Correspondence Prof. Dr. Heiner Wedemeyer, Department of Gastroenterology, Hepatology and Endocrinology, Medizinische Hochschule Hannover, Carl-Neuberg Str. 1, 30625 Hannover, Germany Tel: +49 511 532 6814 Fax: +49 511 532 8662 e-mail: Wedemeyer.Heiner@mh-hannover.de
More than 30 years after Mario Rizzetto and colleagues described a new antigen in livers of HBsAg-positive patients called the “delta antigen”, a re-emerging interest in hepatitis delta is currently observed. The state-of-the art on basic and clinical research on hepatitis delta was presented during a monothematic conference organized by the European Association for the Study of the Liver (EASL) in September 2010. Hepatitis delta is caused by infection with the hepatitis D virus (HDV) which requires presence of HBsAg for complete replication and transmission. Recent data confirmed the severe long-term course of HDV infection with high rates of hepatic decompensation while controversial data on the risk for the development of hepatocellular carcinoma were reported. Pegylated interferon alpha can lead to sustained HDV RNA elimination in about one quarter of patients while HBV polymerase inhibitors are ineffective against HDV. Novel treatment options include prenylation inhibitors and HBV entry inhibitors which are currently in early clinical development.
Mario Rizzetto and colleagues identified a new antigen in the livers of patients infected with the hepatitis B virus (HBV) in 1977 (1, 2). Antibodies against the delta-antigen were found in patients with a severe course of liver disease (3). In the following years, the hepatitis D virus (HDV) was identified as a cause of hepatitis in chimpanzees and humans if HBV was also present (4). Hepatitis D can therefore only occur in individuals who are also infected with HBV as HDV uses theHBsAg as its envelope protein.
Hepatitis D virus RNA is a small genome consisting of 1700 nucleotides (5). Two hepatitis delta antigens (HDAg) exist: the small HDAg (24 kDa) and the large HDAg (27 kDa), which is 214 amino acids long. The small HDAg accelerates genome synthesis, while the large HDAg inhibits HDV RNA synthesis but is necessary for virion morphogenesis (5). Eight HDV genotypes have been described (6), including genotype 1, which is the most frequent genotype throughout the world, especially in Europe, the Middle East, North America and North Africa. In contrast, genotype 2 is mainly prevalent in the Far East and genotype 3 is only found in the Amazonian area of South America. Distinct clinical courses are associated with different HDV genotypes, with more severe disease in genotypes 1 and 3 and milder disease activity in HDV genotype 2 (7).
Several guidelines suggest that all HBsAg-positive patients should be tested for anti-HDV antibodies (8, 9). Testing for anti-HDV antobodies shows a high specificity and evidence suggests that non-HDV genotype 1 infections are identified in currently available assays (Fig. 1). At the same time, anti-HDV antibodies may not be identified after recovery from infection. Hepatitis D should be confirmed by the detection of HDV RNA. If HDV RNA is positive, evaluation of grading and staging of liver disease, monitoring for hepatocellular carcinoma (HCC) and consideration of antiviral treatment is indicated. Although HDV RNA quantification is offered by some laboratories, standardization is poor and few assays have been evaluated carefully (10). There is no evidence that the HDV RNA levels correlate with any clinical marker of activity or stage of liver disease (11) Thus, HDV RNA quantification is only useful if an antiviral treatment is indicated. Anti-HDV-IgM testing may be useful in patients who test HDV RNA negative but have evidence of unexplained liver disease.
Re-emerging epidemiology of hepatitis D?
15–20 million individuals are estimated to be anti-HDV positive worldwide (12). High prevalences of anti-HDV have been described in HBsAg-positive patients in Mediterranean countries, the Middle East, Central Africa and northern parts of South America (6). In contrast, HDV seemed to be restricted to HBsAg-positive intravenous drug addicts in Western countries (13–15). Although the incidence of HDV infections has decreased significantly in various European countries (14), chronic hepatitis D is still a significant health burden in Europe – in particular because of the immigration of individuals from highly endemic areas (13, 16). In our experience, about 8–10% of HBsAg-positive patients test positive for anti-HDV and more than three quarters of our hepatitis D patients were not born in Germany. However, the geographical origin of our patients has changed during the last decade. Until the mid 1990s, most HDV-positive patients were born in Turkey, while the proportion of Eastern European patients and those born in the Soviet Union have increased significantly in recent years (13). HDV infection is also increasingly prevalent in South London. Also, 8.5% of almost 1000 consecutive patients with chronic hepatitis B tested positive for anti-HDV at King's College Hospital between 2000 and 2006. In that study, HDV-infected patients were born mainly in Africa or Eastern Europe (17). Like in Germany and England, among immigrant populations, there was a high anti-HDV prevalence in HBsAg-positive samples tested in France at the ‘Laboratoire Associe au Centre National de Reference de Hepatitis B, C’ (18). However, acute HDV infections may still occur in non-immigrant populations. The main risk factors in Italy have been attributed to promiscuous sexual activity, beauty treatment and injection drug use (19). A recent study in 2006 also showed a high prevalence of hepatitis D in a US-IV Drug cohort, with 50% of HBsAg-positive anti-HDV-positive patients (15).
Immunopathogenesis of hepatitis D
Hepatitis B virus and hepatitis C virus (HCV)-specific immune responses have been investigated in numerous studies and associated with the control of acute and persistent infections. In contrast, knowledge of the pathogenesis of HDV-induced liver disease is limited. HDV genotype 1 is not cytopathic and thus immune responses probably play an important role in the pathogenesis of HDV-induced liver disease. In a study from Italy, patients with inactive but not active-HDV disease displayed HDV-specific proliferative CD4 responses to HDAg, suggesting that cellular immune responses probably controlled HDV infection (20). Preliminary data suggest that the endogenous interferon (IFN) system is highly activated in hepatitis D. This is supported by results showing that perforin-expressing lymphoid cells including CD4+ T cells are more frequent in hepatitis D patients than in HBV or HCV monoinfection (21). However, antigen-specific T cell responses were weak in persistent HDV/HBV infection (22). Peripheral HDV-specific T cell responses declined during IFN-based antiviral therapy and the quality of T cell responses seemed to correlate with the treatment response. Moreover, HDV coinfection may also alter cellular immune responses against HBV, as HBV-specific T cell responses were more frequent in HDV-coinfected than that in HBV-monoinfected individuals. This suggests that the immune network of innate and adaptive immune responses is complex in hepatitis D and both HBV- and HDV-specific responses must be considered. Recently, Negro and colleagues have shown that HDV interferes with IFNα signalling by blocking Tyk2 activation. Thus, HDV impairs activation and translocation to the nucleus of signal transducers and activators of transcription (STAT)1 and STAT2 (23).
Up to one-third of European patients with HDV are co-infected with the hepatitis C virus. HDV not only suppresses HBV replication but also HCV replication in triple infected patients (24) and chronic hepatitis C may even be cleared in patients superinfected with HBV and HDV (25). In our experience, less than one-fifth of anti-HCV/HBsAg/anti-HDV-positive individuals are positive for HCV RNA (26). Nevertheless, it is not clear whether anti-HCV-positive/HCV RNA-negative patients have truly recovered from HCV infection or whether HCV replication is suppressed because of viral co-infections. It is therefore interesting to note that viral dominance changes over time (27). This suggests that triple infected patients should be followed closely and – if indicated – the treatment of the dominant virus should be considered.
Clinical course of hepatitis D
Acute hepatitis B virus/hepatitis D virus coinfection
Acute HBV/HDV co-infection leads to recovery in more than 90% of patients but may cause severe acute hepatitis with a risk of a fulminant course (28). In contrast, HDV is cleared spontaneously in a minority of chronic HBsAg patients with HDV superinfection. The histopathology of simultaneous HBV and HDV infection has been shown to be more severe than in infection with HBV alone in the experimental infection of chimpanzees (29). Several outbreaks of very severe courses of acute hepatitis D in patients have been described in different regions of the world (30–32). However, in the last two decades, acute hepatitis D has become infrequent in Western countries because of the introduction of vaccination programmes.
Natural history of chronic hepatitis D
Several studies have shown that chronic HDV infection leads to more severe liver disease than chronic HBV mono-infection, with an accelerated course of progression to fibrosis, increased risk of HCC and early decompensation of cirrhosis (28, 33–37). HDV accounts for almost half of all cases of cirrhosis and HCC in South-East Turkey (34, 38, 39). Recent studies from Italy have shown that up to 25% of patients with cirrhosis developed HCC and that liver failure was the cause of death in about 60% of patients (40, 41). These are similar to results in Taiwan, where the cumulative survival of HDV-genotype 1-infected patients was low, with only 50% of patients surviving after 15 years (7). HDV infection has also been associated with a higher risk of cirrhosis in human immunodeficiency virus (HIV)-co-infected patients, with 66% of HIV/HBV/HCV/HDV-infected patients but only 6% of HBV/HCV/HIV-infected patients with cirrhosis in a Spanish corhort (42). Similarly, hepatitis D was associated with poorer survival in HIV-infected patients in Taiwan (43).
Treatment of chronic hepatitis D
Nucleoside and nucleotide analogues used for the treatment of HBV infection are ineffective against HDV. This has been shown for famciclovir (44), lamivudine (45) and adefovir (46). Also, ribavirin alone or in combination with IFN did not lead to increased rates of HDV RNA clearance (47–49). Beneficial effects in hepatitis D from the reduction of HBsAg levels in the long term with HBV polymerase treatment such as tenofovir (50) must be confirmed in larger studies. Although clevudine has been shown to inhibit D virus viraemia in woodchucks (51), recent data from Ankara suggest that this drug is not effective in patients with hepatitis D (EASL monothematic conference September 2010).
Several studies have investigated IFNα for the treatment of hepatitis D since the mid 1980s (2). However, data are difficult to compare because the endpoints were different and few studies have studied HDV-RNA levels over time. One randomized Italian study with high doses of IFNα showed a beneficial long-term outcome in hepatitis D patients (52, 53). Recently, pegylated IFNα (PEG-IFN) has also been used in three smaller trials to treat hepatitis D (47, 54, 55). These studies used PEG-IFNα-2b for 48 or 72 weeks. The French study by Castelnau et al. (55) included 14 patients who were treated for 1 year. A sustained response for HDV-RNA was observed in six patients (43%). A German group applied a similar protocol in 12 patients but were only able to cure HDV in two patients (54) while a study performed in Italy showed a cure rate of 21% with 72 weeks of treatment with PEG-IFNα-2b (47). Ribavirin had no beneficial effect in this study.
The Hep-Net International hepatitis D intervention trial (HIDIT-1) included 90 patients in Germany, Turkey and Greece (46). Patients were randomized to receive either 180 μg PEG-IFNα-2a q.w. plus 10 mg adefovir dipivoxil q.d., 180 μg PEG-IFNα-2a q.w. plus placebo or 10 mg adefovir dipivoxil q.d. alone for 48 weeks. Both PEG-IFN groups showed a significantly higher reduction in mean HDV-RNA levels than the adefovir monotherapy group by week 48. HDV-RNA was only negative after treatment with PEG-IFNα-2a but not adefovir. The PEG-IFNα-2a/adefovir combination group showed a 1.1 log 10 IU/ml decline of HBsAg levels by week 48. Overall, the HIDIT-1 study showed that PEG-IFNα-2a displays significant antiviral efficacy against HDV in more than 40% of patients, with 25% becoming HDV-RNA negative after 48 weeks.
Currently, additional trials are ongoing to optimize the efficacy of available treatment, e.g. long-term combination of PEG-IFNα-2a with tenofovir. Moreover, alternative treatment options need to be explored. Prenylation inhibitors may be promising (56) because HDV replication depends on a prenylation step and prenylation inhibitors have been developed for the treatment of malignancies and have been shown to be safe. The efficacy of other compounds with a broad antiviral efficacy including nitazoxanide (57) could be interesting to investigate against HDV. Finally, several alternative IFNs under clinical development including IFNλ should be explored for hepatitis D. Continuous efforts to improve treatment options for hepatitis D are urgently needed as this infection is underestimated and represents a significant health risk in certain countries.
Conflicts of interest
Heiner Wedemeyer receives grant support, travel grants, lecture fees and is on the advisory boards of Roche, Gilead, Merck, BMS and Novartis.