Teresa L. Wright, MD
Professor of Medicine
University of California at San Francisco
The last decade has seen many advances in the management of hepatitis B. There are now three drugs that are FDA-approved as first-line therapy: interferon alfa-2b (Intron®), lamivudine (Epivir-HBV®), and adefovir dipivoxil (Hepsera®). Remarkable strides have also been made with respect to liver transplantation outcomes, which are now excellent largely due to use of these effective medications (particularly lamivudine and adefovir in combination with hepatitis B immune globulin) given pre-, post-, or peritransplantation.1 The tolerability of the oral agents (lamivudine and adefovir) is superior to that of interferon; however, interferon is more frequently associated with loss of surface antigen (HBsAg). Loss of HBeAg is similar between lamivudine and interferon, and may be a little higher with these two drugs than with adefovir. A major difference between the oral agents and interferon is that drug resistance has been observed with the nucleoside/nucleotide analogs (commonly with lamivudine, much less commonly with adefovir), but not with interferon.
Combinations of lamivudine and adefovir are increasingly being used. While there is no clear evidence of synergy in terms of antiviral efficacy, it is likely that combination therapy reduces the development of resistance. This may be particularly important for patients who are likely to require long-term therapy (eg, those with replicative HBeAg-negative disease), and/or who are at high risk for clinical decompensation if resistance should develop (eg, those with advanced liver disease).
Patients with chronic hepatitis B are generally considered candidates for treatment if they have aminotransferase (ALT) greater than twice the upper limit of normal (ULN) and active viral replication (usually defined as HBV DNA >105 copies/mL). In HBeAg-positive patients with compensated liver disease, treatment should be delayed for 3 to 6 months to determine whether spontaneous HBe seroconversion will occur.2
However, there can be some flexibility with regard to the ALT and HBV DNA guidelines. For example, treatment may be indicated for patients with liver enzymes ≤2 x ULN if they have significant liver disease (moderate to severe biopsy findings or clinical evidence of cirrhosis) and active replication.2 In HBeAgnegative patients, active replication at a lower level (HBV DNA ≥104 copies/mL) may be considered an indication for treatment.3
Various treatment endpoints have been used. Most clinical trials have focused on improvement in liver histology, typically defined as a ≥2-point decrease in histological activity index (HAI). Virologic endpoints have also been used-eg, loss of HBeAg (in patients who are initially HBeAg-positive), HBe seroconversion (acquisition of anti-HBe in addition to loss of HBeAg), and suppression of HBV DNA. Traditionally, HBV DNA is considered to be suppressed when it is below the level of detection on various assays; older hybridization assays had sensitivity levels of 0.4 pg/mL or about 105copies/mL; however, newer, more sensitive assays may have a lower detection limit of <400 copies/mL. Comparisons of different HBV DNA assays are summarized in the 2003 American Association for the Study of Liver Diseases (AASLD) Hepatitis B Practice Guidelines.4 Viral suppression may also be measured in terms of log reductions of HBV DNA. (A log reduction corresponds to an order of magnitude; for example, reduction of HBV DNA levels from 105 to 103 is a 2-log reduction.) Biochemical improvement typically follows virologic improvement, and is defined in terms of either reduction or normalization of serum ALT.
The first oral agent to be FDA approved was lamivudine, a synthetic nucleoside analog. A negative enantiomer of 3´-thiacytidine, it is incorporated into viral DNA, resulting in DNA chain termination. Lamivudine inhibits reverse transcriptase activity of both HBV and HIV, preventing synthesis of the negative HBV strand.
In the original lamivudine trials that led to FDA approval, the primary endpoint of treatment was histologic improvement. Figure 1 shows the results from three prospective trials-one in an Asian population,5 one in a western population,6 and another in a group of interferon nonresponders7-comparing lamivudine (100 mg/day) vs placebo. The patients in all three trials were HBeAg-positive. All three trials had similar results: after a year of treatment, >50% of lamivudinetreated patients had improved liver histology compared to ≤25% in the placebo groups.
Lamivudine also very effectively suppresses HBV replication. The phase 2 trials used hybridization assays with limited sensitivity (see above); however, data based on more sensitive polymerase chain reaction (PCR) assays are now also available. Figure 2 shows HBV DNA data based on a PCR assay with a detection limit of approximately 103 copies/mL. In this study, three different lamivudine doses (25 mg/day, 100 mg/day, and 300 mg/day) were compared. At 24 weeks, the two higher doses produced an approximately 4-log reduction in HBV DNA-significantly better than that seen with the low dose. At the 25 mg, 100 mg, and 300 mg doses, the percentages of patients who became HBV DNAnegative were 12%, 29%, and 37%, respectively.8
Figure 3 illustrates aggregate HBV DNA data from three placebo-controlled trials of lamivudine (100 mg/day) over a year of treatment. Although some patients became HBV DNA-negative even with placebo, the percentage of patients who did so was significantly greater with lamivudine. Toward the end of the year, a percentage of patients who had initially become HBV DNA-negative on lamivudine reverted to HBV DNA-positive-largely due to the development of resistance. The lamivudine-resistant YMDD variant has been observed in 16% to 32% of adults after a year of lamivudine therapy.
Lamivudine is also associated with loss of HBeAg and/or HBe seroconversion (acquisition of anti-HBe antibody along with HBeAg loss). Figure 4 shows data from four different trials of lamivudine, 100 mg/day.5-7,9 The percentage of patients who become HBeAg-negative within a year of therapy has been 32% to 33%, while the percentage who seroconvert has been consistent at about 16% to 18%. FIGURE 4
Figure 5 illustrates the durability of HBeAg loss. Data from clinical trials with a median followup of ≥12 months suggest that patients who achieve HBeAg loss during therapy tend to maintain that loss after therapy is discontinued. In clinical trials in both Asian and Caucasian populations, lamivudine-induced HBeAg loss was maintained in 83% to 86% of patients.10,11 In another clinical trial, interferon-induced HBeAg loss was maintained in 87% of patients.12 In a study of spontaneous HBeAg loss, a somewhat lower maintenance rate of 68% was observed.13
Seroconversion rates tend to increase with time; however, the strongest predictor of lamivudine-induced HBeAg loss and HBe seroconversion is probably the baseline serum ALT. Figure 6 illustrates the effects of time and baseline ALT on seroconversion rates in Asian patients (initially HBeAg-positive) who received prolonged lamivudine therapy. Over time, the percentage of all patients who seroconverted increased from 22% at 1 year to 50% at 5 years (although the number of patients followed for the full 5 years was relatively small). Among patients with a high baseline ALT (>2 x ULN), the likelihood of seroconversion was greater than among those with lower baseline ALT (>1 x ULN). Patients with pretreatment ALT >2 x ULN had a 75% chance of seroconverting after 5 years, compared to 63% of those with pretreatment ALT >1 x ULN.14
Lamivudine has also been used to treat replicative HBeAg-negative hepatitis B (the so-called precore variant). As with HBeAg-positive patients, good viral suppression can be achieved. In one trial (Fig. 7), 68% of patients were negative for HBV DNA after a year of treatment. However, the likelihood of lamivudine resistance (due to development of the YMDD mutation) increases over time, so that by 30 months only about 42% of patients were still HBV DNA-negative.15
Resistance due to the development of YMDD variants is the major limitation of prolonged lamivudine therapy. Although HBV DNA levels at 1 and 3 years are somewhat lower than pretreatment levels in these patients,16,17 lamivudine resistance has been associated with decreased HBe seroconversion rates and worsening liver disease.16-18 In rare cases, death from decompensated liver disease has been reported.19 YMDD variants have also been associated with adverse events (such as graft failure and fulminant hepatitis) in liver transplant patients.20,21
Worsening of the histologic activity index (HAI) was reported in association with lamivudine resistance in a small Asian study (Fig. 8). Of four patients who did not develop resistance, all either maintained their initial improvements or continued to improve after the first year. In contrast, of nine patients who did develop resistance, six had deterioration of their HAI scores after the first year; only three continued to improve or maintained their initial improvements.16
In each patient, the benefits of prolonged lamivudine treatment must be weighed against the risk of developing resistance. The best candidates for lamivudine therapy are patients who are likely to achieve early HBe seroconversion and/or HBV DNA suppression-eg, those with high pretreatment ALT levels (especially those with ALT >5 x ULN). It has been suggested that lamivudine should be stopped as soon as the patient achieves seroconversion with HBV DNA loss (by non-PCR methods) in two consecutive measurements at least 1 month apart.22
Fortunately, another oral agent has been approved for hepatitis B management. Adefovir dipivoxil is a prodrug of adefovir, a nucleotide analog of adenosine monophosphate. Adefovir has potent activity against hepadnaviruses, retroviruses, and herpesviruses. It also has in vitro and in vivo activity against both wild-type and lamivudine-resistant HBV.23 The approved dose of adefovir is 10 mg/day.
Figure 9 summarizes results of a placebo-controlled trial of adefovir in HBeAgpositive patients. With 48 weeks of therapy, 48% of the patients on 10 mg adefovir had normalization of serum ALT; 53% had improved histology; and 12% had HBe seroconversion. These responses were significantly better than those seen in the placebo group (16%, 25%, and 6%, respectively). Median HBV DNA reduction was 3.52 logs-comparable to the reductions seen with lamivudine in previous trials-and, again, significantly better than the placebo response (0.55 logs). No adefovir-related resistance mutations were noted during the study.24
Adefovir, 10 mg/day, suppresses HBV DNA in both HBeAg-positive and HBeAgnegative hepatitis B. Figure 10 shows results of the two pivotal trials that led to FDA approval of adefovir for both HBeAg-positive and HBeAg-negative patients.
Adefovir produced a ≥3.5-log reduction in HBV DNA in both populations; this was significantly better than placebo, and was maintained throughout the 48 weeks of therapy.24,25
Three-year data on adefovir use in HBeAg-negative patients were presented at the 2004 European Association for the Study of the Liver (EASL) Conference. Adefovir produced substantial reduction in HBV DNA, which was maintained for the full 144 weeks of therapy (Fig. 11). This HBV DNA reduction was associated with sustained improvement in serum ALT; from a pretreatment median of 99 IU/L, ALT decreased to normal levels in a majority of patients (Fig. 12).26
FIGURE 12 Adefovir resistance has been observed in long-term studies, but it develops later and at a much lower rate than has been observed with lamivudine. Adefovir resistance is associated with specific mutations (rtN236T and rtA181V) in the viral polymerase gene, occurring with a cumulative probability of 3.9% over 144 weeks. In vitro and preliminary clinical data suggest that these mutants remain susceptible to lamivudine.27 This has important implications for the potential use of both drugs in combination.
The lamivudine and adefovir data presented above have focused on intermediate endpoints of treatment: viral suppression, serum ALT, and liver histology. However, it is also important to know whether these drugs can prevent the longterm complications of hepatitis B-specifically, hepatocellular carcinoma (HCC) and other severe or life-threatening clinical outcomes such as variceal bleeding.
Preliminary results of a randomized trial (lamivudine 100 mg/day vs placebo) were presented at the 2003 AASLD Conference. The primary endpoint in this study was the time to disease progression (defined as Child-Pugh score increase ≥2, development of HCC, spontaneous bacterial peritonitis, bleeding upper GI varices, or liver-related death). Figure 13 is a Kaplan-Meier plot showing the time to disease progression. Over 3 years of treatment, there was a significant reduction in disease progression with lamivudine (9% progressed, compared to 21% of the placebo group). Lamivudine was also associated with significantly less deterioration of the Child-Pugh score (3%, compared to 9% in the placebo group) and with significantly less HCC (4%, compared to 7% with placebo).28
In this trial, 49% of lamivudine-treated patients developed lamivudine-resistant YMDD mutations. Among those who developed YMDD variants, 6.7% (14/209) had a ≥2-point increase in Child-Pugh score (not significantly different from the placebo group with 8.9%). In contrast, Child-Pugh score progression occurred in only about 0.5% (1/221) of patients who did not develop YMDD variants-a significant difference compared to placebo.28
The optimal management of lamivudine resistance has not been established; however, adding or switching to adefovir may be an effective strategy. This was demonstrated in a double-blind study of HBeAg-positive patients with compensated liver disease and genotypic evidence of lamivudine resistance. The patients were randomly assigned to one to three treatments: 1) ongoing lamivudine (100 mg/day), 2) adefovir (10 mg/day), or 3) combination therapy with adefovir plus lamivudine. The effects of those 3 treatments on HBV DNA levels are shown in Figure 14. Adefovir (either alone or in combination with lamivudine) produced a rapid reduction in HBV DNA (statistically significant by week 4), with continued reduction over the 48-week study period (a 3- to 4-log reduction by week 48). In contrast, patients who remained on lamivudine alone had no change in their HBV DNA levels. The difference between the adefovir and combination therapy groups was not statistically significant.29 These results suggest that once lamivudine resistance develops, maintaining lamivudine therapy alone will not be beneficial; however, either switching to adefovir or adding adefovir to the regimen results in potent viral suppression.
The use of lamivudine plus adefovir as initial therapy also has been studied. Patients were randomly assigned to receive either lamivudine plus adefovir or lamivudine plus placebo for 1 year. Initially, both treatments resulted in equally effective viral suppression, with 4- to 5-log HBV DNA reduction by week 16.
However, by week 52 the two arms had diverged. The combination therapy group remained virally suppressed, with a median log reduction from baseline of 5.4; but HBV DNA levels were rising in the lamivudine monotherapy group (Fig. 15). This rise was largely due to the development of lamivudine resistance. If patients who developed lamivudine resistance are excluded, viral suppression in the two treatment groups remained comparable.30
The principles of lamivudine and/or adefovir treatment for chronic hepatitis B may be summarized as follows: