Secondary Logo

Share this article on:

Effects of long-term tenofovir-based combination antiretroviral therapy in HIV-hepatitis B virus coinfection on persistent hepatitis B virus viremia and the role of hepatitis B virus quasispecies diversity

Audsley, Jennifera,b,c; Bent, Stephen J.d; Littlejohn, Margarete; Avihingsanon, Anchaleef; Matthews, Gailg; Bowden, Scotte; Bayliss, Juliannee; Luciani, Fabioh; Yuen, Lillye; Fairley, Christopher K.i,j; Locarnini, Stephene; Lewin, Sharon R.a,b,c; Sasadeusz, Joeb

doi: 10.1097/QAD.0000000000001080

Objective: Hepatitis B virus (HBV) can persist in some HIV–HBV coinfected individuals on tenofovir disoproxil fumarate (TDF)-containing combination antiretroviral therapy (cART) but HBV resistance to TDF has not been reported and the source of persistent HBV DNA on TDF is poorly understood. The aims of this study were to assess long-term HBV suppression in HIV–HBV coinfected individuals receiving TDF and investigate quasispecies variation using ultradeep pyrosequencing (UDPS).

Methods: Ninety-two HIV–HBV coinfected participants on, or about to commence, TDF-containing cART were enrolled [Australia (n = 40), Thailand (n = 52)] and followed for 2 years with study visits every 6 months. HBV reverse transcriptase sequencing was performed on samples with HBV DNA more than 400 IU/ml by population-based methods and UDPS. Quasispecies diversity was assessed using Shannon entropy.

Results: Over 24 months, viremia was detected at least once in 17% (n = 16) of the cohort. Novel mutations were not identified in on TDF samples tested by population-based sequencing (n = 19). Using UDPS, the median Shannon entropy value in samples prior to TDF in patients aviremic on TDF was not statistically different from those who were viremic on TDF (n = 50; 8.4 and 9.1, respectively, P = 0.9). Longitudinal Shannon entropy analysis of on TDF samples from five participants showed three individuals with significant changes in viral diversity over time.

Conclusion: Persistent viremia on TDF-containing cART is common but TDF-resistance was not detected. In some individuals, changes in viral diversity over time were observed on TDF which could potentially be active viral replication. Further follow-up will be needed to determine the clinical significance of detectable HBV DNA on TDF-containing cART.

aDepartment of Infectious Diseases, Monash University, Melbourne

bDepartment of Infectious Diseases, The Alfred Hospital, Prahran

cDoherty Institute for Infection and Immunity, University of Melbourne, Parkville

dRobinson Research Institute, University of Adelaide, Adelaide, Australia

eHIV-NAT Research Collaboration, Bangkok, Thailand

fThe Kirby Institute, Sydney

gVictorian Infectious Diseases Reference Laboratory, North Melbourne

hSystems Medicine, School of Medical Sciences, University of NSW, Sydney

iCentral Clinical School, Monash University, Melbourne

jMelbourne Sexual Health Centre, Carlton, Australia.

Correspondence to Jennifer Audsley, PhD, Doherty Institute for Infection and Immunity, University of Melbourne, 4th Floor, 792 Elizabeth Street, Melbourne, Victoria 3010, Australia. Tel: +61 3 83443266; fax: +61 3 93472952; e-mail:

Received 2 December, 2015

Revised 9 February, 2016

Accepted 10 February, 2016

Back to Top | Article Outline


Approximately 33 million people globally are infected with HIV and on average 10–20% are coinfected with hepatitis B virus (HBV) [1,2]. The prevalence of chronic HBV infection in people with HIV varies in accordance with epidemiological patterns of transmission, and it has been estimated that 6–9% of HIV-infected individuals in Australia and in Asia, respectively are chronically infected with HBV [3,4]. HIV infection has a significant impact on the natural history of chronic HBV infection, with increased levels of HBV DNA, accelerated progression of liver disease, and increased liver-associated mortality [5–7].

Tenofovir disoproxil fumarate (TDF), an oral acyclic nucleotide analogue approved for the treatment of both HIV and HBV, is very effective in suppressing both HIV and HBV replication in participants with HIV–HBV coinfection [8,9]. TDF is active against wild-type HBV and HBV strains that contain lamivudine (LMV) resistance polymerase gene mutations [10]. However, approximately 10% of HIV–HBV coinfected individuals on TDF-containing regimens, referred to as HBV-active combination antiretroviral therapy (cART), have persistent detectable HBV DNA or have initial control of HBV DNA, followed by viral rebound [11]. This is higher than levels reported for persistence of HBV DNA in HBV monoinfection, where only 1% of hepatitis B e antigen (HBeAg) positive and 3% HBeAg negative patients had detectable HBV DNA after 5 years of TDF treatment [12].

HBV resistance to TDF has not been reported so the reasons for persistent HBV DNA on HBV-active cART are unclear [13], and the clinical determinants of persistent HBV viremia remain unknown. In addition, it is unclear whether persistent viremia is associated with actively replicating virus, or is because of release from hepatocytes or extrahepatic sources such as plasma or lymphoid cells. Furthermore, the long-term clinical significance of persistent low-level viremia is not known. The main aims of the prospective TDF Surveillance Study were [1] to assess long-term suppression of HBV in HIV–HBV coinfected individuals receiving TDF [2], to detect any novel HBV mutations [3], to determine the clinical and virological risk factors associated with persistent viremia on TDF and [4] to investigate the role of quasispecies variation using ultradeep pyrosequencing (UDPS).

Back to Top | Article Outline

Participants and methods

Study participants

Total 92 HIV–HBV coinfected individuals were enrolled from multiple sites in Australia (total of 40 participants from The Alfred Hospital and Melbourne Sexual Health Centre (Melbourne), and St Vincent's Hospital, Sydney) and Thailand (total of 52 participants from HIV-NAT, Thai Red Cross AIDS Research Centre, Bangkok) over the period July 2008 to November 2009. Eligibility criteria included chronic HBV infection (defined as two positive HBV surface antigen (HBsAg) or HBV DNA tests at least 6 months apart), HIV antibody positive status, currently receiving or about to commence cART, including TDF and the availability of a pre TDF sample. A pre-TDF serum sample was required for study entry. Written informed consent was obtained from all participants, and the study was approved by the relevant Human Research Ethics Committees in Australia and Thailand.

Back to Top | Article Outline

Data abstraction and collection

Clinical and laboratory data were collected or abstracted from medical records at study entry and follow-up visits. Clinical data included demographics, prior and current anti-HIV and anti-HBV therapy, previous/present AIDS-defining illnesses, history/current jaundice, hepatocellular carcinoma, ascites, oesophageal varices, and hepatic encephalopathy. Laboratory measurements included alanine aminotransferase, aspartate aminotransferase, haemoglobin, white blood cell count, platelets, HBeAg, HBe antibody (anti-HBe), hepatitis B surface antibody (anti-HBs), HCV antibody, HIV RNA, current CD4+ cell count, and nadir CD4+ cell count. Data on alcohol intake and compliance with cART were also collected at each visit. Participants were followed for 2 years, with four study visits at six monthly intervals. HBV polymerase sequencing was performed on plasma samples with an HBV DNA more than 400 IU/ml.

Back to Top | Article Outline

Definitions of hepatitis B virus response on tenofovir disoproxil fumarate

Virological suppression on TDF was defined as HBV DNA less than 20 IU/ml. Participants who remained virologically supressed at all study visits while receiving TDF were defined as virological responders. Those who had HBV DNA detected (HBV DNA > 20 IU/ml) on one occasion while receiving TDF were defined as transiently viremic, those with HBV DNA detected (HBV DNA > 20 IU/ml) on more than one but not all occasions were defined as intermittently viremic and those with detectable HBV DNA detected (HBV DNA > 20 IU/ml) at all study visits were defined as persistently viremic.

Back to Top | Article Outline

Hepatitis B virus DNA quantification

HBV DNA quantification was performed using either the RealART HBV Light Cycler PCR (QIAGEN, Hilden, Germany), lower limit of detection (LLOD) 20 IU/ml, or the Abbott real-time HBV assay (Abbott Laboratories), LLOD 15 IU/ml, in accordance with the manufacturer's instructions. Assay used was dependent on local standard of care. Some older, stored pre-TDF samples had HBV DNA quantified using the standard of care assay at the time – Versant HBV DNA 3.0 assay (Bayer HealthCare-Diagnostics, Tarrytown, New York, USA), LLOD 357 IU/ml. No prospectively collected samples were quantified using the Bayer Versant assay.

Back to Top | Article Outline

HIV RNA quantification

HIV RNA quantification was by the standard test used at each site, according to manufacturer's instructions. For analysis, HIV RNA was classified as detectable (≥50 copies/ml) or undetectable (<50 copies/ml).

Back to Top | Article Outline

Hepatitis B virus DNA polymerase sequencing and genotype determination

HBV DNA was extracted from 200 μl of serum using the QIAamp DNA mini kit (QIAGEN), according to the manufacturer's instructions. Extracted DNA was eluted in a final volume of 50 μl of supplied elution buffer. Domains A–E of the HBV polymerase were amplified, sequenced, and analysed as previously described [14]. HBV consensus sequences were constructed, genotyped, and HBV-polymerase mutations were identified using the DNA sequence analysis program SeqScape (ABI Prism; Applied Biosystems, Foster City, California, USA).

Back to Top | Article Outline

Hepatitis B virus DNA ultradeep pyro-sequencing

For UDPS, the HBV reverse transcriptase (domains A–E) was amplified and UDPS was performed with the Roche 454 FLX Titanium as previously described [15].

Back to Top | Article Outline

Hepatitis B virus DNA ultradeep pyrosequencing analysis

Raw sequence output was merged into fastq files and adapter and primer sequence were trimmed using cutadapt [16]. Trimmed reads with length below 365 nucleotides (23% of reads) were excluded from further analyses. The remaining reads were aligned to a set of HBV reference sequences corresponding to the PCR amplicon sequence from a representative of each genotype (A–G) using bowtie2 (99.97% of reads aligned). Full-length (365 nucleotides) reads that mapped to the dominant genotype in each sample without insertions or deletions were used in Binary Alignment Map format for subsequent analysis.

Aligned reads were further analysed for detection of single nucleotide variants (SNV), including low frequency variants using Lofreq [17]. To measure viral diversity in each sample, Shannon entropy per nucleotide position was calculated from SNV files using a custom script in Perl. Shannon entropy provides an estimate of the complexity within a viral population [15]. Haplotype reconstruction of the amplicon pool from each sample was performed with Short Reads Assembly into Haplotypes (ShoRAH, open source software) [18,19]. Haplotypes obtained for this analysis were further curated to reduce systematic errors around homopolymeric regions. The DNA sequence of each inferred haplotype was translated into a protein sequence in each of the two reading frames for the reverse transcriptase and S genes, and nonsynonymous variants were extracted using a custom script.

The haplotype sequences within the amplicon region of viral variants were then utilised to construct phylogenetic trees. Phylogenetic trees were constructed using MEGA version 5.2.2 using the GTR+I+G maximum likelihood model and SPR heuristic search for optimal topology using neighbour joining for the starting tree. Participants who had an HBV DNA positive pre-TDF sample; at least one sample positive on TDF; and haplotypes were present at a frequency greater than 1% were included in the tree, along with reference and outgroup taxa.

Back to Top | Article Outline

Statistical analysis

Categorical variables were compared using the χ2 test, or, where cell numbers were small, Fishers Exact test. Continuous variables were compared using the Mann–Whitney test. A participant was classified as aviremic on TDF if all study plasma samples had undetectable (<20 IU/ml) HBV DNA, and viremic on TDF if at least one study plasma sample was more than 20 IU/ml. Data from the viremic on TDF participants were compared with the data from aviremic on TDF participants. As European Association for the Study of the Liver guidelines define successful virological response as undetectable HBV DNA by 6 months of treatment [20], those participants who commenced TDF at study entry were not classified as viremic for the first 6 months of the study. Multivariate modelling was restricted to two variables because of the size (n = 16) of the viremic on TDF group. Firth logistic regression was used in the multivariate analysis (MVA) as the most appropriate approach to adjust for the small number of cases. Final model derivation process included a model, including all continuous variables identified as significant in univariate analysis to identify possible independently significant variables. Any variable identified as independently significant in this initial model was then tested in a series of two-variable models. This was repeated in models using categorical variables. A further series of models then paired the significant categorical and continuous variables. In all analyses P < 0.05 was considered significant. Analysis was completed using IBM SPSS Statistics software (Version 22.0.0; IBM, Armonk, New York, USA). MVA was performed using the SPSS Firth logistic regression extension bundle (V1.1.0) and the Integration Plug-in for R.

Back to Top | Article Outline


Study participants and clinical features

Demographic and clinical characteristics of the cohort (n = 92) at study entry are displayed in Table 1. The majority of the cohort were men (80%), median age was 42.5 years and the most commonly reported HIV risk factor was sexual transmission (MSM 50% and heterosexual 43%). A total of 61% of the cohort were of Asian ethnicity and 24% Caucasian. HBV DNA was detectable in six (7.6%) and HIV RNA was detected in three (3.8%) participants at enrolment.

Table 1

Table 1

The median duration on TDF was 1.9 years [range 0.0–7.8 years, nterquartile range (IQR) 1.1–4.9 years]. Thirteen (14.1%) of participants were within the first 6 months of a TDF-inclusive regimen at study entry. Clinical signs of liver disease (including jaundice, oesophegeal varices, and splenomegaly) were observed in 16% of the cohort and no participant exhibited ascites, fetor, or metabolic flap. Median alanine aminotransferase was 35 IU/l (IQR 29–49) and aspartate aminotransferase was 34 IU/l (IQR 28–42). Pre-TDF LMV-resistance mutations were identified by population-based sequencing in 18 of 61 samples (30%) with sufficient HBV DNA to enable PCR amplification.

Back to Top | Article Outline

Hepatitis B virus DNA on tenofovir disoproxil fumarate

Almost three-quarters [74%, 95% confidence interval (CI) 64–83%] of pre-TDF samples (n = 68) had detectable HBV DNA with median log10 3.5 (IQR 1.3–7.3). By study entry (after a median 1.91 years of TDF) only six (7.6%) of samples had detectable HBV DNA and HBV DNA had decreased to median log10 1.2 (IQR 1.2–2.5). Over 24 months of follow-up, less than 10% of participants had detectable HBV DNA at any time point while on TDF, with a total of 30 viremic on TDF samples. Cumulatively, 16 (17%, (95% CI 10–27%) of participants had detectable HBV DNA on at least one study visit during follow-up. Only one participant reported missing doses of TDF on more than one study visit and this participant also had detectable HIV RNA. All other episodes of detectable HBV DNA correlated with undetectable HIV RNA.

Back to Top | Article Outline

Patterns of hepatitis B virus response on tenofovir disoproxil fumarate

Over 24 months of follow-up, virological responder was achieved in 76 participants (83%, 95% CI 73–90); nine participants were transiently viremic (10%, 95% CI 5–18), four were intermittently viremic 4% (95% CI 1–11), median number of occasions equals to two (IQR 2–2.75)) and three participants, all adherent to cART, were persistently viremic (3%, 95% CI 0.9–10). There were six participants with detectable HBV DNA at study entry, five of whom became HBV DNA negative during follow-up. Five participants with detectable HBV DNA once during follow-up had undetectable HBV DNA at study entry.

Back to Top | Article Outline

Population sequencing of hepatitis B virus polymerase

All cohort pre-TDF and on TDF samples with sufficient HBV DNA to successfully perform PCR underwent population sequencing and UDPS. There were sufficient levels of HBV DNA in 61 pre-TDF samples to amplify by PCR, all of which were successfully sequenced using population-based methods. LMV-resistance mutations were detected by population sequencing in 30% of these samples, and the presence of pre-TDF LMV-resistance polymerase mutations was not associated with viremia on TDF (P = 0.75). There were 30 viremic on TDF samples from 16 participants, 13 of which were successfully amplified by PCR. One of the 13 PCR products failed population-based sequencing. Novel HBV reverse transcriptase mutations were not identified in any of the 12 on TDF samples that underwent HBV population sequencing.

Back to Top | Article Outline

Clinical factors associated with detectable hepatitis B virus DNA on tenofovir disoproxil fumarate

Categorical variables at study entry which were statistically significantly associated at the univariate level with detectable HBV while on TDF were HBV genotype, HBeAg positive status, country of recruitment, peripheral oedema, and previously HCV antibody positive (Table 2). Continuous variables that were statistically significantly associated with detectable HBV at study entry were lower platelets, urea, phosphate, and albumin; and higher γ glutamyl transferase and log10 HBV DNA (Table 2). Factors that were not significantly associated with on TDF viremia included age, sex, presence of LMV resistance mutations, duration on TDF, duration on cART, duration on LMV, HIV risk factors, known duration HIV positive, HIV RNA and CD4+ cell count at study entry, BMI, and HCV antibody positive at study entry. To derive the final MVA model, the continuous variables that were statistically significant at the univariate level (n = 6) were put together in a Firth logistic regression model, and only log10 HBV DNA remained significant. This was repeated with the statistically significant categorical variables (n = 6), where only HBeAg positive status remained significant. As the model should be restricted to two variables because of the number of cases (n = 16), a series of models were then analysed, testing all variables identified as significant in univariate analyses in pairs against log10 HBV DNA and then in pairs against HBeAg status. The only variables that remained independently associated with detectable HBV DNA on TDF in all MVA combinations were log10 HBV DNA at study entry and HBeAg status (Table 2).

Table 2

Table 2

Back to Top | Article Outline

Hepatitis B virus surface antigen and haemoglobin e antigen serological responses

Seven participants (8%) lost HBsAg by 24 months of follow-up (after a median 3.1 years of TDF), two of whom (2% of the cohort) seroconverted with acquisition of anti-HBs. Both seroconverters had low HBV DNA levels in samples collected prior to TDF (<1.18 and 3.31 log10 IU/ml, respectively). One participant who was anti-HBs positive at study entry became anti-HBs negative at the 24-month visit. Three participants who were HBeAg positive at study entry became HBeAg negative by 24 months of follow-up, two of whom also seroconverted with acquisition of anti-HBe. An additional three participants acquired anti-HBe but were HBeAg negative at both study entry and at the 24-month visit.

Back to Top | Article Outline

Hepatitis B virus quasispecies diversity

There were sufficient levels of HBV DNA in 61 pre-TDF samples to amplify by PCR, and 50 of these samples were successfully sequenced using UDPS. These 50 samples were obtained from 36 participants who subsequently had successful virological response on TDF and 14 participants who subsequently had detectable HBV DNA at least once on TDF (which included transiently viremic, intermittently viremic, and persistently viremic). The distribution of viral quasispecies was analysed by identifying nonsynonymous SNV for each pre TDF sample. SNV distribution was then used to calculate Shannon entropy across the sequenced region for each sample. There was no significant difference in the median pre-TDF Shannon entropy in samples from participants with virological responder compared to those participants with detectable HBV DNA at least once on TDF (transiently viremic, intermittently viremic, and persistently viremic) (Table 3, P = 0.93, median Shannon entropy 8.4 and 9.1, respectively).

Low level (<20%) known LMV resistance mutations in two pre-TDF samples and one on TDF sample were detected in three samples by UDPS that were not detected by population-based sequencing. The mutations were V173L/L190 M/M204 V and rtA181T/V.

There were 30 viremic on TDF samples from 16 participants, 13 of which were successfully amplified by PCR, and all 13 samples successfully underwent UDPS. Viral diversity and phylogenetics analysis of viral sequences using haplotypes (viral sequences covering the full amplicon region thus representing the distribution of viral quasispecies) reconstructed from UDPS data were performed longitudinally on the 13 on TDF samples, which were from a total of five participants. Longitudinal Shannon entropy analysis showed a substantial change in HBV viral diversity over time on TDF in three participants (Fig. 1a). Change in HBV viral diversity did not correspond with change in HBV DNA level of the samples (Fig. 1b).

Fig. 1

Fig. 1

A phylogenetic tree was generated using haplotypes with a frequency more than 1% from these five participants (Fig. 2). Three participants demonstrated notable changes on phylogenetic trees over time: participant 1 (genotype shift), 2 (mutational changes), and 4 (significant increase in viral diversity at 18 months) (Fig. 3).

Fig. 2

Fig. 2

Fig. 3

Fig. 3

Back to Top | Article Outline


The aim of this study was to define the viral and serological responses to TDF, the frequency of persistent HBV DNA and whether TDF resistance emerged in HIV–HBV coinfected participants on TDF-containing cART. We showed that TDF-based cART was very effective in suppressing HBV in the setting of HIV–HBV coinfection, as others have shown [21]. However, 17% of the cohort had detectable HBV DNA whereas on TDF on at least one occasion and 10% at least twice, although TDF resistance was never observed. These findings are similar to other published coinfection cohort studies [22–25].

This is the first study, to our knowledge, to examine HBV quasispecies variation over time on TDF using UDPS. We did not find a significant difference in the median Shannon entropy prior to commencing TDF in patients with detectable HBV DNA on TDF compared to those without detectable HBV DNA on TDF. This suggests that quasispecies variation prior to treatment with TDF did not play a role in persistent viremia. In a limited number of participants, however, longitudinal assessment of HBV sequences in plasma over time on TDF showed increasing quasispecies diversity, potentially consistent with active virus replication.

In univariate analysis, laboratory parameters at study entry that are known to be markers of severe liver disease were factors associated with detectable HBV DNA while on TDF, suggesting that perhaps cirrhosis was associated with impaired TDF efficacy. HBV genotype and country of recruitment were associated with detectable HBV DNA on TDF, but these were related factors in our cohort, as participants from Australian sites were predominantly genotype A, whereas participants from Bangkok were either genotype C or genotype B. Although HBV DNA levels were lower in genotype C participants in this cohort, higher in-vitro levels of HBV DNA expression have been observed with genotype C and lower levels with genotype A, in both cell lysates and culture media [26]. Genotype G has previously been reported as a factor associated with nonresponse to long-term TDF in the setting of HIV–HBV coinfection; however, none of our cohort had HBV genotype G [27]. We identified HBeAg positive status as associated with detectable HBV DNA on TDF, and this is a well known surrogate marker for HBV DNA in HIV–HBV coinfected participants [28]. Higher HBV DNA and HBeAg status have been previously associated with partial, suboptimal and delayed response to TDF [11,25,29–31]. The only variables that remained statistically significant in multivariate analyses were higher HBV DNA at study entry and HBeAg positive status.

Little data have been published using deep sequencing in HIV–HBV coinfection. One of the earliest published studies of UDPS in HBV infection included four HIV–HBV coinfected participants and showed that UDPS was sensitive and could quantify minor HBV variants [32]. Recently, another study utilized UDPS to analyse LMV-associated mutations in HIV–HBV coinfected participants from Malawi and again, as expected, showed that UDPS was more sensitive than Sanger sequencing and the A181S mutation was identified in one participant [33]. Another study also reported evidence of quasispecies variation in the HBV rt region during TDF treatment by clonal analysis [34]. This study presented proportions of residue substitutions at baseline between ‘delayed’ TDF responders (failed to achieve 2 log IU/ml decline in HBV DNA at 24 weeks of therapy) and ‘good TDF responders (HBV DNA < 2.3 log IU/ml at 24 weeks of therapy), and observed a number of significant differences in the frequency of individual site changes at a number of loci in the HBV reverse transcriptase. They found that delayed responders had a higher incidence of LMV-resistant mutations than good responders (71 and 35%, respectively, P = 0.03). In contrast, we did not find a significant difference in the presence of LMV-resistant mutations (28 and 33%, respectively, P = 0.75) between those with virological responder and those with detectable HBV DNA at least once on TDF (transiently viremic, intermittently viremic, and persistently viremic); however, there are differences in the groups being compared. Our cohort compared always aviremic participants with those who were viremic at least once on TDF over 2 years, whereas Lada et al. compared virological responder with delayed responders as determined at 24 weeks of therapy. In addition, they found an association between delayed response and genotype G, whereas none of the participants tested in our cohort were genotype G.

Although persistent viremia on treatment has been rarely studied for HBV, this is an area of significant research in HIV-infected patients on cART, however. the explanation for this phenomenon still remains unclear [35]. A number of potential mechanisms driving persistent HIV RNA in plasma following cART have been proposed, including on-going replication despite cART or release of virus from a stable latent reservoir [36,37]. Sequencing of virus from the plasma of patients on suppressive cART when HIV RNA was less than 50 copies/ml did not show novel drug resistance mutations, and viral evolution studies have reported conflicting results regarding genetic changes in this setting [35–37]. Persistent viremia in treated HBV could potentially have similar origins – either residual virus replication or release of virus from hepatocytes and/or extrahepatic sites that have been previously shown to persist indefinitely following nucleos(t)ide analog treatment [38,39]. Our data using longitudinal UDPS analysis suggest that in some individuals there may indeed be ongoing plasma virus replication although why drug resistance does not appear in these patients in the presence of drug is unclear. It is also possible that replication could be occurring in sites in the liver with suboptimal drug penetration as recently reported in the setting of HIV infection [40]. More work is clearly needed to fully understand the pathogenesis and origin of persistent HBV DNA in plasma on TDF.

This is the first prospective cohort study of HIV–HBV coinfection that includes a mix of genotypes (A, B, C, and D) and UDPS analysis in a relatively large number of samples. This study, however, also had a number of limitations. The cohort included participants who were on TDF prior to study entry; hence some data prior to TDF were retrospective. Although a sample collected and stored prior to TDF was required for study inclusion, HBV DNA was too low to perform HBV polymerase sequencing and genotyping in 30% of these samples. In addition, the low number of viremic samples on TDF did not allow us to separately analyse different patterns of viremia, such as transiently viremic, intermittently viremic, and persistently viremic. It is possible that the different types of viremia may have different pathogenic origins.

In conclusion, while detection of HBV DNA on more than one occasion in HIV–HBV coinfected participants on TDF-containing cART was common, drug resistance to TDF did not occur in this setting. HBV DNA level at enrolment, positive HBeAg status, markers of severe liver disease, HBV genotype and country of recruitment were associated with detectable HBV DNA on TDF at the univariate level. Only HBV DNA at study entry and HBeAg positive status remained independently associated with detectable HBV DNA while on TDF in multivariate analyses. HBV quasispecies diversity prior to commencing TDF was not associated with detectable HBV DNA on TDF. There was significant quasispecies variation in some participants over time while on TDF, suggesting that ongoing viral replication could be in part responsible for persistent residual HBV viremia on TDF.

Back to Top | Article Outline


The authors thank the participants in the study, clinical research nursing staff at all sites, and the assistance of the HIV database team in the Department of Infectious Diseases, Alfred Hospital, Melbourne.

We acknowledge funding from Gilead Sciences. S.R.L. is an NHMRC practitioner fellow and has received support from the Alfred Foundation.

Back to Top | Article Outline

Conflicts of interest

G.M. has received research funding from Gilead Sciences and MSD, travel support from BMS and Roche; funding from AbbVie for consultancy and advisory board involvement. S.A.L. has received royalties and intellectual property rights/patent holder from Melbourne Health; consulting fees from Gilead Sciences; consulting fees and fees for non-CME services from BMS. S.R.L.'s institution has received investigator-initiated funding for clinical trials from Merck and Gilead Sciences; and her institution has received funding from Viiv Healthcare and Gilead Sciences for her involvement in education activities. J.S. has received funding from Gilead Sciences and BMS for Advisory board involvement, research funding from Gilead Sciences and AbbVie and speaker fees from Gilead Sciences and BMS.

Back to Top | Article Outline


1. Soriano V, Puoti M, Bonacini M, Brook G, Cargnel A, Rockstroh J, et al. Care of patients with chronic hepatitis B and HIV co-infection: recommendations from an HIV-HBV International Panel. AIDS 2005; 19:221–240.
2. UNAIDS. Report on the global AIDS epidemic: 2006: Geneva, Switzerland: UNAIDS; 2006.
3. Lincoln D, Petoumenos K, Dore G. Australian HIV Observational Database. HIV/HBV and HIV/HCV coinfection, and outcomes following highly active antiretroviral therapy. HIV Med 2003; 4:241–249.
4. Ruxrungtham K, Brown T, Phanuphak P. HIV/AIDS in Asia. Lancet 2004; 364:69–82.
5. Colin J, Cazals-Hatem D, Loriot M, Martinot-Peignoux M, Pham B, Auperin A, et al. Influence of human immunodeficiency virus infection on chronic hepatitis B in homosexual men. Hepatology 1999; 29:1306–1310.
6. Brook MG, Gilson R, Wilkins EL. BHIVA guidelines: coinfection with HIV and chronic hepatitis B virus. HIV Medicine 2003; 4:42–51.
7. Thio CL, Seaberg EC, Skolasky R Jr, Phair J, Visscher B, Munoz A, Thomas DL. HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter Cohort Study (MACS). Lancet 2002; 360:1921–1926.
8. Audsley J, Arrifin N, Yuen L, Ayres A, Crowe S, Bartholomeusz A, et al. Prolonged use of tenofovir in HIV/hepatitis B virus (HBV)-coinfected individuals does not lead to HBV polymerase mutations and is associated with persistence of lamivudine HBV polymerase mutations. HIV Med 2009; 10:229–235.
9. de Vries-Sluijs TE, Reijnders JG, Hansen BE, Zaaijer HL, Prins JM, Pas SD, et al. Long-term therapy with tenofovir is effective for patients co-infected with human immunodeficiency virus and hepatitis B virus. Gastroenterology 2010; 139:1934–1941.
10. Dore GJ, Cooper DA, Pozniak AL, DeJesus E, Zhong L, Miller MD, et al. Efficacy of tenofovir disoproxil fumarate in antiretroviral therapy-naive and -experienced patients coinfected with HIV-1 and hepatitis B virus. J Infect Dis 2004; 189:1185–1192.
11. Matthews G, Seaberg E, Avihingsanon A, Bowden S, Dore G, Lewin S, Sasadeusz J, et al. Patterns and causes of suboptimal response to tenofovir based therapy in HIV-HBV infected individuals. Clin Infect Dis 2013; 56:e87–e94.
12. Marcellin P, Gane E, Buti M, Afdhal N, Sievert W, Jacobson IM, et al. Regression of cirrhosis during treatment with tenofovir disoproxil fumarate for chronic hepatitis B: a 5-year open-label follow-up study. Lancet 2013; 381:468–475.
13. Snow-Lampart A, Chappell B, Curtis M, Zhu Y, Myrick F, Schawalder J, et al. No resistance to tenofovir disoproxil fumarate detected after up to 144 weeks of therapy in patients monoinfected with chronic hepatitis B virus. Hepatology 2011; 53:763–773.
14. Davies J, Littlejohn M, Locarnini SA, Whiting S, Hajkowicz K, Cowie BC, et al. Molecular epidemiology of hepatitis B in the Indigenous people of northern Australia. J Gastroenterol Hepatol 2013; 28:1234–1241.
15. Ayres A, Yuen L, Jackson KM, Manoharan S, Glass A, Maley M, et al. Short duration of lamivudine for the prevention of hepatitis B virus transmission in pregnancy: lack of potency and selection of resistance mutations. J Viral Hepat 2013; 21:809–817.
16. Martin M. Cutadapt removes adapter sequences from high-throughput sequencing reads. In: EMBnet.journal. 10–12.
17. Wilm A, Aw PPK, Bertrand D, Yeo GHT, Ong SH, Wong CH, et al. LoFreq: a sequence-quality aware, ultra-sensitive variant caller for uncovering cell-population heterogeneity from high-throughput sequencing datasets. Nucleic Acids Res 2012; 40:11189–11201.
18. McElroy K, Zagordi O, Bull R, Luciani F, Beerenwinkel N. Accurate single nucleotide variant detection in viral populations by combining probabilistic clustering with a statistical test of strand bias. BMC Genomics 2013; 14:501.
19. Zagordi O, Klein R, Daumer M, Beerenwinkel N. Error correction of next-generation sequencing data and reliable estimation of HIV quasispecies. Nucleic Acids Res 2010; 38:7400–7409.
20. European Association For The Study Of The Liver. EASL clinical practice guidelines: management of chronic hepatitis B virus infection. J Hepatol 2012; 57:167–185.
21. Price H, Bansi L, Sabin CA, Bhagani S, Burroughs A, Chadwick D, et al. Hepatitis B Virus Infection in HIV-Positive Individuals in the UK Collaborative HIV Cohort (UK CHIC) Study. PLoS One 2012; 7:e49314.
22. Boyd A, Gozlan J, Maylin S, Delaugerre C, Peytavin G, Girard PM, et al. Persistent viremia in human immunodeficiency virus/hepatitis B coinfected patients undergoing long-term tenofovir: virological and clinical implications. Hepatology 2014; 60:497–507.
23. Kosi L, Reiberger T, Payer BA, Grabmeier-Pfistershammer K, Strassl R, Rieger A, Peck-Radosavljevic M. Five-year on-treatment efficacy of lamivudine-, tenofovir- and tenofovir + emtricitabine-based HAART in HBV-HIV-coinfected patients. J Viral Hepat 2012; 19:801–810.
24. Matthews GV, Seaberg E, Dore GJ, Bowden S, Lewin SR, Sasadeusz J, et al. Combination HBV therapy is linked to greater HBV DNA suppression in a cohort of lamivudine-experienced HIV/HBV coinfected individuals. AIDS 2009; 23:1707–1715.
25. Plaza Z, Aguilera A, Mena A, Vispo E, Sierra-Enguita R, Tome S, et al. Influence of HIV infection on response to tenofovir in patients with chronic hepatitis B. AIDS 2013; 27:2219–2224.
26. Sugiyama M, Tanaka Y, Kato T, Orito E, Ito K, Acharya SK, et al. Influence of hepatitis B virus genotypes on the intra- and extracellular expression of viral DNA and antigens. Hepatology 2006; 44:915–924.
27. Lada O, Gervais A, Branger M, Peytavin G, Roquebert B, Collin G, et al. Long-term outcome of primary nonresponders to tenofovir therapy in HIV/HBV-co-infected patients: impact of HBV genotype G. Liver Int 2012; 32:93–101.
28. Matthews GV, Avihingsanon A, Lewin SR, Amin J, Rerknimitr R, Petcharapirat P, et al. A randomized trial of combination hepatitis B therapy in HIV/HBV coinfected antiretroviral naïve individuals in Thailand. Hepatology 2008; 48:1062–1069.
29. Childs K, Joshi D, Byrne R, Bruce M, Carey I, Agarwal K, Taylor C. Tenofovir-based combination therapy for HIV/HBV co-infection: factors associated with a partial HBV virological response in patients with undetectable HIV viraemia. AIDS 2013; 27:1443–1448.
30. Hafkin JS, Osborn MK, Localio AR, Amorosa VK, Kostman JR, Stern JJ, et al. Incidence and risk factors for incomplete HBV DNA suppression in HIV/HBV-co-infected patients initiating tenofovir-based therapy. J Viral Hepat 2014; 21:288–296.
31. Kim HN, Rodriguez CV, Van Rompaey S, Eron JJ, Thio CL, Crane HM, et al. Factors associated with delayed hepatitis B viral suppression on tenofovir among HBV-HIV coinfected patients in the CNICS cohort. J Acquir Immune Defic Syndr 2014; 66:96–101.
32. Solmone M, Vincenti D, Prosperi MCF, Bruselles A, Ippolito G, Capobianchi MR. Use of massively parallel ultradeep pyrosequencing to characterize the genetic diversity of hepatitis B virus in drug-resistant and drug-naive patients and to detect minor variants in reverse transcriptase and hepatitis B S antigen. J Virol 2009; 83:1718–1726.
33. Aoudjane S, Chaponda M, Gonzalez Del Castillo AA, O’Connor J, Noguera M, Beloukas A, et al. Hepatitis B virus (HBV) sub-genotype A1 infection is characterized by high replication levels and rapid emergence of drug resistance in HIV-positive adults receiving first-line antiretroviral therapy (ART) in Malawi. Clin Infect Dis 2014; 59:1618–1626.
34. Lada O, Gervais A, Branger M, Peytavin G, Roquebert B, Collin G, et al. Quasispecies analysis and in vitro susceptibility of HBV strains isolated from HIV-HBV-coinfected patients with delayed response to tenofovir. Antivir Ther 2012; 17:61–70.
35. Palmer S. Advances in detection and monitoring of plasma viremia in HIV-infected individuals receiving antiretroviral therapy. Curr Opin HIV AIDS 2013; 8:87–92.
36. Siliciano RF, Greene WC. HIV latency. Cold Spring Harb Perspect Med 2011; 1:a007096.
37. Van Lint C, Bouchat S, Marcello A. HIV-1 transcription and latency: an update. Retrovirology 2013; 10:67.
38. Coffin CS, Mulrooney-Cousins PM, Peters MG, van Marle G, Roberts JP, Michalak TI, Terrault NA. Molecular characterization of intrahepatic and extrahepatic hepatitis B virus (HBV) reservoirs in patients on suppressive antiviral therapy. J Viral Hepat 2011; 18:415–423.
39. Lucifora J, Xia Y, Reisinger F, Zhang K, Stadler D, Cheng X, et al. Specific and nonhepatotoxic degradation of nuclear hepatitis B virus cccDNA. Science 2014; 343:1221–1228.
40. Fletcher CV, Staskus K, Wietgrefe SW, Rothenberger M, Reilly C, Chipman JG, et al. Persistent HIV-1 replication is associated with lower antiretroviral drug concentrations in lymphatic tissues. Proc Natl Acad Sci U S A 2014; 111:2307–2312.

HIV–hepatitis B virus coinfection; persistent viremia; tenofovir; ultradeep pyrosequencing

Copyright © 2016 Wolters Kluwer Health, Inc.