The ultimate goal of treating a chronic infectious disease is the eradication of the infectious agent to prevent organ damage or death. The natural history of untreated chronic hepatitis B may result in the development of cirrhosis, followed by hepatic decompensation and death. Hepatocellular carcinoma (HCC) can also occur in patients with chronic hepatitis B virus (HBV) infection, with or without the presence of cirrhosis. Antiviral therapy with oral nucleos(t)ide analog (NUC) agents suppress viral replication but do not directly act on the covalently closed circular DNA that resides within infected hepatocytes; thus, current oral therapy rarely eradicates HBV infection. Although these oral antiviral agents are well-tolerated with minimal side effects, prolonged viral suppression runs the risk of selecting for antiviral drug–resistant mutations and causing virologic breakthrough; decreased patient adherence with antiviral treatment may also be associated with breakthrough. Therefore, routine “cure” or eradication of HBV infection is unrealistic, at least with current therapy. Nevertheless, preventing the clinical outcomes, or end points, of this disease remains important and are likely achievable with available treatment.
The development of complications of chronic hepatitis B occur over the long natural history of HBV infection, but are not necessarily uniform as demonstrated by the occurrence of HCC in patients without cirrhosis. Furthermore, cirrhosis may take from years to decades to develop, and patients with chronic HBV infection are often asymptomatic before these clinical end points of cirrhosis or HCC are discovered. Judging the efficacy of antiviral therapy by waiting to see if these clinical end points are avoided is neither practical nor ethical. To satisfy the need to assess the impact of antiviral therapy, reliable intermediate or surrogate end points have been used as therapeutic goals with the treatment of chronic hepatitis B. According to the National Institutes of Health, surrogate end points are “biomarkers intended to substitute for a clinical endpoint.” These surrogate end points should occur on the causal pathway of disease, be highly associated with the long-term clinical end point of interest, and be easier to assess than the ultimate outcome. Furthermore, the effect of an intervention upon the surrogate end point should also have a downstream effect on the prevention of the clinical end point.
The National Institutes of Health–proposed surrogate end points for chronic HBV infection include histologic criteria [decrease in histology activity index (HAI) by ≥2 points with no worsening of fibrosis], biochemical changes [normalization of alanine aminotransferase (ALT)] levels, virologic response (maintained suppression of serum HBV DNA to undetectable levels by a sensitive assay), and serologic changes [loss of hepatitis B e antigen (HBeAg), with or without seroconversion to antibody to HBeAg (anti-HBe), and potentially loss of hepatitis B surface antigen (HBsAg); Table 1 ]. Although these surrogate end points are used to assess the efficacy of antiviral therapy, particularly during phase III clinical trials of HBV treatment and by regulatory authorities for licensing new treatments, they do not evaluate the various aspects of disease status associated with chronic hepatitis B. Therefore, such end points may not be sufficient in individual patients for determination of the overall course of treatment such that combination of end points may be necessary. Given their intermediate status along the causal pathway of disease, these surrogate end points may still require further clarification of their pathogenic role and impact on ultimate outcomes, although much information has been learned from cohort studies involving both treated and untreated patients. In addition, these surrogate end points could be subject to confounding owing to undefined interactions between each other. Last, cost-effectiveness analyses of these various end points are currently unavailable.
| End Point | Criteria |
|---|---|
| Histologic | Reduction of fibrosis stage (≥2-point decrease in HAI) and no worsening of fibrosis Reduction of inflammatory activity |
| Biochemical | Normalization of ALT levels |
| Virologic | Sustained decrease in serum HBV DNA to low levels, ideally to undetectable |
| Serologic | HBeAg loss, with seroconversion to anti-HBe HBsAg loss, with or without seroconversion to anti-HBs |
Histologic End Point: The Role Of Liver Biopsy
Traditionally, liver biopsy has been considered the most direct and accurate method of assessing liver inflammation and fibrosis, and liver histology has been used as the main end point of clinical trials evaluating the efficacy of antiviral therapies. Traditionally, regulatory approval required comparison of pretreatment and end-of-treatment liver biopsies showing histologic improvement, as defined by a 2-point or greater decrease in the HAI and no worsening of fibrosis. With respect to NUC therapies that have been approved (including lamivudine, adefovir, telbivudine, entecavir, and tenofovir), the end-of-treatment achievement of histologic improvement has ranged between 49% and 74% for HBeAg-positive patients and 61% and 72% for HBeAg-negative patients after 1 year of antiviral therapy. Monotherapy with peginterferon, or combination therapy with peginterferon plus lamivudine, have shown much lower rates of histologic improvement for HBeAg-positive patients, namely, 22% and 33%, respectively. Longer follow-up (>3–5 years of treatment) and determination of whether histologic improvement has been sustained is not required for pivotal clinical trials, but a few small studies have shown persistent histologic improvement as well as regression of fibrosis and even early cirrhosis.
Although liver biopsies have provided much important information on the course of liver disease, there remain several limitations, such as sampling variability and error. When biopsies are obtained by percutaneous or transjugular access, the operator is unable to identify or target specific areas of the liver, and sampling is based on the assumption that liver disease uniformly affects the whole liver. In addition, liver biopsies that yield tissue samples less than 2.5 cm in size have been shown to underestimate hepatic fibrosis. Once samples are obtained, interobserver and intra-observer variability with regard to assessment of histologic activity and fibrosis may also occur. The development of various scoring systems, including the HAI (also known as the Knodell score) are intended to standardize evaluation of liver biopsies, but interobserver and intra-observer variability can still be quite variable and affect assessment. Last, inherent in this invasive procedure is the small risk of serious complications, including pneumothorax, hemoperitoneum, and death, which may outweigh the benefit of an individual biopsy or serial repeat biopsies.
Thus, although liver biopsies have provided useful and accurate assessment on the disease activity of chronic HBV infection, they are not currently considered a practical end point outside of clinical trials for antiviral therapies. On the other hand, liver biopsies could still be used to provide not only a histologic end point, but also potentially novel virologic end points, such as direct quantification of intrahepatic HBV DNA or covalently closed circular DNA using polymerase chain reaction.
Biochemical End Point: Serum Alanine Aminotransferase Levels
Serum ALT and aspartate aminotransferase are biochemical markers used to indirectly measure the activity of underlying liver disease. During chronic HBV infection, elevated aminotransferase levels reflect ongoing T-cell–directed cytotoxic activity against HBV-infected hepatocytes, and are associated with an increased risk of liver complications and death. Changes in both serum aminotransferase and HBV DNA levels over time can also help to differentiate between the various stages of chronic HBV infection. According to standard guidelines, the presence of an elevated ALT, in association with an elevated HBV DNA level (depending on HBeAg status), is the major indication for antiviral therapy. With NUC antiviral treatment, ALT levels can decrease and eventually normalize over time, whereas the sudden increase in ALT while undergoing therapy or after discontinuing treatment may herald viral reactivation with biochemical and possibly clinical relapse. Hence, normalization of ALT has been included as the biochemical end point in clinical trials of HBV treatment.
Because these hepatic biomarkers are inexpensive and easy to obtain, they would seem to be a satisfactory alternative end point to the invasive histologic assessment of liver disease. In combination with other biomarkers, in the form of the aspartate aminotransferase-to-platelet ratio index, these biomarkers can also be used to assess underlying hepatic fibrosis, although this noninvasive approach has only been validated in patients with chronic hepatitis C. Additionally, the “normal” ranges of ALT and aspartate aminotransferase may not only vary with use of different laboratory assays, but also even in the biopsy-proven absence of underlying liver disease. Conversely, up to 23% of patients with chronic hepatitis B and serum aminotransferase levels within the so-called normal range may have serious liver disease, including significant hepatic fibrosis and cirrhosis, with or without hepatic decompensation. Such major discrepancies between serum aminotransferase levels and underlying liver biopsy findings calls into question whether this biomarker end point should still be utilized.
Although the measurement of serum ALT levels may be a useful component in helping to elucidate underlying disease activity in chronic hepatitis B, this biomarker cannot stand alone, especially without being accompanied by HBV DNA levels. There is some potential for combining this biomarker with others to provide a noninvasive assessment of underlying necroinflammation, but this approach has been primarily validated in patients with chronic hepatitis C. Additionally, there are already accurate models that reliably predict advanced liver disease, such as the Child–Turcotte–Pugh and Model for End-Stage Liver Disease scores. Last, the lack of sensitivity and specificity for chronic hepatitis B limits the utility of ALT levels as both indicators for antiviral treatment and their normalization as a sole end point.
Biochemical End Point: Serum Alanine Aminotransferase Levels
Serum ALT and aspartate aminotransferase are biochemical markers used to indirectly measure the activity of underlying liver disease. During chronic HBV infection, elevated aminotransferase levels reflect ongoing T-cell–directed cytotoxic activity against HBV-infected hepatocytes, and are associated with an increased risk of liver complications and death. Changes in both serum aminotransferase and HBV DNA levels over time can also help to differentiate between the various stages of chronic HBV infection. According to standard guidelines, the presence of an elevated ALT, in association with an elevated HBV DNA level (depending on HBeAg status), is the major indication for antiviral therapy. With NUC antiviral treatment, ALT levels can decrease and eventually normalize over time, whereas the sudden increase in ALT while undergoing therapy or after discontinuing treatment may herald viral reactivation with biochemical and possibly clinical relapse. Hence, normalization of ALT has been included as the biochemical end point in clinical trials of HBV treatment.
Because these hepatic biomarkers are inexpensive and easy to obtain, they would seem to be a satisfactory alternative end point to the invasive histologic assessment of liver disease. In combination with other biomarkers, in the form of the aspartate aminotransferase-to-platelet ratio index, these biomarkers can also be used to assess underlying hepatic fibrosis, although this noninvasive approach has only been validated in patients with chronic hepatitis C. Additionally, the “normal” ranges of ALT and aspartate aminotransferase may not only vary with use of different laboratory assays, but also even in the biopsy-proven absence of underlying liver disease. Conversely, up to 23% of patients with chronic hepatitis B and serum aminotransferase levels within the so-called normal range may have serious liver disease, including significant hepatic fibrosis and cirrhosis, with or without hepatic decompensation. Such major discrepancies between serum aminotransferase levels and underlying liver biopsy findings calls into question whether this biomarker end point should still be utilized.
Although the measurement of serum ALT levels may be a useful component in helping to elucidate underlying disease activity in chronic hepatitis B, this biomarker cannot stand alone, especially without being accompanied by HBV DNA levels. There is some potential for combining this biomarker with others to provide a noninvasive assessment of underlying necroinflammation, but this approach has been primarily validated in patients with chronic hepatitis C. Additionally, there are already accurate models that reliably predict advanced liver disease, such as the Child–Turcotte–Pugh and Model for End-Stage Liver Disease scores. Last, the lack of sensitivity and specificity for chronic hepatitis B limits the utility of ALT levels as both indicators for antiviral treatment and their normalization as a sole end point.
Virologic End Point: Serum Hbv Dna Levels
During the natural history of chronic HBV infection, serum HBV DNA levels reflects underlying intrahepatic HBV replication and can be quite dynamic, with very high levels (>2 × 10 IU/mL or >10 copies/mL) during the immune tolerant phase, lower levels with chronic HBeAg-negative disease (<10 5 IU/mL), and even lower levels during the residual inactive carrier phase (generally <2000 IU/mL. Although HBV DNA levels vary greatly, during the immune clearance phase or a reactivation flare that follows the immune tolerance phase, the host immune response against the virus will lead to T-cell–directed liver injury. However, despite the absence of obvious liver injury, large (>3600 patients) prospective cohort studies by the community-based Taiwanese Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer–Hepatitis B Virus Study Group have shown that very high HBV DNA levels (>10 6 copies/mL) are significantly associated with the development of cirrhosis and HCC, independent of HBeAg status or serum ALT levels. Furthermore, there does not seem to be a risk-free threshold, as even low viral loads (<10 3 copies/mL) have been found to be associated with cirrhosis or HCC. Therefore, suppression of HBV replication to an undetectable serum level by a sensitive polymerase chain reaction assay has been used as an end point in clinical trials of HBV treatment, and high levels in association with elevated ALT are an indication for treatment.
On-therapy virologic response has been demonstrated with both peginterferon and NUC agents. Depending on the NUC, loss of serum HBV DNA ranges between 21% and 76% for HBeAg-positive patients, and a higher range of 51% to 93% in HBeAg-negative patients. These rates were achieved (in increasing order) by adefovir, lamivudine, telbivudine, entecavir, and tenofovir. Treatment with standard interferon and peginterferon (with or without lamivudine) achieved viral suppression rates of 25% to 69% and 60% to 87% for HBeAg-positive and HBeAg-negative patients, respectively. The clinical benefits of antiviral treatment in reducing the risks of cirrhosis and HCC have been clearly demonstrated, including the potential regression of fibrosis and early cirrhosis. However, off-treatment viral response may not be sustainable without long-term treatment, particularly in patients with HBeAg-negative disease, whereas long-term treatment may also induce the development of viral resistance. In patients whose antiviral treatment was discontinued after 1 year of treatment, the viral relapse rate was very rapid (within 6–12 months after discontinuing treatment), with recurrent detectable viral loads in 95% of HBeAg-positive patients and 92% to 97% of HBeAg-negative patients. On the other hand, a longer duration of antiviral therapy (5 years) has increased virologic response rates by about 27% in patients receiving entecavir for HBeAg-positive disease. Hence, the high relapse rates after short-term therapy has encouraged extending the duration of therapy to more than 1 year, at least until achieving both undetectable HBV DNA and HBeAg seroconversion in HBeAg-positive chronic hepatitis B (with a consolidation therapy for 6 months after the appearance of anti-HBe), and long-term in patients with HBeAg-negative chronic hepatitis B, until achievement of HBsAg clearance.
Given the long-term therapy necessary to achieve seroconversion of HBeAg or the clearance of HBsAg, the risk of antiviral resistance becomes a serious concern and has shifted clinical practice toward the use of NUCs with a high genetic barrier to resistance. The first NUC, lamivudine, was used widely around the world and demonstrated the development of resistance in just 1 year of therapy at a rate of 24%, which increased to 70% after 5 years of therapy. Subsequently developed NUC agents showed lower rates of long-term antiviral resistance; adefovir demonstrated a risk of drug resistance of 29% after 5 years of therapy, telbivudine a 25% rate after 2 years of therapy, entecavir a 1.2% rate after 6 years of therapy, and tenofovir so far no resistance after 4 years of therapy ( Table 2 ). Therefore, tenofovir or entecavir should be considered first-line oral agents for antiviral treatment of chronic hepatitis B, especially in patients who are naïve to HBV treatment and in those who prolonged or indefinite therapy is anticipated.
Related posts:
Noninvasive Tools to Assess Hepatic Fibrosis: Ready for Prime Time?
Acute Liver Failure: Current Practice and Recent Advances
Long-Term Management of the Liver Transplant Recipient: Pearls for the Practicing Gastroenterologist
Noninvasive Tools to Assess Hepatic Fibrosis: Ready for Prime Time?
Liver Transplantation in the 21st Century: Expanding the Donor Options
Alcoholic Hepatitis: Prognostic Models and Treatment
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
