Natural History, Therapy and Pathogenesis of Chronic Viral Hepatitis B

National Institute Of Diabetes And Digestive And Kidney Diseases

Investigators

Linked publications, trials & patents

Abstract

Summary: Globally there are an estimated 254 million persons infected with hepatitis B virus (HBV). In the United States, there are between 0.8 and 2 million individuals with chronic HBV infection. Chronic HBV infection is a major cause of chronic liver disease, cirrhosis and hepatocellular carcinoma (HCC) worldwide. The natural history of chronic hepatitis B (CHB) appears to be changing with an increasing prevalence of HBeAg negative CHB and recognition of a group of patients with moderate levels of HBV DNA and ALT levels with undetermined natural history. Knowledge of the rate of disease progression among individuals with moderate levels of HBV DNA and ALT levels is unknown. Despite the availability of safe and effective oral nucleoside analogues for treatment of CHB, therapy remains problematic due to the need for prolonged therapy and limited effectiveness and tolerability of the alternate treatment, interferon. Clearance of hepatitis B surface antigen (HBsAg) is the desired surrogate endpoint of therapy but is rarely achieved with current therapy. Identifying the optimal regimen, defining when to treat, for how long and when to stop therapy to achieve this endpoint are major unresolved issues. In addition, defining the best parameters to monitor patients both on and off therapy are not clear. Hypotheses/problems addressed: 1) Define the host, viral and environmental factors that determine the natural history and outcome of HBV infection. Two forms of CHB with markedly different outcomes are recognized based on hepatitis B e antigen (HBeAg) status – either HBeAg positive or negative. A substantial proportion of patients in the HBeAg-negative phase of CHB cannot be phenotyped as either immune active (IA) or inactive carrier (IC) and are classified as “indeterminant” (IND). Their clinical outcome is unclear. We conducted an analysis of all HBeAg negative patients followed at the NIH Clinical Center liver clinic to assess outcomes of HBeAg negative IND CHB with the goal of identifying factors associated with a good outcome (maintenance of HBV DNA <2000 IU/mL and HBsAg loss). We showed that a majority of HBeAg negative patients had an IND phenotype (75%). No predictors of sustained HBV DNA<2,000 were identified, and HBV genotype A predicted HBsAg loss. Despite frequent HBV DNA and ALT fluctuations resulting in multiple phenotype transitions, clinical outcomes and need for treatment were low, and short-term (7 year) outcomes were excellent. We performed a similar analysis of adults and children with an HBeAg negative IND phenotype enrolled in the Hepatitis B Research Network to evaluate phenotype transitions, rates of HBsAg loss, treatment initiation and clinical outcome of the IND phenotype. We utilized multi-state Markov transition models to estimate percentages of patients transitioning to IA, IC, and IND groups within a 1- and 5-year period. IND participants with an HBV DNA >2000 IU/mL had nearly double the rate for experiencing a phenotype transition compared to participants with an HBV DNA <2000 IU/mL, 55.2 versus 31.8 per 100PYs, and initiating treatment, 7.6 versus 4.2 per 100PYs, respectively. Notably, HBsAg loss occurred only among participants with an HBV DNA <2000 IU/mL. The rate of clinical outcomes was 0.44 PYs. Our findings highlight the generally good short-term prognosis for persons with HBeAg negative IND CHB and contrast sharply with studies from Asia. This is important information for counselling patients of expected outcome. 2) Develop and evaluate novel, safer and more effective therapies for chronic viral hepatitis. Current therapy for CHB remains less than optimal. Only two classes of drugs are approved for use: nucleos(t)ide analogues and peginterferon. Relapse is common if nucleos(t)ides are discontinued in the absence of HBsAg loss. Consequently, they must often be administered long-term or indefinitely. However, long-term use is associated with increased risk of side effects and higher costs. Therefore, the focus of current studies is to develop strategies to induce HBsAg loss (functional cure) to permit discontinuation of therapy and improve outcome of the infection. CHB is characterized by immune exhaustion with upregulation of inhibitory receptors on immune cells. We postulate that the immune exhaustion is driven by high levels of circulating HBsAg levels which dampen the host immune response. Therefore, reducing viral antigen burden might be a promising strategy. We are using an siRNA that targets the HBsAg gene to reduce HBsAg levels. However, because the siRNA does not target the viral covalently closed circular DNA (cccDNA), the template for the viral transcripts, it is predicted that HBsAg levels would rebound after the siRNA is stopped. Combining the siRNA with peginterferon, which targets cccDNA as well as inducing an immune response, may result in a more sustained effect or even cure. We have initiated a clinical trial to test a strategy of lead-in siRNA followed by combination siRNA plus peginterferon followed by peginterferon alone (Protocol IRB001606). The primary endpoint will be reduction in HBsAg levels and/or loss of HBsAg. Secondary analyses will investigate the changes in virological profiles including cccDNA using fine needle aspirates and digital droplet PCR to quantify cccDNA (in collaboration with Dr. Tu, Sydney, Australia) and to investigate knockdown of HBsAg levels on immunological profiles (in collaboration with Dr. Rehermann, LDB) during siRNA administration. 3) Elucidate the mechanisms of action of therapy and assess novel markers of HBV disease activity. Two therapies, nucleos(t)ide analogues (NAs) and peginterferon, are approved for treatment of CHB. There has been renewed interest in using peginterferon in combination with novel therapies in development for CHB. While it is known that peginterferon has both antiviral and immunomodulatory actions, its mechanism of action remains uncertain and in-vivo human data is limited. High viral load and HBsAg levels were shown to be associated with a lower response to peginterferon. Therefore, we postulated that in patients who have achieved long-term viral suppression on NAs, the addition of peginterferon may result in a higher rate of HBsAg seroconversion (the endpoint of treatment). We conducted a clinical trial to test this hypothesis and investigate the mechanisms of action of peginterferon. The trial design was an open label, single arm study in which peginterferon was added for 24 weeks to ongoing NA therapy in patients with viral suppression. Paired liver biopsies pre- and 6 hours post-peginterferon were obtained. Selecting a cohort with viral suppression allowed us to investigate if intrahepatic HBsAg affected the host response to peginterferon. Notably, within 6 hours of administration, peginterferon resulted in robust induction of interferon stimulated genes (ISGs) both in patients with and without a reduction in HBsAg levels. However, peginterferon was ineffective at inducing HBsAg loss. One subject experienced transient HBsAg loss and 3 (including the 1 with transient HBsAg loss) achieved a >0.5 log reduction in HBsAg levels, all with HBV genotype A. Using digital spatial profiling, we analyzed the transcriptome of HBsAg positive and negative regions of interest, identified by immunostaining. We found that induction of ISGs was not inhibited by the presence of intrahepatic HBsAg. Rather, response to peginterferon was associated with a greater and more varied ISG profile suggesting the type of ISG induced may an important factor in response. Moreover, through digital spatial proteomic and peripheral cytokine/chemokine analyses we showed that induction of a hepatic and peripheral immune response characterized by influx of inflammatory cell (macrophages, monocytes, NK cell and activated T-cells) contributed to the action of peginterferon in-vivo. The findings suggest that the NK-NKT cell response is important for the action of peginterferon and may play a greater role in mediating HBsAg clearance than the host antiviral response. Our novel findings provide important mechanistic information on peginterferon that may drive rational design of novel immunotherapeutic strategies and to permit selection of patients who benefit from peginterferon-based therapy. Functional cure (hepatitis B surface antigen (HBsAg) loss and HBV DNA negative) 6 months after stopping treatment is the primary therapeutic endpoint of chronic hepatitis B (CHB). However, HBsAg can be derived from both cccDNA (the template for transcription and replication) and integrated HBV DNA (iDNA). Therefore, there may be persons who fail to achieve functional cure because HBsAg is derived predominantly from iDNA. Such persons may have a partial cure (PC i.e. HBsAg positive but HBV DNA negative). We hypothesized that HBV RNA, a marker of cccDNA activity, may be used to identify such subjects. We examined virological profiles of subjects in different phases of HBeAg(-)CHB with a primary aim of differentiating patients with PC from inactive carriers (IC) and comparing patients with PC to ones with HBsAg loss. We studied 7 groups of patients with HBeAg(-)CHB based on HBV DNA and ALT levels at the time of sera collection: 1) HBsAg loss, 2) partial cure (PC) 3) inactive carrier (IC) (HBV DNA <2000 IU/mL and normal ALT), 4) grey zone 1 (GZ1) (HBV DNA <2000 IU/mL with elevated ALT), 5) GZ2 (HBV DNA >2000 IU/mL with normal ALT), 6) GZ3 (HBV DNA >2000 IU/mL with ALT >ULN-< 2X ULN), and 7) immune active (IA) (HBV DNA >2000 and ALT >2xULN). The proportion testing HBV RNA negative was 73% in the HBsAg loss group, 43% in PC, 17% in IC, 5% in GZ 1, 0% in GZ 2, 10% in GZ 3 and 0% in IA. Interestingly, 4/30 (13%) HBsAg loss subjects tested HBsAg positive by qHBsAg assay and 10/30 were positive using the ultrasensitive HBsAg Next assay. The proportion with qHBsAg levels <100 IU/mL was 97% in the HBsAg loss group, 43% in PC, 30% in IC, 15% in GZ1, 5% in GZ 2, 15% in GZ 3, and 0% in IA groups. A virological profile of HBV RNA negative and qHBsAg<100 was observed in 33% of PC, 10% of IC, 5% of GZ1, 0% of GZ 2 and 5% of GZ3 compared to the 73% HBsAg group and could differentiate the PC group from the IA groups. Therefore, testing negative for HBV DNA and HBV RNA together with a qHBsAg level <100 IU/mL is useful for differentiating PC from IA. Our results demonstrated that up to one third of untreated subjects in the PC phase of CHB have virological profiles similar to those with HBsAg loss and may be functionally cured.