In HIV-positive individuals, untreated coinfection with hepatitis B virus (HBV) leads to accelerated liver fibrosis and higher rates of hepatocellular carcinoma and end-stage liver disease . The recommended treatment for chronic HBV in HIV-positive individuals is currently antiretroviral therapy (ART) including a potent nucleos(-t)ide analogue, such as tenofovir (TDF) or tenofovir alafenamide, both of which have dual activity against circulating HBV and HIV . The ultimate goal of treating HBV infection is the clearance of hepatitis B surface antigen (HBsAg), which is associated with histological improvement, reduced risk of hepatocellular carcinoma and prolonged survival . Nevertheless, very few treated HIV–HBV-coinfected individuals are expected to achieve HBsAg-seroclearance [4,5].
In patients with serologic evidence of HBsAg-seroclearance and acquiring HBs antibodies, covalently closed circular (ccc)DNA can still be detected in infected hepatocytes , suggesting continued viral activity despite achieving this endpoint. The formation of cccDNA, an episomal minichromosome that serves as a transcriptional template for the production of new HBV progeny, is a crucial step in the HBV life cycle. As such, its presence indicates active HBV replication in the liver and is responsible for viral persistence during chronic hepatitis B [6–8]. However, as the assessment of cccDNA requires an invasive liver biopsy, the development of novel serum biomarkers that accurately assess the size of the intrahepatic cccDNA pool and intrahepatic transcriptional activity are needed.
Lower levels of quantified hepatitis B core-related antigen (qHBcrAg) have been shown to bear a strong association with hepatitis B ‘e’ antigen (HBeAg)-seroclearance for both HBV-monoinfected and HIV–HBV-coinfected patients undergoing nucleos(-t)ide analogue treatment [9,10]. This novel surrogate marker has also been found to strongly correlate with the size of the cccDNA pool [11–17]. Recent research has shown that qHBcrAg reflects cccDNA transcriptional activity more strongly than quantified HBsAg (qHBsAg) . Nevertheless, these studies were conducted in HBV-monoinfected patients, mainly from Asian countries, where HBV genotypes B and C predominate. Considering that immunological control of intrahepatic HBV is impaired in HIV–HBV-coinfection and stronger degrees of immunosuppression are associated with higher cccDNA levels [1,8], the correlation between qHBcrAg and intrahepatic replication could be different in coinfected individuals.
On the contrary, no previous study has investigated to what extent qHBcrAg is able to reflect intrahepatic HBV replication for HIV–HBV-coinfection. The aim of this study was then to examine the correlation of qHBcrAg and intrahepatic HBV viral loads, including total intrahepatic-DNA, cccDNA and the ratio between cccDNA, and total intrahepatic-DNA, in HIV-positive patients coinfected with chronic HBV.
Using data from the French HIV–HBV Cohort Study [19,20], we selected patients included in a substudy on novel markers of HBV replication . Inclusion criteria were as follows: HIV-positive serology confirmed by western blot, HBsAg-positive serology for at least 6 months, and available quantification of serum HBcrAg and HBV intrahepatic markers from at least one liver biopsy. All patients provided written informed consent and the protocol was approved by the Hôpital Pitié-Salpêtrière and Hôpital Saint-Antoine Ethics Committees (Paris, France) in accordance with the Helsinki Declaration.
Serum HBV-DNA was quantified using a real-time PCR assay (COBAS AmpliPrep/COBAS TaqMan, detection limit: 12 IU/ml; or COBAS Amplicor HBV Monitor, detection limit: 60 IU/ml; Roche Diagnostics, Meylan, France). qHBsAg was performed using the ARCHITECT HBsAg assay (Abbott Laboratories, Rungis, France) . qHBcrAg (U/ml) was measured using a commercially available, automated HBcrAg chemiluminescence enzyme immunoassay (Lumipulse G System, FujiRebio Europe, Gent, Belgium) . Liver biopsies were obtained based on concomitant guidelines from the European Association for the Study of the Liver . DNA was extracted from snap-frozen biopsy specimens using the MasterPure DNA purification kit (Epicentre, Le-Perray-en-Yvelines, France). cccDNA and total intrahepatic-DNA levels were quantified by real-time PCR using a LightCycler instrument (Roche Diagnostics, Mannheim, Germany) as described previously [6,8].
In statistical analysis, HBV-DNA, qHBcrAg, qHBsAg, total intrahepatic-DNA, and cccDNA were log10 transformed. Spearman's rank correlation coefficients were calculated comparing each intrahepatic marker of HBV replication (total intrahepatic-DNA, cccDNA, or cccDNA : total intrahepatic-DNA ratio) to each serum marker (HBV-DNA, qHBcrAg, or qHBsAg). Analysis was stratified on HBeAg-status and detection of serum HBV-DNA (≥60 IU/ml, <60 IU/ml). The Kruskal–Wallis test was used to compare median levels of qHBcrAg at different stages of liver fibrosis (Metavir F0–F1, F2, and F3–F4). Scatterplots were used to illustrate the decline of qHBcrAg, cccDNA and total intrahepatic-DNA in individuals with paired biopsies during TDF-containing-ART. All statistical analysis was performed using STATA statistical software (v15.1, College Station, Texas, USA) and significance was defined as a P value less than 0.05.
In total, 31 patients (with 38 liver biopsies) were included. Patients were predominately male (90.3%) with median age of 42 years [interquartile range (IQR) = 37–53]. Only two patients were ART-naïve and six (19.4%) had HIV-RNA more than 50 copies/ml. Median CD4+ cell count was fairly high at 448/μl (IQR = 331–641), yet eight patients (25.8%) ever had an AIDS-defining illness and nadir CD4+ cell count was a median 262/μl (IQR = 150–326). 22 (71.0%) patients were HBeAg-positive and 22 (71.0%) had detectable plasma HBV-DNA (median = 3.1 log10 IU/ml, IQR = 2.7–7.1). At biopsy, 23 (74.2%) patients were on an anti-HBV-containing-ART regimen: lamivudine (LAM), n = 4 (17.4%); TDF, n = 2 (8.7%); LAM + TDF, n = 15 (65.2%); LAM + adefovir (ADV), n = 2 (8.7%). Previous exposure to an active anti-HBV treatment was observed in 28 (90.3%) patients, with a cumulative median months as follows: LAM, 73.2 (IQR = 50.0–91.7); ADV, 11.7 (range = 8.4–33.6); TDF, 24.7 (IQR = 9.1–31.0); interferon, 6.5 (IQR = 3.1–14.3); and pegylated-interferon, 23.2 (range = 13.2–33.2). Of those with previous exposure to LAM, 7/28 (25%) patients had developed resistance. Approximately 30% of participants (N = 9) had advanced liver fibrosis or cirrhosis (Metavir F3–F4). A complete description of the study population at the time of liver biopsy is provided in Supplemental Digital Content Table S1, https://links.lww.com/QAD/B806.
In the samples taken at the time of biopsy, median qHBcrAg was 5.5 log10 U/ml (IQR = 3.1–7.0, n = 38) and median qHBsAg 4.0 log10 IU/ml (IQR = 3.2–4.5; n = 30). HBV cccDNA was available for 27 patients (in 34 liver biopsies) and was a median 0.26 copies/cell (IQR = 0.04, 2.89) or −0.59 log10 copies/cell (IQR = −1.46, 0.46). Total intrahepatic-DNA was available for all 31 patients (in 38 liver biopsies) and was a median 2.38 copies/cell (IQR = 0.58, 207.9) or 0.38 log10 copies/cell (IQR = −0.24, 2.32). Median cccDNA : total intrahepatic-DNA ratio was 0.05 (IQR = 0.01, 0.12, n = 34).
As shown in Table 1, there was a significant and strong correlation between qHBcrAg and cccDNA in all patients (Rho = 0.65, P < 0.001; Fig. 1a), while a moderate correlation was observed between qHBcrAg and total intrahepatic-DNA (Rho = 0.57; P < 0.001; Fig. 1b) or cccDNA : total intrahepatic-DNA ratio (Rho = −0.45; P = 0.012; Fig. 1c). Similar findings were observed for HBeAg-positive patients and those with detectable HBV-DNA, with the exception of no significant correlation between qHBcrAg and cccDNA or cccDNA : total intrahepatic-DNA ratio. In contrast, qHBcrAg had no significant correlation with any intrahepatic marker when HBeAg was negative or plasma HBV-DNA was undetectable.
qHBsAg was also significantly and strongly correlated with cccDNA (Rho = 0.74, P < 0.001), total intrahepatic-DNA (Rho = 0.68, P < 0.001) and cccDNA : total intrahepatic-DNA ratio (Rho = −0.49, P = 0.009) in all patients (Table 1). In contrast, the correlation between qHBsAg and total intrahepatic-DNA remained moderate for HBeAg-positive patients, and strong between qHBsAg and total intrahepatic-DNA as well as cccDNA for those who were HBeAg-negative. Moreover, in analysis stratified on plasma HBV-DNA, qHBsAg was only strongly correlated with total intrahepatic-DNA (Rho = 0.80, P = 0.002) when plasma HBV-DNA was undetectable.
In five of the seven patients who had two liver biopsies during TDF-containing-ART, a moderately faster rate of qHBcrAg decline was observed in the first 3 years of treatment and became remarkably slower thereafter (Fig. 1d). Similarly, median cccDNA and total intrahepatic-DNA declined from 2.89 copies/cell (range = 0.02–8.36) and 45.08 copies/cell (range = 0.58–918.95) to 0.04 copies/cell (range = 0.01–0.305) and 1.40 copies/cell (range = 0.32–2.44), respectively, during roughly the first 3 years of TDF (median 40.2 months, IQR = 34.9–41.6).
Although a lower median level of qHBcrAg was observed in patients with none or mild liver fibrosis at biopsy (Fig. 1e), no significant difference in median qHBcrAg levels was observed across liver fibrosis stages (F0–F1: 4.3 U/ml, IQR = 2.6–7.1; F2: 6.4 U/ml, IQR = 4.6–7; F3–F4: 5.2 U/ml, IQR = 5–7.5; P = 0.5).
In our study, we demonstrated a significant and strong correlation overall between qHBcrAg and intrahepatocellular replication, namely levels of cccDNA and total intrahepatic-DNA. This would suggest the usefulness of qHBcrAg as a surrogate marker to assess the size of the cccDNA pool and transcriptional activity in HIV–HBV-coinfected patients. This result also corroborates previous findings in HBV-monoinfected population, either untreated [17,18] or during treatment with pegylated-interferon [13,14] or nucleos(-t)ide analogues [11,12,15,16,22].
Nevertheless, other studies in HBV-monoinfected patients have observed significant correlations between qHBcrAg and intrahepatic replication during low-active phases of HBV infection, that is, when HBV-DNA is undetectable and/or HBeAg is negative [12,17,18,22]. We were unable to confirm these findings in our cohort of HIV–HBV-coinfected patients with HBeAg-negative serology or undetectable serum HBV-DNA. The reasons for these discrepancies are unclear. The majority of studies within the context of HBV-monoinfection were conducted in Asia, where there are substantial differences in viral sequences, replication levels, and disease activity compared with patients from Europe or Africa . HBV genotypes have also been reported to influence the correlation between many markers of HBV replication [15,24] and since our study included mostly patients harboring HBV genotypes A, D, and E (as opposed to B and C in many of the Asian, HBV-monoinfection studies), it could be the reason for lack of strong correlation. Alternatively, the lack of correlation could simply be due the small numbers of patients analyzed, contributing to the failure of detecting a significant correlation. It should be noted, however, that the correlations, being between 0.05 and 0.22, were still quite low.
Significantly, declines in qHBcrAg appeared to tightly coincide with declines in both cccDNA and total intrahepatic-DNA during the first 3 years of TDF-containing-ART. Although we did not have data on intracellular replication thereafter, no further decline in qHBcrAg was noted. This finding is similar to other markers of replication, such as qHBsAg, and assuming that qHBcrAg remains significantly correlated with cccDNA over time, reinforces that very few patients undergoing long-term treatment with anti-HBV nucleos(-t)ide analogues are expected to clear intracellular HBV replication .
Our study has some limitations. The cross-sectional design makes it difficult to infer on correlation of these markers over time and the small sample sizes prohibits further stratification, especially with respect to genotype, precore mutations , and levels of CD4+ cell count. Large and multicenter prospective studies would help confirm the present findings, yet as liver biopsies are becoming increasingly rarer in clinical practice, may be unfeasible. In addition, HBcrAg is a composite biomarker whose assessment may be biased by HBeAg positivity and limited sensitivity, especially at low levels of viral replication. Although qHBcrAg is mainly correlated with cccDNA transcriptional activity , our study could not analyze other intrahepatic viral RNAs because of the lack of samples. Notwithstanding these limitations, we conclude that serum qHBcrAg could be useful in assessing levels of cccDNA, a marker denoting HBV persistence and stability. Nevertheless, the low correlations of this novel surrogate marker with cccDNA levels when serum HBV-DNA is undetectable or HBeAg is negative could limit its clinical practicality. Since the correlations with qHBcrAg observed herein do not seem to surpass those with qHBsAg, it remains debatable whether qHBcrAg provides further clinical utility over qHBsAg.
The current study was sponsored by the Institut de Médecine et d’Epidémiologie Appliquée (IMEA). L.N.C.D. was awarded a postdoctoral fellowship from the France REcherche Nord&sud Sida-hiv Hépatites (ANRS).
Source of funding: The current work was supported by SIDACTION (AO 19) and the ANRS. Gilead Sciences, Inc. provided an unrestricted grant for the French HIV–HBV cohort and was not involved in any part of the data collection, analysis, and article writing.
Authors’ contribution: L.N.C.D. was responsible for the statistical analysis, interpretation of the data, and drafting the article. S.M. was responsible for HBcrAg, HBsAg, and HBeAg quantification, interpretation of the data, and drafting the article. A.G. and C.D. were responsible for HBcrAg, HBsAg, and HBeAg quantification and drafted parts of the article. H.R., P.-M.G., C. L-C., and J.C. acquired data for the cohort, assisted in interpreting data, and gave critical revisions of the article. F.Z. gave technical support and provided all biological measurements from liver biopsies, drafted parts of the article, and provided critical revision of the article. P.-M.G. and K.L. helped design, conceptualize, and obtain funding for the French HIV–HBV cohort study, coordinated data collection, and drafted the article. A.B. coordinated data analysis, gave important comments on data interpretation, drafted the article, and provided critical revisions of the article. All authors approved the final version.
Conflicts of interest
All authors have no potential conflicts of interest, be it financial, professional or personal, that are relevant to the article.
1. Singh KP, Crane M, Audsley J, Avihingsanon A, Sasadeusz J, Lewin SR. HIV-hepatitis B virus coinfection: epidemiology, pathogenesis, and treatment
3. European Association for the Study of the Liver. EASL 2017 clinical practice guidelines on the management of hepatitis B virus infection
. J Hepatol
4. Boyd A, Maylin S, Moh R, Mahjoub N, Gabillard D, Eholié SP, et al. Hepatitis B surface antigen quantification as a predictor of seroclearance during treatment in HIV-hepatitis B virus coinfected patients from sub-Saharan Africa: seroclearance during treatment
. J Gastroenterol Hepatol
5. Boyd A, Maylin S, Gozlan J, Delaugerre C, Simon F, Girard P-M, et al. Use of hepatitis B surface and ‘e’ antigen quantification during extensive treatment with tenofovir in patients co-infected with HIV–HBV
. Liver Int
6. Werle-Lapostolle B, Bowden S, Locarnini S, Wursthorn K, Petersen J, Lau G, et al. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy
7. Levrero M, Pollicino T, Petersen J, Belloni L, Raimondo G, Dandri M. Control of cccDNA function in hepatitis B virus infection
. J Hepatol
8. Boyd A, Lacombe K, Lavocat F, Maylin S, Miailhes P, Lascoux-Combe C, et al. Decay of ccc-DNA marks persistence of intrahepatic viral DNA synthesis under tenofovir in HIV–HBV co-infected patients
. J Hepatol
9. Dezanet LNC, Maylin S, Gabassi A, Rougier H, Miailhes P, Lascoux-Combe C, et al. Kinetics of hepatitis B core-related antigen and antihepatitis B core antibody and their association with serological response in HIV-hepatitis B co-infection
. J Inf Dis
10. Wang B, Carey I, Bruce M, Montague S, Dusheiko G, Agarwal K. HBsAg and HBcrAg as predictors of HBeAg seroconversion in HBeAg-positive patients treated with nucleos(t)ide analogues
. J Viral Hepat
11. Wong DK-H, Tanaka Y, Lai C-L, Mizokami M, Fung J, Yuen M-F. Hepatitis B virus core-related antigens as markers for monitoring chronic hepatitis B infection
. J Clin Microbiol
12. Suzuki F, Miyakoshi H, Kobayashi M, Kumada H. Correlation between serum hepatitis B virus core-related antigen and intrahepatic covalently closed circular DNA in chronic hepatitis B patients: HBcrAg and cccDNA in Chronic HB Patients
. J Med Virol
13. Chuaypen N, Posuwan N, Payungporn S, Tanaka Y, Shinkai N, Poovorawan Y, et al. Serum hepatitis B core-related antigen as a treatment predictor of pegylated interferon in patients with HBeAg-positive chronic hepatitis B
. Liver Int
14. Chuaypen N, Posuwan N, Chittmittraprap S, Hirankarn N, Treeprasertsuk S, Tanaka Y, et al. Predictive role of serum HBsAg and HBcrAg kinetics in patients with HBeAg-negative chronic hepatitis B receiving pegylated interferon-based therapy
. Clin Microbiol Infect
15. Wang L, Cao X, Wang Z, Gao Y, Deng J, Liu X, et al. Correlation of HBcrAg with intrahepatic hepatitis B virus total DNA and covalently closed circular DNA in HBeAg-positive chronic hepatitis B patients
. J Clin Microbiol
16. Chen E-Q, Feng S, Wang M-L, Liang L-B, Zhou L-Y, Du L-Y, et al. Serum hepatitis B core-related antigen is a satisfactory surrogate marker of intrahepatic covalently closed circular DNA in chronic hepatitis B
. Sci Rep
17. Chen E, Wang M, Tao Y, Wu D, Liao J, He M, et al. Serum HBcrAg is better than HBV RNA and HBsAg in reflecting intrahepatic covalently closed circular DNA
. J Viral Hepat
18. Testoni B, Lebossé F, Scholtes C, Berby F, Miaglia C, Subic M, et al. Serum hepatitis B core-related antigen (HBcrAg) correlates with covalently closed circular DNA transcriptional activity in chronic hepatitis B patients
. J Hepatol
19. Boyd A, Gozlan J, Miailhes P, Lascoux-Combe C, Cam MS-L, Rougier H, et al. Rates and determinants of hepatitis B ‘e’ antigen and hepatitis B surface antigen seroclearance during long-term follow-up of patients coinfected with HIV and hepatitis B virus
20. Lacombe K, Massari V, Girard P-M, Serfaty L, Gozlan J, Pialoux G, et al. Major role of hepatitis B genotypes in liver fibrosis during coinfection with HIV
21. Kimura T, Rokuhara A, Sakamoto Y, Yagi S, Tanaka E, Kiyosawa K, et al. Sensitive enzyme immunoassay for hepatitis B virus core-related antigens and their correlation to virus load
. J Clin Microbiol
22. Wong DK-H, Seto W-K, Cheung K-S, Chong C-K, Huang F-Y, Fung J, et al. Hepatitis B virus core-related antigen as a surrogate marker for covalently closed circular DNA
. Liver Int
23. Lin C-L, Kao J-H. Natural history of acute and chronic hepatitis B: the role of HBV genotypes and mutants
. Best Pract Res Clin Gastroenterol
24. Maasoumy B, Wiegand SB, Jaroszewicz J, Bremer B, Lehmann P, Deterding K, et al. Hepatitis B core-related antigen (HBcrAg) levels in the natural history of hepatitis B virus infection in a large European cohort predominantly infected with genotypes A and D
. Clin Microbiol Infect
25. Boyd A, Lacombe K, Lavocat F, Miailhes P, Lascoux-Combe C, Girard P-M, et al. Low incidence of precore W28∗ mutant variants in treated hepatitis B virus and human immunodeficiency virus co-infected patients
. Antiviral Res