The Natural History and Treatment of Chronic Hepatitis B: A Critical Evaluation of Standard Treatment Criteria and End Points

Approximately 350 to 400 million people worldwide have chronic hepatitis B virus (HBV) infection, with roughly 1.25 million cases in the United States and 180000 in Britain. More than 70% of all people with chronic HBV infection are Asian. Although white persons in the U.S. and British populations are usually infected in adolescence or adulthood (by means of sexual contact or sharing of intravenous needles), nearly every Asian person with this infection (whether born in or outside Asia) acquires it at birth from an infected mother or within 2 years of birth by means of close contact with infected relatives. There are important differences between these 2 groups. Patients infected in adolescence or adulthood immediately enter an immune clearance phase; their disease is of short duration and tends to become quiescent after they seroconvert from hepatitis B e antigen (HBeAg) to antibodies against HBeAg (anti-HBe). Such patients have been termed healthy carriers (1). In contrast, patients infected early in life have a prolonged immune tolerance phase followed by a prolonged immune clearance phase before HBeAg seroconversion, and strong evidence indicates that their disease progresses after HBeAg seroconversion. Guidelines for treatment and treatment end points from 3 regional liver associations (the American Association for the Study of Liver Diseases, the European Association for the Study of the Liver, and the Asian Pacific Association for the Study of the Liver) suggest that treatment should be initiated for patients who are HBeAg seropositive if their HBV DNA levels are 105 copies/mL or greater (20000 IU/mL) and their alanine aminotransferase (ALT) levels are greater than 2 times the upper limit of normal (Table) (24). Liver biopsy may be considered for patients with ALT levels between 1 and 2 times the upper limit of normal, particularly those older than 40 years, and treatment may be initiated if there are abnormal histologic findings. They also suggest that treatment can be stopped after seroconversion to anti-HBe and a decrease in HBV DNA to less than 105 copies/mL (<20000 IU/mL) and in ALT to normal levels. Table. Guidelines for Treatment and Treatment End Points from the 3 Regional Liver Associations* These guidelines assume that chronic HBV infection is likely to progress to cirrhosis and hepatocellular carcinoma only in patients with these markers of active infection and that treatment can be stopped once a patient achieves a healthy hepatitis B carrier state (5), defined as seronegativity for HBeAg (or seropositivity for anti-HBe) and HBV DNA and ALT levels below the indicated thresholds. This concept of a healthy hepatitis B carrier state may apply to patients infected in adolescence or adulthood, but the following evidence suggests that it does not apply to patients infected earlier in life. Hepatitis B e Antigen In a 30-year follow-up of 296 blood donors from northern Italy, there was no clinically significant liver illness or mortality compared with 157 uninfected control participants (6). The investigators attribute this favorable prognosis to the absence of comorbid conditions, inactive HBV infection, and abstinence from alcohol. Another study, a long-term follow-up of white, HBeAg-positive, interferon-treated patients and untreated control participants, showed that patients who cleared HBeAg survived much longer than those who remained HBeAg positive (7). The essential contributor to the favorable outcome in these groups is probably that the patients were infected in adolescence or adulthood. In contrast, other studies show that the disease continues to progress after HBeAg seroconversion in patients infected early in life. A study from Taiwan of 683 patients showed that new cirrhosis developed at similar annual rates (1.3% vs. 2.4%) in HBeAg-positive and anti-HBepositive patients (8). In a recent study of 3233 Asian patients without cirrhosis, the median age at HBeAg seroconversion was 35 years, whereas the age at development of cirrhosis complications was 57.2 years (9). More than 73% of the patients were anti-HBe positive when cirrhosis complications and hepatocellular carcinoma developed. In a follow-up of 92 Chinese carriers of hepatitis B surface antigen (HBsAg) with HBsAg seroclearance, 37% still had intrahepatic HBV DNA. Furthermore, 5.4% of these patients developed hepatocellular carcinoma during follow-up compared with 8.7% of 92 control patients who did not lose HBsAg (10), a statistically insignificant difference. These findings establish that hepatocellular carcinoma can develop even after HBsAg seroclearance. A study from Taiwan of 11893 men in which HBeAg positivity was associated with increased risk for hepatocellular carcinoma (11) seems to contradict this conclusion; however, the participants' HBeAg and anti-HBe status were examined only at the time of enrollment. Because the participants were between 30 and 65 years of age, a substantial proportion of them would have HBeAg seroconversion during the 9 to 10 years of follow-up, and the effect of HBeAg seroconversion subsequent to enrollment on the development of hepatocellular carcinoma is unknown. In addition, the relative risk for hepatocellular carcinoma was still 9.6 times higher in participants who were anti-HBe positive at enrollment than in noncarriers. Hepatitis B Virus DNA In a retrospective study of 165 Chinese patients, 45% of HBeAg-negative patients with active hepatitis (defined as abnormal ALT levels) had HBV DNA levels less than 105 copies/mL (12). In another study, 67.1% of 79 Chinese carriers with cirrhosis-related complications were anti-HBe positive, with 37.7% having HBV DNA levels less than 105 copies/mL (<20000 IU/mL) and 24.5% having levels less than 104 copies/mL (<2000 IU/mL) (13). A study of 3653 carriers from Taiwan concludes that an HBV DNA level of at least 104 copies/mL (2000 IU/mL) is a strong risk predictor for hepatocellular carcinoma (14). During follow-up, the hazard ratio for the development of hepatocellular carcinoma was 10.1 (95% CI, 6.3 to 16.2) for persons whose HBV DNA level remained greater than 105 copies/mL (>20000 IU/mL) compared with those who had levels less than 104 copies/mL (<2000 IU/mL) at enrollment. However, the hazard ratio was still substantial at 3.8 (CI, 1.7 to 8.4) for persons whose HBV DNA levels had decreased from greater than 105 copies/mL (>20000 IU/mL) on enrollment to less than 104 copies/mL (<2000 IU/mL) during follow-up. These studies suggest that a clinically significant proportion of Asian patients continue to be at risk for active hepatitis and its complications after their HBV DNA levels decrease to less than 105 copies/mL (<20000 IU/mL) or even 104 copies/mL (<2000 IU/mL). Alanine Aminotransferase In a prospective cohort study of 94533 men and 47522 women with no known liver disease, mortality from liver disease was greater in patients with ALT levels between 0.5 and 1 times the upper limit of normal than in those with ALT levels less than 0.5 times the upper limit of normal (15). We suggest that adjustment of the normal range for ALT may be necessary. For Asian patients with chronic hepatitis B, the previously mentioned study of 3233 patients found that patients with ALT levels less than 0.5 times the upper limit of normal had the lowest risk for complications, patients with ALT levels of 0.5 to 1 times the upper limit of normal had a statistically significantly increased risk for complications, and patients with ALT levels greater than 1 to 2 times the upper limit of normal had the highest risk. Implications These studies suggest that the current treatment criteria and end points are not satisfactory for most patients with chronic hepatitis B who are infected early in life. The disease continues to progress in a substantial proportion of patients who achieve standard milestones, such as HBeAg seroconversion, HBsAg clearance, HBV DNA levels less than 105 copies/mL (<20000 IU/mL) or even 104 copies/mL (<2000 IU/mL), and ALT levels between 0.5 and 2 times the upper limit of normal. For these patients, HBeAg seroconversion may be an inappropriate end point (16), because immune damage to the liver may continue after HBeAg seroconversion even with relatively low levels of viral replication. To prevent the development of complications, patients with ALT levels between 0.5 and 2 times the upper limit of normal should be treated; placebo-controlled trials are needed to confirm effectiveness. In addition, HBV DNA suppression should continue until the level decreases to less than 104 copies/mL (<2000 IU/mL). Two studies demonstrated the effectiveness of long-term suppression of HBV replication with nucleoside or nucleotide treatment. A study of 63 HBeAg-positive patients who took lamivudine for 3 years (17) and a study of 185 HBeAg-negative patients who took adefovir for 96 weeks (18) showed reversal of fibrosis and even cirrhosis. However, the emergence of the tyrosine, methionine, aspartate, aspartate (YMDD) mutation in the virus during lamivudine therapy blunts the improvements (17). Two other studies investigated the effects of nucleoside and nucleotide analogues on the development of cirrhosis complications and hepatocellular carcinoma. In 651 patients with confirmed advanced fibrosis or cirrhosis, lamivudine treatment reduced cirrhosis complications and hepatocellular carcinoma compared with placebo (by 7.8% and 17.7%, respectively; P= 0.001) after a median of 32.4 months (19). Patients who developed YMDD viral mutations were more likely than those without these mutations to have an increase in their ChildPugh score (P< 0.001) but were still less likely than those receiving placebo to develop complications (P> 0.05). A similar reduction in the development of cirrhosis or hepatocellular carcinoma was found in 142 HBeAg-positive Asian patients without cirrhosis who received lamivudine for a median of 89.9 months compared with 124 untreated control patients (P=