- Email: [email protected]
Clinical Events After Cessation of Lamivudine Therapy in Patients Recovered From Hepatitis B Flare With Hepatic Decompensation
Clinical Gastroenterology and Hepatology 2015;13:979–986
Clinical Events After Cessation of Lamivudine Therapy in Patients Recovered From Hepatitis B Flare With Hepatic Decompensation Ming-Ling Chang, Wen-Juei Jeng, and Yun-Fan Liaw Liver Research Unit, Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Taipei, Taiwan BACKGROUND & AIMS:
Before guidelines were issued, many patients with hepatitis B flare and hepatic decompensation had discontinued lamivudine therapy instead of indefinite therapy. We investigated their outcomes.
METHODS:
We performed a retrospective cohort study of 263 consecutive patients with chronic hepatitis B (94 with cirrhosis) who recovered from a flare of hepatitis with hepatic decompensation and were followed after cessation of lamivudine therapy. Clinical events that occurred during the follow-up period were assessed by chart review and analysis of results from retrospective assays.
RESULTS:
The mean duration of lamivudine therapy was 12.1 – 8.6 months; data were collected from patients for 89.1 – 38.7 months after therapy ended. In the first year off therapy, 29.9% of patients had clinical relapse, 16.2% had hepatitis flares, and 8.2% had hepatic decompensation. There was no significant difference in the incidence of hepatic decompensation between patients with and without cirrhosis. Hepatocellular carcinoma developed in 14 patients 20–109 months after cessation of therapy, with 5-year cumulative incidence of 5.2% in patients with cirrhosis. Three patients with cirrhosis died of hepatic decompensation 38–76 months after cessation of therapy (5year cumulative mortality, 2.9%). Multivariate analyses showed that men were more likely than women to have recurrence of hepatic decompensation (hazard ratio [HR], 4.339; P [ .014). Liver cirrhosis (HR, 2.766; P [ .041) and age (HR, 1.054; P [ .023) increased risk for hepatocellular carcinoma.
CONCLUSIONS:
Cessation of lamivudine therapy after recovery from hepatitis B flare with decompensation was safe for most patients. However, 8.2% develop decompensation within 1 year and can be rescued by timely retreatment. With close monitoring, the stopping strategy could be a feasible alternative to indefinite therapy, especially in low resource settings.
Keywords: Antiviral Therapy; Cirrhosis; Liver Cancer; Nucleos(t)ide Analog.
hronic hepatitis B virus (HBV) infection has potential adverse outcomes including hepatic decompensation (HD), cirrhosis, and hepatocellular carcinoma (HCC).1 Hepatitis flare with HD in chronic hepatitis B (CHB) patients is a serious complication with high mortality,2 so that requires immediate nucleos(t)ide analog (Nuc) therapy3–6; most experts recommend indefinite or even lifelong therapy.4,6 Of note, indefinite or lifelong Nuc therapy has problems of concern, including cost, lack of long-term safety data beyond 5–10 years, negligible rate of hepatitis B surface antigen (HBsAg) loss,7 and emergence of drug resistance.8 Furthermore, the 1-year Nuc persistent rate was only 73.4% in new patients,9 and the medication possession rate 80% was only 53.7% in patients treated with entecavir (ETV) or tenofovir (TDF).10 This may bring CHB patients to unpredictable hazards. Weighted against all these potential problems, Asian Pacific guidelines have
C
recommended that cessation of Nuc therapy can be considered in hepatitis B e antigen (HBeAg) negative patients if undetectable HBV DNA has been documented on 3 occasions at least 6 months apart.5 By using this stopping rule, we have shown that ETV therapy can be safely discontinued in HBeAg-negative CHB patients, even with compensated cirrhosis, under proper off-therapy Abbreviations used in this paper: ADV, adefovir; AFP, alpha fetoprotein; ALT, alanine aminotransferase; anti-HBe, hepatitis B e antibody; cccDNA, covalently closed circular DNA; CHB, chronic hepatitis B; dPT, prolongation of prothrombin time; ETV, entecavir; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HD, hepatic decompensation; HDV, hepatitis delta virus; HR, hazard ratio; LAM, lamivudine; Nuc, nucleos(t)ide analogue; TDF, tenofovir; ULN, upper limit of normal; US, ultrasonography. © 2015 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.10.023
980
Chang et al
monitoring.11 Whether this stopping strategy is applicable to patients with HD remains to be studied. Before the Asian Pacific stopping rule was issued, lamivudine (LAM) was widely used and was discontinued in most of our patients after recovery from HD because of the concerns about drug resistance and cost/reimbursement issues. These patients were followed up off-therapy. Their medical records were available for reviewing what happened after cessation of LAM therapy. Considering that the dogma of indefinite/lifelong Nuc therapy in patients with HD was based on expert opinion rather than solid scientific evidence, we conducted a retrospective cohort study to examine the off-LAM events with survival analyses in CHB patients recovered from hepatitis flare with HD.
Patients Excluding patients with other causes of liver disease such as hepatitis C virus (HCV) or hepatitis delta virus (HDV) infection, alcohol, autoimmune, and malignancy and those with pregnancy or breastfeeding, severe pancreatitis, prior liver transplantation, or need for mechanical ventilation, 318 consecutive CHB patients (18 years old) who had hepatitis flare with HD were treated with 100 mg LAM daily in our unit. Forty-one of them died, mostly within 6 months after starting LAM therapy, and 14 underwent liver transplantation. The remaining 263 patients recovered during LAM therapy and were followed up after cessation of therapy (Figure 1). HD was defined as a severe clinical syndrome with hepatic function impairment as indicated by jaundice and a prolonged prothrombin time (dPT) and/or occurrence of ascites/encephalopathy in patients with or without cirrhosis, as described elsewhere since 1980s.12,13 LAM treatment was discontinued in all patients with clinical recovery (normalization of prothrombin time and bilirubin level) and at a time point at the discretion of their physician, usually within reimbursement allowance (18 months before November 1, 2009, 36 months thereafter). Then the patients were followed up every 1–6 months, more frequently within the first 3 months and when alanine aminotransferase (ALT) became abnormal. Follow-up studies included clinical assessment, liver biochemical tests, and serum HBV DNA when appropriate. Alpha fetoprotein (AFP) and abdominal ultrasonography (US) were performed every 3–6 months for HCC surveillance. The primary events of concern were life-threatening safety issues including HD and mortality. The secondary events of concern included clinical relapse, hepatitis flare, or HCC. Retreatment was instituted when indicated. Clinical relapse was defined as virologic relapse (serum HBV DNA >2000 IU/mL) plus ALT elevation greater than 2 times upper limit of normal (ULN).5 Hepatitis flare was defined as abrupt rise of serum ALT to level greater than 5 ULN.14 Cirrhosis was diagnosed by histologic findings or repeated US findings consistent with cirrhosis, supplemented with clinical features such as varices and
Clinical Gastroenterology and Hepatology Vol. 13, No. 5
thrombocytopenia, as described elsewhere.15 HCC was diagnosed by histology/cytology or US findings plus high AFP level or imaging findings of enhanced arterial contrast uptake followed by washout in the portal venous phase and equilibrium phase (dynamic computed tomography or magnetic resonance imaging), as described in generally accepted guidelines.16 HCC that developed within 1 year after cessation of therapy was considered as preexisting and therefore was excluded as an off-therapy event.17
Methods Liver biochemical tests were performed by using routine automated techniques. Serum HBsAg, HBeAg/ anti-HBe, and anti-HDV were assayed by using radioimmunoassay kits (Abbott Diagnostics, North Chicago, IL). Anti-HCV was assayed by using a commercial enzyme immunoassay (Abbott Diagnostics). HBV genotype was determined retrospectively by using polymerase chain reaction–restriction fragment length polymorphism of the surface gene of HBV. Serum HBV DNA level was measured by using the hybrid Capture II assay (Digene Corp, Gaithersburg, MD; lower limit of detection, 1.4 105 copies/mL) before April 2007. Because Digene assay reported HBV DNA level in copies/mL, we divided the levels by a factor of 5 to become IU/mL in the analyses of this study. Ultrasensitive polymerase chain reaction assay (Cobas Amplicor HBV Monitor, lower limit of detection, 300 copies/mL; or the Roche COBAS TaqMan HBV Test, lower limit of detection, 69 copies/mL, or 12 IU/mL; Roche Diagnostics, Pleasanton, CA) was used for stored serum samples. Gender, age, HBV DNA level, HBeAg status, HBV genotype, presence of cirrhosis, ALT, bilirubin, dPT, albumin, creatinine, as well as treatment duration were used for comparisons where appropriate.
Statistics The follow-up period was calculated from cessation of LAM therapy until the last visit of the patient or when an event occurred. The follow-up period was censored when treatment was resumed in the secondary analyses. Events in patients with and without cirrhosis were compared. All statistical analysis was performed by using Statistics Package for Social Science software (version 18.0; SPSS Inc, Chicago, IL). To compare between groups, continuous variables were analyzed by using Student t test or Mann– Whitney U test, whereas categorical variables were analyzed by using c2 test or Fisher exact test when appropriate. Kaplan–Meier and univariate Cox regression were used to assess the relationship of various variables to the events. Variables found to be significant were included in multivariate Cox regression models. Statistical significance was defined at the 5% level on the basis of two-tailed test of the null hypothesis. The study was approved by the hospital institutional review board.
May 2015
Stop Nuc in CHB With Decompensation
981
Figure 1. Flow chart of study subjects in current study (263 CHB with hepatitis flare and HD). Durations of LAM treatment for early mortality were listed in left upper panel.
Results Of the 263 patients included in the study, 94 had evidence of cirrhosis. The mean and medium durations of LAM therapy were 12.1 8.6 and 10.0 months (1.0–50 months), respectively. The baseline features at the start of LAM therapy are compared between patients with and without cirrhosis in Table 1. The patients with cirrhosis were older and less frequently HBeAg positive and had lower baseline serum HBV DNA, albumin, and ALT levels and longer dPT. The off-therapy followed-up duration was 3–169 months (median, 90 months; mean, 89.1 38.7). The end of LAM treatment features and off-therapy events are compared between patients with and without cirrhosis in Table 2. Patients with cirrhosis had longer treatment
duration, more clinical relapse, more frequent retreatment, and more HCC development. The cumulative incidences of clinical relapse, hepatitis flare, HD, HCC, and mortality are compared between patients with and without cirrhosis in Figure 2. Most events occurred within 2 years. When the follow-up period was not censored on retreatment, the 1-year and 2-year cumulative incidences of clinical relapse were higher in patients with cirrhosis than in those without cirrhosis (35.1% vs 27.1% and 57.5% vs 47.9%, respectively; P ¼ .013). The 1-year cumulative incidence of hepatitis flare, HD, and mortality was 16.2%, 8.2%, and 0%, respectively. The 2-year and 5-year cumulative incidence of HD increased to 12.5% and 19.8%, respectively, and were similar in cirrhotic and non-cirrhotic patients (P ¼ .626). The median interval between the onset of HD and prior
982
Chang et al
Clinical Gastroenterology and Hepatology Vol. 13, No. 5
Table 1. Comparisons of Baseline Characteristics Between Those With and Without Cirrhosis
Total (N ¼ 263) Male, N (%)a Age (y)b HBeAg (positive)a HBV DNA (log10 IU/mL)b Genotype Ba ALT (U/L)b dPT(sec)b Albumin (g/dL)b Bilirubin (mg/dL)b Creatinine (mg/dL)b
213 41.6 12.0, 41 116 6.2 1.5, 6.2 229 1457.9 978.7, 1284 8.5 8.0, 6.2 3.3 0.6, 3.3 13.8 9.7, 11.7 0.89 0.32, 0.9
(81) (18–83) (44%) (2.6–9.7) (87.9%) (187–4815) (1.8–88.9) (1.0–5.1) (2.1–44.3) (0.2–2.5)
Non-cirrhosis (n ¼ 169) 133 (78.7) 39.7 12.2, 38 (18–83) 86 (50.8%) 6.4 1.4, 6.5 (3.6–9.7) 145 (86%) 1610.0 1018.8, 1557 (187–4815) 7.9 8.8, 5.8 (2.0–88.9) 3.5 0.6, 3.4 (1.0–5.0) 13.5 9.4, 11.8 (2.3–38.7) 0.9 0.3, 0.85 (0.4–2.2)
Cirrhosis (n ¼ 94) 80 44.9 11.0, 45 30 5.89 1.63, 5.64 98 1184.3 841.7, 930 9.5 3.1 14.5 0.9
6.9, 7.9 0.5, 3.1 10.4, 11.6 0.4, 0.9
P values of Student t or c2 tests
(85.1) (24–72) (31.9%) (2.57–8.82) (89.3%) (192–9750)
.189 <.001 .003 .013 .186 .001
(1.9–50.1) (2–5.1) (2.1–44.3) (0.2–2.5)
.033 <.001 .399 .815
a
Data expressed as N (%). Data expressed as mean standard deviation, median (range).
b
follow-up assessment was 65 days (20–213), 31% greater than 3 months and 67% greater than 2 months. LAM was restarted in 112 patients, and ETV was used for retreatment after 2008 in 27 patients. The retreatment was effective and discontinued again in 105 patients, as shown in Figure 1. Of the 14 patients who developed HCC, only 1 occurred within 2 years (20–109 months). Of the 3 patients who died, all had cirrhosis at baseline and died of HD 38–79 months after cessation of LAM therapy. The 5-year cumulative incidence of HD, HCC, and mortality was 21.2% vs 18.1% (P ¼ .626), 5.6% vs 1.6% (P ¼ .038), and 2.9% vs 0% (P ¼ .021), respectively, in patients with and without cirrhosis (Figure 2C–E). The hazard ratio (HR), 95% confidence interval (CI), and the P values of HD and HCC are listed in Table 3. After adjusting for the covariances, male gender was the only factor for HD, whereas age and cirrhosis were the factors for HCC development. When follow-up period was censored on retreatment, severe events decreased. The 1-year cumulative incidence of hepatitis flare, HD, and mortality was 16.2%, 6.1%, and 0%, respectively. The 2-year and 5-year cumulative incidence of HD increased to 10.6% and 16.6%, respectively, and was similar in cirrhotic and non-cirrhotic patients (P ¼ .57). The 5-year cumulative incidence of HD, HCC, and mortality was 18.5% vs 15.5% (P ¼ .57), 3.5% vs 0.7% (P ¼ .313), and 1.6% vs 0% (P ¼ .177), respectively, in patients with and without cirrhosis. After adjusting for the covariances, male gender (95% CI of HR, 1.024–10.73) was the only factor for HD, whereas age (95% CI of HR, 1.019–1.127) and albumin (95% CI of HR, 0.026–0.588) were the factors for HCC development.
Discussion The most important concern of stopping Nuc therapy is the short-term (within 6–12 months) urgent adverse
outcomes such as HD or mortality. The results of the present study showed that the off-therapy 1-year HD incidence was 8.2% and no mortality in the first 3 years. The 1-year incidence of HD was 6.1% if the follow-up duration was censored on retreatment. Also important is that 43% of the patients were able to enjoy Nuc-free life for a median of 90 months. Of note is that the median duration of LAM therapy was only 10 months, and 67% of the HDs occurred in patients who were monitored at an interval >2 months (30% >3 months). The need for retreatment might be even lower if the patients had followed the more stringent stopping rule, and a closer monitoring schedule of Asian Pacific guidelines would detect and treat clinical relapse or hepatitis flare in time to prevent HD.5 The late mortality in 3 of the 263 patients after cessation of LAM therapy is superior to that in the compensated anti-HBe–positive patients on long-term LAM therapy, of whom 7.4% died during 6 3 years of treatment.18 In well-developed countries, LAM has now been largely replaced by the more potent ETV and TDF, which have high genetic barrier to drug resistance and are safe and effective in HD.19,20 Because offETV therapy is associated with a significantly lower clinical relapse rate as compared with off-LAM event in compensated HBeAg-negative CHB patients,11 it is anticipated that cessation of ETV or TDF therapy after stringent stopping rule of Asian Pacific Association for the Study of the Liver in such patients may have better outcomes. All of these add support to the notion that cessation of Nuc therapy can be considered in patients who had recovered from HD, even in patients with cirrhosis, if proper off-therapy monitoring is provided. This is particularly relevant in less developed countries where ETV or TDF is not yet available, or indefinite longterm therapy is not affordable. Even in a rich country like Singapore, a recent survey showed that 27% of the treated patients were taking LAM, and only 25% were willing to pay $56 US weekly for lifelong therapy.21 The
May 2015
Stop Nuc in CHB With Decompensation
983
Table 2. Comparisons of Lamivudine Treatment and Off-therapy Events Between Those With and Without Cirrhosis
Treatment durationa (mo) EOT status HBeAg (þ)b ALT (U/L) Albumin (g/dL) Bilirubin (mg/dL) Creatinine (mg/dL) Off-therapy follow-up (mo)a Off-therapy retreatmentb Clinical relapseb Hepatitis flareb HDb HCCb Mortalityb
Total (N ¼ 263)
Non-cirrhotic (n ¼ 169)
Cirrhosis (n ¼ 94)
P values of Student t, c2, or Mann–Whitney U tests
12.1 8.6, 10.0 (1.0–50)
11.3 8.6, 9.0 (1.0–50.0)
13.5 8.4, 12.0 (1.0–43.0)
.047
113 (42.9%) 32.1 21.1 4.2 0.5 1.4 2.4 1.0 0.4 89.1 38.7, 90.0 (3.0–169) 139 (52.8%) 175 (66.5%) 96 (36.5%) 48 (18.3%) 14 (5.3%) 3 (1.1%)
83 (49.1%) 29.4 23.3 4.2 0.6 0.99 0.5 1.1 0.5 88.6 39.8, 90.0 (3.0–165) 79 (46.7%) 101 (59.7%) 62 (36.7%) 29 (17.2%) 4 (2.3%) 0 (0%)
30 (31.9%) 34.1 18.9 4.2 0.5 1.6 3.4 1.0 0.1 90.1 36.9, 88.5 (18.0–169) 60 (63.8%) 74 (75.5%) 34 (36.2%) 19 (20.2%) 10 (10.6%) 3 (3.2%)
.03 .787 .473 .23 .57 .775 .016 .020 .721 .541 .036 .093
EOT, end of treatment. a Data expressed as mean standard deviation, median (range). b Data expressed as N (%).
cost issue and the aforementioned potential problems of indefinite long-term Nuc therapy are worth consideration to modify the recommendation for HD patients by experts in the major liver associations.4–6,22 In the present cohort of HD survivors, the 1-year offLAM clinical relapse incidence of 27.1% in patients without cirrhosis or 34.2% in the whole cohort (Figure 2) tends to be lower than or at least comparable to that in compensated CHB treated with LAM or adefovir (ADV) or ETV, although various definitions in different studies might account partly for the different relapse rates among studies with compensated HBeAgnegative patients.11,23,24 Because earlier studies had much longer duration of Nuc therapy and used much more stringent stopping rule,11,23,24 it is likely that the relapse rate in our HD survivors would be lower if the duration of LAM therapy and the stringent stopping rule of Asian Pacific guidelines were followed. It is known that covalently closed circular DNA (cccDNA) in the hepatocyte nuclei is usually maintained for a lifetime25 and may be reduced by cell killing.26 Perhaps the rigorous immune-mediated hepatocyte killing in HD1,14 might have reduced cccDNA and therefore decreased or delayed clinical relapse. Further studies are needed to prove this line of speculation. Cirrhosis is considered to be the most important risk factor for HCC development. Only 10 of our 94 patients with cirrhosis (10.6%) developed HCC during mean duration of 90 months. The 5-year cumulative HCC incidence in our patients with cirrhosis was 5.6%, lower than half of the calculated 17% 5-year incidence in untreated East Asia CHB patients with cirrhosis.27 It is also lower than 7% in Japanese patients with cirrhosis treated with continuous ETV for mean of 3.3 years and much lower than
22% in the LAM-treated counterpart during 6.8 years without rescue therapy,17 although these are not head-tohead comparisons, and the study populations differ in factors such as age and gender, which may influence the results. Our finding again confirms our earlier observation that stopping a course of successful Nuc therapy will not compromise the beneficial effect in reducing HCC in patients with cirrhosis.28 However, it remains to be determined whether longer-term Nuc therapy will reduce more HCC than that observed in this off-therapy study. Obviously, HCC surveillance with AFP and US every 3–6 months is mandatory for early detection and timely curative therapy of HCC.2 The current study has several limitations. First, this retrospective study involved patients in an era when no guidelines for antiviral therapy in CHB patients with HD were available, so that the patients stopped LAM therapy after clinical recovery at a time point at the discretion of their physician. Inadequate duration of consolidation therapy and monitoring may have increased the offtherapy incidence of HD. Because of the widely accepted dogma that Nuc therapy should continue indefinitely/lifelong in HD patients, it is almost impossible to conduct a prospective randomized controlled study to compare the outcomes of patients who continue or discontinue Nuc therapy. Using information from a reasonably large number of earlier patients may be the only way to address this issue. The findings of the present study may provide a basis for designing future controlled studies. In addition, the results of the current study suggest that the optimal monitoring interval in the first 2 years after cessation of Nuc therapy is at longest every 2 months. Actual adherence to frequent monitoring may not be easy in clinical practice but is of
984
Chang et al
Clinical Gastroenterology and Hepatology Vol. 13, No. 5
Figure 2. Comparisons of cumulative incidences of (A) clinical relapse, (B) hepatitis flare, (C) HD, (D) HCC, and (E) mortality between patients with cirrhosis (0) and without cirrhosis (1). LC, cirrhosis.
paramount importance. This must be stressed when physicians/patients consider stopping therapy. Second, most cirrhosis was diagnosed according to clinical and US findings because liver biopsy is contraindicated in patients with coagulopathy. Although the sensitivity and specificity of US findings in the detection of early cirrhosis were estimated to be >80%,29 it may have underestimated the prevalence of cirrhosis and may have overestimated the off-therapy events in patients with cirrhosis, not biased in favor of our conclusion. Third, serum samples were not available for retrospective assays of baseline HBsAg level and naturally
occurring HBV mutations. Serum HBsAg quantity correlates with transcriptionally active cccDNA and is considered a surrogate marker of infected hepatocytes.30 Perhaps the changes in HBsAg level may support or refute our speculation that the cccDNA level was suppressed during HD. As to natural HBV mutations, studies showed that these mutations played no role in the occurrence of HD during hepatitis flare of CHB31,32 or in clinical relapse after cessation of ETV therapy.11 Fourth, the emergence of LAM resistance and whether HBV replication was fully suppressed under LAM treatment were not explored. This does not seem to impact on the
.21
.92 0.958 (0.424–2.16)
.118 .041 0.716 (0.471–1.088) 2.766 (1.021–7.49)
1.23 (0.997–1.53)
.025 .014
.066
4.339 (1.343–14.02)
1.979 (0.955–4.103)
0.981 1.060 1.396 0.639 12.26 2.23 0.99 0.99 0.98 0.997 0.394 1.76
(0.28–3.46) (1.021–1.096) (0.778–2.50) (0.445–0.918) (1.615–93.26) (0.72–6.89) (0.92–1.07) (0.98–1.000) (0.92–1.06) (0.98–1.059) (0.18–0.863) (0.014–25.98)
.98 .002 .263 .015 .015 .172 .918 .01 .73 .672 .02 .59
Multivariate analysis HR (95% CI) Multivariate analysis HR (95% CI)
P values
Univariate analysis HR (95% CI)
P values
HCC
985
feasibility of discontinuing ETV or TDF in this setting because ETV or TDF may maintain full HBV suppression at a rate >90% with virtually no drug resistance after 2–3 years of therapy.33 Fifth, LAM is no longer a recommended first-line Nuc for CHB in guidelines of major liver associations.4–6 However, LAM is still used not only in Asian countries including Singapore21,34 but also in some European countries such as Poland and Turkey in up to 45% of the patients.35 In conclusion, the present study has demonstrated that cessation of LAM therapy after recovery from hepatitis flare with HD was safe for most patients. However, 8.2% developed HD within 1 year, although all were rescued by timely retreatment. Hence, the stopping strategy can be considered as a feasible alternative to indefinite therapy only if proper offtherapy monitoring is provided and adhered to. This strategy is especially relevant in low resource countries where patients are unable to afford the cost or are not willing to keep on unlimited daily medication. It is also important to note that cessation of antiviral therapy with properly scheduled monitoring would be much safer than stopping therapy by patients themselves against the advice of their physician for whatever reason(s), because such patients may encounter unpredictable untoward outcomes. In real-world practice, it is not uncommon to see serious liver-related complication(s) in patients who defaulted on follow-up after stopping Nuc therapy. Along this line, measures and effort are required to ensure adherence to close monitoring, which is the key to a safe off-therapy course. Of note is that the data in this study do not apply to patients with decompensated cirrhosis unrelated to hepatitis flare. Controlled trials of ETV or TDF in CHB patients with HD are required to confirm our findings and conclusions.
.021 .717 .268 .382 .627 .025 .597 .124 .164 .289 .24 .247
3. Peng CY, Chien RN, Liaw YF. Hepatitis B virus-related decompensated liver cirrhosis: benefits of antiviral therapy. J Hepatol 2012;57:442–450.
3.958 0.996 1.392 1.94 1.154 1.932 0.989 1.000 0.961 0.983 1.374 1.934
(1.23–12.74) (0.972–1.02) (0.775–2.498) (0.894–1.34) (0.647–2.159) (1.03–3.988) (0.951–1.029) (0.999–1.000) (0.909–1.016) (0.953–1.014) (0.809–2.33) (0.634–5.90)
References 1. Liaw YF, Chu CM. Hepatitis B virus infection. Lancet 2009; 373:582–592. 2. Chu CM, Liaw YF. Hepatitis B virus-related cirrhosis: natural history and treatment. Semin Liver Dis 2006;26:142–152.
Gender (male) Age (y) HBeAg (positive) HBV DNA (log10 IU/mL) Cirrhosis Genotype (C) LAM duration (mo) ALT (U/L) dPT (sec) Bilirubin (mg/dL) Albumin (g/dL) Creatinine (mg/dL)
Factors
Univariate analysis HR (95% CI)
P values
HD
Table 3. Logistic Regression Analyses of Factors for HD and HCC Occurred During Off-therapy Follow-up
1.0541 (1.006–1.003)
Stop Nuc in CHB With Decompensation P values
May 2015
4. Lok AS, McMahon BJ. Chronic hepatitis B: update 2009. Hepatology 2009;50:661–662. 5. Liaw YF, Kao JH, Piratvisuth T, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2012 update. Hepatol Int 2012;6:531–561. 6. European Association for the Study of the Liver. EASL clinical practice guidelines: management of chronic hepatitis B virus infection. J Hepatol 2012;57:167–185. 7. Lampertico P, Viganò M, Colombo M. Why do I treat HBeAgnegative chronic hepatitis B patients with pegylated interferon? Liver Int 2013;33(Suppl 1):157–163. 8. Locarnini S, Zoulim F. Molecular genetics of HBV infection. Antivir Ther 2010;15(Suppl 3):3–14.
986
Chang et al
9. Chotiyaputta W, Peterson C, Ditah FA, et al. Persistence and adherence to nucleos(t)ide analogue treatment for chronic hepatitis B. J Hepatol 2011;54:12–18.
Clinical Gastroenterology and Hepatology Vol. 13, No. 5 stringent cessation criteria for adefovir. Arch Virol 2012; 157:285–290. 25. Locarnini S, Zoulim F. Molecular genetics of HBV infection. Antivir Ther 2010;3(Suppl):3–14.
10. Han SH, Jing W, Mena E, et al. Adherence, persistence, healthcare utilization, and cost benefits of guidelinerecommended hepatitis B pharmacotherapy. J Med Econ 2012;15:1159–1166.
26. Levrero M, Pollicino T, Petersen J, et al. Control of cccDNA function in hepatitis B virus infection. J Hepatol 2009; 51:581–592.
11. Jeng WJ, Sheen IS, Chen YC, et al. Off therapy durability of response to Entecavir therapy in hepatitis B e antigen negative chronic hepatitis B patients. Hepatology 2013;58:1888–1896.
27. Fattovich G, Bortolotti F, Donato F. Natural history of chronic hepatitis B: special emphasis on disease progression and prognostic factors. J Hepatol 2008;48:335–352.
12. Sheen IS, Liaw YF, Tai DI, et al. Hepatic decompensation associated with hepatitis B e antigen clearance in chronic type B hepatitis. Gastroenterology 1985;89:732–735.
28. Chen YC, Jeng WJ, Chien RN, et al. Hepatocellular carcinoma development in HBe-negative hepatitis B cirrhotic patients with entecavir treatment. J Hepatol 2013;58:S214.
13. Chien RN, Lin CH, Liaw YF. The effect of lamivudine therapy in hepatic decompensation during acute exacerbation of chronic hepatitis B. J Hepatol 2003;38:322–327.
29. Lin DY, Sheen IS, Chiu CT, et al. Ultrasonographic changes of early liver cirrhosis in chronic hepatitis B: a longitudinal study. J Clin Ultrasound 1993;21:303–308.
14. Liaw YF. Hepatitis flares and hepatitis B e antigen seroconversion: implication in anti-hepatitis B virus therapy. J Gastroenterol Hepatol 2003;18:246–252.
30. Liaw YF. Clinical utility of hepatitis B surface antigen quantitation in patients with chronic hepatitis B: a review. Hepatology 2011;53:2121–2129.
15. Liaw YF, Tai DI, Chu CM, et al. The development of cirrhosis in patients with chronic type B hepatitis: a prospective study. Hepatology 1988;8:493–496.
31. Chu CM, Yeh CT, Lee CS, et al. Procore stop mutant in HBeAgpositive patients with chronic hepatitis B: clinical characteristics and correlation with the course of HBeAg-to-anti-HBe seroconversion. J Clin Microbiol 2002;40:16–21.
16. Bruix J, Sherman M, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020–1022. 17. Hosaka T, Suzuki F, Kobayashi M, et al. Long-term entecavir treatment reduces hepatocellular carcinoma incidence in patients with hepatitis B virus infection. Hepatology 2013;58:98–107. 18. Niro GA, Ippolito AM, Fontana R, et al. Long-term outcome of hepatitis B virus-related chronic hepatitis under protracted nucleos(t)ide analogues. J Viral Hepat 2013;20:502–509. 19. Liaw YF, Raptopoulou-Gigi M, Cheinquer H, et al. Efficacy and safety of entecavir versus adefovir in chronic hepatitis B patients with hepatic decompensation: a randomized open-label study. Hepatology 2011;54:91–100. 20. Liaw YF, Sheen IS, Lee CM, et al. Tenofovir disoproxil fumarate (TDF), emtricitabine/TDF and entecavir in patients with decompensated chronic hepatitis B liver disease. Hepatology 2011; 53:62–72. 21. Lim SG, Aung MO, Chung SW, et al. Patient preferences for hepatitis B therapy. Antivir Ther 2013;18:663–670. 22. Hadziyannis SJ, Sevastianos V, Rapti I, et al. Sustained responses and loss of HBsAg in HBeAg-negative patients with chronic hepatitis B who stop long-term treatment with adefovir. Gastroenterology 2012;143:629–636. 23. Liu F, Wang L, Li XY, et al. Poor durability of lamivudine effectiveness despite stringent cessation criteria: a prospective clinical study in hepatitis B e antigen-negative chronic hepatitis B patients. J Gastroenterol Hepatol 2011;26:456–460. 24. Ha M, Zhang G, Diao S, et al. A prospective clinical study in hepatitis B e antigen-negative chronic hepatitis B patients with
32. Tsai WL, Lo GH, Hsu PI, et al. Role of genotype and precore/ basal core promoter mutations of hepatitis B virus in patients with chronic hepatitis B with acute exacerbation. Scand J Gastroenterol 2008;43:196–201. 33. Pol S, Lampertico P. First-line treatment of chronic hepatitis B with entecavir or tenofovir in ‘real-life’ settings: from clinical trials to clinical practice. J Viral Hepat 2012;19:377–386. 34. Liaw YF. Antiviral therapy of chronic hepatitis B: opportunities and challenges in Asia. J Hepatol 2009;51:403–410. 35. Arama V, Leblebicioglu H, Simon K, et al. Chronic hepatitis B monitoring and treatment patterns in five European countries with different access and reimbursement policies. Antivir Ther 2014;19:245–257. Reprint requests Address requests for reprints to: Yun-Fan Liaw, MD, 199, Tung Hwa North Road, Taipei 105, Taiwan. e-mail: liveryfl@gmail.com; fax: 886-3-3282824. Acknowledgments The authors thank Ms Li-Hua Lu for laboratory work, Ms Yu-Ju Lan for data collection, Ms Shu-Chun Chen for data mining, and Ms Su-Chiung Chu for assistance in preparing the manuscript. Conflicts of interest This author discloses the following: Y. F. Liaw has been involved with clinical trials and served as a global advisory board member of Roche. The remaining authors disclose no conflicts. Funding Supported by grants from Chang Gung Medical Research Fund (SMRPG1005, OMRPG380061) and the Prosperous Foundation, Taipei, Taiwan.
Clinical Events After Cessation of Lamivudine Therapy in Patients Recovered From Hepatitis B Flare With Hepatic Decompensation Ming-Ling Chang, Wen-Juei Jeng, and Yun-Fan Liaw Liver Research Unit, Chang Gung Memorial Hospital, Chang Gung University College of Medicine, Taipei, Taiwan BACKGROUND & AIMS:
Before guidelines were issued, many patients with hepatitis B flare and hepatic decompensation had discontinued lamivudine therapy instead of indefinite therapy. We investigated their outcomes.
METHODS:
We performed a retrospective cohort study of 263 consecutive patients with chronic hepatitis B (94 with cirrhosis) who recovered from a flare of hepatitis with hepatic decompensation and were followed after cessation of lamivudine therapy. Clinical events that occurred during the follow-up period were assessed by chart review and analysis of results from retrospective assays.
RESULTS:
The mean duration of lamivudine therapy was 12.1 – 8.6 months; data were collected from patients for 89.1 – 38.7 months after therapy ended. In the first year off therapy, 29.9% of patients had clinical relapse, 16.2% had hepatitis flares, and 8.2% had hepatic decompensation. There was no significant difference in the incidence of hepatic decompensation between patients with and without cirrhosis. Hepatocellular carcinoma developed in 14 patients 20–109 months after cessation of therapy, with 5-year cumulative incidence of 5.2% in patients with cirrhosis. Three patients with cirrhosis died of hepatic decompensation 38–76 months after cessation of therapy (5year cumulative mortality, 2.9%). Multivariate analyses showed that men were more likely than women to have recurrence of hepatic decompensation (hazard ratio [HR], 4.339; P [ .014). Liver cirrhosis (HR, 2.766; P [ .041) and age (HR, 1.054; P [ .023) increased risk for hepatocellular carcinoma.
CONCLUSIONS:
Cessation of lamivudine therapy after recovery from hepatitis B flare with decompensation was safe for most patients. However, 8.2% develop decompensation within 1 year and can be rescued by timely retreatment. With close monitoring, the stopping strategy could be a feasible alternative to indefinite therapy, especially in low resource settings.
Keywords: Antiviral Therapy; Cirrhosis; Liver Cancer; Nucleos(t)ide Analog.
hronic hepatitis B virus (HBV) infection has potential adverse outcomes including hepatic decompensation (HD), cirrhosis, and hepatocellular carcinoma (HCC).1 Hepatitis flare with HD in chronic hepatitis B (CHB) patients is a serious complication with high mortality,2 so that requires immediate nucleos(t)ide analog (Nuc) therapy3–6; most experts recommend indefinite or even lifelong therapy.4,6 Of note, indefinite or lifelong Nuc therapy has problems of concern, including cost, lack of long-term safety data beyond 5–10 years, negligible rate of hepatitis B surface antigen (HBsAg) loss,7 and emergence of drug resistance.8 Furthermore, the 1-year Nuc persistent rate was only 73.4% in new patients,9 and the medication possession rate 80% was only 53.7% in patients treated with entecavir (ETV) or tenofovir (TDF).10 This may bring CHB patients to unpredictable hazards. Weighted against all these potential problems, Asian Pacific guidelines have
C
recommended that cessation of Nuc therapy can be considered in hepatitis B e antigen (HBeAg) negative patients if undetectable HBV DNA has been documented on 3 occasions at least 6 months apart.5 By using this stopping rule, we have shown that ETV therapy can be safely discontinued in HBeAg-negative CHB patients, even with compensated cirrhosis, under proper off-therapy Abbreviations used in this paper: ADV, adefovir; AFP, alpha fetoprotein; ALT, alanine aminotransferase; anti-HBe, hepatitis B e antibody; cccDNA, covalently closed circular DNA; CHB, chronic hepatitis B; dPT, prolongation of prothrombin time; ETV, entecavir; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HD, hepatic decompensation; HDV, hepatitis delta virus; HR, hazard ratio; LAM, lamivudine; Nuc, nucleos(t)ide analogue; TDF, tenofovir; ULN, upper limit of normal; US, ultrasonography. © 2015 by the AGA Institute 1542-3565/$36.00 http://dx.doi.org/10.1016/j.cgh.2014.10.023
980
Chang et al
monitoring.11 Whether this stopping strategy is applicable to patients with HD remains to be studied. Before the Asian Pacific stopping rule was issued, lamivudine (LAM) was widely used and was discontinued in most of our patients after recovery from HD because of the concerns about drug resistance and cost/reimbursement issues. These patients were followed up off-therapy. Their medical records were available for reviewing what happened after cessation of LAM therapy. Considering that the dogma of indefinite/lifelong Nuc therapy in patients with HD was based on expert opinion rather than solid scientific evidence, we conducted a retrospective cohort study to examine the off-LAM events with survival analyses in CHB patients recovered from hepatitis flare with HD.
Patients Excluding patients with other causes of liver disease such as hepatitis C virus (HCV) or hepatitis delta virus (HDV) infection, alcohol, autoimmune, and malignancy and those with pregnancy or breastfeeding, severe pancreatitis, prior liver transplantation, or need for mechanical ventilation, 318 consecutive CHB patients (18 years old) who had hepatitis flare with HD were treated with 100 mg LAM daily in our unit. Forty-one of them died, mostly within 6 months after starting LAM therapy, and 14 underwent liver transplantation. The remaining 263 patients recovered during LAM therapy and were followed up after cessation of therapy (Figure 1). HD was defined as a severe clinical syndrome with hepatic function impairment as indicated by jaundice and a prolonged prothrombin time (dPT) and/or occurrence of ascites/encephalopathy in patients with or without cirrhosis, as described elsewhere since 1980s.12,13 LAM treatment was discontinued in all patients with clinical recovery (normalization of prothrombin time and bilirubin level) and at a time point at the discretion of their physician, usually within reimbursement allowance (18 months before November 1, 2009, 36 months thereafter). Then the patients were followed up every 1–6 months, more frequently within the first 3 months and when alanine aminotransferase (ALT) became abnormal. Follow-up studies included clinical assessment, liver biochemical tests, and serum HBV DNA when appropriate. Alpha fetoprotein (AFP) and abdominal ultrasonography (US) were performed every 3–6 months for HCC surveillance. The primary events of concern were life-threatening safety issues including HD and mortality. The secondary events of concern included clinical relapse, hepatitis flare, or HCC. Retreatment was instituted when indicated. Clinical relapse was defined as virologic relapse (serum HBV DNA >2000 IU/mL) plus ALT elevation greater than 2 times upper limit of normal (ULN).5 Hepatitis flare was defined as abrupt rise of serum ALT to level greater than 5 ULN.14 Cirrhosis was diagnosed by histologic findings or repeated US findings consistent with cirrhosis, supplemented with clinical features such as varices and
Clinical Gastroenterology and Hepatology Vol. 13, No. 5
thrombocytopenia, as described elsewhere.15 HCC was diagnosed by histology/cytology or US findings plus high AFP level or imaging findings of enhanced arterial contrast uptake followed by washout in the portal venous phase and equilibrium phase (dynamic computed tomography or magnetic resonance imaging), as described in generally accepted guidelines.16 HCC that developed within 1 year after cessation of therapy was considered as preexisting and therefore was excluded as an off-therapy event.17
Methods Liver biochemical tests were performed by using routine automated techniques. Serum HBsAg, HBeAg/ anti-HBe, and anti-HDV were assayed by using radioimmunoassay kits (Abbott Diagnostics, North Chicago, IL). Anti-HCV was assayed by using a commercial enzyme immunoassay (Abbott Diagnostics). HBV genotype was determined retrospectively by using polymerase chain reaction–restriction fragment length polymorphism of the surface gene of HBV. Serum HBV DNA level was measured by using the hybrid Capture II assay (Digene Corp, Gaithersburg, MD; lower limit of detection, 1.4 105 copies/mL) before April 2007. Because Digene assay reported HBV DNA level in copies/mL, we divided the levels by a factor of 5 to become IU/mL in the analyses of this study. Ultrasensitive polymerase chain reaction assay (Cobas Amplicor HBV Monitor, lower limit of detection, 300 copies/mL; or the Roche COBAS TaqMan HBV Test, lower limit of detection, 69 copies/mL, or 12 IU/mL; Roche Diagnostics, Pleasanton, CA) was used for stored serum samples. Gender, age, HBV DNA level, HBeAg status, HBV genotype, presence of cirrhosis, ALT, bilirubin, dPT, albumin, creatinine, as well as treatment duration were used for comparisons where appropriate.
Statistics The follow-up period was calculated from cessation of LAM therapy until the last visit of the patient or when an event occurred. The follow-up period was censored when treatment was resumed in the secondary analyses. Events in patients with and without cirrhosis were compared. All statistical analysis was performed by using Statistics Package for Social Science software (version 18.0; SPSS Inc, Chicago, IL). To compare between groups, continuous variables were analyzed by using Student t test or Mann– Whitney U test, whereas categorical variables were analyzed by using c2 test or Fisher exact test when appropriate. Kaplan–Meier and univariate Cox regression were used to assess the relationship of various variables to the events. Variables found to be significant were included in multivariate Cox regression models. Statistical significance was defined at the 5% level on the basis of two-tailed test of the null hypothesis. The study was approved by the hospital institutional review board.
May 2015
Stop Nuc in CHB With Decompensation
981
Figure 1. Flow chart of study subjects in current study (263 CHB with hepatitis flare and HD). Durations of LAM treatment for early mortality were listed in left upper panel.
Results Of the 263 patients included in the study, 94 had evidence of cirrhosis. The mean and medium durations of LAM therapy were 12.1 8.6 and 10.0 months (1.0–50 months), respectively. The baseline features at the start of LAM therapy are compared between patients with and without cirrhosis in Table 1. The patients with cirrhosis were older and less frequently HBeAg positive and had lower baseline serum HBV DNA, albumin, and ALT levels and longer dPT. The off-therapy followed-up duration was 3–169 months (median, 90 months; mean, 89.1 38.7). The end of LAM treatment features and off-therapy events are compared between patients with and without cirrhosis in Table 2. Patients with cirrhosis had longer treatment
duration, more clinical relapse, more frequent retreatment, and more HCC development. The cumulative incidences of clinical relapse, hepatitis flare, HD, HCC, and mortality are compared between patients with and without cirrhosis in Figure 2. Most events occurred within 2 years. When the follow-up period was not censored on retreatment, the 1-year and 2-year cumulative incidences of clinical relapse were higher in patients with cirrhosis than in those without cirrhosis (35.1% vs 27.1% and 57.5% vs 47.9%, respectively; P ¼ .013). The 1-year cumulative incidence of hepatitis flare, HD, and mortality was 16.2%, 8.2%, and 0%, respectively. The 2-year and 5-year cumulative incidence of HD increased to 12.5% and 19.8%, respectively, and were similar in cirrhotic and non-cirrhotic patients (P ¼ .626). The median interval between the onset of HD and prior
982
Chang et al
Clinical Gastroenterology and Hepatology Vol. 13, No. 5
Table 1. Comparisons of Baseline Characteristics Between Those With and Without Cirrhosis
Total (N ¼ 263) Male, N (%)a Age (y)b HBeAg (positive)a HBV DNA (log10 IU/mL)b Genotype Ba ALT (U/L)b dPT(sec)b Albumin (g/dL)b Bilirubin (mg/dL)b Creatinine (mg/dL)b
213 41.6 12.0, 41 116 6.2 1.5, 6.2 229 1457.9 978.7, 1284 8.5 8.0, 6.2 3.3 0.6, 3.3 13.8 9.7, 11.7 0.89 0.32, 0.9
(81) (18–83) (44%) (2.6–9.7) (87.9%) (187–4815) (1.8–88.9) (1.0–5.1) (2.1–44.3) (0.2–2.5)
Non-cirrhosis (n ¼ 169) 133 (78.7) 39.7 12.2, 38 (18–83) 86 (50.8%) 6.4 1.4, 6.5 (3.6–9.7) 145 (86%) 1610.0 1018.8, 1557 (187–4815) 7.9 8.8, 5.8 (2.0–88.9) 3.5 0.6, 3.4 (1.0–5.0) 13.5 9.4, 11.8 (2.3–38.7) 0.9 0.3, 0.85 (0.4–2.2)
Cirrhosis (n ¼ 94) 80 44.9 11.0, 45 30 5.89 1.63, 5.64 98 1184.3 841.7, 930 9.5 3.1 14.5 0.9
6.9, 7.9 0.5, 3.1 10.4, 11.6 0.4, 0.9
P values of Student t or c2 tests
(85.1) (24–72) (31.9%) (2.57–8.82) (89.3%) (192–9750)
.189 <.001 .003 .013 .186 .001
(1.9–50.1) (2–5.1) (2.1–44.3) (0.2–2.5)
.033 <.001 .399 .815
a
Data expressed as N (%). Data expressed as mean standard deviation, median (range).
b
follow-up assessment was 65 days (20–213), 31% greater than 3 months and 67% greater than 2 months. LAM was restarted in 112 patients, and ETV was used for retreatment after 2008 in 27 patients. The retreatment was effective and discontinued again in 105 patients, as shown in Figure 1. Of the 14 patients who developed HCC, only 1 occurred within 2 years (20–109 months). Of the 3 patients who died, all had cirrhosis at baseline and died of HD 38–79 months after cessation of LAM therapy. The 5-year cumulative incidence of HD, HCC, and mortality was 21.2% vs 18.1% (P ¼ .626), 5.6% vs 1.6% (P ¼ .038), and 2.9% vs 0% (P ¼ .021), respectively, in patients with and without cirrhosis (Figure 2C–E). The hazard ratio (HR), 95% confidence interval (CI), and the P values of HD and HCC are listed in Table 3. After adjusting for the covariances, male gender was the only factor for HD, whereas age and cirrhosis were the factors for HCC development. When follow-up period was censored on retreatment, severe events decreased. The 1-year cumulative incidence of hepatitis flare, HD, and mortality was 16.2%, 6.1%, and 0%, respectively. The 2-year and 5-year cumulative incidence of HD increased to 10.6% and 16.6%, respectively, and was similar in cirrhotic and non-cirrhotic patients (P ¼ .57). The 5-year cumulative incidence of HD, HCC, and mortality was 18.5% vs 15.5% (P ¼ .57), 3.5% vs 0.7% (P ¼ .313), and 1.6% vs 0% (P ¼ .177), respectively, in patients with and without cirrhosis. After adjusting for the covariances, male gender (95% CI of HR, 1.024–10.73) was the only factor for HD, whereas age (95% CI of HR, 1.019–1.127) and albumin (95% CI of HR, 0.026–0.588) were the factors for HCC development.
Discussion The most important concern of stopping Nuc therapy is the short-term (within 6–12 months) urgent adverse
outcomes such as HD or mortality. The results of the present study showed that the off-therapy 1-year HD incidence was 8.2% and no mortality in the first 3 years. The 1-year incidence of HD was 6.1% if the follow-up duration was censored on retreatment. Also important is that 43% of the patients were able to enjoy Nuc-free life for a median of 90 months. Of note is that the median duration of LAM therapy was only 10 months, and 67% of the HDs occurred in patients who were monitored at an interval >2 months (30% >3 months). The need for retreatment might be even lower if the patients had followed the more stringent stopping rule, and a closer monitoring schedule of Asian Pacific guidelines would detect and treat clinical relapse or hepatitis flare in time to prevent HD.5 The late mortality in 3 of the 263 patients after cessation of LAM therapy is superior to that in the compensated anti-HBe–positive patients on long-term LAM therapy, of whom 7.4% died during 6 3 years of treatment.18 In well-developed countries, LAM has now been largely replaced by the more potent ETV and TDF, which have high genetic barrier to drug resistance and are safe and effective in HD.19,20 Because offETV therapy is associated with a significantly lower clinical relapse rate as compared with off-LAM event in compensated HBeAg-negative CHB patients,11 it is anticipated that cessation of ETV or TDF therapy after stringent stopping rule of Asian Pacific Association for the Study of the Liver in such patients may have better outcomes. All of these add support to the notion that cessation of Nuc therapy can be considered in patients who had recovered from HD, even in patients with cirrhosis, if proper off-therapy monitoring is provided. This is particularly relevant in less developed countries where ETV or TDF is not yet available, or indefinite longterm therapy is not affordable. Even in a rich country like Singapore, a recent survey showed that 27% of the treated patients were taking LAM, and only 25% were willing to pay $56 US weekly for lifelong therapy.21 The
May 2015
Stop Nuc in CHB With Decompensation
983
Table 2. Comparisons of Lamivudine Treatment and Off-therapy Events Between Those With and Without Cirrhosis
Treatment durationa (mo) EOT status HBeAg (þ)b ALT (U/L) Albumin (g/dL) Bilirubin (mg/dL) Creatinine (mg/dL) Off-therapy follow-up (mo)a Off-therapy retreatmentb Clinical relapseb Hepatitis flareb HDb HCCb Mortalityb
Total (N ¼ 263)
Non-cirrhotic (n ¼ 169)
Cirrhosis (n ¼ 94)
P values of Student t, c2, or Mann–Whitney U tests
12.1 8.6, 10.0 (1.0–50)
11.3 8.6, 9.0 (1.0–50.0)
13.5 8.4, 12.0 (1.0–43.0)
.047
113 (42.9%) 32.1 21.1 4.2 0.5 1.4 2.4 1.0 0.4 89.1 38.7, 90.0 (3.0–169) 139 (52.8%) 175 (66.5%) 96 (36.5%) 48 (18.3%) 14 (5.3%) 3 (1.1%)
83 (49.1%) 29.4 23.3 4.2 0.6 0.99 0.5 1.1 0.5 88.6 39.8, 90.0 (3.0–165) 79 (46.7%) 101 (59.7%) 62 (36.7%) 29 (17.2%) 4 (2.3%) 0 (0%)
30 (31.9%) 34.1 18.9 4.2 0.5 1.6 3.4 1.0 0.1 90.1 36.9, 88.5 (18.0–169) 60 (63.8%) 74 (75.5%) 34 (36.2%) 19 (20.2%) 10 (10.6%) 3 (3.2%)
.03 .787 .473 .23 .57 .775 .016 .020 .721 .541 .036 .093
EOT, end of treatment. a Data expressed as mean standard deviation, median (range). b Data expressed as N (%).
cost issue and the aforementioned potential problems of indefinite long-term Nuc therapy are worth consideration to modify the recommendation for HD patients by experts in the major liver associations.4–6,22 In the present cohort of HD survivors, the 1-year offLAM clinical relapse incidence of 27.1% in patients without cirrhosis or 34.2% in the whole cohort (Figure 2) tends to be lower than or at least comparable to that in compensated CHB treated with LAM or adefovir (ADV) or ETV, although various definitions in different studies might account partly for the different relapse rates among studies with compensated HBeAgnegative patients.11,23,24 Because earlier studies had much longer duration of Nuc therapy and used much more stringent stopping rule,11,23,24 it is likely that the relapse rate in our HD survivors would be lower if the duration of LAM therapy and the stringent stopping rule of Asian Pacific guidelines were followed. It is known that covalently closed circular DNA (cccDNA) in the hepatocyte nuclei is usually maintained for a lifetime25 and may be reduced by cell killing.26 Perhaps the rigorous immune-mediated hepatocyte killing in HD1,14 might have reduced cccDNA and therefore decreased or delayed clinical relapse. Further studies are needed to prove this line of speculation. Cirrhosis is considered to be the most important risk factor for HCC development. Only 10 of our 94 patients with cirrhosis (10.6%) developed HCC during mean duration of 90 months. The 5-year cumulative HCC incidence in our patients with cirrhosis was 5.6%, lower than half of the calculated 17% 5-year incidence in untreated East Asia CHB patients with cirrhosis.27 It is also lower than 7% in Japanese patients with cirrhosis treated with continuous ETV for mean of 3.3 years and much lower than
22% in the LAM-treated counterpart during 6.8 years without rescue therapy,17 although these are not head-tohead comparisons, and the study populations differ in factors such as age and gender, which may influence the results. Our finding again confirms our earlier observation that stopping a course of successful Nuc therapy will not compromise the beneficial effect in reducing HCC in patients with cirrhosis.28 However, it remains to be determined whether longer-term Nuc therapy will reduce more HCC than that observed in this off-therapy study. Obviously, HCC surveillance with AFP and US every 3–6 months is mandatory for early detection and timely curative therapy of HCC.2 The current study has several limitations. First, this retrospective study involved patients in an era when no guidelines for antiviral therapy in CHB patients with HD were available, so that the patients stopped LAM therapy after clinical recovery at a time point at the discretion of their physician. Inadequate duration of consolidation therapy and monitoring may have increased the offtherapy incidence of HD. Because of the widely accepted dogma that Nuc therapy should continue indefinitely/lifelong in HD patients, it is almost impossible to conduct a prospective randomized controlled study to compare the outcomes of patients who continue or discontinue Nuc therapy. Using information from a reasonably large number of earlier patients may be the only way to address this issue. The findings of the present study may provide a basis for designing future controlled studies. In addition, the results of the current study suggest that the optimal monitoring interval in the first 2 years after cessation of Nuc therapy is at longest every 2 months. Actual adherence to frequent monitoring may not be easy in clinical practice but is of
984
Chang et al
Clinical Gastroenterology and Hepatology Vol. 13, No. 5
Figure 2. Comparisons of cumulative incidences of (A) clinical relapse, (B) hepatitis flare, (C) HD, (D) HCC, and (E) mortality between patients with cirrhosis (0) and without cirrhosis (1). LC, cirrhosis.
paramount importance. This must be stressed when physicians/patients consider stopping therapy. Second, most cirrhosis was diagnosed according to clinical and US findings because liver biopsy is contraindicated in patients with coagulopathy. Although the sensitivity and specificity of US findings in the detection of early cirrhosis were estimated to be >80%,29 it may have underestimated the prevalence of cirrhosis and may have overestimated the off-therapy events in patients with cirrhosis, not biased in favor of our conclusion. Third, serum samples were not available for retrospective assays of baseline HBsAg level and naturally
occurring HBV mutations. Serum HBsAg quantity correlates with transcriptionally active cccDNA and is considered a surrogate marker of infected hepatocytes.30 Perhaps the changes in HBsAg level may support or refute our speculation that the cccDNA level was suppressed during HD. As to natural HBV mutations, studies showed that these mutations played no role in the occurrence of HD during hepatitis flare of CHB31,32 or in clinical relapse after cessation of ETV therapy.11 Fourth, the emergence of LAM resistance and whether HBV replication was fully suppressed under LAM treatment were not explored. This does not seem to impact on the
.21
.92 0.958 (0.424–2.16)
.118 .041 0.716 (0.471–1.088) 2.766 (1.021–7.49)
1.23 (0.997–1.53)
.025 .014
.066
4.339 (1.343–14.02)
1.979 (0.955–4.103)
0.981 1.060 1.396 0.639 12.26 2.23 0.99 0.99 0.98 0.997 0.394 1.76
(0.28–3.46) (1.021–1.096) (0.778–2.50) (0.445–0.918) (1.615–93.26) (0.72–6.89) (0.92–1.07) (0.98–1.000) (0.92–1.06) (0.98–1.059) (0.18–0.863) (0.014–25.98)
.98 .002 .263 .015 .015 .172 .918 .01 .73 .672 .02 .59
Multivariate analysis HR (95% CI) Multivariate analysis HR (95% CI)
P values
Univariate analysis HR (95% CI)
P values
HCC
985
feasibility of discontinuing ETV or TDF in this setting because ETV or TDF may maintain full HBV suppression at a rate >90% with virtually no drug resistance after 2–3 years of therapy.33 Fifth, LAM is no longer a recommended first-line Nuc for CHB in guidelines of major liver associations.4–6 However, LAM is still used not only in Asian countries including Singapore21,34 but also in some European countries such as Poland and Turkey in up to 45% of the patients.35 In conclusion, the present study has demonstrated that cessation of LAM therapy after recovery from hepatitis flare with HD was safe for most patients. However, 8.2% developed HD within 1 year, although all were rescued by timely retreatment. Hence, the stopping strategy can be considered as a feasible alternative to indefinite therapy only if proper offtherapy monitoring is provided and adhered to. This strategy is especially relevant in low resource countries where patients are unable to afford the cost or are not willing to keep on unlimited daily medication. It is also important to note that cessation of antiviral therapy with properly scheduled monitoring would be much safer than stopping therapy by patients themselves against the advice of their physician for whatever reason(s), because such patients may encounter unpredictable untoward outcomes. In real-world practice, it is not uncommon to see serious liver-related complication(s) in patients who defaulted on follow-up after stopping Nuc therapy. Along this line, measures and effort are required to ensure adherence to close monitoring, which is the key to a safe off-therapy course. Of note is that the data in this study do not apply to patients with decompensated cirrhosis unrelated to hepatitis flare. Controlled trials of ETV or TDF in CHB patients with HD are required to confirm our findings and conclusions.
.021 .717 .268 .382 .627 .025 .597 .124 .164 .289 .24 .247
3. Peng CY, Chien RN, Liaw YF. Hepatitis B virus-related decompensated liver cirrhosis: benefits of antiviral therapy. J Hepatol 2012;57:442–450.
3.958 0.996 1.392 1.94 1.154 1.932 0.989 1.000 0.961 0.983 1.374 1.934
(1.23–12.74) (0.972–1.02) (0.775–2.498) (0.894–1.34) (0.647–2.159) (1.03–3.988) (0.951–1.029) (0.999–1.000) (0.909–1.016) (0.953–1.014) (0.809–2.33) (0.634–5.90)
References 1. Liaw YF, Chu CM. Hepatitis B virus infection. Lancet 2009; 373:582–592. 2. Chu CM, Liaw YF. Hepatitis B virus-related cirrhosis: natural history and treatment. Semin Liver Dis 2006;26:142–152.
Gender (male) Age (y) HBeAg (positive) HBV DNA (log10 IU/mL) Cirrhosis Genotype (C) LAM duration (mo) ALT (U/L) dPT (sec) Bilirubin (mg/dL) Albumin (g/dL) Creatinine (mg/dL)
Factors
Univariate analysis HR (95% CI)
P values
HD
Table 3. Logistic Regression Analyses of Factors for HD and HCC Occurred During Off-therapy Follow-up
1.0541 (1.006–1.003)
Stop Nuc in CHB With Decompensation P values
May 2015
4. Lok AS, McMahon BJ. Chronic hepatitis B: update 2009. Hepatology 2009;50:661–662. 5. Liaw YF, Kao JH, Piratvisuth T, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2012 update. Hepatol Int 2012;6:531–561. 6. European Association for the Study of the Liver. EASL clinical practice guidelines: management of chronic hepatitis B virus infection. J Hepatol 2012;57:167–185. 7. Lampertico P, Viganò M, Colombo M. Why do I treat HBeAgnegative chronic hepatitis B patients with pegylated interferon? Liver Int 2013;33(Suppl 1):157–163. 8. Locarnini S, Zoulim F. Molecular genetics of HBV infection. Antivir Ther 2010;15(Suppl 3):3–14.
986
Chang et al
9. Chotiyaputta W, Peterson C, Ditah FA, et al. Persistence and adherence to nucleos(t)ide analogue treatment for chronic hepatitis B. J Hepatol 2011;54:12–18.
Clinical Gastroenterology and Hepatology Vol. 13, No. 5 stringent cessation criteria for adefovir. Arch Virol 2012; 157:285–290. 25. Locarnini S, Zoulim F. Molecular genetics of HBV infection. Antivir Ther 2010;3(Suppl):3–14.
10. Han SH, Jing W, Mena E, et al. Adherence, persistence, healthcare utilization, and cost benefits of guidelinerecommended hepatitis B pharmacotherapy. J Med Econ 2012;15:1159–1166.
26. Levrero M, Pollicino T, Petersen J, et al. Control of cccDNA function in hepatitis B virus infection. J Hepatol 2009; 51:581–592.
11. Jeng WJ, Sheen IS, Chen YC, et al. Off therapy durability of response to Entecavir therapy in hepatitis B e antigen negative chronic hepatitis B patients. Hepatology 2013;58:1888–1896.
27. Fattovich G, Bortolotti F, Donato F. Natural history of chronic hepatitis B: special emphasis on disease progression and prognostic factors. J Hepatol 2008;48:335–352.
12. Sheen IS, Liaw YF, Tai DI, et al. Hepatic decompensation associated with hepatitis B e antigen clearance in chronic type B hepatitis. Gastroenterology 1985;89:732–735.
28. Chen YC, Jeng WJ, Chien RN, et al. Hepatocellular carcinoma development in HBe-negative hepatitis B cirrhotic patients with entecavir treatment. J Hepatol 2013;58:S214.
13. Chien RN, Lin CH, Liaw YF. The effect of lamivudine therapy in hepatic decompensation during acute exacerbation of chronic hepatitis B. J Hepatol 2003;38:322–327.
29. Lin DY, Sheen IS, Chiu CT, et al. Ultrasonographic changes of early liver cirrhosis in chronic hepatitis B: a longitudinal study. J Clin Ultrasound 1993;21:303–308.
14. Liaw YF. Hepatitis flares and hepatitis B e antigen seroconversion: implication in anti-hepatitis B virus therapy. J Gastroenterol Hepatol 2003;18:246–252.
30. Liaw YF. Clinical utility of hepatitis B surface antigen quantitation in patients with chronic hepatitis B: a review. Hepatology 2011;53:2121–2129.
15. Liaw YF, Tai DI, Chu CM, et al. The development of cirrhosis in patients with chronic type B hepatitis: a prospective study. Hepatology 1988;8:493–496.
31. Chu CM, Yeh CT, Lee CS, et al. Procore stop mutant in HBeAgpositive patients with chronic hepatitis B: clinical characteristics and correlation with the course of HBeAg-to-anti-HBe seroconversion. J Clin Microbiol 2002;40:16–21.
16. Bruix J, Sherman M, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020–1022. 17. Hosaka T, Suzuki F, Kobayashi M, et al. Long-term entecavir treatment reduces hepatocellular carcinoma incidence in patients with hepatitis B virus infection. Hepatology 2013;58:98–107. 18. Niro GA, Ippolito AM, Fontana R, et al. Long-term outcome of hepatitis B virus-related chronic hepatitis under protracted nucleos(t)ide analogues. J Viral Hepat 2013;20:502–509. 19. Liaw YF, Raptopoulou-Gigi M, Cheinquer H, et al. Efficacy and safety of entecavir versus adefovir in chronic hepatitis B patients with hepatic decompensation: a randomized open-label study. Hepatology 2011;54:91–100. 20. Liaw YF, Sheen IS, Lee CM, et al. Tenofovir disoproxil fumarate (TDF), emtricitabine/TDF and entecavir in patients with decompensated chronic hepatitis B liver disease. Hepatology 2011; 53:62–72. 21. Lim SG, Aung MO, Chung SW, et al. Patient preferences for hepatitis B therapy. Antivir Ther 2013;18:663–670. 22. Hadziyannis SJ, Sevastianos V, Rapti I, et al. Sustained responses and loss of HBsAg in HBeAg-negative patients with chronic hepatitis B who stop long-term treatment with adefovir. Gastroenterology 2012;143:629–636. 23. Liu F, Wang L, Li XY, et al. Poor durability of lamivudine effectiveness despite stringent cessation criteria: a prospective clinical study in hepatitis B e antigen-negative chronic hepatitis B patients. J Gastroenterol Hepatol 2011;26:456–460. 24. Ha M, Zhang G, Diao S, et al. A prospective clinical study in hepatitis B e antigen-negative chronic hepatitis B patients with
32. Tsai WL, Lo GH, Hsu PI, et al. Role of genotype and precore/ basal core promoter mutations of hepatitis B virus in patients with chronic hepatitis B with acute exacerbation. Scand J Gastroenterol 2008;43:196–201. 33. Pol S, Lampertico P. First-line treatment of chronic hepatitis B with entecavir or tenofovir in ‘real-life’ settings: from clinical trials to clinical practice. J Viral Hepat 2012;19:377–386. 34. Liaw YF. Antiviral therapy of chronic hepatitis B: opportunities and challenges in Asia. J Hepatol 2009;51:403–410. 35. Arama V, Leblebicioglu H, Simon K, et al. Chronic hepatitis B monitoring and treatment patterns in five European countries with different access and reimbursement policies. Antivir Ther 2014;19:245–257. Reprint requests Address requests for reprints to: Yun-Fan Liaw, MD, 199, Tung Hwa North Road, Taipei 105, Taiwan. e-mail: liveryfl@gmail.com; fax: 886-3-3282824. Acknowledgments The authors thank Ms Li-Hua Lu for laboratory work, Ms Yu-Ju Lan for data collection, Ms Shu-Chun Chen for data mining, and Ms Su-Chiung Chu for assistance in preparing the manuscript. Conflicts of interest This author discloses the following: Y. F. Liaw has been involved with clinical trials and served as a global advisory board member of Roche. The remaining authors disclose no conflicts. Funding Supported by grants from Chang Gung Medical Research Fund (SMRPG1005, OMRPG380061) and the Prosperous Foundation, Taipei, Taiwan.