Spontaneous hepatitis B e antigen (HBeAg) seroconversion

Transcription

1 GASTROENTEROLOGY 2007;133: Pre-S Deletion and Complex Mutations of Hepatitis B Virus Related to Advanced Liver Disease in HBeAg-Negative Patients CHIEN HUNG CHEN,*, CHAO HUNG HUNG,* CHUAN MO LEE,*,, TSUNG HUI HU,* JING HOUNG WANG,* JYH CHWAN WANG,* SHENG NAN LU,* and CHI SIN CHANGCHIEN* *Division of Hepatogastroenterology, Department of Internal Medicine, Graduate Institute of Clinical Medical Sciences, Chang Gung Memorial Hospital-Kaohsiung Medical Center, and Department of Chinese Medicine, Chang Gung University College of Medicine, Kaohsiung, Taiwan Background & Aims: This longitudinal study investigated the interactions and roles of hepatitis B virus (HBV) genotypes, pre-s deletions, and core promoter and precore mutations on the progression of liver disease in hepatitis B e antigen (HBeAg)-negative patients. Methods: A total of 141 HBeAg-negative patients without liver cirrhosis or hepatocellular carcinoma at study entry were recruited for this study, including 45 inactive HBV carriers and 96 patients with HBeAg-negative chronic hepatitis B. The HBV genotypes and the sequences of pre-s, core promoter, and precore regions were determined. Results: Compared with patients without developing liver cirrhosis, patients with the development of liver cirrhosis had higher rates of genotype C; pre-s deletions; C or G1753, T1762/A1764, T1766, and/or A1768 mutants; and G1799 variant. Cox regression analysis showed that older age, higher total bilirubin and HBV DNA levels, pre-s deletions, and T1766 and/or A1768 mutants were significantly associated with the development of liver cirrhosis. HBV with a complex mutation pattern (pre-s deletion, T1762/A1764, and T1766 and/or A1768 mutants) rather than a single mutation was associated with the development of liver cirrhosis, and the patterns of mutation combinations differed between HBV genotype B and C. Moreover, pre-s deletion was a significant risk factor for hepatocellular carcinoma. Conclusions: This study indicated that pre-s deletion and combined mutations of HBV are useful molecular markers for predicting the clinical outcomes of HBeAg-negative patients. Spontaneous hepatitis B e antigen (HBeAg) seroconversion during the course of chronic hepatitis B virus (HBV) infection usually correlates with the normalization of biochemical markers of hepatitis and drastically reduced viremia in HBV carriers. However, active HBV replication and an increase in serum alanine aminotransferase (ALT) levels recur or persist in 15% 33.2% of patients after HBeAg seroconversion. 1 3 HBeAg-negative chronic hepatitis B (CHB) is a potentially progressive form of chronic liver disease that often progresses to cirrhosis and hepatocellular carcinoma (HCC). 1 5 Mutations in the precore and core promoter regions of the HBV genome are commonly observed in patients with HBeAg-negative CHB. The most common precore and core promoter variants are G to A at nucleotide (nt) 1896 (A1896) and A to T at nt 1762 and G to A at nt 1764 (T1762/A1764, respectively) However, not all patients with HBeAg-negative CHB develop A1896 or T1762/A1764 mutations. Therefore, it would be of interest to clarify whether these virologic mutants contribute to persistent viral replication and ALT flare in HBeAgnegative patients. Several recent cross-sectional studies have suggested that HBV genotype C and the T1762/ A1764 mutants are strongly associated with liver cirrhosis and HCC rather than HBV genotype B and the A1762/ T1764 wild-type virus However, some studies report no significant correlation between HBV genotypes or core promoter mutations 21 and clinical outcome. The major limitation of such cross-sectional designs is the difficulty ascertaining whether these mutations preceded the development of advanced liver disease. Thus, longitudinal studies are needed to clarify these important issues. HBV DNA has 4 open reading frames that encode viral envelope protein: the core and e antigen, the HBx protein, and the viral polymerase. The HBV envelope is composed of 3 forms of hepatitis B surface antigen (HBsAg), the so-called large (L, coded for by the pre-s1/s2/s gene), middle (M, the pres2/s gene), and small (S, the S gene) protein. The small or major peptide is 226 amino acids in length, and the M and L proteins are assembled by amino-terminal extension of 55 amino acids at the pre-s2 domain and of amino acids (depending on the strain) of the pre-s1 domain. The pre-s1 and pre-s2 regions play an essential role in the interaction with immune responses because they contain several epitopes for T or B cells. 22 The pre-s1 protein contains the hepatocyte binding site (amino acids 21 47) and is known to be essential for virion assembly and for the transporting Abbreviations used in this paper: CHB, chronic hepatitis B; CI, confidence interval; nt, nucleotide; PCR, polymerase chain reaction by the AGA Institute /07/$32.00 doi: /j.gastro

2 November 2007 PRE-S DELETION IN HBeAg-NEGATIVE PATIENTS 1467 of virions out of the hepatocyte. 23 The Pr-S2 domain can bind to polymerized human serum albumin (amino acids 3 16), but the significance of this binding is unknown. 24 Previous studies demonstrated that a number of pre- S1/S2 rearrangements, including deletions and initiation codon mutants, have been identified and accumulated in patients at a later stage of persistent HBV infection and fulminant hepatitis. 25,26 Recent cross-sectional studies have also demonstrated that patients with progressive liver diseases have a higher frequency of pre-s deletions. 27,28 However, the effects of pre-s deletions on the progression of liver disease among HBeAg-negative patients are rarely investigated in longitudinal studies. Because previous studies only focused on one of these factors (eg, pre-s deletions, HBV genotype, or core promoter and precore mutations), it is unclear whether these factors are confounding or whether a specific combination of these factors is associated with the development of advanced liver disease in HBeAg-negative patients. This longitudinal study therefore investigated the interactions and roles of pre-s deletions, HBV genotypes, and core promoter and precore mutations on the progression of liver disease in HBeAg-negative patients. Materials and Methods Patients From 1995 to 1998, HBsAg-positive carriers who had undergone long-term follow-up at Chang Gung Memorial Hospital Kaohsiung Medical Center were included in this study if they fulfilled the following criteria: (1) seropositivity for HBsAg for at least 6 months before entry, (2) seronegativity for HBeAg at entry, (3) a regular follow-up for at least 36 months, (4) no evidence of liver cirrhosis or HCC based on the clinical criteria and ultrasound examination at baseline, and (5) no special antiviral treatment (such as interferon, lamivudine, or adefovir) before entry or during follow-up. Patients were excluded if they had any evidence of autoimmune hepatitis or markers of hepatitis C virus, hepatitis D virus, or human immunodeficiency virus. A total of 148 HBeAg-negative patients who fulfilled these criteria were recruited into the study. Of the 148 patients, 98 patients with a history of intermittently or persistently elevated ALT levels ( 60 U/L) and HBV DNA 10 5 copies/ml during this follow-up period (mean SD, months) were defined as HBeAg-negative CHB (group I); 50 consecutive controls with persistent ALT normality ( 40 U/L) during the follow-up period (mean SD, months) were defined as the inactive HBV carriers (group II). Liver ultrasonography was performed every 3 6 months. An ultrasound scoring system for liver surface, parenchyma, vascular structure, and spleen size was used to describe the severity of hepatic parenchymal damage. 29,30 Cirrhosis was diagnosed by liver biopsy histology or repeated ultrasound findings suggestive of cirrhosis at least twice 6 months apart, supplemented with clinical criteria indicating portal hypertension (ie, the presence of ascites, thrombocytopenia, and esophageal varices). Diagnostic criteria for diagnosis of HCC were (1) positive findings from cytologic or pathologic examinations or (2) typical images compatible with HCC with an -fetoprotein level 400 ng/ml. Methods Patients were followed up every 3 6 months. Routine follow-up studies included clinical assessment, conventional liver biochemical tests, -fetoprotein level, and serologic hepatitis B markers (including HBeAg and antibody to hepatitis B e antigen). The earliest available serum sample was assessed to analyze the HBV DNA level, genotype, pre-s sequences, and core promoter and precore sequences. The sera were frozen at 70 C until use. In our judgment, we lacked sufficient information to perform an informative power calculation before conducting our study. Serology The presence of HBsAg, HBeAg, anti-hcv antibodies, and anti hepatitis D virus antibodies was determined using commercial assay kits (HBsAg, enzyme immunoassay, Abbott, North Chicago, IL; HBeAg, enzyme immunoassay, Abbott; anti-hcv, enzyme immunoassay 3.0, Abbott; anti hepatitis D virus, radioimmunoassay, Abbott). The HBV DNA was quantified using Amplicor HBV Monitor (Roche Diagnostic Systems, Branchburg, NJ) with a detection limit of 200 copies/ml. Dilution was performed if HBV DNA levels exceeded 10 6 copies/ml. Nested Polymerase Chain Reaction and Direct Sequencing of the Precore, Core Promoter, and Pre-S Regions The precore and core promoter sequences in sera were determined using nested polymerase chain reaction (PCR) and direct sequencing, as described previously. 13,31 All sequences were interpreted in the precore and core promoter regions, and dual signals (mixed type) were classified as a mutant type. For pre-s sequence analysis, pre-s genes were amplified using nested PCR. First-round PCR primers were 5=- AAAATTAATTATCCTGCTAGG-3= (nt ) and 5=-GAGAAGTCCACCACGAGTC-3= (nt ). The PCR was performed as follows: 95 C for 2 minutes, 95 C for 50 seconds, 50 C for 50 seconds, and 72 C for 2 minutes for 30 cycles, and finally 72 C for 7 minutes. Second-round PCR primers were 5=-TTTACAACTCTGT- GGAAGGC-3= (nt ) and 5=-GAGTCTAGACT- CTGTGGTATTGTG-3= (nt ) and the same PCR condition as the first-round reaction. All necessary precautions to prevent cross-contamination were taken, and negative controls were included in each assay. The nucleotide sequences of the amplified products were directly

3 1468 CHEN ET AL GASTROENTEROLOGY Vol. 133, No. 5 determined by using fluorescent-labeled primers with an ABIPRISM 377 Genetic Analyzer (Applied Biosystems, Foster City, CA). The pre-s deletion mutants were determined by direct sequencing for all patients. Molecular cloning and sequencing was performed only when pre-s deletions were found by direct sequencing and if the pre-s deletion sites were obscure. The HBV pre-s gene PCR products were cloned into the pgem-t Easy Vector System (Promega, Madison, WI) after adding A-tailing procedure. After transformation into Escherichia coli, the individual recombinant clone was selected and purified using a commercially available kit (GFX Micro Plasmid Prep Kit; Amersham Pharmacia Biotech Europe GmbH, Freiburg, Germany) for further sequencing analysis. HBV Genotyping The HBV genotypes in sera were determined using restriction fragment length polymorphism on surfacegene sequence, amplified by PCR with nested primers, as described previously. 32 Data Analysis Data are presented as means SD, proportions, or median (range). To compare the values between the 2 groups, 2 or Fisher exact tests were applied to analyze categorical variables and Student t tests and Mann Whitney U test were used for continuous variables with normal and skewed distributions, respectively. Multiple logistic regression analysis was performed to assess the influence of all baseline factors on HBeAg-negative CHB and pre-s deletions. Cumulative incidences of liver cirrhosis or HCC were analyzed by Kaplan Meier method with a log rank test. Univariate and multivariate analyses were performed to identify baseline factors associated with progression to cirrhosis and HCC using the Cox proportional hazards regression models. All statistical tests were 2 sided. P.05 was considered statistically significant. All statistical analyses were performed using SPSS 13.0 for Windows (SPSS Inc, Chicago, IL). Results Clinical Characteristics at Baseline After the exclusion of 2 patients diagnosed with liver cirrhosis within 12 months after study entry and 5 inactive carriers who were negative for PCR due to very low HBV DNA levels, 141 patients were enrolled in this study (96 in group I and 45 in group II). These patients included 109 men and 32 women, with a mean age of years at entry. Table 1 compares the baseline characteristics between patients with HBeAg-negative CHB (group I) and inactive HBV carriers (group II). Multiple logistic regression analysis was performed to examine age, sex, HBV genotype, pre-s deletions, and precore and core promoter mutations. Older age (odds Table 1. Comparison of Baseline Characteristics Between Inactive Carriers and Patients With HBeAg-Negative CHB HBeAg-negative CHB (n 96) Inactive carriers (n 45) P value Age (y), mean SD Sex (M/F) 83/13 26/ ALT (U/L), median 59 ( ) 22 (7 40).001 (range) HBV DNA (log copies/ml), mean SD HBV genotypes B/C 74/22 37/8.49 Pre-S deletion (%) 19 (19.8) 8 (17.8).78 T or G1753 (%) 16 (16.7) 3 (6.7).12 T1762/A1764 (%) 51 (53.1) 12 (26.7).003 T1766 and/or 12 (12.5) 1 (2.2).062 A1768 (%) G1799 (%) 74 (77.1) 33 (73.3).63 A1896 (%) 67 (69.8) 29 (64.4).53 A1899 (%) 26 (27.1) 1 (2.2).001 ratio, 1.099; 95% confidence interval [CI], ; P.001), male gender (odds ratio, 5.95; 95% CI, ; P.001), and A1899 mutant (odds ratio, 12.2; 95% CI, ; P.02) were independently associated with HBeAg-negative CHB. Of the 141 patients, 111 had genotype B and 30 had genotype C. Patients with HBV genotype C had higher HBV DNA levels and higher frequencies of pre-s deletions; G or C1753, T1762/ A1764, and T1766 and/or A1768 mutations; and lower frequencies of G1752 and G1799 variants and A1896 mutation than those with HBV genotype B. Biochemical Pattern in Patients With HBeAg-Negative CHB During Follow-Up Of the 96 patients in group I, 4 major patterns of biochemical changes were identified. First, ALT flares (ALT 200 U/L) with normal ALT levels ( 40 U/L) between flares were observed in 42 patients (43.8%). Second, ALT flares without normal ALT levels between flares were observed in 14 patients (14.6%). Third, elevated ALT levels ( 60 U/L) without ALT flares ( 200 U/L) were observed in 30 patients (31.3%). Fourth, one-time ALT flare followed by persistently normal ALT levels was observed in 10 patients (10.4%). Patients with ALT flares (first and second biochemical patterns) were older (mean SD, vs years; P.028) and had higher HBV DNA levels (mean SD log copies/ ml, vs ; P.001) at study entry than those without ALT flares (third and forth biochemical patterns). Patients with and without ALT flares did not significantly differ in terms of sex, HBV genotypes, pre-s deletions, and core promoter and precore mutations. Of the 28 patients who developed liver cirrhosis, 15 had the first biochemical pattern, 7 had the second, 5 had the third, and one had the fourth. Patients with development

4 November 2007 PRE-S DELETION IN HBeAg-NEGATIVE PATIENTS 1469 Table 2. Univariate Analysis of Risk Factors for Development of Liver Cirrhosis Total (N 141) Group I (n 96) Variable Comparison Hazard ratio (95% CI) P value Hazard ratio (95% CI) P value Age per 1 year increased ( ) ( ).09 Sex Male vs female 1.78 ( ) ( ).89 ALT 40 vs 40 U/L 3.71 ( ) ( ).17 Total bilirubin per 1 mg/dl increased 1.14 ( ) ( ).001 HBV DNA per 1 log copies/ml increased 1.67 ( ) ( ).005 HBV genotypes B vs C 3.26 ( ) ( ).002 Pre-S deletions Presence vs absence 3.11 ( ) ( ).002 T or G1753 Presence vs absence 2.91 ( ) ( ).031 T1762/A1764 Presence vs absence 3.27 ( ) ( ).016 T1766 and/or A1768 Presence vs absence 6.45 ( ) ( ).0001 G1799 Presence vs absence 0.37 ( ) ( ).003 A1896 Presence vs absence 0.80 ( ) ( ).43 of liver cirrhosis had more frequent ALT flares than those without (22/28 vs 34/68; P.01). Predictive Factors for Liver Cirrhosis Among the 141 patients, 28 developed new cirrhosis after a median of 44 months (range, months) after entry into the study. Of these, 10 were diagnosed by liver biopsy histology. The frequency of ultrasound examinations before liver cirrhosis did not significantly differ between those with and without development of cirrhosis. All cases of cirrhosis were in the patients in group I. Univariate analysis showed that age at entry, ALT level 40 U/L, higher total bilirubin and HBV DNA levels, genotype C, pre-s deletions, C or G1753 mutant, T1762/A1764 mutants, T1766 and/or A1768 mutants, and G1799 variant were risk factors for developing liver cirrhosis in all 141 patients (Table 2); additionally, risk factors for liver cirrhosis in patients in group I were higher total bilirubin and HBV DNA levels; genotype C; pre-s deletions; C or G1753 mutant; T1762/A1764, T1766, and/or A1768 mutants; and G1799 variant (Table 2). Multivariate analysis revealed that older age, higher total bilirubin and HBV DNA levels, pre-s deletions, and T1766 and/or A1768 mutants were significantly associated with increased risk for cirrhosis in all 141 patients (Table 3), and the risk factors for liver cirrhosis in patients in group I were older age, higher total bilirubin and HBV DNA levels, and pre-s deletions (Table 3). Figures 1 and 2 display the cumulative probability of progression to cirrhosis in relation to pre-s deletion and T1766 and/or A1768 mutants, respectively. Moreover, compared with the patients having serum HBV DNA levels of less than 10 3 copies/ml (n 21), patients with serum HBV DNA levels exceeding 10 6 copies/ml (n 33) had a hazard ratio of (95% CI, ; P.008) for liver cirrhosis. The risk for liver cirrhosis did not significantly differ between patients with serum HBV DNA levels of (n 21), 10,000 99,999 (n 25), and 100, ,999 (n 41) copies/ml and those with HBV DNA levels of less than 10 3 copies/ml (Figure 3). A statistical analysis of the 3 mutation combinations (pre-s deletion, T1762/A1764, and T1766 and/or A1768) (Table 4) showed that any 2 or 3 combinations rather than a single mutation were significantly associated with liver cirrhosis. However, in HBV genotype B, only T1762/ A1764 mutations combined with pre-s deletions was associated with liver cirrhosis. In contrast, in HBV genotype C, T1762/A1764 and T1766 and/or A1768 mutations, T1766 and/or A1768 and pre-s deletions, and T1762/A1764 and T1766 and/or A1768 and pre-s deletions statistically differed between patients with and without liver cirrhosis. However, for genotype B, the prevalence of T1766 and/or A1768 mutations was too low to determine whether that mutation was associated with cirrhosis. Table 3. Multivariate Analysis of Risk Factors for Development of Liver Cirrhosis Total (N 141) Group I (n 96) Factors Comparison Hazard ratio (95% CI) P value Hazard ratio (95% CI) P value Age per 1 year increased ( ) ( ).046 Total bilirubin per 1 mg/dl increased ( ) ( ).001 HBV DNA per 1 log copies/ml increased ( ) ( ) and/or 1768 mutant Presence vs absence ( ) ( ).063 Pre-S deletions Presence vs absence ( ) ( ).016

5 1470 CHEN ET AL GASTROENTEROLOGY Vol. 133, No. 5 Figure 1. Comparison of cumulative incidence of liver cirrhosis between patients with and without pre-s deletion mutants. Predictive Factors for HCC Among the 141 patients, 7 developed HCC (median, 84 months; range, months). One case of HCC was in group I (no cirrhosis in nontumor area after operation), and 6 were in group II. Univariate analysis indicated that pre-s deletions (hazard ratio, 11.26; 95% CI, ; P.004) and G1896 wild-type (hazard ratio, 5.81; 95% CI, ; P.036) were factors significantly predictive of HCC. However, HBV DNA level (per 1 log copies/ml increased) (hazard ratio, 1.60; 95% CI, ; P.059) was of borderline statistical significance. Multivariate analysis showed that only pre-s deletions (hazard ratio, 11.25; 95% CI, ; P.004) were independent determinants for HCC. Figure 4 shows the cumulative probability of HCC development in pre-s deletions. Figure 2. Comparison of cumulative incidence of liver cirrhosis between patients with and without T1766 and/or A1768 mutants. Figure 3. Cumulative incidence of liver cirrhosis by serum HBV DNA levels (copies/ml) at study entry. P.001 for log-rank test. Risk Factors for Pre-S Deletions Of the 27 patients with pre-s deletions, 23 patients were coinfected with wild-type pre-s subgenomes (mixed infection). Table 5 compares baseline characteristics of patients with and without pre-s deletions. Patients with pre-s deletions had higher rates of HBV genotype C; T1762/A1764, T1766, and/or A1768 mutations; and G1799 wild-type virus than those without pre-s deletions. Based on multiple logistic regression analysis for all baseline variants, only T1766 and/or A1768 mutation (odds ratio, 6.3; 95% CI, ; P.002) was significantly associated with pre-s deletions. Most of the pre-s deletion regions encompassed T- and B-cell epitopes and important functional sites. No specific site of pre-s deletions was predictive of liver cirrhosis or HCC development. Discussion In this longitudinal study, pre-s deletion and T1766 and/or A1768 mutants of HBV were independent factors significantly predictive of cirrhosis in HBeAgnegative patients with HBV infection. Univariate analysis indicated that HBV genotype C and T1762/A1764 mutations were associated with cirrhosis. However, they were not independently associated with liver cirrhosis after adjusting for pre-s deletions and T1766 and/or A1768 mutations. Mutation of T1766/A1768 has been reported previously in patients with fulminant hepatitis 33 and associated with a high level of replication and enhanced encapsidation of pregenomic RNA in in vitro studies. 33,34 Moreover, the presence of T1766/A1768 mutations reduced cell viability due to the induction of apoptosis. 34 This study demonstrated that T1766 and/or A1768 mutations were independently associated with higher HBV DNA levels (data not shown). Thus, T1766 and/or A1768 mutants associated with high levels of HBV DNA may induce more severe liver injury and increase the risk of liver cirrhosis.

6 November 2007 PRE-S DELETION IN HBeAg-NEGATIVE PATIENTS 1471 Table 4. Association Between Liver Cirrhosis and the Presence of Specific Mutation Patterns of HBV All patients HBV genotype B HBV genotype C LC Non-LC LC Non-LC LC Non-LC Mutation pattern (n 28) (n 113) P value (n 16) (n 94) P value (n 12) (n 19) P value No mutation 6 (21.4) 62 (54.9) (37.5) 59 (62.8) (15.8).26 Any mutation 22 (78.6) 51 (45.1) (62.5) 35 (37.2) (100) 16 (84.2).26 T1762/A1764 alone or in any 20 (71.4) 43 (38.1) (56.3) 27 (28.7) (91.7) 16 (84.2) 1.0 combination T1762/A1764 alone 8 (28.6) 32 (28.3).98 5 (31.3) 23 (24.5).55 3 (25) 9 (47.3).27 T1762/A1764 and T1766 and/or 8 (28.6) 4 (3.5) (6.3) 1 (1.1).27 7 (58.3) 3 (15.8).021 A1768 T1766 and/or A1768 alone or in 9 (32.1) 4 (3.5) (6.3) 1 (1.1).27 8 (66.7) 3 (15.8).032 any combination T1766 and/or A1768 alone 1 (3.6) (8.3) 0.39 Pre-S deletions alone or in any 11 (39.3) 16 (14.2) (31.3) 10 (10.6) (50) 6 (31.8).31 combination Pre-S deletions alone 1 (3.6) 8 (7.1).69 1 (6.3) 8 (8.5) Pre-S deletions and T1762/A (35.7) 8 (7.1) (25) 2 (2.1) (50) 6 (31.6).31 Pre-S deletions and T1766 and/or 6 (21.4) 1 (0.9) (6.3) (41.7) 1 (5.3).022 A1768 Pre-S deletions and T1762/A (21.4) 1 (0.9) (6.3) (41.7) 1 (5.3).022 and T1766 and/or A1768 NOTE. All values are expressed as n (%) unless otherwise noted. Previous studies showed that pre-s deletions were more frequent in HBV carriers with genotype C than in those with genotype B. 27,28 The current findings are compatible with these results. However, the data showed that T1762/A1764 and T1766 and/or A1768 mutants and the G1799 variant were significantly associated with pre-s deletions. Multiple logistic regression analysis indicated that T1766 and/or A1768 mutations were independent factors associated with pre-s deletions. Recent cross-sectional studies have shown that patients with later stages of persistent HBV infection and progressive liver diseases have an increased incidence of pre-s deletion. 25,27,28 However, the role of pre-s deletions in advanced liver diseases has not been satisfactorily confirmed by longitudinal studies. The current study showed that pre-s deletion was significantly associated with the development of liver cirrhosis and HCC, and most of the deletion regions encompassed T- and B-cell epitopes and important functional sites. The pre-s regions play an essential role in the interaction with the immune responses because they contain several epitopes for T or B cells. 22 In persistent HBV infection, immune epitope deletion mutants may escape the host immune surveillance and lose important functional sites. These deletion mutants might result in the Table 5. Comparison of Baseline Characteristics Between Patients With and Without Pre-S Deletion Figure 4. Comparison of cumulative incidence of HCC between patients with and without pre-s deletion mutants. No Pre-S deletion (n 114) Pre-S deletion (n 27) P value Age (y), mean SD Sex (M/F) 87/27 22/5.8 ALT (U/L), median 35.5 ( ) 36 (7 535).62 (range) HBV DNA (log copies/ml), mean SD HBV genotype B/C 96/18 15/ T or G1753 (%) 14 (12.3) 5 (18.5).37 T1762/A1764 (%) 46 (40.4) 17 (63.0).034 T1766 and/or 6 (5.3) 7 (25.9).001 A1768 (%) G1799 (%) 93 (81.5) 14 (51.8).001 A1896 (%) 77 (67.5) 19 (70.4).78 A1899 (%) 20 (17.5) 7 (25.9).32

7 1472 CHEN ET AL GASTROENTEROLOGY Vol. 133, No. 5 intracellular retention of HBV envelope proteins and viral particles, as well as the formation of ground glass appearing hepatocytes. 25,35 In a transgenic mice model, overexpression of the HBV large surface protein has been shown to be cytopathic and could cause cell death, regeneration in the liver, oxidative DNA damage, and eventually hepatocellular carcinogenesis. 36 Recent research has shown that pre-s mutant HBsAg contributes to 2 patterns of ground glass hepatocyte : types I and II with deletion over the pre-s1 and pre-s2 regions, respectively. 25,37 These deletion mutants accumulate in the endoplasmic reticulum and cause endoplasmic reticulum stress signals. Through endoplasmic reticulum stress signaling pathways, large amounts of reactive oxygen species are generated and lead to oxidative DNA damage of HBV-infected hepatocytes, such as the 8-oxoguanine DNA lesion. 38 Presence of the oxidative DNA lesions stimulates DNA repair activity; the induced mutagenesis occurs in the genome. The genomic instability is therefore likely to occur and may lead to liver damage and HCC. 38 In the present study, HBV with a complex mutation pattern (pre-s deletions, T1762/A1764 and T1766 and/or A1768 mutations) rather than a single mutation was associated with the development of liver cirrhosis, and the complex mutation patterns differed between HBV genotype B and C. Chen et al 28 found that complex viral mutants with precore/basal core promoter mutations and pre-s deletion were apparently associated with the development of advanced liver diseases. Preikschat et al 39 reported that HBV variant populations characterized by deletions/insertions in core promoter plus deletions in C gene and/or deletions in the pre-s region were observed more frequently in patients with development of liver cirrhosis than those without during follow-up of immunosuppressed renal transplant recipients. The characteristic phenotype of complex variants was a combination of increased genome replication, reduced splicing of pregenomic RNA, and defect in HBeAg, core, and surface protein expression and secretion. 40 Thus, the accumulation of HBV complex variants may have a synergistic role in advanced liver diseases. In addition, a previous study indicated that deletions in the core promoter emerge first but are unrelated to development of liver cirrhosis if present alone. Apparently, core promoter mutations provide the basis for accumulated deletions in the C gene and pre-s region. 39 Results were similar in the current study because pre-s deletions and T1766 and/or A1768 mutations were observed nearly exclusively in virus populations with T1762/A1764 mutations. Also, the HBV genotype C had higher frequencies of pre-s deletions, T1762/A1764, and T1766, and/or A1768 mutations than HBV genotype B. These combined mutations had a synergistic role in the development of liver cirrhosis in HBV genotype C. Thus, the close association between HBV genotype C and these mutation combinations suggests an association between genotype C and more severe liver disease. This study suggested that higher HBV DNA level was an independent predictor for the development of liver cirrhosis. This is consistent with previous community studies showing that cirrhosis and HCC risk increased significantly with increasing HBV DNA levels is independent of HBeAg status. Significantly increased risk was noted at a serum HBV DNA level of 10 4 copies/ml. 41,42 However, the current study showed that the risk of cirrhosis increased significantly at a serum HBV DNA level exceeding 10 6 copies/ml. Moreover, HBV DNA levels were not significantly related to development of HCC. The small number of cases developing HCC and the short follow-up period in this study may explain this discrepancy. The present study is limited by the absence of histologic assessment for all patients. Although liver biopsy remains the gold standard for assessing fibrosis stage, this invasive diagnosis is associated with potential complications, sampling errors, and interobserver and intraobserver variability. 43 In this study, the diagnosis of cirrhosis for some patients was based on clinical criteria and ultrasound scoring system, which is reportedly helpful for diagnosing cirrhosis, particularly in patients with HBV infection. 29,30 Lin et al 29 demonstrated an 82.8% sensitivity and 80% specificity for diagnosing of liver cirrhosis with an ultrasound score of 7 in patients with CHB. Our previous study showed that ultrasound scores were significantly correlated with hepatic fibrosis scores, and an ultrasound score of 7 was the best cut-off point for predicting HBV-related cirrhosis, with a sensitivity and specificity of 77.8% and 92.5%, respectively. 30 Recently, Nishiura et al 44 reported that the accumulated ultrasound scores of 3 parameters (edge, surface, and parenchymal texture of the liver) was the most reliable indicator of hepatic fibrosis. A score of 6.5 was found to identify fibrosis stage 4 with 100% sensitivity. However, despite the ability to accurately predict cirrhosis by ultrasonography, interobserver and intraobserver variability may influence the diagnosis of the severity of hepatic parenchymal damage. 30 Moreover, ultrasonography is more likely to underestimate early cirrhosis. In summary, older age, higher total bilirubin and HBV DNA levels, pre-s deletion, and T1766 and/or A1768 mutations were major independent determinants of progression to cirrhosis in HBeAg-negative patients. HBV with a complex mutation pattern rather than a single mutation was associated with the development of liver cirrhosis. Furthermore, pre-s deletion mutant may be significantly associated with the development of HCC. Thus, pre-s deletion and combined mutations of HBV were useful viral genomic markers for predicting the clinical outcomes of HBeAg-negative patients.

8 November 2007 PRE-S DELETION IN HBeAg-NEGATIVE PATIENTS 1473 References 1. Chu CM, Hung SJ, Lin J, et al. Natural history of hepatitis B e antigen to antibody seroconversion in patients with normal serum aminotransferase levels. Am J Med 2004;116: Hsu YS, Chien RN, Yeh CT, et al. Long-term outcome after spontaneous HBeAg seroconversion in patients with chronic hepatitis B. Hepatology 2002;35: Hadziyannis SJ, Vassilopoulos D. Hepatitis B e antigen-negative chronic hepatitis B. Hepatology 2001;34: Brunetto MR, Giarin M, Oliveri F, et al. Wild type and e antigenminus hepatitis B virus in course of chronic hepatitis. Proc Natl Acad Sci U S A 1991;88: Naoumov NV, Schneider R, Grotzinger T, et al. Precore mutant hepatitis B virus infection and liver disease. Gastroenterology 1992;102: Carman WF, Jacyna MR, Hadziyannis S, et al. Mutation preventing formation of hepatitis B e antigen in patients with chronic hepatitis B infection. Lancet 1989;2: Akahane Y, Yamanaka T, Suzuki H, et al. Chronic active hepatitis with hepatitis B virus DNA and antibody against e antigen in the serum. Disturbed synthesis and secretion of e antigen from hepatocytes due to a point mutation in the precore region. Gastroenterology 1990;99: Hamasaki K, Nakata K, Nagayama Y, et al. Changes in the prevalence of HBeAg-negative mutants hepatitis B virus during the course of chronic hepatitis B. Hepatology 1994;20: Okamoto H, Tsuda F, Akahane Y, et al. Hepatitis B virus with mutations in the core promoter for an e antigen-negative phenotype in carriers with antibody to e antigen. J Virol 1994;68: Kurosaki M, Enomoto N, Asahina Y, et al. Mutations in the core promoter region of hepatitis B virus in patients with chronic hepatitis B. J Med Virol 1996;49: Buckwold VE, Xu Z, Chen M, et al. Effects of a naturally occurring mutation in the hepatitis B virus basal core promoter on precore gene expression and viral replication. J Virol 1996;70: Gűnther S, Piwon N, Will H. Wild-type levels of pregenomic RNA and replication but reduced pre-c RNA and e-antigen synthesis of hepatitis B virus with C(1653) T,A(1762) T and G (1764) A mutations in the core promoter. J Gen Virol 1998;79: Chen CH, Lee CM, Lu SN, et al. Clinical significance of hepatitis B virus (HBV) genotypes and precore and core promoter mutations affecting HBV e antigen expression in Taiwan. J Clin Microbiol 2005;43: Kao JH, Chen PJ, Lai MY, et al. Basal core promoter mutations of hepatitis B virus increase the risk of hepatocellular carcinoma in hepatitis B carriers. Gastroenterology 2003;124: Lindh M, Hannoun C, Dhillon AP, et al. Core promoter mutations and genotypes in relation to viral replication and liver damage in East Asian hepatitis B virus carriers. J Infect Dis 1999;179: Lee CM, Chen CH, Lu SN, et al. Prevalence of clinical implications of hepatitis B virus genotypes in southern Taiwan. Scand J Gastroenterol 2003;1: Kao JH, Chen PJ, Lai MY, et al. Hepatitis B genotypes correlate with clinical outcomes in patients with chronic hepatitis B. Gastroenterology 2000;118: Yuen MF, Sablon E, Yuan HJ, et al. Significance of hepatitis B genotype in acute exacerbation, HBeAg seroconversion, cirrhosis-related complications, and hepatocellular carcinoma. Hepatology 2003;37: Sumi H, Yokosuka O, Seki N, et al. Influence of hepatitis B virus genotypes on the progression of chronic type B liver disease. Hepatology 2003;37: Lin CL, Liao LY, Wang CS, et al. Basal core-promoter mutant of hepatitis B virus and progression of liver disease in hepatitis B e antigen-negative chronic hepatitis B. Liver Int 2005; 25: Chun YK, Kim JY, Woo HJ, et al. No significant correlation exists between core promoter mutations, viral replication, and liver damage in chronic hepatitis B infection. Hepatology 2000;32: Chisari FV. Hepatitis B virus immunopathogenesis. Annu Rev Immunol 1995;13: Ganem D. Assembly of hepadnaviral virions and subparticles. Curr Top Microbiol Immunol 1991;168: Sobotta D, Sominskaya I, Jansons J, et al. Mapping of immunodominant B-cell epitopes and the human serum albumin-binding site in natural hepatitis B virus surface antigen of defined genosubtype. J Gen Virol 2000;81: Fan YF, Lu CC, Chen WC, et al. Prevalence and significance of hepatitis B virus (HBV) pre-s mutants in serum and liver at different replicative stages of chronic HBV infection. Hepatology 2001;33: Pollicino T, Zanetti AR, Cacciola I, et al. Pre-S2 defective hepatitis B virus infection in patients with fulminant hepatitis. Hepatology 1997;26: Sugauchi F, Ohno T, Orito E, et al. Influence of hepatitis B virus genotypes on the development of pres deletions and advanced liver disease. J Med Virol 2003;70: Chen BF, Liu CJ, Jow GM, et al. High prevalence and mapping of pre-s deletion in hepatitis B virus carriers with progressive liver diseases. Gastroenterology 2006;130: 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: Hung CH, Lu SN, Wang JH, et al. Correlation between ultrasonographic and pathologic diagnoses of hepatitis B and C virusrelated cirrhosis. J Gastroenterol 2003;38: Chen CH, Lee CM, Lu SN, et al. Comparison of sequence changes of precore and core promoter regions in HBeAg-positive chronic hepatitis B patients with and without HBeAg clearance in lamivudine therapy. J Hepatol 2006;44: Mizokami M, Nakano T, Orito E, et al. Hepatitis B virus genotype assignment using restriction fragment length polymorphism patterns. FEBS Lett 1999;450: Baumert TF, Rogers SA, Hasegawa K, et al. Two core promoter mutations identified in a hepatitis B virus strain associated with fulminant hepatitis result in enhanced viral replication. J Clin Invest 1996;98: Baumert TF, Yang C, Schurmann P, et al. Hepatitis B virus mutations associated with fulminant hepatitis induce apoptosis in primary Tupaia hepatocytes. Hepatology 2005;41: Wang HC, Wu HC, Chen CF, et al. Different types of ground glass hepatocytes in chronic hepatitis B virus infection contain specific pre-s mutants that induce endoplasmic reticulum stress. Am J Pathol 2003;163: Chisari FV, Klopchin K, Moriyama T, et al. Molecular pathogenesis of hepatocellular carcinoma in hepatitis B virus transgenic mice. Cell 1989;59: Fan YF, Lu CC, Chang YC, et al. Identification of a pre-s2 mutant in hepatocytes expressing a novel margin pattern of surface antigen in advanced disease of chronic hepatitis B virus infection. J Gastroenterol Hepatol 2000;15: Hsieh YH, Su IJ, Wang HC, et al. Pre-S mutant surface antigens in chronic hepatitis B virus infection induce oxidative stress and DNA damage. Carcinogenesis 2004;25: Preikschat P, Gunther S, Reinhold S, et al. Complex HBV populations with mutations in core promoter, C gene, and pre-s region are associated with development of cirrhosis in long-term renal transplant recipients. Hepatology 2002;35:

9 1474 CHEN ET AL GASTROENTEROLOGY Vol. 133, No Marschenz S, Endres AS, Brinckmann A, et al. Function analysis of complex hepatitis B virus variants associated with development of liver cirrhosis. Gastroenterology 2006;131: Iloeje UH, Yang HI, Lu J, et al. Predicting cirrhosis risk based on the level of circulating hepatitis B viral load. Gastroenterology 2006;130: Chen CJ, Yang HI, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA 2006;295: Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation on liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002;97: Nishiura T, Watanabe H, Ito M, et al. Ultrasound evaluation of the fibrosis stage in chronic liver disease by the simultaneous use of low and high frequency probes. Br J Radiol 2005;78: Received October 24, Accepted August 9, Address requests for reprints to: Chuan-Mo Lee, MD, Division of Hepatogastroenterology, Department of Internal Medicine, Chang Gung Memorial Hospital-Kaohsiung Medical Center, 123 Ta Pei Road, Kaohsiung, Taiwan. chmolee@ms15.hinet.net; fax: (886) Supported by grant CMRPG 8066 from Chang Gung Memorial Hospital and grant NSC B from the National Council of Science, Taiwan. The authors have no conflict of interest to disclose.