Jules L. Dienstag, MD
Massachusetts General Hospital
Harvard Medical School
Chronic hepatitis B is a major global healthcare problem. Affecting an estimated 350 to 400 million people worldwide,1,2 chronic hepatitis B results in approximately one million deaths each year,2 making it the world's tenth leading cause of death. In the US, 1 to 1.25 million people have chronic hepatitis B, and approximately 4000 to 5000 die each year of its complications.3 The progression of chronic hepatitis B is variable, ranging from mild asymptomatic infection to severe chronic liver disease. In about 15% to 25% of patients, ongoing viral replication leads to serious complications including hepatic decompensation, cirrhosis, and hepatocellular carcinoma (HCC).2,3
Hepatitis B virus (HBV) is a complex virus. Classified as a hepatotropic DNA virus, or hepadnavirus, it replicates primarily in the liver but can also be found in other organs and in lymphocytes. Three morphological forms can be seen on electron microscopy: double-shelled spheres (intact virions), smaller spherical particles, and tubular structures. The outer surface of the intact virion is composed of viral envelope protein or surface antigen (HBsAg). The internal part of the virion consists of a nucleocapsid-composed of two proteins, the core antigen (HBcAg) and the e-antigen (HBeAg)-that surrounds the genome (Fig. 1). The smaller spherical particles and tubular structures are composed of excess HBsAg and may outnumber the intact virions by a factor of 100 to 1000.
The infectious virion has an incomplete (partially single stranded), open circular DNA genome of 3200 base pairs (3.2 kb), comprising four genes: S (that encodes the envelope or surface antigen); C (that encodes the core protein and e-antigen); P (that encodes a DNA polymerase that also has reverse transcriptase activity); and X (a transactivating protein that can enhance the replication of HBV as well as HIV and that appears to be more frequently expressed in patients with severe liver disease and HCC; in clinical practice, this viral protein is of limited relevance or application). Although the genome is small, it can produce these four large proteins because the genes overlap. For example, the S gene overlaps the P gene completely, whereas the C and P genes partially overlap (Fig 2).
Figure 3 is a simplified diagram of HBV replication. After the virion enters the host cell, it is uncoated and its genome is delivered to the nucleus. There the viral genome is converted to covalently closed circular (ccc) DNA-a minichromosome that serves as the viral transcription template. From this template, RNA is translated to yield the viral proteins. In addition, the 3.5 kb RNA serves as a pregenomic template for reverse transcription to negative-strand DNA, which, in turn, becomes the template for transcription of positive-strand DNA. Nucleocapsid and envelope proteins are assembled around the viral genome to form capsids that bud from the cell membrane. From a clinical perspective, the importance of this reverse transcription replicative strategy is that some antiretroviral drugs, originally developed for treatment of HIV infection, also are active against HBV and are currently used to treat chronic hepatitis B. Adding to the complexity of HBV is the fact that it has multiple serotypes or genotypes, which vary in frequency among different populations (Table 1).4,5
At least seven genotypes (A-G) have been recognized; among patients born in the US, genotype A is the most prevalent.5 The clinical significance of genotype will be discussed later.
HBV is not cytotoxic but destroys liver cells indirectly by provoking an immunologic response (Fig. 4). Kupffer cells endocytose viral antigens and present them bound to MHC class II molecules to T-helper cells. These CD4+ cells recognize the antigens and release cytokines that direct B-cell and cytolytic T-cell (CTL) activity. Stimulated B cells produce specific antibodies, including neutralizing antibodies. CTLs recognize viral peptides bound to MHC class I molecules on hepatocyte surfaces, leading to destruction of infected hepatocytes. In persons who fail to mount a sufficiently vigorous immune response to HBV during acute infection, chronic infection develops, and the persistent, ineffective immune response results in progressive liver damage and fibrosis.6
For clinical diagnostic purposes, we rely on an understanding of serologic events that appear during acute and chronic hepatitis B. Figure 5 illustrates the serologic course of a typical case of acute, self-limited hepatitis B.7 The first viral protein to appear (within several weeks of infection) is the envelope protein or surface antigen, HBsAg, which increases in titer over the next several weeks, even before symptoms begin. During clinically apparent illness, levels of HBsAg are high, symptoms occur, and aminotransferase levels are elevated. Once symptoms appear, HBsAg begins to decline and is ultimately replaced by anti-HBs (HBsAg seroconversion). Anti-HBs is recognized to be a neutralizing antibody, coinciding temporally with recovery from infection.
Within a week or two after the appearance of HBsAg, antibody to the core protein, anti-HBc, can be detected. Early during acute infection, anti-HBc is primarily of the IgM class; however, after approximately 6 months, IgM anti-HBc is replaced by anti-HBc of the IgG class (reported by clinical laboratories as total anti-HBc reactivity with undetectable IgM anti-HBc).
HBeAg, a nonparticulate, circulating protein derived from translation of the core gene, typically appears during the period of peak active virus replication. Because HBeAg is always present during acute HBV infection, testing for it during acute infection is not indicated. Ultimately, HBeAg is replaced by anti- HBe (HBeAg seroconversion); when recovery from acute hepatitis B is delayed beyond 3 to 4 months, the persistence of HBeAg may be a harbinger of chronic infection. Thus, a person who recovers from hepatitis B generally has all three antibodies-anti-HBs, anti-HBc, and anti-HBe.
Figure 6 illustrates the serologic course when acute hepatitis B fails to resolve and, instead, progresses to chronic hepatitis B.7 In such cases, HBsAg remains reactive indefinitely, and anti-HBc is of the IgG class. Two relative phases of chronic hepatitis B are recognized. A relatively highly replicative phase occurs early during the natural history of chronic infection and is associated with HBeAg and high levels of HBV DNA (usually >106 copies/mL). About 10% to 15% of patients per year convert spontaneously to a relatively nonreplicative phase-associated with HBeAg seroconversion-in which HBV DNA circulates at levels ≤103 copies/mL.
Patients with highly replicative hepatitis B tend to have chronic liver disease, while those with relatively nonreplicative disease tend to be inactive carriers with normal liver histology. Patients with highly replicative disease also tend to be much more infectious. For example, a mother with hepatitis B and HBeAg has approximately a 90% chance of transmitting the infection to her offspring, while a mother with relatively nonreplicative hepatitis B has about a 10% chance of transmitting the infection. Similarly, the risk of acquiring hepatitis B infection from a needle stick is about 30% if the needle is contaminated with blood from a person in the replicative phase but only 0.1% if the needle is contaminated with blood from a person in the relatively nonreplicative phase.
The level of replication in hepatitis B is much higher than that seen with other bloodborne viral diseases such as hepatitis C and HIV. In hepatitis C, complete eradication of the virus can be achieved with antiviral therapy. In chronic hepatitis B, however, complete eradication of virus replication is rarely achieved; instead, the goal of therapy is to maintain suppression of virus levels to relatively nonreplicative levels.
As just reviewed, HBeAg is generally a marker of active replication; however, a subgroup of patients with chronic hepatitis B exists who lack HBeAg and yet have high levels of replication and infectivity. This variant often results from a mutation in the precore region of the C gene (Fig. 7). The C gene can be transcribed starting at two different loci: when transcription starts at nucleotide 1901, the core protein is produced; when transcription starts at nucleotide 1814, HBeAg is produced. An important mutation can occur at nucleotide 1896 (in the precore region), converting guanine to adenine. This changes a TGG codon to TAG, which is a stop codon; when TAG appears in the precore region, the virus is unable to produce HBeAg.
The frequency of precore stop-codon mutations varies among HBV genotypes. The reason for this is shown in Figure 8. Nucleotide 1896 (normally guanine) is located in a stem loop opposite to nucleotide 1858. Nucleotide 1858 can be either cytosine or thymidine, depending on the HBV genotype. In genotype A, cytosine occurs at position 1858. Because cytosine forms a very strong bond with guanine, two mutations would be required in order to produce a stop-codon mutation. On the other hand, in non-A genotypes, nucleotide 1858 is thymidine, which forms a relatively unstable bond with guanine-requiring only one mutation to produce a stop-codon mutation. Therefore, precore stop codons are much more likely to occur in non-A genotypes. Precore stop-codon mutations are relatively uncommon in North America and Western Europe (where genotype A predominates), but occur more often in Mediterranean countries and Asia (where non-A genotypes are more frequent).8,9
Mutations in the core promoter region also can interfere with HBeAg production. These mutations occur as easily in genotype A as in other genotypes (Table 2). In the US, replicative HBeAg-negative hepatitis B is more often due to core promoter than to precore variants.10
HBeAg-negative chronic hepatitis B can be severe and progressive, despite the fact that HBV DNA levels tend to be lower (on the order of 105 virions/ml) than in patients with HBeAg-reactive chronic hepatitis B. In addition, although ALT levels are persistently elevated in a proportion of patients with HBeAg-negative chronic hepatitis B, in others, ALT levels wax and wane (sometimes with intervening periods of normal ALT activity), reflecting fluctuations in hepatic necroinflammatory activity.
Despite the known relationship between HBV genotype and precore mutation frequency, the impact of genotype on the natural history of HBV infection is not yet well understood. Preliminary studies comparing genotypes B and C suggest that genotype B is associated with a lower prevalence of HBeAg, resulting from earlier and more sustained HBeAg seroconversion (Fig. 9). This observation may explain the less active liver disease and slower progression to cirrhosis reported to occur in patients with genotype B.11 Preliminary evidence suggests that genotype B also is associated with slower development of HCC,12 although results from different reports are conflicting.5,13
The clinical expression of hepatitis B in different parts of the world depends not only on the prevalent genotypes but also on the prevalent modes of transmission. In fact, the clinical expression of hepatitis B is influenced dramatically by host and epidemiologic factors. Figure 10 compares areas of low endemicity (such as the US, with a prevalence of 0.1% to 0.2%) versus areas of high endemicity (such as Asia and Sub-Saharan Africa, with prevalences of 10% to 15% or higher). In Asia, for example, the high prevalence of hepatitis B in the general population is reflected by a high prevalence among women of childbearing age. Thus, many infections are transmitted perinatally from mother to offspring. Neonatal infection is associated with a high level of immunological tolerance, which is very difficult to overcome. As a result, the likelihood of chronicity exceeds 90%.14,15,16
In contrast, the US has relatively few hepatitis B carriers and, because the prevalence of HBV infection in women of childbearing age is comparably low, perinatal infections are rare. Most infections, instead, are acquired in early adulthood-via occupational exposure (eg, in healthcare workers), via injection drug use and, most commonly, via sexual activity (ie, via human behaviors that foster the spread of bloodborne viruses). The adult immune system mounts a robust response to the presence of the virus in liver cells, resulting in immunologic destruction of virus-infected liver cells and in clinically apparent acute hepatitis B, which resolves with elimination of the virus. The likelihood of chronicity in such clinically apparent acute hepatitis B is low (probably ≤1%).15,17 Moreover, the clinical expression of disease is reflected by another dichotomy between areas of high and low HBV prevalence. In highly endemic areas, the lifetime risk of succumbing to complications (cirrhosis and HCC) has been estimated to be as high as 40%, whereas in low prevalence areas the risk of HCC in those with adult-acquired chronic hepatitis B is very low.
To summarize: In the West, clinically apparent acute hepatitis B is usually a self-limited disease acquired in adulthood with a low rate of progression tochronicity, rarely, if ever, leading to HCC. In the East, hepatitis B is usually a chronic disease acquired at birth, associated with a high rate of progression to cirrhosis and cancer. The difference in natural course is mediated by the interaction between virus and host, which is largely determined by the age at which the infection is acquired.16
Among patients with chronic HBV infection, the level of HBV replication may be an important determinant of the risk of adverse outcomes, such as HCC. In a prospective study of Asian adults with chronic hepatitis B, those who were HBeAg-positive (and who therefore continued to have high levels of viral replication) had a high risk of HCC over a decade, whereas in those who were HBeAg-negative, the risk was substantially lower (Fig. 11).18
Finally, among patients with cirrhosis, some will have compensated disease for many years; however, once decompensation begins, the 5-year survival rate may be as low as 14%, as shown in a Dutch study19.
Now that antiviral agents are available for treatment of hepatitis B, preventing (and in some cases even reversing) decompensation can be achieved by timely treatment.