DNA 2 91% (185/203), 82% (75/91), 75% (47/63), 62% (26/42) 200,000, 20, ,999, 2,000-19,999, <2,000 ( P

Transcription

1 Is There a Meaningful Serum Hepatitis B Virus DNA Cutoff Level for Therapeutic Decisions in Hepatitis B e Antigen Negative Chronic Hepatitis B Virus Infection? George V. Papatheodoridis, 1 Emanuel K. Manesis, 1 Spilios Manolakopoulos, 1 Ioannis S. Elefsiniotis, 2 John Goulis, 3 John Giannousis, 1 Antonios Bilalis, 4 Georgia Kafiri, 5 Dimitrios Tzourmakliotis, 4 and Athanasios J. Archimandritis 1 The diagnosis of hepatitis B e antigen (HBeAg)-negative chronic hepatitis B indicating therapeutic intervention currently requires serum hepatitis B virus (HBV) DNA >2,000 IU/mL. We evaluated the severity of liver histology and the presence of histological indication for treatment in patients with HBeAg-negative chronic HBV infection focusing on those with low viremia and/or normal alanine aminotransferase (ALT). In total, 399 patients with increased ALT and detectable serum HBV DNA (chronic hepatitis B patients) and 35 cases with persistently normal ALT and HBV DNA >2,000 IU/mL (inactive carriers) were included. Histological indication for treatment (grading score >7 and/or stage >2 in Ishak s classification) was found in 91% (185/203), 82% (75/91), 75% (47/63), and 62% (26/42) of chronic hepatitis B patients with HBV DNA >200,000, 20, ,999, 2,000-19,999, and <2,000 IU/mL, respectively (P < 0.001). Histological indication for treatment was more frequent in chronic hepatitis B patients with persistently elevated ALT (86% or 275/321), but it was also found in 74% (58/78) of those with transiently normal ALT (P 0.025). All inactive carriers had HBV DNA <20,000 IU/mL. Histological indication for treatment was present in 17% (6/35) of inactive carriers always due to moderate (stage 2) fibrosis without active necroinflammation. Conclusion: HBeAg-negative chronic HBV patients with persistently or transiently increased ALT and HBV DNA >20,000 IU/mL almost always require therapeutic intervention, but histological indications for treatment are also present in the majority of such cases with HBV DNA <20,000 and even <2,000 IU/mL. In contrast, minimal histological lesions are observed in the majority of HBeAg-negative patients with persistently normal ALT and HBV DNA >2,000 IU/mL, who may not require immediate liver biopsy and treatment but only close follow-up. (HEPATOLOGY 2008;48: ) Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CI, confidence interval; cp/ml, copies per milliliter; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HDV, hepatitis delta virus; HIV, human immunodeficiency virus; OR, odds ratio; ULN, upper limit of normal. From the 1 2nd Department of Internal Medicine, Athens University Medical School, Hippokration General Hospital of Athens, Athens, Greece; the 2 University Department of Internal Medicine, Helena Venizelou Hospital, Athens, Greece; the 3 4th Department of Medicine, Aristotelian University of Thessaloniki, Hippokration General Hospital of Thessaloniki, Thessaloniki, Greece; the 4 Department of Gastroenterology, Polyclinic General Hospital of Athens, Athens, Greece; and the 5 Department of Pathology, Hippokration General Hospital of Athens, Athens, Greece. Received February 13, 2008; accepted June 29, Supported in part by a research grant from the Center for Control and Prevention of Diseases, Ministry of Health, Greece. Address reprint requests to: George V. Papatheodoridis, M.D., 2nd Department of Internal Medicine, Athens University Medical School, Hippokration General Hospital of Athens, 114 Vas. Sophias Ave., Athens, Greece. gepapath@med.uoa.gr; fax: (30) Copyright 2008 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience ( DOI /hep Potential conflict of interest: Dr. Papatheodoridis is a consultant for, advises, is on the speakers bureau of, and received grants from Gilead Sciences, Roche, and Novartis. He is a consultant and advises Bristol-Myers Squibb. Dr. Manolakopoulos is a consultant, advises, is on the speakers bureau of, and received grants from Gilead. He is a consultant for and received grants from Roche. He is a consultant for, advises, and received grants from Schering-Plough. He advises Bristol-Myers Squibb. 1451

2 1452 PAPATHEODORIDIS ET AL. HEPATOLOGY, November 2008 Chronic hepatitis B virus (HBV) infection with negative hepatitis B e antigen (HBeAg) is becoming the predominant type of chronic HBV infection worldwide. 1 It develops after HBeAg loss and can be separated into the low- or non-replicative phase of inactive chronic HBV carrier state and the active phase of HBeAg-negative chronic hepatitis B. 2 The differential diagnosis between these two phases is mandatory, as inactive chronic HBV carriers need only be followed up regularly, whereas patients with HBeAg-negative chronic hepatitis B require therapeutic intervention. 2 Such a differential diagnosis is based largely on alanine aminotransferase (ALT) activity, serum HBV DNA levels, and liver histology. Ideally, all HBeAg-negative chronic hepatitis B patients should have elevated ALT activity, high HBV DNA levels, and active liver histological lesions, and all inactive carriers should present with persistently normal ALT values and low or undetectable viremia avoiding liver biopsy anticipated to reveal minimal or no liver injury. However, because patients with HBeAg-negative chronic hepatitis B may present with fluctuating biochemical activity, including variable duration periods of completely normal ALT levels, 2 and because liver biopsy is an invasive procedure that cannot be easily repeated, the determination of serum HBV DNA levels is becoming of great importance. The development of commercially available, sensitive, and reliable assays for serum HBV DNA has resulted in an increasing demand for such determinations. Although the old insensitive assays were often yielding negative results in patients with clinically evident HBeAg-negative chronic hepatitis B, the new sensitive assays have resulted in problems of interpretation of low viremia levels. 3 In a consensus conference for hepatitis in 2000, the HBV DNA level of 100,000 copies/ml was arbitrarily suggested to be the cutoff point for the differentiation between inactive carriers and HBeAg-negative chronic hepatitis B patients. 4 A previous report by our group suggested that 30,000 copies/ml might be a more appropriate cutoff point, while more recent guidelines lowered this cutoff point to approximately 10,000 copies/ml or 2,000 IU/mL. 5,6 In our center, the diagnosis of HBeAg-negative chronic hepatitis B is primarily based on the close follow-up of ALT and aspartate aminotransferase (AST) values, with liver biopsies being performed in all patients with elevated ALT/AST and detectable HBV DNA regardless of HBV DNA levels. In addition, after the recent suggestions that 2,000 IU/mL may represent an appropriate viremia cutoff level for the differentiation between inactive carrier state and HBeAg-negative chronic hepatitis B, 5 we began performing liver biopsies in our HBeAgnegative chronic HBV patients with persistently normal ALT/AST values and serum HBV DNA above 2,000 IU/ ml. The aim of this study was to evaluate the severity of liver histological lesions and the presence of histological indication for treatment in our patients with HBeAgnegative chronic HBV infection in relation to their serum HBV DNA and ALT levels, focusing on those with low viremia and/or normal ALT activity. Patients and Methods Patient Population. Four hundred thirty-four treatment-naïve patients with HBeAg-negative chronic HBV infection who underwent liver biopsy at two hospitals in Athens, Greece, between September 2001 and December 2007 were included in this study. All patients had positive hepatitis B surface antigen (HBsAg) and negative HBeAg for at least 6 months and detectable serum HBV DNA. Until August 2006, only patients who had elevated ALT values on at least two monthly occasions within the last 6 months underwent liver biopsy. However, after September 2006, HBeAg-negative chronic HBV patients with persistently normal ALT and serum HBV DNA above 2,000 IU/mL were also advised to undergo liver biopsy. For this study, the inclusion of patients with elevated ALT and normal ALT ended in August 2007 and December 2007, respectively. Patients with hepatitis delta virus (HDV), hepatitis C virus (HCV), or human immunodeficiency virus (HIV) coinfection were excluded, as were those with decompensated liver disease (evidence of ascites, variceal bleeding, hepatic encephalopathy, jaundice) and/or hepatocellular carcinoma at presentation. All patients gave informed consent to undergo liver biopsy. The study was approved by the local ethics committee. Follow-up. At baseline, detailed medical history, including epidemiological data, and physical examination findings were recorded for all patients. Current alcohol abuse was defined as a mean daily alcohol consumption during the previous year of 30 g in males or 20 g in females. All patients were followed with sequential determinations of ALT/AST. Patients with abnormal ALT values that is, values higher than the upper limit of normal (ULN) at baseline or during follow-up were followed with monthly ALT determinations and underwent liver biopsy in case of increased ALT activity on at least two occasions within the last 6 months. Patients with normal baseline ALT values were followed with ALT determinations at least every 3 months for the first year and at least every 6 months thereafter. Patients were considered to have persistently normal ALT activity if they had ALT values below the ULN during the total follow-up period, which was at least 1 year. No patient received any kind of antiviral or immunosuppressive therapy during the study period.

3 HEPATOLOGY, Vol. 48, No. 5, 2008 PAPATHEODORIDIS ET AL Based on ALT and viremia levels, HBeAg-negative chronic hepatitis B was considered to be present in 399 patients with elevated ALT values on at least two occasions and any level of detectable HBV DNA, whereas an inactive chronic HBV carrier state was considered to be present in 35 patients with persistently normal ALT and HBV DNA levels between 2,000 and 20,000 IU/mL. There was no case with persistently normal ALT and HBV DNA levels higher than 20,000 IU/mL. Laboratory Methods. Liver biochemistries were performed in commercially available autoanalyzers. The ULN was 40 IU/L for both ALT and AST. Hepatitis A, B, D, C virus and HIV 1,2 serological markers (anti-hav, IgM anti-hav, HBsAg, anti-hbs, anti-hbc, HBeAg, anti-hbe, IgG anti-hdv, anti-hcv, anti-hiv 1,2 ) were evaluated via commercially available enzyme immunoassays at baseline in all patients. Serum HBV DNA was measured via a sensitive, quantitative, commercially available polymerase chain reaction assay (Amplicor HBV Monitor test; Roche Diagnostic Systems, Inc., Branchburg, NJ) with a sensitivity of 400 copies per milliliter (cp/ml) or approximately 80 IU/mL. Serum HBV DNA levels were always expressed in IU/mL (1 IU/mL 5.6 cp/ml). 7 In particular, serum HBV DNA levels were determined within the first month of follow-up at baseline and close (within 1 month) to the day of liver biopsy in patients who underwent liver biopsy away from the onset of follow-up. The HBV DNA levels determined close to the liver biopsy were taken into account for this study. All liver biopsies were evaluated by a single liver histopathologist (G. K.) who was blind to the ALT values and serum HBV DNA levels. The histological lesions were graded according to the classification proposed by Ishak et al. 8 Histological lesions compatible with chronic hepatitis B were considered to be present in cases with a grading score 4 and/or stage 1, while a histological indication for treatment was considered to be present in cases with a grading score 7 and/or stage 2. Statistical Analysis. Data entry and statistical analysis were performed with the statistical package SPSS version 13.0 (SPSS Inc., Chicago, IL). The corrected chisquare test or two-sided Fisher s exact test was used to compare categorical data. The Student t test or one-way analysis of variance was used for group comparisons of parametric quantitative data, and the Mann-Whitney or Kruskal-Wallis test was used for similar comparisons of nonparametric data. Logistic regression models were used for multivariate analysis. Results are presented as the mean standard deviation or median (range) when appropriate. In all cases, tests were two-tailed, and a P value of 0.05 was considered statistically significant. Results HBeAg-Negative Chronic Hepatitis B Patients. Of the 399 chronic hepatitis B patients, serum HBV DNA levels were 200,000 IU/mL in 203 (51%) (group A), between 20,000 and 199,999 IU/mL in 91 (30%) (group B), between 2,000 and 19,999 IU/mL in 63 (21%) (group C), and detectable but below 2,000 IU/mL in 42 (14%) (group D). Histological lesions compatible with chronic hepatitis B were present in all 294 patients with HBV DNA 20,000 IU/mL (groups A and B) as well as in 94% (59/63) of group C and 83% (35/42) of group D patients. The main characteristics of all patients in relation to their viremia levels are presented in Table 1. Age, sex distribution, body mass index, and prevalence of reported alcohol abuse did not differ among the patients groups. Up to 7% of cases in most groups were considered current alcohol abusers according to our definition, with the mean daily alcohol consumption being moderate ( 60 g) in the majority of them (19/25 or 76%). ALT and AST levels as well as necroinflammatory activity were greater in group A than group B than group C (or D) (always P 0.05), while they did not significantly differ between group C and group D. The severity of fibrosis did not differ between group A and B, while it was worse in group A than C (P 0.05), group A than D (P 0.001), group B than D (P 0.001) and tended to be worse in group B than C (P 0.055) and group C than D (P 0.087). In particular, severe fibrosis or cirrhosis (stage 4-6) was present in 81 (40%) patients in group A, 45 (49%) patients in group B, 23 (37%) patients in group C, and eight (19%) patients in group D (P 0.001). Of the 42 group D patients, 25 (60%) had at least moderate fibrosis (stage 2). The severity of fibrosis in relation to the severity of necroinflammation in the four groups of chronic hepatitis B patients is presented in Table 2. Minimal necroinflammation (grading score 0-4) without severe fibrosis or cirrhosis (stage 0-3) was observed in 7% (14/203) of patients in group A, 18% (16/91) of patients in group B, 30% (19/63) of patients in group C, and 38% (16/42) of patients in group D (P 0.001). Although all 399 patients with HBeAg-negative chronic hepatitis B had elevated ALT activity on 2 recent determinations, 78 (20%) of them had normal ALT on the liver biopsy day. Because HBeAg-negative chronic hepatitis B patients with transiently normal ALT activity represent the cases that may be easily misclassified as inactive HBV carriers, we also analyzed separately the characteristics of this particular subgroup (Table 3). HBeAgnegative chronic hepatitis B patients with transiently normal ALT, compared with patients with persistently elevated ALT, were less frequently males (67% versus 80%, P 0.038) and had significantly lower serum HBV

4 1454 PAPATHEODORIDIS ET AL. HEPATOLOGY, November 2008 Table 1. Main Characteristics of Patients with HBeAg-Negative Chronic HBV infection in relation to their ALT and Serum HBV DNA Levels HBeAg-Negative Chronic Hepatitis B Patients Inactive Chronic HBV Carriers Patient Characteristics on the Day of Liver Biopsy Group A: DNA >200,000 IU/mL (n 203) Group B: DNA 20, ,999 IU/mL (n 91) Group C: DNA 2,000 19,999 IU/mL (n 63) Group D: DNA 80 1,999 IU/mL (n 42) HBV DNA 2,000 19,999 IU/mL (n 35) Age, years Male sex 156 (77%) 71 (78%) 48 (76%) 35 (83%) 22 (63%) Body mass index, kg/m Alcohol abuse 15 (7%) 6 (7%) 4 (6%) 0 2 (7%) ALT, IU/L 121 (12 784) 75 (14 656) 61 (12 387) 52 (13 565) 28 (13 39) Normal ALT 18 (9%) 14 (15%) 15 (24%) 12 (29%) 35 (100%) AST, IU/L 78 (18 592) 66 (20 449) 41 (15 222) 38 (16 242) 23 (16 40) Necroinflammation* Minimal (0 4)* 16 (8%) 19 (21%) 20 (32%) 16 (38%) 34 (97%) Mild (5 8)* 86 (42%) 41 (45%) 33 (52%) 20 (48%) 1 (3%) Moderate (9 12)* 85 (42%) 28 (31%) 8 (13%) 4 (10%) 0 Severe (13 18)* 16 (8%) 3 (3%) 2 (3%) 2 (5%) 0 Fibrosis* No/Mild (0 1)* 23 (11%) 17 (19%) 16 (25%) 17 (40%) 29 (83%) Moderate (2 3)* 99 (49%) 29 (32%) 24 (38%) 17 (40%) 6 (17%) Severe (4)* 28 (14%) 12 (13%) 10 (16%) 3 (7%) 0 Cirrhosis (5 6)* 53 (26%) 33 (36%) 13 (21%) 5 (12%) 0 Histological indication for treatment 185 (91%) 75 (82%) 47 (75%) 26 (62%) 6 (17%) Patients with elevated ALT on at least two occasions were characterized as HBeAg-negative chronic hepatitis B cases and patients with persistently normal ALT as inactive chronic HBV carriers (only inactive carriers with HBV DNA 2,000 IU/mL were included). *Grading score, stage according to Ishak s classification. 8 Histological indication for treatment was considered to be present in cases with grading score 7 and/or stage 2. DNA levels (P 0.001) and milder necroinflammation (P 0.038), but similar severity of fibrosis. Of the 78 chronic hepatitis B patients with transiently normal ALT activity, 21 (27%) cases had at least moderate necroinflammation, while 28 (34%) patients had advanced fibrosis (stage 4), including 21 (27%) cases with cirrhosis. Inactive Chronic HBV Carriers. According to our inclusion criterion, serum HBV DNA levels were between 2,000 and 20,000 IU/mL in all 35 inactive carriers. Seven (20%) of 35 inactive carriers had mild ( 2 ULN) ALT elevations on just one occasion without significant changes in serum HBV DNA levels and with ALT values returning and remaining within the normal range during several subsequent monthly determinations. These seven cases were not classified into HBeAg-negative chronic hepatitis B, because they did not fulfill its definition requiring elevated ALT on at least two occasions. Inactive chronic HBV carriers did not differ from HBeAg-negative chronic hepatitis B patients in age, body mass index, or prevalence of reported alcohol abuse. They were less frequently males (63% versus 78%), but the difference did not reach statistical significance (P 0.08). As expected, they had significantly lower ALT and AST levels than the patients of any chronic hepatitis B group (P 0.001) (Table 1). Histological lesions compatible with chronic hepatitis B, according to the definition of our study (grading score 4 and/or stage 1), were present in 30 (86%) of the 35 inactive carriers (Table 2). However, necroinflammatory activity was minimal in all but one (97%), whereas no fibrosis was present in five (14%), mild fibrosis (stage 1) was present in 24 (69%), and moderate fibrosis (always stage 2) was present in six (17%) of them. Of course, both necroinflammatory and fibrosis severity were significantly milder in inactive carriers than in chronic hepatitis B patients (P 0.001) (Table 1). There was no significant difference between the 30 patients with and the five patients without histological lesions of chronic hepatitis B in serum HBV DNA levels (2,732 [2,000-18,700] versus 3,013 [2,143-5,000] IU/mL, P 0.91) or any other characteristic (data not shown). Similarly, there was no difference between the 29 cases with fibrosis score of 0-1 and the five cases with fibrosis score of 2 in serum HBV DNA levels (2,857 [2,000-18,700] versus 2,630 [2,107-14,321] IU/mL, P 0.86) or any other characteristic (data not shown). Histological Indication for Treatment. Histological indication for treatment was present in 185 (91%) patients in group A, 75 (82%) patients in group B, 47 (75%) patients in group C, and 26 (62%) patients in group D (P 0.001) (Table 1). In particular, histological indication for treatment was present in 188 (96%) of the 196 chronic hepatitis B patients with serum HBV DNA 20,000 IU/mL (group A and B) and ALT 2 ULN. The proportion of patients with histological indication

5 HEPATOLOGY, Vol. 48, No. 5, 2008 PAPATHEODORIDIS ET AL Table 2. Severity of Fibrosis (Stage) in Relation to the Severity of Necroinflammatory Activity (Grading Score) in Patients with Chronic HBV Infection According to ALT and Serum HBV DNA levels (IU/mL) Patient Stage, n (%) P Value CHB HBV DNA 200,000 (n 203) Grading score (4.4) 5 (2.5) 1 (0.5) 1 (0.5) (0.5) 8 (3.9) 31 (15.3) 1 (0.5) 4 (2.0) (1.5) 23 (11.3) 7 (3.4) 8 (3.9) (1.0) 36 (17.7) 15 (7.4) 32 (15.8) (2.0) 4 (2.0) 8 (3.9) CHB HBV DNA 20, ,000 (n 91) Grading score (1.1) 4 2 (2.2) 9 (9.9) 5 (5.5) 2 (2.2) (5.5) 10 (11.0) 1 (1.1) 2 (2.2) (1.1) 7 (7.7) 4 (4.4) 11 (12.1) (6.6) 5 (5.5) 17 (18.7) (1.1) 0 2 (2.2) CHB HBV DNA 2,000 19,000 (n 63) Grading score (6.3) 3 (4.8) 2 (3.2) (3.2) 4 (6.3) 4 (6.3) 1 (1.6) (4.8) 10 (15.9) 0 6 (9.5) (9.5) 5 (7.9) 3 (4.8) (1.6) 3 (4.8) 4 (6.3) (1.6) 1 (1.6) 0 CHB HBV DNA 2,000 (n 42) Grading score (16.7) 2 (4.8) (9.5) 3 (7.1) (7.1) 5 (11.9) 0 1 (2.4) (2.4) 6 (14.3) 2 (4.8) 2 (4.8) (4.8) 1 (2.4) 1 (2.4) (2.4) 0 1 (2.4) Inactive carriers HBV DNA 2,000 19,000 (n 35) 0.62 Grading score (14.3) 16 (45.7) 4 (11.4) (20.0) 2 (5.7) (2.9) Patients with elevated ALT on at least two occasions were characterized as HBeAg-negative chronic hepatitis B cases and patients with persistently normal ALT as inactive chronic HBV carriers (only inactive carriers with HBV DNA 2,000 IU/mL were included). Abbreviation: CHB, chronic hepatitis B. for treatment was higher in group A than any other group (A versus B, P 0.05; A versus C, P 0.006; A versus D, P 0.001) and in group B than D (P 0.019), but it did not differ significantly between group B and C or between group C and D. Among the patients without a histological indication for treatment, the necroinflammatory activity score was (4-6) in those of group A, (3-6) in those of group B, (2-6) in those of group C and (2-6) in those of group D. Histological indication for treatment was more frequent in chronic hepatitis B patients with persistently elevated ALT (86%), but it was also found in 74% of those with transiently normal ALT activity (P 0.025) (Table 3). Among cases with transiently normal ALT, there was no difference in the presence of histological indication for treatment between those with transiently low normal ( 20 IU/mL) and transiently high normal ALT values (20-39 IU/ ml) (7/10 or 70% versus 51/68 or 75%, P 0.71). Histological indication for treatment according to our definition was present in six (17%) of the 35 inactive carriers, and this was always due to the presence of stage 2 fibrosis without active necroinflammation (grading score 2-4 in all six cases) (Table 2). Again, there was no difference in the presence of histological indication for treat-

6 1456 PAPATHEODORIDIS ET AL. HEPATOLOGY, November 2008 Table 3. Main Characteristics of Patients with HBeAg-Negative Chronic HBV Infection in Relation to ALT Levels Patient Characteristics on Day of Liver Biopsy HBeAg-Negative Chronic Hepatitis B Elevated ALT (n 321) Normal ALT (n 78) Inactive Chronic HBV Carriers (n 35) Age, years Male sex 258 (80%) 1 52 (67%) 2 22 (63%) Body mass index, kg/m Alcohol abuse 22 (7%) 3 (4%) 3 (9%) ALT, IU/L 116 (41 784) 3 33 (12 40) 4 28 (13 39) AST, IU/L 76 (22 592) 3 31 (15 66) 4 23 (16 40) 3 Serum HBV DNA, IU/mL 290,703 ( ) 3 32,628 (93 1, ) 4 2,768 (2,000 18,700) 3 Patients with HBV DNA 5 6, ,999 IU/mL 30 (9%) 12 (15%) 0 2,000 19,999 IU/mL 44 (14%) 19 (24%) 35 (100%) 20, ,999 IU/mL 71 (22%) 20 (26%) 0 200,000 IU/mL 176 (55%) 27 (35%) 0 Grading score* , Patients with necroinflammation 1 2,4 3 Minimal (0 4)* 54 (17%) 18 (23%) 34 (97%) Mild (5 8)* 140 (44%) 39 (50%) 1 (3%) Moderate (9 12)* 107 (33%) 18 (23%) 0 Severe (13 18)* 20 (6%) 3 (4%) 0 Stage* Patients with fibrosis 4 3 No/mild (0 1)* 52 (16%) 21 (27%) 29 (83%) Moderate (2 3)* 140 (44%) 29 (37%) 6 (17%) Severe (4) * 46 (14%) 7 (9%) 0 Cirrhosis (5 6) * 83 (26%) 21 (27%) 0 Patients with histological indication for treatment 275 (86%) 9 58 (74%) 10,4 6 (17%) 3 Patients with elevated ALT on at least two occasions were characterized as having HBeAg-negative chronic hepatitis B and patients with persistently normal ALT as inactive chronic HBV carriers (only carriers with HBV DNA 2,000 IU/mL were included). HBeAg-negative chronic hepatitis B patients were divided according to their ALT levels on the day of their liver biopsy. At the same row: 1 versus 2, P 0.038; 3 versus 4, P 0.001, 5 versus 6, P 0.007; 7 versus 8, P 0.071; 9 versus 10, P *Grading score, stage according to Ishak s classification. 8 Histological indication for treatment was considered to be present in cases with grading score 7 and/or stage 2. ment between those with low normal ( 20 IU/mL) and high normal ALT values (20-39 IU/mL) (1/8 [12.5%] versus 5/27 [18.5%], P 1.00). Older age was significantly associated with the presence of histological indication for treatment (mean age, versus years in patients with versus without histological indication for treatment, P 0.001). The proportion of patients with histological indication for treatment in relation to their age groups and viremia levels are shown in Fig. 1. Histological indication for treatment was found significantly more frequently in older age groups in chronic hepatitis B patients with serum HBV DNA 20,000 IU/mL, but it did not seem to be associated with age in chronic hepatitis B patients with lower viremia levels or in inactive carriers. Histological indication for treatment was almost always present (97% or 189/195) in HBeAg-negative chronic hepatitis B patients with HBV DNA 20,000 IU/mL who were older than or equal to 45 years and in 62% (38/61) of them who were younger than 45 years (P 0.001). In the 399 chronic hepatitis B patients, multivariate analysis revealed that histological indication for treatment was independently associated with age 45 years (odds ratio [OR] 6.203, 95% confidence interval [CI] ; P 0.001), higher ALT levels (OR 1.004, 95% CI ; P 0.049), and serum HBV DNA 20,000 IU/mL (OR 2.235, 95% CI ; P 0.010). When all 434 patients were included in the multivariate analysis, histological indication for treatment was independently associated with presence of HBeAg-negative chronic hepatitis B than inactive carrier state (OR , 95% CI ; P 0.001) or with abnormal ALT on liver biopsy (OR 3.663, 95% CI ; P 0.001), but also with older age (P 0.001) and higher serum HBV DNA levels (P 0.001) (Table 4). Discussion Our findings suggest that serum HBV DNA levels alone cannot always differentiate patients with HBeAg-negative chronic hepatitis B from inactive chronic HBV carriers and, most importantly, cannot represent an absolute indication for therapeutic intervention. Of course, serum HBV DNA levels 20,000 IU/mL in patients with HBeAg-negative chronic HBV infection appear to safely diagnose HBeAgnegative chronic hepatitis B, since all or at least the vast

7 HEPATOLOGY, Vol. 48, No. 5, 2008 PAPATHEODORIDIS ET AL Fig. 1. Histological indication for treatment in 434 patients with HBeAg-negative chronic hepatitis B virus infection in relation to their age group and serum HBV DNA levels. Histological indication for treatment was considered to be present in cases with Ishak s grading score 7 and/or stage 2. HBeAg-negative chronic hepatitis B was considered to be present in patients with elevated ALT on 2 occasions and any level of detectable HBV DNA, whereas inactive chronic HBV carrier state was considered to be present in cases with persistently normal ALT for at least 12 months and HBV DNA levels 20,000 IU/mL. Only inactive chronic HBV carriers with HBV DNA 2,000 IU/mL were included. majority of such cases also have persistently or transiently elevated ALT values. Our data show that a liver biopsy in patients with viremia levels 20,000 IU/mL will not usually affect the decision for treatment, although it will offer valuable information on the severity of liver lesions and the patients prognosis. Widely accepted histological indications for therapeutic intervention (Ishak s grading score 7 and/or stage 2) were found in 90% of our patients with serum HBV DNA 200,000 IU/mL and in 80% of them with serum HBV DNA between 20,000 and 200,000 IU/ ml. In addition, marginally fewer severe histological lesions (grading score 4-6 and usually stage 1) were present in all patients with viremia levels 20,000 IU/mL who did not fulfill the previous histological criteria for treatment (Table 2). Given the possibility of biopsy sampling errors, the presence of more severe histological lesions in patients of the latter subgroup cannot be definitely excluded. Moreover, it is difficult and potentially harmful to defer treatment in HBeAg-negative chronic HBV patients with high viremia and elevated aminotransferase activity, in whom liver histological lesions tend to worsen over time. 9 Although the decision for treatment seems to be straightforward in HBeAg-negative chronic HBV patients with viremia levels 20,000 IU/mL, it is not so simple in cases with lower serum HBV DNA levels. Histological indication for treatment was found in 75% of our patients with elevated ALT values and serum HBV Table 4. Risk Factors Associated with Presence of Histological Indication for Treatment (Ishak s Grading Score >7 and/or Stage >2) in Patients with HBeAg-Negative Chronic HBV Infection All Patients (n 434) CHB Patients (n 399) OR (95% CI) P Value OR (95% CI) P Value Chronic hepatitis B* (versus inactive carrier state) or 18.7 ( ) Abnormal ALT on liver biopsy 3.7 ( ) ( ) Age, years * * ( ) ( ) ( ) ( ) ( ) ( ) Serum HBV DNA, IU/mL 80 1, * * 2,000 19, ( ) ( ) , , ( ) ( ) , ( ) ( ) Results of logistic regression analysis are shown. *For trend.

8 1458 PAPATHEODORIDIS ET AL. HEPATOLOGY, November 2008 DNA levels between 2,000 and 20,000 IU/mL. The presence of histological lesions requiring therapeutic intervention in three-fourths of such patients and the detection of histological lesions of chronic hepatitis in almost 95% of patients in this group clearly support the recent recommendations for reduction of the HBV DNA cutoff level as indication for treatment from 20,000 IU/mL (or 100,000 cp/ml) to 2,000 IU/mL. 5,6 However, the more recent but again arbitrary lower limit of 2,000 IU/mL might also be a reason for excluding HBeAg-negative patients from the correct diagnosis and thus from the necessary therapeutic intervention. In fact, histological indication for treatment was found in the majority (62%) and histological lesions of chronic hepatitis in more than 80% of our patients with elevated ALT values and serum HBV DNA below 2,000 IU/mL. These findings challenge the 2,000 IU/mL serum HBV DNA cutoff level, because such patients who definitely require treatment 10 represent a sizeable proportion of patients with HBeAg-negative chronic HBV infection. Approximately one out of seven (14%) patients with HBeAg-negative chronic hepatitis B in this crosssectional study had serum HBV DNA below 2,000 IU/ ml. In another prospective study from our group, 5% of patients with HBeAg-negative chronic hepatitis B and 16% of their serum samples had HBV DNA levels below 2,000 IU/mL, while 18% of such patients occasionally had viremia below 2,000 IU/mL. 11 In an older study from Lok s group, 42% of chronic hepatitis B patients were reported to have occasionally serum HBV DNA levels below a similar level of 10,000 cp/ml. 12 Despite these data, HBeAg-negative patients with increased ALT/AST activity and viremia levels initially less than 100,000 cp/ml (approximately 18,000 IU/mL) and recently less than 2,000 IU/mL have been ignored from all the past or recent guidelines. 4,5,13-16 Because the majority of such patients may have significant liver lesions, liver biopsy should be recommended in all HBeAg-negative chronic HBV patients with elevated ALT activity regardless of viremia levels and treatment should be given in cases with appropriate histological findings. Using the current sensitive assays, serum HBV DNA is detected in the majority of HBeAg-negative chronic HBV patients, even those with persistently normal ALT/AST activity. 11,17 In fact, a substantial proportion of HBeAgnegative patients with persistently normal ALT/AST values may have serum HBV DNA levels 2,000 IU/mL. In our recent prospective study including 85 inactive chronic HBV carriers followed very closely for a median of 3 years, serum HBV DNA was 2,000 IU/mL (always 20,000 IU/mL) in 22% of cases at baseline or in 23% of cases at any time point or in 28% of 228 serum samples tested. 11 Whether such patients should undergo liver biopsy and whether they require therapeutic intervention are still questionable. Minimal necroinflammatory lesions were found in the vast majority ( 90%) of the 35 such cases included in this study, while fibrosis was absent or mild in 83% of them. Approximately one-sixth of these 35 cases were found to have moderate fibrosis (Ishak s stage 2) with minimal necroinflammatory activity and were therefore considered to have a histological indication for treatment, according to our predetermined definition. In another recent study, significant necroinflammation and/or fibrosis were reported to be present in more than one-third of 33 HBeAg-negative patients with persistently normal ALT. 18 The patients of the latter study, however, may be different from our closely followed patients with persistently normal ALT/AST, as they could have only two normal ALT values at least 6 months apart to fulfill the persistently normal ALT definition. 18 Moreover, the HBeAg-negative patients with normal ALT in the latter study had rather high serum HBV DNA levels ( 200,000 IU/mL in 50% and 80,000 IU/mL in 95% of them) and often ( 33%) moderate or severe necroinflammation, 18 which are in contrast with our findings and might be at least partly related to the differences in the definition of normal ALT activity. Differences in the HBV genotypes or other viral characteristics might also be responsible for such discrepant findings. Genotypes were not determined in our study, but it is well known that genotype D predominates in our country being responsible for 90% of chronic HBV cases. 1 Whether patients with some degree of moderate fibrosis and minimal necroinflammation require immediate therapeutic intervention is unclear. Based on reasonable clinical judgment, treatment should be given to chronic HBV patients with some degree of both necroinflammation and fibrosis. Such an approach would prevent unnecessary treatment in chronic HBV patients who do not actually require therapy, such as inactive chronic HBV carriers without hepatitis activity but with some fibrosis accumulated during the HBeAg seroconversion phase in the past. On the other hand, all patients with HBV DNA 2,000 IU/mL definitely require close follow-up, because such viremia levels and/or high ALT activity within the traditional normal range seem to represent strong predictors of progression from the inactive carrier state to HBeAg-negative chronic hepatitis B in the near future. 11,19 Except for the inactive carriers, dilemmas about the most appropriate histological indications for treatment may be also raised for patients with elevated ALT and relatively low viremia (HBV DNA 20,000 IU/mL). Histological indication for treatment according to our criteria was found in 70% (73/105) of such cases in our study, but this was due

9 HEPATOLOGY, Vol. 48, No. 5, 2008 PAPATHEODORIDIS ET AL only to presence of moderate fibrosis (stage 2-3) in 33% (24/73) of these patients who had minimal (grading score 0-4 [12%]) or very mild necroinflammation (grading score 5-6 [21%]) (Table 2). Although any degree of necroinflammation probably requires therapeutic intervention in patients with severe fibrosis or cirrhosis, it is questionable whether patients with increased ALT activity (particularly with mild ALT elevations), relatively low viremia levels and moderate fibrosis but minimal necroinflammation should be immediately treated. Older age has been considered as a factor associated with worse histological lesions in patients with chronic HBV infection; therefore, liver biopsy may be more easily recommended in patients older than 40 years. 6 Our data support the independent association of older age with worse liver histology, but only in patients with serum HBV DNA 20,000 IU/mL (Fig. 1). In fact, histological indication for treatment was present in almost all (97%) of our 195 HBeAg-negative chronic hepatitis B patients with HBV DNA 20,000 IU/mL who were 45 years old. However, if our findings are confirmed by others, age will not represent a good predictor of significant histological lesions in the patients with low viremia and/or normal ALT levels, in whom such predictors are mostly needed. In conclusion, the differential diagnosis of HBeAg-negative chronic HBV infection should be initially based on the combination of ALT/AST activity and serum HBV DNA levels. Patients, at least in Greece, presenting with high ( 20,000 IU/mL) serum HBV DNA levels almost always have elevated ALT/AST values if followed closely. They can be safely classified as HBeAg-negative chronic hepatitis B cases requiring prompt therapeutic intervention. 10 In patients with high viremia, liver biopsy would offer valuable prognostic information, but it is not expected to affect the decision to treat. In contrast, patients presenting with low ( 20,000 IU/mL) serum HBV DNA levels can be either inactive chronic HBV carriers or HBeAg-negative chronic hepatitis B cases. 20 Although it would have been simpler and more practical, it seems that there is no clear cutoff serum HBV DNA level to differentiate these two groups. The recently suggested HBV DNA cutoff level of 2,000 IU/mL would not correctly diagnose a substantial proportion of inactive chronic HBV carriers, perhaps leading to unnecessary liver biopsies, but also more than one-tenth of HBeAgnegative chronic hepatitis B patients who may not receive the appropriate management. Thus, for HBeAg-negative chronic HBV patients with low viremia, the correct diagnosis should be initially based on close follow-up with sequential ALT/AST determinations. 20 All patients with persistently or transiently abnormal ALT/AST values of any level may benefit from a liver biopsy regardless of viremia levels, while patients with persistently normal ALT/AST values and serum HBV DNA below 20,000 IU/mL, who are usually characterized as inactive carriers, should be followed for life. In the latter setting, baseline HBV DNA levels above 2,000 IU/mL warrant closer follow-up. References 1. Funk ML, Rosenberg DM, Lok ASF. World-wide epidemiology of HBeAg-negative chronic hepatitis B and associated precore and core promoter variants. J Viral Hepat 2002;9: Hadziyannis SJ, Papatheodoridis GV. Hepatitis Be antigen negative chronic hepatitis B natural history and treatment. Semin Liver Dis 2006;26: Pawlotsky J-M, Bastie A, Lonjon I, Remire J, Darthuy F, Soussy C-J, et al. What techniques should be used for routine detection and quantification of HBV DNA in clinical samples? J Virol Methods 1997;65: Lok AS, Heathcote EJ, Hoofnagle JH. Management of hepatitis B: 2000 summary of a workshop. Gastroenterology 2001;120: Keeffe EB, Dieterich DT, Han S-HB, Jacobson IM, Martin P, Schiff ER, et al. A treatment algorithm for the management of chronic hepatitis B virus infection in the United States: an update. Clin Gastroenterol Hepatol 2006;4: Lok ASF, McMahon BJ. Chronic hepatitis B. HEPATOLOGY 2007;45: Saldanha J, Gerlich W, Lelie N, Dawson P, Heermann K, Heath A. An international collaborative study to establish a World Health Organization international standard for hepatitis B virus DNA nucleic acid amplification techniques. Vox Sanguinis 2001;80: Ishak K, Baptista A, Bianchi L, Callea F, De Groote J, Gudat F, et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995;22: Brunetto MR, Oliveri F, Coco B, Leandro G, Colombatto P, Gorin JM, et al. Outcome of anti-hbe positive chronic hepatitis B in alpha-interferon treated and untreated patients: a long-term cohort study. J Hepatol 2002;36: Papatheodoridis GV, Manolakopoulos S, Dusheiko G, Archimandritis AJ. Therapeutic strategies in the management of patients with chronic hepatitis B. Lancet Infect Dis 2007;8: Papatheodoridis GV, Chrysanthos N, Hadziyannis E, Cholongitas E, Manesis EK. Longitudinal changes in serum HBV DNA levels and predictors of progression during the natural course of HBeAg-negative chronic hepatitis B virus infection. J Viral Hepat 2008;15: Chu C-J, Hussain M, Lok ASF. Quantitative serum HBV DNA levels during different stages of chronic hepatitis B infection. HEPATOLOGY 2002;36: Lok ASF, McMahon BJ. Chronic hepatitis B. HEPATOLOGY 2001;34: Keeffe EB, Dieterich DT, Han SH, Jacobson IM, Martin P, Schiff ER, et al. A treatment algorithm for the management of chronic hepatitis B virus infection in the United States. Clin Gastroenterol Hepatol 2004;2: Lok AS, McMahon BJ. Chronic hepatitis B: update of recommendations. HEPATOLOGY 2004;39: Gish RG, Locarnini SA. Chronic hepatitis B: current testing strategies. Clin Gastroenterol Hepatol 2006;4: Manesis E, Papatheodoridis GV, Sevastianos V, Cholongitas E, Papaioannou C, Hadziyannis SJ. Significance of hepatitis B viremia levels determined by a quantitative polymerase chain reaction assay in patients with HBeAg-negative chronic hepatitis B virus infection. Am J Gastroenterol 2003;98: Lai M, Hyatt BJ, Nasser I, Curry M, Afdhal NH. The clinical significance of persistently normal ALT in chronic hepatitis B infection. J Hepatol 2007;47: Feld JJ, Ayers M, El-Ashry D, Mazzulli T, Tellier R, Heathcote EJ. Hepatitis B virus DNA prediction rules for hepatitis B e antigen-negative chronic hepatitis B. HEPATOLOGY 2007;46: Papatheodoridis GV, Hadziyannis SJ. Diagnosis and management of precore mutant chronic hepatitis B. J Viral Hepat 2001;8: