV-HBV coinfection, since HBV reactivation has been observed in coinfected patients with serological indicators of past infection after temporarily discontinuing ART with potent anti-HBV nucleoside/nucleotide analogues (Seang et al., 2013). Seeing that routine HBsAg and HBV-DNA testing are not commonly available in SSA, reducing false-negative tests would help clinicians assess this etiology with minimal repeat or follow-up testing. From an epidemiological standpoint, many studies rely on only one HBsAg result to determine HBV status and no other markers (Hawkins et al., 2013). If patients with very inactive forms of HBV infection, exhibiting low risk of serious clinical events, are identified as HBsAg positive in these studies, the effects of HBV infection on liver-related or overall mortality could risk becoming diluted (Hoffmann et al., 2008). One limitation of our study is that patients with physiciandiagnosed liver disease were excluded, comprising a patient population likely to have more active forms of viral infection. Furthermore, HBsAg testing was not performed on the exact same sample at the same time, which could have induced further variability. However, our results are in the context of a practical application that is commonplace for many conducting research on stored samples. Notwithstanding these limitations, our experience highlights the problem of HBsAg testing encountered during nested cohort studies from SSA. As a single HBsAg test might not be sufficient in epidemiological studies, researchers should strive to fully evaluate the phase of HBV infection of their participants. Statement of interest This study was supported in full by funds from the Agence Nationale de Recherche sur le Sida et les Hépatites (ANRS 12240). The Trivacan and Temprano studies also received funding from the ANRS (ANRS 1269/ANRS 12104 and ANRS 12136, respectively). Finally, A.B. was awarded a postdoctoral fellowship from the ANRS for some of the work presented in this manuscript and is currently receiving postdoctoral funding from SIDACTION. Acknowledgements and disclosures We thank all patients who participated in both trials. We also gratefully acknowledge the valuable contributions of the SMIT, CeDReS, CEPREF, USAC, CIRBA, CNTS, La Pierre Angulaire, Hôpital Général Abobo, Formation Sanitaire Anonkoua Kouté, Centre de santé El Rapha, the Programme PACCI team, as well as the INSERM exU593 and U897 teams (Abanou Matthieu, Adou Isabelle, Aman Adou, Anasthasie Yapo, Bakayoko Ibouraîma, Bombo Léontine, Célestin N’chot, Christian Kouadio, Cissé Edwidge, Coulibaly Ali, Djédjé Lucien, Djetouan Hugues, Djobi-Djo Edouard, Goly Jocelyn, Kassi Marie-Cécile, Koffi Justine, Koffi-N’Dri Aholi, Konan Sylvie, Konaté Mamadou, Kouadio Bertin, Kouamé Martin, Kouadio Cheftin, Kouadio Martin, Kouadio Victoire, Kouakou-Aboua Adrienne, Kouakou Yao, Kouamé Antoine, Kouamé Ferdinand, Kouamé Gérald, Kouamé Justine, Labibi Georgette, Lehou Jean, Lokou Benjamin, Martin Marie-Pierre, Moh Jules, MoussaDoumbia Mariam, N’Chot Célestin, N’Dri Marie Julie, Nalourgou Tuo, N’Goran Brou, Nogbout Marie-Pascale, Orne-Gliemann Joanna, Ouattara Bakary, Ouattara Minata, Oupoh Joséphine, Sidibé Abdelh, Siloué Bertine, Soro Adidiata, Tchehy Amah-Cécile, Yao Emile, Yao Juliette, Yoro Guei, and Zaho Marcel); and Bristol-Myers Squibb for providing Zerit and Videx; Gilead Sciences, for the donation of Truvada; and Merck Sharp & Dohme, for the donation of Stocrin. We would finally like to thank Paul Dény and Mariama Issoufou for the quantification of HBV-DNA viral loads and HBV sequencing. References Boyd A, Moh R, Gabillard D, le Carrou J, Danel C, Anglaret X, et al. Low risk of lamivudineresistant hepatitis B virus and hepatic flares in treated HIV-HBV co-infected patients from Côte d’Ivoire. Antivir Ther 2015. http://dx.doi.org/10.3851/IMP2959. [in press]. Chen J-D, Yang H-I, Iloeje UH, You S-L, Lu S-N, Wang L-Y, et al. Carriers of inactive hepatitis B virus are still at risk for hepatocellular carcinoma and liver-related death. Gastroenterology 2010;138(5):1747–54. Chudy M, Scheiblauer H, Hanschmann K-M, Kress J, Nick S, Wend U, et al. Performance of hepatitis B surface antigen tests with the first WHO international hepatitis B virus genotype reference panel. J Clin Virol 2013;58(1):47–53. Danel C, Moh R, Minga A, Anzian A, Ba-Gomis O, Kanga C, et al. CD4-guided structured antiretroviral treatment interruption strategy in HIV-infected adults in west Africa (Trivacan ANRS 1269 trial): a randomised trial. Lancet 2006;367(9527):1981–9. European Association For The Study Of The Liver. EASL clinical practice guidelines: management of chronic hepatitis B virus infection. J Hepatol 2012;57(1):167–85. Ferlay J, Shin H-R, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010;127(12): 2893–917. Fung J, Lai C-L, Young J, Wong DK-H, Yuen J, Seto W-K, et al. Stability of hepatitis B surface antigen over time: implications for studies using stored sera. J Med Virol 2011; 83(11):1900–4. Geretti AM, Patel M, Sarfo FS, Chadwick D, Verheyen J, Fraune M, et al. Detection of highly prevalent hepatitis B virus coinfection among HIV-seropositive persons in Ghana. J Clin Microbiol 2010;48(9):3223–30. Hawkins C, Christian B, Ye J, Nagu T, Aris E, Chalamilla G, et al. Prevalence of hepatitis B co-infection and response to antiretroviral therapy among HIV-infected patients in Tanzania. AIDS 2013;27(6):919–27. Hoffmann CJ, Charalambous S, Martin DJ, Innes C, Churchyard GJ, Chaisson RE, et al. Hepatitis B virus infection and response to antiretroviral therapy (ART) in a South African ART program. Clin Infect Dis 2008;47(11):1479–85. Jaroszewicz J, Calle Serrano B, Wursthorn K, Deterding K, Schlue J, Raupach R, et al. Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective. J Hepatol 2010;52(4):514–22. Jean K, Gabillard D, Moh R, Danel C, Fassassi R, Desgrées-du-Loû A, et al. Effect of early antiretroviral therapy on sexual behaviors and HIV-1 transmission risk among adults with diverse heterosexual partnership statuses in Côte d’Ivoire. J Infect Dis 2014; 209(3):431–40. Lemoine M, Eholié S, Lacombe K. Reducing the neglected burden of viral hepatitis in Africa: strategies for a global approach. J Hepatol 2015;62(2):469–76. Liu C, Chen T, Lin J, Chen H, Chen J, Lin S, et al. Evaluation of the performance of four methods for detection of hepatitis B surface antigen and their application for testing 116,455 specimens. J Virol Methods 2014;196:174–8. Lou SC, Pearce SK, Lukaszewska TX, Taylor RE, Williams GT, Leary TP. An improved Abbott ARCHITECT assay for the detection of hepatitis B virus surface antigen (HBsAg). J Clin Virol 2011;51(1):59–63. N’Dri-Yoman T, Anglaret X, Messou E, Attia A, Polneau S, Toni T, et al. Occult HBV infection in untreated HIV-infected adults in Côte d’Ivoire. Antivir Ther 2010;15(7):1029–34. A. Boyd et al. / Diagnostic Microbiology and Infectious Disease 83 (2015) 248–251 Njai HF, Shimakawa Y, Sanneh B, Ferguson L, Ndow G, Mendy M, et al. Validation of rapid point-of-care (POC) tests for detection of hepatitis B surface antigen in field and laboratory settings in the Gambia, Western Africa. J Clin Microbiol 2015;53(4): 1156–63. Ott JJ, Stevens GA, Groeger J, Wiersma ST. Global epidemiology of hepatitis B virus infection: new estimates of age-specific HBsAg seroprevalence and endemicity. Vaccine 2012;30(12):2212–9. Scheiblauer H, El-Nageh M, Diaz S, Nick S, Zeichhardt H, Grunert H-P, et al. Performance evaluation of 70 hepatitis B virus (HBV) surface antigen (HBsAg) assays from around the world by a geographically diverse panel 251 with an array of HBV genotypes and HBsAg subtypes. Vox Sang 2010;98(3 Pt 2): 403–14. Seang S, Thibault V, Valantin MA, Katlama C. Adjustment of antiretroviral regimen may lead to HBV reactivation even in patients with past HBV infection serological profile. J Infect Chemother 2013;19(5):987–9. Servant-Delmas A, Mercier-Darty M, Ly TD, Wind F, Alloui C, Sureau C, et al. Variable capacity of 13 hepatitis B virus surface antigen assays for the detection of HBsAg mutants in blood samples. J Clin Virol 2012;53(4):338–45. Weber B. Diagnostic impact of the genetic variability of the hepatitis B virus surface antigen gene. J Med Virol 2006;78(Suppl. 1):S59–65.