| |
Clinical outcome of HBeAg-negative chronic hepatitis B in relation to virological response to lamivudine
|
| |
| |
Hepatology
Volume 40, Issue 4, October 2004
Vito Di Marco 1 *, Alfredo Marzano 2, Pietro Lampertico 3, Pietro Andreone 4, Teresa Santantonio 5, Piero Luigi Almasio 1, Mario Rizzetto 2, Antonio Craxì 1, fot the Italian Association for the Study of the Liver (AISF) Lamivudine Study Group, Italy
1Cattedra e U.O.C. di Gastroenterologia, Clinica Medica, Università di Palermo, Palermo, Italy
2Dipartimento di Gastroenterologia, Ospedale San Giovanni Battista, Turin, Italy
3Divisione di Epatologia, IRCCS, Ospedale Maggiore, Università di Milano, Milan, Italy
4Dipartimento di Medicina Interna, Cardioangiologia ed Epatologia, Università di Bologna, Bologna, Italy
5Dipartimento di Clinica Medica, Immunologia e Malattie Infettive, Università di Bari, Bari, Italy
Peginterferon Alfa-2a Alone, Lamivudine Alone, and the Two in Combination in Patients with HBeAg-Negative Chronic Hepatitis B
http://www.natap.org/2004/HBV/091604_01.htm
"...After 24 weeks of follow-up, the percentage of patients with normalization of alanine aminotransferase levels or hepatitis B virus (HBV) DNA levels below 20,000 copies per milliliter was significantly higher with peginterferon alfa-2a monotherapy (59 percent and 43 percent, respectively) and peginterferon alfa-2a plus lamivudine (60 percent and 44 percent) than with lamivudine monotherapy (44 percent, P=0.004 and P=0.003, respectively; and 29 percent, P=0.007 and P=0.003, respectively)..."
ABSTRACT/SUMMARY
The effect of lamivudine treatment on the outcome of patients with hepatitis B e antigen (HBeAg)-negative chronic hepatitis is unclear. In a retrospective multicenter study, we have analyzed the virological events observed during lamivudine therapy in patients with HBeAg-negative chronic hepatitis and evaluated the correlation between virological response and clinical outcomes.
Among 656 patients (mean age 49.1 years) included in the database, 54% had chronic hepatitis, 30% had Child-Turcotte-Pugh (CTP) A cirrhosis, and 16% had CTP B/C cirrhosis. On therapy (median 22 months, range 1-66), a virological response was obtained in 616 patients (93.9%). The rate of maintained virological response was 39% after 4 years.
During follow-up, 47 (7.2%) patients underwent liver transplantation, liver disease worsened in 31 (4.7%), hepatocellular carcinoma (HCC) developed in 31 (4.7%), and 24 patients (3.6%) died of liver-related causes. Patients who had cirrhosis and who maintained virological response were less likely than those with viral breakthrough to develop HCC (P < .001) and disease worsening (P < .001).
Survival was better in CTP A patients with cirrhosis and maintained virological response (P = .01 by rank test).
Multivariate analysis revealed that presence of cirrhosis and viral breakthrough were independently related to mortality and development of HCC.
Hepatitic Flares-
Overall, 37 of 209 (17.7%) patients with virological breakthrough had a hepatitis flare with ALT levels 10 or more times the ULN after the emergence of YMDD mutants. Seventeen patients had chronic hepatitis, 15 patients had CTP A cirrhosis and 5 patients had CTP B/C cirrhosis. The incidence of hepatitic flare after virological breakthrough was comparable in patients with chronic hepatitis and in patients with cirrhosis. The hepatitis flare did not cause liver disease worsening in patients with chronic hepatitis. Six of 15 (40%) patients with CTP A cirrhosis who developed hepatitis flares after virological breakthrough had a disease worsening, and 2 (13%) died. Among 5 patients with CTP B/C cirrhosis and hepatitis flares, 3 (60%) died after developing a decompensation of liver disease. The risk of disease worsening after a virological breakthrough was higher in patients with cirrhosis.
Patients with chronic hepatitis did not have a worsening of liver disease after the appearance of YMDD mutants, even when accompanied by an ALT flare. Withdrawal of lamivudine after HBV-DNA reappearance due to YMDD mutant selection did not result in further ALT elevations or disease worsening. When YMDD mutants are selected in patients with cirrhosis, liver worsening and development of HCC are clearly more frequent than in patients with a maintained virological response. Maintained reduction of HBV replication to subliminal levels thus reduces disease progression - even in these patients with advanced disease - and conversely, the abrupt onset of cytolysis due to YMDD mutation or treatment withdrawal may cause rapid deterioration of the residual liver function.
In conclusion, lamivudine is highly effective in reducing viral load in HBeAg-negative patients. After 4 years of therapy, 39% of patients maintain a virological and biochemical response. Loss of virological response may lead to clinical deterioration in patients with cirrhosis.
AUTHOR DISCUSSION
See Results below.
Data on prolonged lamivudine therapy in HBeAg-negative patients with chronic hepatitis or cirrhosis in a real life situation outside clinical trials are scarce. It is not yet clear whether suppressive prolonged lamivudine treatment can slow down or halt the progression of cirrhosis toward liver failure or if it can reduce the incidence of HCC. A major concern is the lack of data on clinical events after YMDD mutant-related breakthrough.
To investigate these issues, the Italian Association for the Study of the Liver has fostered a program of cooperation between liver units throughout Italy. The aim of this program is to collect the large amount of data available from the widespread use of lamivudine both in the premarketing period, when the drug had been made available for compassionate use, and in the postmarketing phase. This approach is obviously fraught with problems, including the retrospective collection of data with potential selection and retrieval biases. In addition, the huge variability in the primary indicator of effectiveness (i.e., HBV-DNA quantification) is an issue because of the use of different tests with a large range of variability in sensitivity and performance from many different laboratories scattered throughout Italy. Nonetheless, it was believed that such sources of potential error could be overcome at least partially by the large cohort size, as already evaluated in large-scale, real-life data collections for hepatitis C treatment.
The present analysis evaluates 656 patients with HBeAg-negative chronic hepatitis, the most frequent type of HBV-related liver disease in Italy. The pretreatment profile of these patients confirms the epidemiological data observed in the Mediterranean area in recent years in terms of sex, mean age, and overall prevalence of cirrhosis (46.2%). In this cohort, a histological diagnosis was available for 515 patients, while cirrhosis had been diagnosed clinically in 141. Approximately one third of the subjects with cirrhosis had advanced disease as evaluated using CTP score. During treatment virological response (HBV-DNA <105 copies/mL or negative non-PCR test) was reported in 94% of patients, and the overall rate of patients with maintained virological response at 4 years was 39%.
Patients with chronic hepatitis did not have a worsening of liver disease after the appearance of YMDD mutants, even when accompanied by an ALT flare. Withdrawal of lamivudine after HBV-DNA reappearance due to YMDD mutant selection did not result in further ALT elevations or disease worsening. Therefore, in these patients lamivudine can be discontinued without any risks of liver disease exacerbation. The rate of development of cirrhosis in these patients cannot be extrapolated from this dataset, because control biopsies at fixed intervals were not performed. Only 4 of 353 (1.3 %) patients without cirrhosis developed HCC, and none of them had maintained a virological response.
When YMDD mutants are selected in patients with cirrhosis, liver worsening and development of HCC are clearly more frequent than in patients with a maintained virological response. Maintained reduction of HBV replication to subliminal levels thus reduces disease progression - even in these patients with advanced disease - and conversely, the abrupt onset of cytolysis due to YMDD mutation or treatment withdrawal may cause rapid deterioration of the residual liver function. This difference is restricted to patients with well-compensated, CTP A cirrhosis, because among those with more advanced liver disease (CTP B or C cirrhosis), short-term survival is more strongly influenced by other factors such as liver failure or portal hypertensive bleeding. The rate of development of HCC in cirrhosis is influenced by lack of virological response: whether this is due to further progression of cirrhosis or to direct viral oncogenic mechanisms cannot be derived from this analysis.
Lamivudine can provide a strong reduction in viral load within the first 2 months of treatment, thus allowing a liver transplant in patients with advanced cirrhosis and high-level viral replication early after starting therapy. Although in some of them virological response induced an improvement in the functional status of liver disease, OLT should not be delayed further than necessary due to the high likelihood of YMDD mutations over time with the ensuing risk of liver decompensation and death.
In conclusion, our data show that, in the presence of advanced cirrhosis, lamivudine should be started when OLT is contemplated over the next few months or if it is possible to add or switch to adefovir therapy. Patients with CTP A cirrhosis may receive long-term lamivudine, but a close biochemical and virological follow-up is needed to diagnose HBV-DNA relapses - which are almost always related to YMDD mutations - as soon as possible, ideally before ALT reactivation. If a virological relapse is evident, these patients should be promptly switched to adefovir or the latter added.
Article Text
Hepatitis B e antigen (HBeAg)-negative chronic hepatitis is caused by hepatitis B virus (HBV) strains with mutations in the pre-core or basic core promoter regions. Such variants are generally common in Southern Europe, but in the last few years the rate of HBeAg-minus mutants has increased worldwide.
Patients with HBeAg-negative chronic hepatitis have a higher risk of disease progression and, when cirrhosis is present, a high incidence of hepatocellular carcinoma (HCC) and liver failure. Suppression of HBV replication obtained by antiviral drugs may lead to clinical improvement and may reduce the incidence of HCC.
The rate of sustained virological response induced by interferon in patients with HBeAg-negative chronic hepatitis is low, and interferon therapy is contraindicated in patients with advanced disease. A short-term course of lamivudine, a nucleoside analogue inhibitor of HBV-DNA polymerase, effectively suppresses serum HBV-DNA levels but induces a low rate of sustained virological response. Prolonged lamivudine therapy obtains suppression of HBV-DNA in patients with chronic hepatitis or cirrhosis and seems to be effective in improving liver function in some patients with decompensated cirrhosis. During long-term lamivudine therapy, HBV strains with mutations in the highly conserved YMDD motif of the HBV-DNA polymerase gene appear at a rate that increases over time. The emergence of YMDD mutants is followed by an increase of HBV-DNA levels in blood and can be associated with sudden exacerbations of hepatitis. It is not known whether prolonged lamivudine therapy could reduce the incidence of HCC or the progression of cirrhosis. More crucially, the relationship between severe reactivation of disease after the appearance of YMDD mutations and the risk of liver failure has yet to be established outside the strictly controlled environment of clinical trials.
To acquire more data on these issues, 20 Italian liver units involved in the management of HBV-related chronic liver disease have joined their databases of patients treated with lamivudine over prolonged periods. The objective of the study was to analyze the relationship between virological response and clinical outcomes in patients with HBeAg-negative chronic hepatitis on lamivudine treatment.
Patient Selection
This retrospective analysis encompassed 656 patients with HBeAg-negative chronic hepatitis with or without cirrhosis on lamivudine monotherapy. Clinical features from 980 HBsAg-positive patients with chronic liver diseases who started lamivudine therapy between 1995 and 2002 in 20 Italian liver units were recorded in a centralized database. The diagnosis of chronic hepatitis was based on a liver biopsy in all patients, while that of cirrhosis was based on liver biopsy features or, if unavailable, on clinical, laboratory, and ultrasound data. Three hundred twenty-four patients were excluded from the analysis because of one or more of the following reasons: HBeAg positivity (n = 151); age below 16 years (n = 2); possession of anti-hepatitis D virus antibodies (n = 52); possession of anti-hepatitis C virus antibodies (n = 43); diagnosis of HCC (histological diagnosis or ultrasound evidence of hepatic lesions and --a-fetoprotein levels >400 ng/mL) (n = 47); or reception at some time of combination therapy with interferon and lamivudine (n = 92). Patients with HIV coinfection were excluded a priori from the database. All patients were determined to be HBV-DNA-positive via non-polymerase chain reaction (PCR) assay prior to treatment. Patients received open-label lamivudine administered orally in single daily doses of 100 mg - or, in a minority of cases, 150 mg. Because this was a retrospective analysis of treated patients, there were no predefined criteria for treatment withdrawal. Treatment withdrawal because of virological nonresponse was based on the persistence of serum HBV-DNA using non-PCR methods. Criteria for withdrawal after virological response were dependent upon the strategy used by each center.
Data Collection
The database recorded the demographic features (age, sex), previous courses of interferon-a or lamivudine therapy, diagnosis of liver disease, Child-Turcotte-Pugh (CTP) score in patients with cirrhosis, quantitative HBV-DNA and alanine aminotransferase (ALT) values at the start of therapy and the duration of lamivudine therapy. We recorded all virological events (time of virological response, virological breakthrough, and the emergence of YMDD mutations) and biochemical events (time of ALT normalization, ALT breakthrough, hepatitis flare with ALT levels >10 times the upper limit of normal[ULN]) during therapy and follow-up observations. The clinical events recorded were the worsening of liver disease defined as an increase of 2 or more points of CTP score or clinical decompensation (ascites, portal hypertensive bleeding, and hepatic encephalopathy), de novo diagnosis of HCC, liver-related death, or liver transplantation during the period of observation. The diagnosis of HCC during lamivudine therapy was confirmed at all centers in all cases with ultrasound-guided fine needle biopsies.
Results
Patient Features
Three hundred fifty-three (53.8%) patients had a histological diagnosis of chronic hepatitis, with a median time since the last liver biopsy of 14 months (range 1-43); 162 (24.7%) patients had a histological diagnosis of cirrhosis, with a median time since liver biopsy of 39 months (range 1-210). In 141 (21.5%) patients the diagnosis of cirrhosis was based on clinical features. At the start of therapy, 198 patients had CTP A cirrhosis, 66 patients had CTP B cirrhosis, and 39 patients had CTP C cirrhosis.
Most patients were males (83.4%), and the patients with cirrhosis were older than those without cirrhosis (mean age 52.1 vs. 46.6 years). Two hundred sixty-three (40%) patients had been treated with interferon-a in the previous years without obtaining a maintained virological response. Seventy-nine patients (12.1%) had received a course of lamivudine not longer than 12 months in the previous years. Those patients had a virological relapse after discontinuation of lamivudine therapy without evidence of YMDD mutations. The median time on lamivudine therapy was 22 months (range 1-66) and treatment had been given on average for longer periods in patients with chronic hepatitis (median time 23.5 months) and CTP A cirrhosis (median time 24 months) than in patients with CTP B/C cirrhosis (median time 11 months).
Efficacy of Therapy and Virological Breakthrough
Lamivudine induced a virological and a biochemical response in 616 of 656 patients (93.9%). The median time of virological response was 2 months (range 1-20), and the median time of ALT normalization was 3 months (range 1-27).
Forty patients (6.1%) did not achieve a virological response during treatment. Ten patients (5 patients with chronic hepatitis and 5 patients with CTP A cirrhosis) withdrew from treatment (mean time 14.3 months, range 8-18) without reaching a virological response, and 2 patients with CTP A cirrhosis developed HCC and died after 16 and 17 months. Four patients with CTP C cirrhosis died from liver failure, and 3 patients with CTP C cirrhosis underwent a liver transplant during the first 6 months of therapy. Twenty-one patients (12 with chronic hepatitis and 9 with cirrhosis) were treated for less than 6 months and were still HBV-DNA positive at the last available assessment.
During treatment, a virological breakthrough associated with ALT increase occurred in 209 of the 616 patients who achieved a virological response. The median time of virological breakthrough was 18 months (range 6-45). Mutations of the YMDD motif were found in 97% of 209 patients with virological breakthrough.
Virological response was maintained after 1 year of lamivudine therapy in 546 of the 616 (88.6 %) patients who had achieved an initial response. This rate decreased to 63.3% after 2 years, to 48% at 3 years and to 39% at 4 years, without differences to the initial diagnosis.
Disease Course
During the observation period (median 23.5 months, range 1-81), 31 of 656 (4.7%) patients experienced worsening of liver disease (increase of >=2 points of CTP score or liver decompensation), and 31 (4.7%) patients developed HCC. Forty-seven patients (7.2%) underwent a liver transplant and 26 patients (3.9%) died from liver-related causes.
Hepatocellular Carcinoma
Thirty-one patients (4.7%) developed HCC during follow-up. Four patients had a histological diagnosis of chronic hepatitis before starting lamivudine; 3 of them were males (mean age 56 years). One of them did not achieve virological response, and 3 patients developed HCC after emergence of YMDD mutants. Twenty-seven patients (23 males, mean age 56 years) had cirrhosis. Ten of them developed HCC on virological response (mean time of therapy 15 months, range 2-39), and 17 patients developed HCC after emergence of YMDD mutations (mean time of therapy 23.5 months, range 7-36). The overall incidence of HCC among patients with cirrhosis was 4.6%, 8.7%, 19.8%, and 22.1% at 1, 2, 3, and 4 years of therapy, respectively, without significant differences in CTP class. The likelihood of developing HCC was significantly less for patients with cirrhosis with maintained virological response than patients with virological breakthrough (P < .001).
Disease Worsening
A decrease of 2 or more points of CTP score was observed in 1 patient with CTP A cirrhosis and 3 patients with CTP B cirrhosis. One or more disease events (other than HCC) occurred in 27 patients (4.1%) during follow-up. Disease events did not occur among patients with chronic hepatitis. Among 198 patients with CTP A cirrhosis, 8 patients developed ascites and 1 patient had an episode of portal hypertensive bleeding. Among 105 patients with CTP B/C cirrhosis, 14 patients developed ascites and 4 patients had an episode of portal hypertensive bleeding. Disease worsening was seen in 21 of 88 (23.9%) patients with cirrhosis with virological breakthrough and in 7 of 194 (3.6%) patients with cirrhosis with maintained virological response (P < .001 by rank test).
Survival
During the period of observation, 4 patients died from unrelated liver disease causes. Twenty-four patients (3.6%) died from liver-related causes during the observation. One patient with chronic hepatitis died after development of HCC. In the group with CTP A cirrhosis, 7 patients died after development of HCC, 1 patient died after portal hypertensive bleeding, and 4 patients died of liver failure. In the group with CTP B/C cirrhosis, 1 patient died after development of HCC, 2 patients died after portal hypertensive bleeding, and 9 patients died of liver failure.
The probability of survival in CTP A patients with cirrhosis (Fig. 4A) was higher if the virological response was maintained (P = .01 by rank test). In contrast, among patients with CTP B/C cirrhosis the probability of survival was not affected by virological response.
Liver Transplantation
Orthotropic liver transplantation (OLT) was performed in 47 patients (7.2%). At the time of OLT, 5 patients were in CTP class A, 22 were in CTP class B, and 20 were in CTP class C. The mean interval between starting lamivudine and OLT was 6.3 months (range 1-31); after OLT all patients continued lamivudine therapy associated with hepatitis B immunoglobulin. Two of 20 patients with CTP C cirrhosis died after liver transplant under maintained virological response.
Hepatitic Flares
Overall, 37 of 209 (17.7%) patients with virological breakthrough had a hepatitis flare with ALT levels 10 or more times the ULN after the emergence of YMDD mutants. Seventeen patients had chronic hepatitis, 15 patients had CTP A cirrhosis and 5 patients had CTP B/C cirrhosis. The incidence of hepatitic flare after virological breakthrough was comparable in patients with chronic hepatitis and in patients with cirrhosis (Fig. 5). The hepatitis flare did not cause liver disease worsening in patients with chronic hepatitis. Six of 15 (40%) patients with CTP A cirrhosis who developed hepatitis flares after virological breakthrough had a disease worsening, and 2 (13%) died. Among 5 patients with CTP B/C cirrhosis and hepatitis flares, 3 (60%) died after developing a decompensation of liver disease. The risk of disease worsening after a virological breakthrough was higher in patients with cirrhosis.
Liver Transplantation
Orthotropic liver transplantation (OLT) was performed in 47 patients (7.2%). At the time of OLT, 5 patients were in CTP class A, 22 were in CTP class B, and 20 were in CTP class C. The mean interval between starting lamivudine and OLT was 6.3 months (range 1-31); after OLT all patients continued lamivudine therapy associated with hepatitis B immunoglobulin. Two of 20 patients with CTP C cirrhosis died after liver transplant under maintained virological response.
Treatment Withdrawal
Lamivudine therapy was withdrawn in 82 patients (12.5%). None of the 58 patients with chronic hepatitis who withdrew from treatment developed clinical events after stopping therapy (mean time of follow-up 16.8 months). Among patients with cirrhosis, 5 withdrew from treatment (mean time 14.3 months) because they did not achieve a virological response, and 5 patients stopped treatment (mean time 14.6 months) under virological and biochemical response. None of them developed liver disease worsening during further follow-up (mean time of follow-up 12.5 months).
Fourteen of 88 patients with cirrhosis and virological breakthrough stopped lamivudine after a mean time of 22.1 months of treatment; their mean time of follow-up after treatment withdrawal was 16.9 months. Five of them developed a decompensation after treatment withdrawal, and 4 died. The course of liver disease was comparable to 74 of 88 patients who continued lamivudine therapy despite evidence of virological breakthrough (mean follow-up after virological breakthrough 19.8 months).
Multivariate Analysis
To assess the likelihood of development of HCC and of death, variables were fitted into a Cox regression model to identify potential predictors of clinical events. Among predicting variables, patient's age (RR 1.068), CTP A cirrhosis (RR 8.768), CTP B/C (RR 11.921) cirrhosis, and virological breakthrough (RR 2.962) were significantly and independently related to development of HCC. Mortality was linked to diagnosis of CTP A (RR 18.712) and CTP B/C (RR 85.386) cirrhosis and virological breakthrough (RR 2.902).
|
|
| |
| |
|
|
|
