Virus and Host Testing to Manage Chronic Hepatitis B

Abstract

Chronic hepatitis B virus (HBV) infection is a major cause of cirrhosis and hepatocellular carcinoma worldwide. The past 50 years have seen rapid developments in HBV testing. Beginning from traditional serologic tests, the availability of sensitive HBV DNA assays allows a thorough understanding of the virology and natural history of chronic HBV infection. Quantification of hepatitis B surface antigen levels reflects the amount and transcriptional activities of covalently closed circular DNA in the liver and may be used to evaluate the stage of disease and guide antiviral therapy. The natural history of chronic HBV infection is also a manifestation of the interaction between the host and the virus, and recent genomic works have shed light on the host–virus relationship and may provide novel tests in the future. This review highlights recent advances in the application of HBV tests in the management of chronic hepatitis B.

Chronic hepatitis B is a global health problem affecting >350 million people [1]. Chronic hepatitis B is an immune-mediated disease; the great variation in disease activity from person to person results from the diverse interactions between host immune clearance and the hepatitis B virus (HBV). Some patients who achieve immune control and clear hepatitis B surface antigen (HBsAg) will have excellent long-term prognosis. Other patients who have prolonged, unsuccessful immune control will develop progressive liver fibrosis, which may eventually lead to cirrhosis and hepatocellular carcinoma (HCC). Treatment of chronic hepatitis B can halt disease progression and reduce the risk of liver-related complications [2].

HBV is a partially double-stranded DNA virus with 4 open reading frames: surface, precore/core, polymerase, and X [3]. Three viral antigens are produced by HBV. HBsAg is produced from the surface reading frame. Hepatitis B e antigen (HBeAg) and hepatitis core antigen (HBcAg) are produced from the precore/core reading frame. HBsAg and HBeAg can be detected in the serum, whereas HBcAg is located inside the nucleus and not readily detected in the serum. The host may generate antibodies against each of these 3 antigens, namely, anti-HBs, anti-HBe, and anti-HBc, respectively.

With advances in antiviral therapy and improved understanding on the interaction between the host and the virus, the clinical meanings of different viral and host markers have changed dramatically since they were first discovered. This review summarizes the viral and host tests that are useful for clinicians to manage patients with chronic hepatitis B. Clinical use of various tests on the assessment of disease severity, prediction of prognosis, decision on the need of antiviral therapy, and monitoring of treatment response will be described. Recommendations on the clinical use of these tests by various regional HBV management guidelines will also be discussed.

NATURAL HISTORY OF CHRONIC HBV INFECTION

In Asia and Africa, vertical transmission has been the main source of HBV infection before the era of universal vaccination programs. The natural history of perinatally acquired chronic HBV infection is complex. Because infants and children have immature immune systems and have suppression of HBV-specific immune response due to in utero exposure to HBV antigens, they have a prolonged period of immune tolerance, characterized by high levels of HBV DNA and consistently normal alanine aminotransferase (ALT) levels (Table 1). In spite of individual variations, patients typically enter the immune clearance phase by the third to fifth decades of life. This phase is characterized by increased ALT level. The outcome of this phase is variable. Some patients can achieve HBeAg seroconversion and HBV DNA suppression and enter the inactive chronic HBsAg carrier phase, and the prognosis is usually good. Other patients may fail to control HBV and stay in the immune clearance phase for many years. This results in cumulative liver injury, progressive liver fibrosis, and development of cirrhosis. Moreover, 10%–20% of HBeAg-negative patients may still have significant viral load and increased ALT. This is often described as the reactivation phase.

In adults with horizontal transmission of HBV infection via sexual contact or other parenteral exposure, there is no immune tolerance phase. Because of potent immune response, 95% of patients can clear the virus altogether, but 5% still go on to develop chronic HBV infection.

It is, however, important to note that the concept of immune tolerance is not as simple as described above. HBV-specific T-cell function is preserved in young patients in the immune tolerance phase [4]. The mechanisms that lead to unchecked viral replication and minimal hepatic necroinflammation in such patients remain unclear. Thus, the terminologies for the phases of chronic HBV infection have been challenged. Some have proposed to describe the phases as HBeAg-positive inactive disease, HBeAg-positive active disease, HBeAg-negative low viremic phase, and HBeAg-negative high viremic phase instead. These neutral terms may more accurately describe the disease status. Nonetheless, before the nomenclature is agreed upon and finalized, to avoid confusion, this review will use the original terminologies that have been used for decades.

Furthermore, it is potentially misleading to define the immune tolerance phase just by age, HBeAg, HBV DNA, and ALT level. Some cases may actually be in the immune clearance phase and have elevated ALT in between clinic visits. In subsequent sections, we will further discuss the role of liver fibrosis assessment in this context.

ANTIVIRAL TREATMENT

Currently, 7 antiviral drugs in 2 broad classes have been registered for the treatment of chronic hepatitis B. Interferon and pegylated interferon (peginterferon) have both immunomodulatory and weak direct antiviral actions to control the virus. The 5 oral nucleos(t)ide analogues (NUCs) include lamivudine, adefovir dipivoxil, entecavir, telbivudine, and tenofovir disoproxil fumarate; these are direct inhibitors of HBV polymerase, which is the essential enzyme at the reverse transcription step of HBV replication.

The primary aim of antiviral treatment is to prevent cirrhosis, HCC, and liver-related mortality. Therefore, treatment should be given to patients at risk of these complications. As hepatic necroinflammation secondary to the host immune response against the virus is the main driver of disease progression, patients in the immune clearance phase and reactivation phase are the main targets of treatment (Tables 1 and 2). For patients with HBeAg-positive chronic hepatitis B and high ALT, there is a chance of spontaneous HBeAg seroconversion and subsequent entry into the inactive chronic HBsAg carrier phase; it is thus reasonable to observe for 3–6 months before embarking on antiviral treatment [5–7]. In contrast, HBeAg-negative patients in the reactivation phase seldom achieve spontaneous immune control; antiviral treatment can be started if the HBV DNA and ALT levels are high (Table 2).

Moreover, the presence of significant histological hepatic necroinflammation, significant fibrosis, cirrhosis, or even liver decompensation already signifies significant liver injury. The use of potent immunosuppressants and cancer chemotherapy also increases the risk of hepatitis flares. Patients in these categories should also be treated (Table 2).

VIRAL TESTS

HBsAg: From Qualitative to Quantitative

HBsAg, also known as the Australian antigen, is the hallmark of HBV infection. Positive HBsAg indicates presence of HBV infection; persistence of HBsAg for >6 months indicates chronic infection. Among patients with chronic HBV infection, seroclearance of HBsAg (ie, loss of HBsAg with or without appearance of antibodies [anti-HBs]) is regarded as a functional cure. Since the discovery of HBsAg in 1965, physicians have been interested only in its positivity or negativity, although standard assays for quantification of HBsAg have been available for a long time. Enthusiasm for the use of quantitative HBsAg started when studies demonstrated the correlation between lower levels of HBsAg and lower covalently closed circular DNA (cccDNA) content in the liver, which appears to be associated with an increased likelihood of HBsAg seroclearance and reduced risk of HCC. The correlation of HBsAg with cccDNA is good in HBeAg-positive patients, but the correlation in HBeAg-negative patients is controversial [8–10].

There are currently 2 commercially available quantitative HBsAg assays. Both are 2-step immunoassays based on chemiluminescent microparticles. The ARCHITECT HBsAg assay (Abbott Diagnostics, Lake Forest, Illinois) has a linear range of 0.05–250 IU/mL. For samples with higher HBsAg levels, manual dilution is required. The Elecsys HBsAg II assay (Roche Diagnostics, Risch-Rotkreuz, Switzerland) has an automatic on-board 1:400 dilution with a range of measurements of 20–52 000 IU/mL. For samples with HBsAg levels <20 IU/mL, the samples should be retested undiluted, resulting in a lower limit of detection of 0.05 IU/mL.

In the natural history of chronic hepatitis B, the level of HBsAg decreases with immune clearance of HBV; it is highest in the immune tolerance phase, decreases with HBeAg seroconversion during the immune clearance phase, and is lowest at the inactive HBsAg carrier phase (Table 1) [11–13]. The temporal change in HBsAg level is much slower than that of HBV DNA [11]. HBsAg level has only borderline predictive value for HCC among HBeAg-positive patients or patients having high viral load (>2000 IU/mL) [14]. In HBeAg-negative patients, a low HBsAg level, particularly among patients with low HBV DNA, can predict subsequent inactive disease. In white patients, HBsAg <1000 IU/mL accompanied with HBV DNA <2000 IU/mL in HBeAg-negative patients can predict inactive hepatitis, at least in the subsequent 1 year of follow-up [15]. In a cohort of 1068 HBeAg-negative Taiwanese patients with HBV DNA <2000 IU/mL followed up for >14 years, serum HBsAg <1000 IU/mL was associated with a significantly lower risk of HCC (58.2 per 100 000 person-years) compared with those with serum HBsAg ≥1000 IU/mL (326.1 per 100 000 person-years) [14]. In Asian HBeAg-negative patients, HBsAg <100 IU/mL is predictive of spontaneous HBsAg seroclearance in 6–8 years [16, 17]. In the Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer in HBV (REVEAL-HBV) cohort in Taiwan, integration of HBsAg <100 IU/mL into a model with age, sex, ALT level, HBV DNA, and HBV genotype can predict the risk of HCC at 5, 10, and 15 years [18].

As decrease in serum HBsAg level can reflect the clearance of cccDNA in the liver, the change in HBsAg level has been found useful to predict and monitor response to peginterferon therapy [19]. In HBeAg-positive patients, a low baseline HBsAg level has modest association with HBeAg seroconversion after peginterferon treatment, but its clinical use is limited [20]. In contrast, patients who respond to peginterferon usually have a significant decline of HBsAg to a low level at weeks 12 and 24. In a combined analysis of 803 HBeAg-positive patients treated with peginterferon in 3 global studies, patients who had HBsAg >20 000 IU/mL at week 24 had a 96% negative predictive value in genotype A HBV and 100% negative predictive values in genotype B, C, and D HBV infection for a response, which was defined as HBeAg loss and HBV DNA <2000 IU/mL 6 months after stopping treatment [21]. In HBeAg-negative patients, the best validated prediction model is a combination of HBV DNA and HBsAg decline at week 12; the negative predictive value of a response (HBV DNA <2000 IU/mL and normal ALT 24 weeks after treatment) for patients who fail to achieve a 2-log decline in HBV DNA and any decline in HBsAg is 95%–100% [22]. This model is best studied in genotype D HBV–infected patients. As the on-treatment HBsAg kinetics in HBeAg-negative patients vary in different HBV genotypes, the use of HBsAg to predict response in non–genotype D HBV infections may require further validation (Table 3) [23].

In contrast to peginterferon treatment, NUC treatment is associated with a very slow decline in HBsAg despite strong HBV DNA suppression, indicating weak immune clearance of the virus [24, 25]. In HBeAg-positive patients, those who have HBeAg loss tend to have a faster decline in HBsAg level, but the value of serum HBsAg to determine treatment cessation is uncertain [26, 27]. In HBeAg-negative patients, small studies in Hong Kong and Taiwan suggested that serum HBsAg <100–200 IU/mL can predict sustained response and HBsAg seroclearance after stopping lamivudine, but these results need to be validated in bigger studies and other patient populations (Table 4) [28, 29]. Today, clearance of HBsAg is still the recommended marker of when stopping NUCs should be considered [5–7].

HBeAg: A Test Through History

HBeAg has long been regarded as a marker of viral replication and infectivity since its discovery in the early 1970s. With the availability of HBV DNA testing, infectivity is much better reflected by the level of HBV DNA. Nowadays, HBeAg status is primarily used to define a patient's phase in the natural history of chronic HBV infection (Table 1) [1]. HBeAg-positive patients are either in the immune tolerance phase or early immune clearance phase, which is associated with high HBV DNA levels. HBeAg seroconversion marks an important step of immune clearance. The majority of patients become inactive chronic HBsAg carriers after HBeAg seroconversion, with low HBV DNA and normal ALT levels. Nonetheless, some HBeAg-negative patients can have active hepatitis and high HBV DNA, often described as the reactivation phase. Most HBeAg-negative patients with active viremia have immune escape mutations at the basal core promoter and/or the precore stop codon that can be detected with HBV genotype testing [30].

As patients at different stages of chronic HBV infection have different clinical characteristics and disease progression, treatment guidelines make recommendations according to the HBeAg status (Table 2) [5–7]. HBeAg-positive patients who are young (<30 years old) and have normal ALT should not be treated even if the HBV DNA is very high, as they are probably in the immune tolerance phase and liver injury is often minimal. On the other hand, HBeAg-negative patients who have high HBV DNA (2000–20 000 IU/mL) should be considered for liver fibrosis assessment even if ALT is normal, as they may be in the reactivation phase and liver cirrhosis is not uncommon. It is important to note that chronic hepatitis B has a complex natural history. Patients may move from one phase to the next over time, and apparently inactive disease may reactivate spontaneously or secondary to immunosuppression [31, 32]; a one-off assessment may therefore be deceiving. All regional guidelines recommend regular monitoring for patients not yet fulfilling treatment criteria [5–7]. For HBeAg-positive patients with normal ALT, the ALT level should be checked once every 3–6 months, and HBeAg every 6–12 months. For HBeAg-negative patients with persistently normal ALT levels, ALT and HBV DNA can be checked once every 6–12 months.

Even with regular monitoring, it is possible that patients with apparently inactive disease have elevated ALT in between visits and are therefore accumulating liver injury. Liver biopsy is therefore recommended in patients with significant viral load (HBV DNA ≥20 000 IU/mL in HBeAg-positive disease and ≥2000 IU/mL in HBeAg-negative disease), particularly in those aged >40 years with mild ALT elevation [5–7]. Nonetheless, liver biopsy is invasive and poorly accepted by patients. In recent years, a number of noninvasive tests of liver fibrosis have been developed and validated against liver histology [33, 34]. They are noninferior to liver histology in predicting future HCC development and mortality [35, 36]. As a result, current European and Latin American guidelines endorse the use of transient elastography and serum tests as initial assessment of liver fibrosis in patients with chronic hepatitis B [37]. In general, transient elastography is more accurate than serum tests based on routine clinical parameters. On the other hand, serum tests such as the aspartate aminotransferase–to-platelet ratio index and Fibrosis-4 index are inexpensive and can be applied to virtually all patients [38]. Liver biopsy can be reserved for cases with indeterminate results.

HBeAg seroconversion is an important marker for treatment response in HBeAg-positive patients (Table 3). A rapid on-treatment decline in HBeAg level can predict HBeAg seroconversion on peginterferon, but there is no commercial assay for quantitative HBeAg on the market [39]. Among interferon-treated patients, HBeAg seroconversion is associated with improved long-term clinical outcome [40]. However, a significant proportion of patients who achieve HBeAg seroconversion on interferon therapy may still have active viremia on long-term follow-up (ie, HBeAg-negative chronic hepatitis B) [41]. Among NUC-treated patients, treatment cessation can be considered after HBeAg seroconversion for ≥12 months; approximately 40%–80% of patients will have durable off-treatment response (Table 4) [6, 7].

Anti-HBc: A Marker of Occult Infection

Anti-HBc is available as commercial enzyme immunoassays for total anti-HBc (which detects both immunoglobulins M [IgM] and G [IgG]) and anti-HBc IgM, which is the hallmark of acute hepatitis B and may be the only clue to the diagnosis during the window period when HBsAg has already turned negative. It is also positive in some cases of severe acute exacerbation of chronic hepatitis B [42]. In contrast, total anti-HBc is a marker of previous or current HBV infection. In patients with negative HBsAg, positive anti-HBc suggests previous HBV infection and the possibility of occult HBV infection (OBI).

At the Taormina expert meeting, OBI was defined as the presence of HBV DNA in the liver (with or without detectable HBV DNA in the serum) of individuals testing HBsAg negative by currently available assays [43]. OBI is further divided into seropositive OBI (anti-HBc and/or anti-HBs positive) and seronegative OBI (anti-HBc and anti-HBs negative). Even when detectable, the serum HBV DNA level is usually <200 IU/mL. A patient with negative HBsAg but high serum HBV DNA is considered to have “false” OBI, which is usually due to mutations in the S gene (escape mutants) [44]. Therefore, although routine measurement of HBV DNA would unlikely alter the management in the majority of cases, it should be checked in patients with unexplained hepatitis activity or cryptogenic cirrhosis, or in patients at risk of reactivation of OBI (eg, those receiving potent immunosuppressants or cancer chemotherapy).

In real-life practice, because liver biopsy is almost never performed in patients with negative HBsAg and many patients with OBI have low or undetectable HBV DNA, positive total anti-HBc is the most commonly used marker for possible OBI. OBI is important in a number of clinical situations. Positive HBV DNA is detected in up to 15% of blood donors with positive anti-HBc [45]. Transmission of HBV by blood and blood-component transfusion is rare but possible. The transmission risk is reduced by 5-fold if the donor has positive anti-HBs [46]. Likewise, OBI may be transmitted to other patients and staff in hemodialysis centers and during liver transplantation [47, 48]. OBI has been reported in patients with cryptogenic HCC. Nevertheless, as substantial liver injury may have occurred before HBsAg seroclearance, it is unclear if OBI contributes further to disease progression in this situation [49].

OBI may also reactivate when a patient receives potent immunosuppression and cancer chemotherapy. B-cell depletion therapy such as rituximab, which is used in the treatment of lymphoma and rheumatological conditions, is commonly associated with reactivation of OBI [50]. In such situations, prophylactic antiviral therapy or careful monitoring of HBV DNA, followed by prompt antiviral therapy in case of reactivation, can effectively prevent hepatitic activity and liver failure [51]. In contrast, the risk of reactivation is low when less-potent immunosuppressants are used [52].

Nucleic Acid Testing: The Treatment Target

The development of polymerase chain reaction–based serum HBV DNA assays has revolutionized diagnostics for HBV. Current commercial assays typically have a lower limit of quantification of ≤20 IU/mL. Because HBV DNA is an indicator of viral replication and the primary target of antiviral therapy, current guidelines support antiviral therapy in patients with elevated ALT and an HBV DNA level of 2000–20 000 IU/mL, and in those with cirrhosis and detectable HBV DNA (with slight difference in cutoffs across guidelines) (Table 2) [5–7]. Such recommendations were based on observational studies showing a strong correlation between high HBV DNA and future risk of cirrhosis and HCC [53, 54].

Patients who have lower baseline HBV DNA levels tend to respond better to peginterferon therapy [55, 56]. Among HBeAg-positive patients, HBV DNA <9 log copies/mL, ALT >2 times the upper limit of normal (ULN), and genotype A HBV infection predict favorable response to peginterferon treatment [55]. HBeAg-positive patients who respond to peginterferon tend to have a faster drop in HBV DNA than nonresponders, but the clinical use of on-treatment HBV DNA monitoring is superseded by quantitative HBsAg [20]. In HBeAg-negative patients, failure of HBV DNA decline of >2 log and HBsAg decline at week 12 is used as a stopping rule for peginterferon treatment (Table 3) [22].

HBV DNA is one of the most important tests to monitor and determine treatment response in patients on NUC treatment (Table 4). When drugs with a low genetic barrier to resistance are used, failure of suppression of HBV DNA to undetectable levels at month 6 (for lamivudine and telbivudine) or month 12 (for adefovir) is associated with increased risk of drug resistance, and a change of antiviral therapy to a more potent agent without cross-resistance is recommended [5]. Suppression of HBV DNA to undetectable levels is associated with reduced cirrhotic complications and HCC [57, 58]. Virological breakthrough, defined as a 10-fold increase in HBV DNA during NUC treatment, is the first signal of drug resistance or poor drug adherence. Although the Asian-Pacific guideline recommends stopping NUCs among HBeAg-negative patients who have undetectable HBV DNA for at least 3 times at 6 months apart [6], 43%–91% of patients developed virological relapse (HBV DNA >2000 IU/mL) and 45% developed clinical relapse (HBV DNA >2000 IU/mL and ALT >2 times the ULN) within 1 year after stopping treatment [29, 59, 60]. It is therefore better to continue NUC treatment until HBsAg seroclearance in HBeAg-negative patients [5, 7]. The optimal duration of consolidation treatment after HBsAg seroclearance is currently unclear.

Because HBV DNA is completely suppressed in the majority of patients on NUC therapy, this marker can no longer reflect the amount of virus in the liver or the intrahepatic replicative activity. Besides serum HBsAg quantification, other groups have explored the role of HBV RNA as a novel biomarker [61]. Because messenger RNA is unstable in the serum, the majority of HBV RNA detected in the serum is believed to be pregenomic RNA. Early studies suggest that the HBV RNA level may be used to predict HBeAg seroconversion in patients on NUC treatment [62]. The clinical use of this biomarker deserves further evaluation.

HOST TESTS

ALT is one of the most common tests used in patients with chronic liver disease. In patients with chronic hepatitis B, ALT level correlates with hepatic necroinflammation. Traditionally, the ULN of ALT was set at 40–70 IU/L based on the reference levels of apparently healthy subjects; however, clinical evidence suggests that such levels may be too high. This is because apparently healthy subjects may harbor conditions that increase the ALT level. For instance, nonalcoholic fatty liver disease affects 15%–40% of the general population [63]. If not screened carefully, the reference cohort can even include patients with drug-induced liver injury, alcoholic liver disease, and viral hepatitis. When the risk factors of liver diseases are excluded, it appears that the optimal ALT cutoffs should be lowered to 30 IU/L for men and 19 IU/L for women [64]. Along the same line, high-normal ALT levels are also associated with cirrhosis and liver-related mortality [65, 66]. Therefore, the current American Association for the Study of Liver Diseases guidelines have already adopted the new ALT cutoffs [7].

As viral persistence and clinical outcomes following chronic HBV infection depend on various host factors that determine a host's immune mechanisms, identifying host factors that interfere with the HBV replication cycle could provide additional therapeutic targets to reduce the antigenic and viral loads [67]. Human zinc finger antiviral protein [68] and the APOBEC3 family of proteins [69] have been studied for their roles in suppressing HBV DNA replication, but these markers are not yet ready for clinical use until more human data are available.

The interleukin 28B (IL28B), recently renamed as interferon lambda 3 (IFN-λ3), polymorphism in the human leukocyte antigen (HLA) locus is the most widely studied host factor predicting virologic response to peginterferon in patients with chronic hepatitis B [70]. A recent systematic review with meta-analysis of 4 genome-wide association studies demonstrated an association between the IFN-λ3 polymorphism and spontaneous resolution of HBV infection in Asian but not white patients [71]. Nonetheless, the preliminary observations of an association between IFN-λ3 and virological and serological responses to interferon in both HBeAg-positive and HBeAg-negative patients were not consistently reproducible [71].

Apart from the IFN-λ3 polymorphism, other HLA loci have also been studied. The HLA-DPA1 and HLA-DPB1 polymorphisms, together with the IFN-λ3 polymorphism and other viral factors (eg, precore stop codon/basal core promoter mutations), were studied in HBeAg-positive patients; HLA-DPA1 was found to be associated with HBeAg seroconversion in addition to the viral factors [72].

Interferon-inducible protein 10 (IP-10), also named chemokine ligand 10 (CXCL10), has recently been found to play an important role in the treatment response in patients with chronic hepatitis B [73, 74]. IP-10 is a chemotactic CXC chemokine that is secreted by hepatocytes and hepatic sinusoidal endothelium in patients with hepatitis [75]. It has been shown that the G-201A allele in the promoter region of the IP-10 gene was associated with liver disease progression in Chinese Han chronic hepatitis B patients through upregulation of IP-10 expression [76]. The role of the IP-10 polymorphism in patients of other ethnicities is yet to be ascertained.

CONCLUSIONS

HBsAg and HBeAg are the traditional serological tests in the assessment of chronic hepatitis B. The availability of sensitive HBV DNA assays allows a thorough understanding of the virology and natural history of chronic HBV infection. Discovery of the association between serum HBV DNA and risk of HCC is a major breakthrough, providing clearer guidance on the indications for antiviral treatment. Quantitative HBsAg, which reflects the amount and transcriptional activities of cccDNA in the liver, revitalizes the use of this traditional test in the modern era. Although the spectrum of viral tests for chronic hepatitis B is getting richer, the development of host tests is still in its infancy. Viral tests should be used in the context of clinical assessment, particularly assessment of liver fibrosis, in the management of patients with chronic hepatitis B.

Notes

Supplement sponsorship. This article appears as part of the supplement “Hepatitis B,” sponsored by the CDC Foundation and Gilead.

Potential conflicts of interest. G. L.-H. W. has served as an advisory committee member for Otsuka and Gilead, and has also served as a speaker for AbbVie, Bristol-Myers Squibb, Echosens, Furui, Gilead, Janssen, and Otsuka. V. W.-S. W. has served as an advisory committee member for AbbVie, Roche, Novartis, Gilead, and Otsuka; has served as a consultant for Merck and NovaMedica; and has served as a speaker for AbbVie, Roche, Novartis, Abbott Diagnostics, and Echosens. H. L.-Y. C. is a consultant for Bristol-Myers Squibb, Gilead, Merck, Novartis, and Roche; has received honoraria for lectures for Abbott, AbbVie, Bristol-Myers Squibb, Echosens, Gilead, GlaxoSmithKline, Merck, Novartis, and Roche; and has received an unrestricted grant from Roche for hepatitis B research. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References