| |
"Chemoprevention of hepatocellular carcinoma in chronic hepatitis B with lamivudine?"
|
| |
| |
Gastroenterology
Volume 129, Issue 6, Pages 2120-2122 (December 2005)
Lei Y. Lim, MD, Kris V. Kowdley, MD
".....This landmark study is the first controlled trial demonstrating that long-term suppression of HBV with lamivudine can delay disease progression and reduce the incidence of HCC....
"......this important study demonstrates unequivocal evidence that treatment to reduce viral replication decreases the incidence of complications and HCC in chronic HBV. It deserves recognition as the first prospective, randomized, controlled clinical trial to demonstrate a favorable effect of viral suppression on disease progression......
.... Patients with YMDD mutations were more likely to have an increase in Child-Pugh score when compared to those who did not.... Because the greatest benefit was obtained among patients who did not develop resistance, it is exciting to speculate that these results could be further improved with newer antiviral agents with higher potency and lower resistance profiles than lamivudine.....
..... We note that 22% of the patients in this study had normal serum ALT levels; furthermore, serum ALT was not predictive of a response. Based on these findings, it is reasonable to conclude that treatment with effective antiviral therapy is indicated in patients with advanced hepatic fibrosis and active viral replication related to chronic HBV regardless of serum ALT level....."
Article Outline
Liaw Y-F, Sung JJY, Chow WC, Farrell G, Lee C-Z, Yuen H, Tanwandee T, Tao Q-M, Shue K, Keene ON, Dixon JS, Gray DS, Sabbat J, for the Cirrhosis Asian Lamivudine Multicentre Study Group (Chang Gung University, Taipei, Taiwan; Prince of Wales Hospital, Hong Kong; Singapore General Hospital, Singapore; University of Sydney, Sydney, Australia; National Taiwan University Hospital, Taipei, Taiwan; Princess Margaret Hospital, Hong Kong; Siriraj Hospital, Bankok, Thailand; People's Hospital, Beijing, China; and GlaxoSmithKline, Singapore and UK). Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004;351:1521-1531.
It is estimated that 2 billion people have been exposed to hepatitis B virus (HBV) worldwide and 350 million people have chronic HBV infection (Vaccine 1990;8[Suppl]:S18-S20, WHO Fact Sheet 204.2000). Despite the utilization of HBV immunization since 1981, chronic HBV remains a problem in areas where the infection is endemic (Bull WHO 1988;66:443-455, Lancet 1996;348:696). In populations where chronic HBV infection is endemic, the virus is primarily transmitted during childbirth or early childhood. In contrast to adults among whom only 3%-5% develop chronic infection, 95% of children infected with HBV will go on to develop chronic HBV and are at risk for cirrhosis or hepatocellular carcinoma (HCC) (J Infect Dis 1985;151:599-603). HCC can develop in patients with chronic HBV even in the absence of cirrhosis, although the risk is much higher among cirrhotic patients (Lancet 1981;2:1129-1133, Semin Liver Dis 1984;4:113-121). Compared with controls, lamivudine has shown efficacy to improve liver histologic response (52% vs 23%; P < .001), seroconversion from HBeAg to HBeAb (17% vs 6%; P = .04), and normalization of serum ALT (41% vs 7%; P < .001) in a year (N Engl J Med 1999;341:1256-1263). However, previous studies have not shown a reduction in the incidence of HCC or clinical disease progression. The aim of the present study was to determine whether treatment with lamivudine was associated with improvement in the above-noted endpoints in patients with advanced hepatic fibrosis (Ishak stage 4-6) secondary to chronic HBV.
Liaw et al conducted a randomized, double-blinded, placebo-controlled, parallel group multicenter trial in which patients were randomized in a 2:1 ratio to receive 100 mg of lamivudine or placebo. Treatment during the double-blind phase of the study was continued until one of the following clinical endpoints developed: an increase of 2 points in Child-Pugh score; spontaneous bacterial peritonitis with sepsis; renal insufficiency (CrCl <50 mL/min); gastric or esophageal variceal bleeding; development of HCC and liver disease-related death; or HBeAg seroconversion. Patients who reached a clinical endpoint point as well as those with HBeAg seroconversion were offered open-label lamivudine for 1 year. The study design allowed for patients to receive open-label lamivudine therapy if the trial was terminated early (before 60 months) according to predefined criteria. Exclusion criteria included: a known diagnosis of HCC; serum ALT >10 times upper limit of normal (ULN); autoimmune hepatitis; coinfection with hepatitis C or D or human immunodeficiency virus. Those with a hemoglobin level <8 g/dL; elevated serum creatinine; white-cell count <1500 per cubic millimeter; platelet count <50,000; immunomodulatory treatment or chronic antiviral therapy within the past 6 months; any investigational drug within the past 30 days; any previous therapy with lamivudine; pregnant women were also excluded.
A total of 651 patients from 41 centers in Australia, China, Hong Kong, Malaysia, New Zealand, the Philippines, Singapore, Taiwan, and Thailand were enrolled; 85% of the patients were male and 98% were Asian. The two groups were comparable with regard to Ishak fibrosis score, baseline laboratory values, and age; 58% in both groups were HBeAg positive at baseline. HBV DNA was determined by a branched-chain hybridization assay (versant HBV DNA Quantitative Assay Bayer Diagnostics, with a lower limit of detection of 0.7 mEq/mL). YMDD mutations were evaluated by PCR and restriction fragment length polymorphism assay at baseline, annually or when a clinical endpoint was reached. A Cox proportional-hazards model was used to compare treatments allowing for the covariates of country, sex, baseline ALT, Child-Pugh, and fibrosis scores. The first interim analysis took place at 18 months and the subsequent analysis occurred between 6 and 12 months later.
The study was stopped early after a median follow-up of 32.4 months after the second interim analysis because of significant improvement in the treatment group. At the end of the study, 67 patients achieved HBeAg seroconversion, 52 stopped therapy for other reasons, and there were 68 endpoints, of which 34 of 436 (7.8%) were in the treatment group, and 38 of 215 (17.7 %) were in the placebo arm. The majority of the end points consisted of increase in Child-Pugh score and development of HCC. Two patients developed renal insufficiency and 5 developed bleeding varices, but there were no cases of spontaneous bacterial peritonitis or death related to liver disease. Fifteen patients (3.4%) in the lamivudine group compared to 19 patients (8.8%) in the placebo group (P = .02) had decompensation of liver disease; 17 patients (3.9 %) in the lamivudine group and 16 patients (7.4%) in the placebo group developed HCC (P = .047). YMDD mutations were detected in 49% of the treatment group versus 5% of the placebo group, the majority of whom had evidence of rebound HBV viremia. Patients with YMDD mutations were more likely to have an increase in Child-Pugh score when compared to those who did not, although fewer endpoints were reached when compared to those in the placebo arm (P > .05). The authors concluded that lamivudine reduces the risk of hepatic decompensation and HCC among patients with chronic HBV and advanced hepatic fibrosis.
Comment
This landmark study is the first controlled trial demonstrating that long-term suppression of HBV with lamivudine can delay disease progression and reduce the incidence of HCC. Previous studies have shown that chronic HBV patients treated with interferon-alpha who lose HBeAg are significantly less likely to have hepatic decompensation compared with patients who remain HBeAg positive (N Engl J Med 1996;334:1422-1427). A retrospective analysis of HBeAg-negative chronic HBV patients treated for 4 years with lamivudine found a significantly lower incidence of HCC among patients with sustained suppression of HBV DNA compared to those who experienced a virologic rebound (Hepatology 2004;40:883-891).
There are several therapies currently approved for the treatment of chronic HBV in this country including interferon-alfa (standard and pegylated), lamivudine, adefovir dipivoxil, and entecavir. There are many criteria for determining candidacy for treatment in patients with chronic HBV including serum ALT level, HBV DNA level, liver function, and degree of hepatic necroinflammation and fibrosis on histology. Historically, the goal of therapy in chronic HBV had been to improve liver histology and was the main endpoint used by the Food and Drug Administration to evaluate response to therapy in several phase III trials of antiviral agents. More recently, it has been recognized that the level of viral replication, as measured by serum HBV DNA titer, is a marker for liver disease progression (Hepatology 2003;37:1309-1319). Therefore, the primary goal of treatment in chronic HBV has been increasingly focused on suppression of viral replication. However, until the current study, there has been a lack of controlled data confirming that suppression of viral replication is in fact associated with a reduction in the incidence of disease progression. Another current controversy regarding the treatment of chronic HBV is whether patients with normal or modest elevation in serum ALT should be treated with antiviral therapy. It is recognized that such patients have a lower likelihood of HBeAg seroconversion (Hepatology 2002;36:186-194, Hepatology 1999;30:770-774). We note that 22% of the patients in this study had normal serum ALT levels; furthermore, serum ALT was not predictive of a response. Based on these findings, it is reasonable to conclude that treatment with effective antiviral therapy is indicated in patients with advanced hepatic fibrosis and active viral replication related to chronic HBV regardless of serum ALT level.
An interesting and possibly unexpected finding in this study was the significant reduction in the incidence of HCC. The hazard ratio for development of HCC was 0.47 (P = .047) in the lamivudine-treated patients compared with placebo-treated patients. After exclusion of patients who developed HCC within the first year after randomization (to eliminate possible "lead-time bias"), the hazard ratio for development of HCC among lamivudine-treated patients was 0.49 (P = .052).
The optimal duration of treatment with antiviral agents in patients with chronic HBV is unknown. Some practice guidelines and consensus statements from learned societies have suggested that treatment in HBeAg-positive patients need only be continued for a short time period (3-6 months) past the point of seroconversion regardless of baseline and have not specifically recommended long-term, indefinite therapy for patients with cirrhosis (Liver Int 2005;25:472-489, Hepatology 2001;34:1225-1241, Hepatology 2004;39:857-861), whereas others have recommended that treatment should be long-term and indefinite even among patients with compensated cirrhosis (Clin Gastroenterol Hepatol 2004;2:87-106). By contrast, long-term treatment has been recommended for HBeAg-negative patients regardless of the presence or absence of cirrhosis (Hepatology 2004;39:857-861).
In the present study, 376 HBeAg-positive patients were enrolled at baseline. Among patients in the placebo group, likelihood of disease progression was significantly higher compared with the lamivudine group (6% vs 20%), resulting in an absolute risk reduction of 14%. The authors reported that overall, 67 patients achieved seroconversion but did not clarify whether there was a difference in the rate of seroconversion between those who responded and those who did not; in addition, they did not describe whether the response to treatment was independent of seroconversion. If, in fact, most of the benefit from lamivudine therapy is related to seroconversion, it would be difficult to advocate for long-term therapy among HBeAg-positive patients even after achievement of HBeAg seroconversion. On the other hand, if HBeAg seroconversion was unrelated to treatment response, this observation would provide support to advocates of long-term, indefinite therapy among cirrhotic patients with active HBV replication, even after HBeAg seroconversion has occurred.
Another important finding in this study was the observation that after the median duration of treatment was 32.4 months, 49% of the patients in the lamivudine group developed a YMDD mutation, 62% of whom had evidence of virologic rebound. Patients who developed YMDD mutation were also more likely to experience hepatic decompensation albeit at a lower rate than those patients in the placebo group.
The incidence of YMDD mutations increases over time with lamivudine treatment at a rate of 17%, 39%, and 57% after 1, 2, and 3 years of treatment, and approaching 70% after 4 years of therapy among Asian patients (Hepatology 2001;33:1527-1532, N Engl J Med 1998;339:61-68, Gastroenterology 1998;114:A1289, Hepatology 1999;30:420A, J Gastroenterol Hepatol 2004;19:1276-1282). The initial benefits of treatment with lamivudine seem to be lost over time among patients who develop YMDD mutations, as an increase in serum ALT and HBV DNA viral loads to pretreatment levels often follows. Patients with cirrhosis are also more likely to develop liver failure after appearance of YMDD mutations (Antivir Ther 2005;10:431-439). Emergence of YMDD mutations may also cause reversal of initial histological improvement (Gastroenterology 2003;124:105-117). Given the significant differences in treatment response between lamivudine-treated patients who did not develop YMDD mutations compared with those who did, it is interesting to speculate that use of newer antiviral agents such as adefovir or entecavir may have improved the treatment response even further in the Liaw study.
A criticism of the Liaw study has been the possibility that "lead-time bias" associated with screening may have led to a favorable result in the lamivudine-treated group. HCC was found in 3.9% who received lamivudine and 7.4% who received placebo with hazard ratio of 0.49 (P = .047). Five patients developed HCC during the first year of the study, 3 in the treatment arm and 2 in the placebo arm. Excluding these patients, the hazard ratio with treatment changed from 0.49 to 0.47 (P = .052). Some have faulted the authors for terminating the study prematurely given this finding and have suggested that a longer treatment period may have provided more definitive support for the role of lamivudine therapy; however, we suspect that most clinicians and practicing physicians would have agreed with the authors' decision to stop the study on the basis of achievement of efficacy.
In summary, despite some lingering unanswered questions, this important study demonstrates unequivocal evidence that treatment to reduce viral replication decreases the incidence of complications and HCC in chronic HBV. It deserves recognition as the first prospective, randomized, controlled clinical trial to demonstrate a favorable effect of viral suppression on disease progression. Because the greatest benefit was obtained among patients who did not develop resistance, it is exciting to speculate that these results could be further improved with newer antiviral agents with higher potency and lower resistance profiles than lamivudine.
Note from Jules Levin: here are several key points I excerpted from the original published study in NEJM.
AUTHOR CONCLUSION: Continuous treatment with lamivudine delays clinical progression in patients with chronic hepatitis B and advanced fibrosis or cirrhosis by significantly reducing the incidence of hepatic decompensation and the risk of hepatocellular carcinoma..... The most important finding of this study is that lamivudine reduces the risk of liver complications for patients with chronic hepatitis B and cirrhosis or advanced fibrosis. The magnitude of protection conferred by lamivudine is substantial, with a reduction of approximately 50 percent in disease progression during a median period of 32 months of treatment..... patients with YMDD mutations were more likely to have an increase in the Child-Pugh score and to die for reasons related to clinical end points than were those patients who did not have YMDD mutations. This may be because the resumption of viral replication restores the potential for facilitating disease progression. The long-term effects of lamivudine on disease progression are not known. Since the present trial was started, treatment with a combination of adefovir dipivoxil and lamivudine has been shown to suppress replication of YMDD mutations and improve liver function in patients with hepatic decompensation.
Hepatocellular carcinoma developed in five patients during the first year of the study, two in the placebo group and three in the lamivudine group. Even if these tumors had existed but had not been detected before study entry, the exclusion of the patients would not have affected the result of the primary analysis of time to disease progression. However, for the time to a diagnosis of hepatocellular carcinoma, the hazard ratio changed from 0.49 (P=0.047) to 0.47 (P=0.052).
Covariate modeling of time to disease progression showed that the factors other than treatment that significantly affected outcome were the Child-Pugh score at baseline and the Ishak fibrosis score at baseline. In both instances, higher scores were associated with a greater frequency of end points.
YMDD Mutations
Two patients had evidence of YMDD mutations at baseline, and 5 patients had no samples after baseline, so data on the emergence of YMDD mutations during therapy were available for 644 patients. Genotypic resistance YMDD mutations developed in 49 percent of the patients treated with lamivudine, and the Child-Pugh score was more likely to increase in patients with these mutations than in the other patients treated with lamivudine (7 percent vs. <1 percent). After baseline, at least one sample with evidence of YMDD mutations was found in 209 of 430 patients (49 percent) in the lamivudine group and 11 of 214 patients (5 percent) in the placebo group. Only 5 percent of patients without YMDD mutations had detectable HBV DNA breakthrough, as compared with 62 percent of patients with YMDD mutations in the lamivudine group.
Patients in the lamivudine group who had YMDD mutations were more likely to have an increased Child-Pugh score than those without YMDD mutations (P<0.001), but they were less likely to reach an end point than were patients in the placebo group (P>0.05).
Overall, 12 percent of the patients in the lamivudine group and 18 percent of the patients in the placebo group reported serious adverse events.
|
|
| |
| |
|
|
|
