Partial virological response to entecavir in treatment-naïve patients with chronic hepatitis B

Transcription

1 Partial virological response to entecavir in treatment-naïve patients with chronic hepatitis B Young Eun Chon Department of Medicine The Graduate School, Yonsei University

2 Partial virological response to entecavir in treatment-naïve patients with chronic hepatitis B Directed by Professor Sang Hoon Ahn The Master's Thesis submitted to the Department of Medicine the Graduate School of Yonsei University in partial fulfillment of the requirements for the degree of Master of Medical Science Young Eun Chon December 2010

3 This certifies that the Master's Thesis Of Young Eun Chon is approved Thesis Supervisor : Sang Hoon Ahn [Thesis Committee Member : Jeon Han Park) [Thesis Committee Member : Ja Kyung Kim) The Graduate School Yonsei University December 2010

4 ACKNOWLEDGEMENTS I feel so happy to publish my precious Master s thesis. First of all, I m deeply grateful to my thesis supervisor Professor Sang Hoon Ahn, for all of the things he has been taken care of,including academic advices and encouragement. I also thank very much to professors Jeon Han Park and Ja Kyung Kim who gave me experienced advices and warm supports. I also appreciate Dr. Seung up Kim who taught me with enthusiasm during this work. Thanks ever so much to my family members, especially my father and mother who lead me to this medical road and gave me the best support. I do give the greatest love and respect for them.

5 <TABLE OF CONTENTS> ABSTRACT 1 I. INTRODUCTION 3 II. MATERIALS AND METHODS 5 1. Study population 5 2. Study design and definitions 6 3. Laboratory assay 7 4. Endpoint 7 5. Statistical analysis 8 III. RESULTS 9 1. Patient basline characteristics 9 2. Treatment outcomes HBV DNA level vs. HBV DNA to define PVR The optimal cut-off HBV DNA level to define PVR The flow of patients according to treatment response Factors predicting VR at week 96 in patients with PVR 14 IV. DISCUSSION 17 V. CONCLUSION 23 REFERENCES 24 ABSTRACT (IN KOREAN) 28

6 LIST OF FIGURES Figure 1. ROC curve analysis using HBV DNA level (left) and HBV DNA reduction from the baseline (right) at week 24, week 48 to predict virologic response at week Figure 2. The flow of patients according to treatment response. 14 Figure 3. A practical management guideline in treatment-naïve patients with CHB receiving ETV. 16

7 LIST OF TABLES Table 1. Baseline characteristics 10 Table 2. Treatment outcomes during 2 years of ETV therapy 11 Table 3. Distribution of patients at week 96 according to optimal PVR 13 Table 4. Factors predicting VR at week 96 in patients with PVR 15

8 ABSTRACT Partial virological response to entecavir in treatment-naïve patients with chronic hepatitis B Young Eun Chon Department of Medicine The Graduate School, Yonsei University (Directed by Professor Sang Hoon Ahn) Background: The proposed definition of a partial virological response (PVR) to nucleos(t)ide analogue in the 2009 European Association for the Study of the Liver (EASL) guidelines is based on limited evidence, especially in terms of the cut-off hepatitis B virus (HBV) DNA level and the time-point at which to judge it. This study assessed optimal PVR criteria for predicting virologic response (VR) at week 96 in treatment-naïve patients with chronic hepatitis B (CHB) receiving entecavir (ETV). Methods: A total of 175 patients (126 men, 49 women) who completed 96 weeks of first-line ETV therapy were prospectively recruited. For predicting VR at week 96, the area under the receiver operating characteristics curve (AUC) was used to find the optimal time-point, and the Youden index was used to calculate the optimal cut-off HBV DNA level. 1

9 Results: After 96 weeks of ETV therapy, 139 (79.4%) patients achieved VR. The AUC was significantly greater at week 48 than that at week 24 for predicting VR at week 96 (P=0.023). The optimal cut-off HBV DNA level at week 48 was 35 IU/mL. Forty-one (23.4%) patients met this PVR criteria of ETV (HBV DNA level 35 IU/mL at week 48). Conclusions: An HBV DNA level exceeding 35 IU/mL at week 48 is the optimal PVR criteria for predicting non-vr at week 96 in treatment-naïve patients with CHB who are receiving ETV. This study supports the proposed EASL PVR for ETV based on scientific evidence Key words chronic hepatitis B, entecavir, HBV DNA, nucleos(t)ide analogue, partial virological response, virologic response 2

10 Partial virological response to entecavir in treatment-naïve patients with chronic hepatitis B Young Eun Chon Department of Medicine The Graduate School, Yonsei University (Directed by Professor Sang Hoon Ahn) I. INTRODUCTION Approximately 400 million people are chronically infected with hepatitis B worldwide, and about 1 million suffer from complications related to hepatitis B virus (HBV), such as cirrhosis and hepatocellular carcinoma (HCC). 1,2 Because the risk for development of HCC or liver cirrhosis increases with an increase in HBV DNA level, 3,4 suppressing HBV replication and maintaining a sustained reduction of HBV DNA level is important to stop or reverse the progression of liver diseases by reducing the hepatic necroinflammatory response. 5-7 Hence, the main aim of chronic hepatitis B (CHB) treatment is a complete and persistent suppression of viral replication. 8 Recently, the European Association for the Study of the Liver (EASL) HBV Clinical Practice Guidelines proposed a definition for the response to antiviral therapy. 9 According to these guidelines, partial virological response (PVR) to nucleos(t)ide analogue (NUC) in CHB was defined as more than one log decline of viremia compared to baseline but still detectable serum levels of 3

11 HBV DNA by the real-time polymerase chain reaction (PCR) assay ( IU/ml) at week 24 or 48, depending on the genetic barrier of the anti-hbv drugs. However, this PVR definition did not suggest a clear evidence for the proposed cut-off HBV DNA level and the time-point at which judge it. With first- (lamivudine [LAM]) and second- (telbivudine [LDT] and adefovir [ADF]) generation anti-hbv drugs, it has been clearly demonstrated that patients who show PVR to these NUCs are at high risk for developing resistance to antiviral therapy Therefore, early detection of patients with PVR and appropriate intervention (changing into or adding other potent drugs) for achieving sustained viral suppression have been emphasized. However, the relationship between the residual viral load and the treatment response has not been fully evaluated for entecavir (ETV), a third generation NUC with potent activity against HBV and a low risk of resistance. 18 Recent data have revealed that 10 33% of ETV-treated patients show 300 HBV DNA copies/ml at week Moreover, ETV-resistant strains were detected in two NUC-naïve patients with incomplete suppression. 20 Therefore, it is necessary to establish PVR criteria and subsequent management guidelines for treatment-naïve patients with CHB who are receiving ETV. This study aimed to determine the optimal PVR criteria for predicting virologic response (VR) at week 96 of ETV therapy in treatment-naïve patients with CHB. 4

12 II. MATERIALS AND METHODS 1. Study population Between March 2007 and December 2009, 420 treatment-naïve patients with CHB starting antiviral therapy with 0.5 mg ETV at Sinchon Severance Hospital (Yonsei University College of Medicine, Seoul, Korea), Gangnam Severance Hospital (Yonsei University College of Medicine, Seoul, Korea), National Health Insurance Corporation Ilsan Hospital (Koyang, Korea), and Pusan National University Hospital (Pusan University College of Medicine, Busan, Korea) were prospectively enrolled. The inclusion criteria for this study were the following: age >16 years, persistent serum hepatitis B surface antigen (HBsAg) levels for more than 6 months, and HBV genotype C of which the prevalence reaches nearly up to 98~100% of HBV in Korea The exclusion criteria were the following: decompensated liver cirrhosis (evidence of ascites, hepatic encephalopathy, varices, serum total bilirubin <2.5 mg/dl, serum albumin <3 mg/dl, or prothrombin time >3 s longer than normal), previous antiviral treatment for CHB (either interferon or NUC), current use of corticosteroid or immunomodulatory drugs, co-infection with hepatitis C or D virus or human immunodeficiency virus, serious concurrent medical illness, evidence of HCC, or prior organ transplantation. 25 Written informed consent was obtained from all patients. Our study protocol was consistent with the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the independent institutional review boards of each institute. 5

13 2. Study design and definitions ETV (0.5 mg) was given once daily according to the clinician s judgment to initiate antiviral therapy. Hepatitis B e antigen (HBeAg) and the antibody to HBeAg (anti-hbe) were checked at baseline and every 6 months. HBV DNA and alanine aminotransferase (ALT) levels were checked at baseline and every 3 months during follow-up. VR was defined as undetectable HBV DNA by PCR (<12 IU/mL) at any time point during ETV therapy. 9 Patients with a decrease in HBV DNA level less than 1 log 10 IU/mL from baseline at week 12 (3 months) was defined as a primary non-responder. 9 Virological breakthrough (VB) was defined as an increase in HBV DNA of more than 1 log 10 IU/mL compared to the nadir HBV DNA level during ETV therapy. 9 Biochemical response (BR) was defined as normalization of serum ALT levels from abnormal baseline levels, and a threshold of 40 IU/L was used as the upper normal range Biochemical breakthrough (BB) was defined as an increase in serum ALT level above the upper limit of normal after achieving normalization during ETV therapy. 29 HBeAg seroclearance was defined as a persistent absence of antigenemia on at least two consecutive follow-ups. 30 HBeAg seroconversion was defined as the loss of HBeAg with the development of anti-hbe on at least two consecutive follow-ups. 31 Serological response (SR) was defined as HBeAg seroclearance with or without HBeAg seroconversion in HBeAg positive patients. Genotypic resistance was defined as the appearance of viral populations bearing amino acid substitutions in the reverse transcriptase region of the HBV genome that conferred resistance to antiviral drugs in patients with VB. 29 Liver cirrhosis was defined as follows: (1) liver cirrhosis confirmed by biopsy, (2) either current or past evidence of cirrhosis-related complications (e.g., ascites, 6

14 varices, or hepatic encephalopathy), or (3) platelet count <100,000/mm 3 with ultrasonographic findings suggestive of cirrhosis, including blunted or nodular liver edge accompanied by splenomegaly (>12 cm). 32,33 Compliance with therapy was assessed when patients visited the outpatient clinic and phone calls were made to those who did not make their scheduled appointment. 3. Laboratory assay HBV DNA levels were measured by quantitative PCR assay (Amplicor HBV Monitor Test, Roche Diagnostics, Basel, Switzerland; detection limit, approximately 12 IU/mL). HBV serologic tests for HBsAg, HBeAg, and anti-hbe were conducted with enzyme-linked immunoassays (Dade Behring, Marburg, Germany). Biochemical data including ALT, aspartate aminotransferase (AST), total bilirubin, albumin, creatinine, and prothrombin time were analyzed using a sequential multiple auto-analyzer (Hitachi Ltd, Tokyo, Japan). For genotypic analysis, HBV DNA was extracted from serum samples using commercial kits (Qiagen, Velencia, CA), and the HBV polymerase reverse transcriptase domain, encoding aminoacids 1 to 344, was amplified by polymerase chain reaction methods and sequenced directly. 4. Endpoint The primary endpoint was to investigate the optimal PVR criteria in terms of cut-off HBV DNA level and time point to predict VR at week 96 in treatment-naïve patients with CHB who were receiving ETV 7

15 5. Statistical analysis Baseline characteristics of the patients with and without VR were compared by student t-test for continuous variables and chi-square test (or Fisher s exact test) for categorical variables. To define the optimal time point for PVR, receiver-operating characteristics (ROC) curves and the area under the ROC curve (AUC) using HBV DNA levels at weeks 24 and 48 were calculated, and their overall accuracies were compared. After the optimal time point was established, the optimal cut-off HBV DNA level for PVR was subsequently calculated using the Youden index (defined as sensitivity+specificity 1). 34 Statistical analysis was performed using SAS software version (SAS, Cary, NC, USA). Two-sided P values less than 0.05 were considered statistically significant. 8

16 III. RESULTS 1. Patient baseline characteristics A total of 420 patients with CHB from four institutes were consecutively treated with ETV as the first-line therapy during the study period. Of these, 22 (5.2%) patients were lost to follow-up, 20 (4.8%) showed poor compliance, 8 (1.9%) were primary non-responders and 195 (46.4%) were followed for less than 2 years. Finally, 175 patients who completed 2 years of ETV therapy were enrolled for analysis. Table 1 shows the baseline characteristics of the patients. The mean age was 45 years (range, 34 56), and male gender was predominant (n=126, 72.0%). HBeAg was positive in 121 (69.1%) patients and cirrhosis accompanied in 61 (34.9%). The baseline median serum ALT and HBV DNA levels were 166 IU/L (range, ) and 7.03 log 10 copies/ml (range, ), respectively. VR at week 96 was identified in 139 (79.4%) patients. Age, gender, and the presence of liver cirrhosis were not significantly different between patients with VR at week 96 and those without (all P >0.05). HBeAg positivity and baseline HBV DNA level were significantly higher in patients without VR (P=0.004 and P=0.001, respectively), whereas baseline ALT level was significantly higher in patients with VR at week 96 (P=0.008). 9

17 Table 1. Baseline characteristics Total population Patients with VR Patients without VR P value n=175 (100) n=139 (79.4) n=36 (20.6) Age (years) 45 ± ± ± Male gender 126 (72.0) 100 (71.9) 26 (72.2) HBeAg positivity 121 (69.1) 89 (64.0) 32 (88.9) Cirrhosis 61 (34.9) 52 (37.4) 9 (25.0) Baseline ALT (IU/L) 166 ± ± ± Baseline HBV DNA (log 10 copies/ml) 7.03 ± ± ± Variables are expressed as mean ± SD or n (%). Virologic response indicates undetectable HBV DNA (<12 IU/mL). VR, virologic response; ALT, alanine aminotransferase. 2. Treatment outcomes Table 2 shows the treatment outcomes following ETV therapy. After 2 years of ETV therapy, 139 (79.4%) patients achieved VR (89 [73.6%] HBeAg positive and 50 [92.6%] HBeAg negative). BR was achieved in 164 (93.7%) patients after 2 years of ETV therapy (113 [93.4%] HBeAg positive and 51 [94.4%] HBeAg negative). The VR and BR rate were always higher in HBeAg negative patients than in HBeAg positive patients at the same time point. The 24-, 48-, and 96-weeks cumulative SR rates in HBeAg positive patients (n=121) were 7.4%, 13.2%, and 18.2%, respectively. The cumulative occurrence of VB were two (1.1%) patients at week 48, four (2.3%) at week 72, and eight (4.6%) at week 96. All of these eight patients underwent genotypic resistance tests, but no one showed genotypic resistance. BB occurred in three (1.7%) patients at week 48 and in six (3.4%) at week 96. Of the six patients with BB, three showed VB. 10

18 Table 2. Treatment outcomes during 2 years of ETV therapy ETV therapy duration 24 week 48 week 96 week Treatment response * Virologic response Total, n = (38.9) 120 (68.6) 139 (79.4) HBeAg (+), n = (29.8) 74 (61.2) 89 (73.6) HBeAg (-), n = (59.3) 46 (85.2) 50 (92.6) Biochemical response Total, n= (80.0) 150 (85.7) 164 (93.7) HBeAg (+), n = (77.7) 101 (83.5) 113 (93.4) HBeAg (-), n = (85.2) 49 (90.7) 51 (94.4) Serological response in HBeAg(+) 9 (7.4) 16 (13.2) 22 (18.2) Seroclearance 7 (5.8) 11(9.1) 14 (11.6) Seroconversion 2 (1.6) 5 (4.1) 8 (6.6) Antiviral resistance Virological breakthrough 0 (0.0) 2 (1.1) 8 (4.6) Biochemical breakthrough 0 (0.0) 3 (1.7) 6 (3.4) Genotypic resistance 0 (0.0) 0 (0.0) 0 (0.0) Variables are expressed as n (%). Virologic response * is defined as undetectable HBV DNA (<12 IU/mL), Biochemical response as normalization of serum ALT levels (<40 IU/L), Serological response as HBeAg seroclearance with or without HBeAg seroconversion in HBeAg positive patients, Virological breakthrough as an increase in HBV DNA more than 1 log 10 IU/mL compared to the nadir HBV DNA, Biochemical breakthrough as an increase in serum ALT level above upper limit of normal after achieving normalization, and Genotypic resistance as the appearance of viral populations bearing amino acid substitutions in the reverse transcriptase region of the HBV genome in patients with virological breakthrough. ETV, entecavir 3. HBV DNA level vs. HBV DNA reduction to define PVR The ROC curves of the HBV DNA level or the HBV DNA reduction from baseline to predict VR at week 96 and their corresponding AUC values are shown in Figure 1. The AUCs of the HBV DNA level at weeks 24 and 48 were consistently higher than those of HBV DNA reduction from baseline (AUC=0.824 vs at week 24; AUC=0.908 vs at week 48). Therefore, HBV DNA level was adopted to define PVR. Subsequent 11

19 comparative analysis showed that the AUC for the HBV DNA level at week 48 was significantly higher than that at week 24 (AUC=0.908 vs ; P=0.023). Figure 1. ROC curve analysis using HBV DNA level (left) and HBV DNA (HBV DNA reduction from the baseline) (right) at week 24, week 48 to predict VR at 2 years. 4. The optimal cut-off HBV DNA level to define PVR Because HBV DNA level at week 48 was more useful to predict VR at week 96, we investigated the optimal cut-off HBV DNA level at week 48 using the Youden index (sensitivity+specificity 1). According to the maximal Youden index, HBV DNA 35 IU/mL (2.24 log 10 copies/ml, 174 copies/ml) was the optimal cut-off to predict VR at week 96 with a sensitivity of 92.8%, a specificity of 86.1%, a positive predictive value of 96.3% and a negative predictive value of 75.6% (Table 3). 12

20 Table 3. Distribution of patients at week 96 according to optimal PVR Optimal cut-off Patients Patients Sensitivity Specificity PPV NPV * HBV DNA with VR (n) without VR (n) (%) (%) (%) (%) 35 IU/mL < 35 IU/mL Variables are expressed as number of patients or %. Optimal cut-off HBV DNA * was determined by the maximal Youden index (sensitivity+specificity-1). Virologic response is defined as undetectable HBV DNA (<12 IU/ml). PVR, partial virological response; VR, virologic response; PPV, positive predictive value; NPV, negative predictive value. 5. The flow of patients according to treatment response According to the cut-off HBV DNA level at week 48 (35 IU/mL), 41 (23.4%) patients met the PVR criteria (HBV DNA level 35 IU/ml at week 48) whereas 134 (76.6%) exhibited HBV DNA level <35 IU/ml at week 48 (defined as favorable VR in this study). Figure 2 shows the treatment response during 2 years of ETV therapy. Of 41 patients with PVR, 31 (75.6%) still had detectable HBV DNA levels at week 96. Of 134 patients with a favorable VR, 129 (96.3%) achieved VR at week 96. Finally, 139 patients (79.4%) achieved VR at week 96. Patients with PVR at week 48 showed a significantly higher risk for detectable HBV DNA levels at week 96 than those with favorable VR at week 48 (Odds ratio 79.9). 13

21 Figure 2. The flow of patients according to treatment response. PVR, HBV DNA 35 IU/ml at week 48; Favorable VR, HBV DNA <35 IU/ml at week 48; VR, HBV DNA <12 IU/mL; Non VR, HBV DNA 12 IU/mL. 6. Factors predicting VR at week 96 in patients with PVR Although 41 (23.4%) patients were PVR at week 48, 10 (24.4%) achieved undetectable HBV DNA levels at week 96 (Fig. 2). We conducted further analysis to identify which of these PVR patients would finally achieve VR at week 96 (Table 4). When we compared baseline and on-treatment variables between patients with VR and those without at week 96, none of the variables differed except the baseline HBV DNA level, which was significantly lower in patients who later achieved VR at week 96 (6.88 ± 1.20 log 10 copies/ml in 14

22 patients with VR vs ± 1.35 log 10 copies/ml in patients without VR, P=0.045). In patients with PVR at week 48, the best cut-off to predict VR at week 96 was a baseline HBV DNA 1.77E + 08 IU/mL (7.85 log 10 copies/ml, 8.84E + 08 copies/ml), according to the maximal Youden index (a sensitivity 93.6%, a specificity 83.5%, a positive predictive value 94.2% and a negative predictive value 89.8%). 34 Table 4. Factors predicting VR at week 96 in patients with PVR Total PVR Patients with VR Patients without VR P value 41 (100) 10 (24.4) 31 (75.6) Baseline variables Age (years) 45 ± ± ± Male gender 31 (75.6) 8 (80.0) 23 (74.3) HBeAg positivity 35 (85.4) 7 (70.0) 28 (90.3) Cirrhosis 9 (22.0) 3 (40.0) 6 (19.4) Baseline ALT (IU/L) 120 ± ± ± Baseline HBV DNA (log 10 copies/ml) 7.64 ± ± ± On-treatment variables ALT level at week 48(IU/L) 31 ± ± 14 31± BR at week (78.0) 8 (80.0) 24 (77.4) HBV DNA level at 4.02 ± ± ± week 48 (log 10 copies/ml) HBV DNA reduction at ± ± ± week 48 (log 10 copies/ml) Variables are expressed as mean ± SD or n (%). Partial virological response is defined as HBV DNA 35 IU/mL at week 48, virologic response as undetectable HBV DNA (<12 IU/mL) and biochemical response as normalization of serum ALT levels. VR, virologic response; PVR, partial virological response; ALT, alanine aminotransferase. 15

23 Figure 3. A practical management guideline in treatment-naïve patients with CHB receiving ETV. 16

24 IV. DISCSSION Suppressing HBV replication is a principal goal of CHB therapy. 6 According to the EASL HBV Clinical Practice Guidelines, the endpoint of CHB therapy is to reduce HBV DNA level as low as possible to ensure virological suppression and biochemical remission, ultimately resulting in histological improvement and prevention of two major complications including cirrhosis and hepatocellular carcinoma. 9 Several strategies may be considered in view of viral replication suppression, as reflected by the HBV DNA level. The first strategy is to continue current NUC until complete viral suppression is achieved. However, this strategy can raise issues of high cost and appearance of resistance. 35 The second strategy is to predict when and under what conditions complete viral suppression might be attained in the future. If this is possible, then time and costs on ineffective NUC would not be wasted. Considering these issues, the EASL guidelines recently suggested a new PVR criteria concept for the use of NUCs, which was defined as more than one log decline of viremia compared to baseline but still detectable serum levels of HBV DNA by real-time PCR assay ( IU/ml) at week 24 or 48, depending on the genetic barrier of the anti-hbv drug. However, the cut-off HBV DNA level and the time-point of this PVR criterion were not based on scientific evidence but on experts perceptions. Several investigations have focused on the issue of PVR in patients treated with LAM or ADF. However, even though ETV has been thought as a potent NUC with high genetic barrier, viral resistance to ETV has emerged. Zhonghua et al. 36 reported that three patients during 96 weeks and additional two patients 17

25 during the following 48 weeks developed ETV resistance among 160 patients who received first-line ETV therapy for 144 weeks. Suzuki et al. 20 reported an ETV-resistant strain in a NUC-naïve patient with genotype H. In the same context, PVR in ETV-treated subjects has been demonstrated by several investigators. Chang et al. 16 reported 33% PVR (HBV DNA level <300 copies/ml at week 48) in HBeAg positive patients and Lai et al. 15 reported 10% of PVR in HBeAg negative patients. To prevent treatment failure and the emergence of ETV-resistant strains, it is crucial to define optimal PVR criteria and change treatment strategies. Therefore, this study investigated the cut-off HBV DNA level and the time point of optimal PVR to predict VR at week 96 in treatment-naïve patients with CHB who were receiving ETV by scientific method. To establish optimal PVR criteria, we determined whether the HBV DNA level or the HBV DNA reduction from baseline was more useful for predicting VR at week 96 during ETV therapy. Based on our data, the AUCs of the HBV DNA level were consistently superior to those of HBV DNA reduction from the baseline both at week 24 and 48, and the AUC of the HBV DNA level at week 48 was more predictable than that at week 24 with statistical significance. Furthermore, the AUC of the HBV DNA at week 48 was significantly higher than that of the baseline HBV DNA (0.908 vs ; P=0.036). This implies that the response to ETV therapy may not be totally determined by baseline HBV DNA level, and viruses that show an excellent response to ETV might exist even in a patient with a high baseline viral load. Therefore, a whole genomic analysis of the virus such as revealing the frequency of core promoter mutant should be the focus of future studies to explain this intrinsically different responsiveness. 18

26 Finally, our PVR criteria for ETV therapy were defined as an HBV DNA level exceeding 35 IU/mL (2.24 log 10 copies/ml, 174 copies/ml) at week 48. That is, if patients achieve HBV DNA levels of less than 35 IU/mL at week 48 (favorable VR), ETV should be continued, because the majority of patients (96.3%) may achieve VR at week 96 with high positive predictive value. On the contrary, if patients show an HBV DNA level of more than 35 IU/mL at week 48 (PVR), more than half of patients (75.6%) may not achieve VR at week 96. However, decisions on switching or add-on strategies for patients with PVR at week 48 should be considered cautiously, because 24.4% of patients with PVR at week 48 finally achieved VR after 2 years of ETV therapy. Considering that the kinetics of viral clearance are different between HBeAg positive and negative HBV, we stratified the study population according to the presence or absence of HBeAg positivity. The results from 121 HBeAg positive patients were consistent with those from all patients. AUCs of the HBV DNA level at week 24 and week 48 were consistently higher than those of HBV DNA reduction from baseline (AUC=0.812 vs at week 24; AUC=0.925 vs at week 48). Moreover, AUC of the HBV DNA level at week 48 was significantly higher than that at week 24 (AUC=0.925 vs ; P=0.019). On the contrary, statistically significant time point and cutoff HBV DNA level were not identified in 54 HBeAg negative patients (all P values for AUC >0.05, data not shown). As these unsatisfactory results may be due to a small number of HBeAg negative patients (n=54), future studies with larger number of HBeAg negative patients will overcome this statistical problem. Among the patients with PVR, we performed a subgroup analysis to identify who would achieve VR at week 96. Only baseline HBV DNA level 19

27 was significantly lower in patients who later achieved VR at week 96 (6.88 ± 1.20 log 10 copies/ml vs ± 1.35 log 10 copies/ml, P=0.045). This indicates that in patients with PVR, baseline HBV DNA level can be used as another predictor for VR at week 96. In patients who show high baseline viral load (HBV DNA 1.77E+08 IU/mL) and PVR at week 48, the treatment strategy should be changed, because these patients will not benefit from further use of ETV. On the contrary, patients who show low baseline viral load (HBV DNA <1.77E+08 IU/mL) and PVR at week 48 might need more time to achieve VR. Thus, modification of treatment strategies should be delayed in these patients, because there is a chance for them to achieve VR if ETV is continued for up to 2 years. However, the role of baseline HBV DNA level measurement in patients with PVR should be validated in the future. Additionally, we analyzed the influence of the pattern of HBV DNA level between week 48 and week 72 on the prediction of VR at week 96 for 41 patients with PVR. Of these, 35 (85.4%) patients exhibited decreasing pattern of HBV DNA from week 48 to week 72, whereas the other 6 (14.6%) patients had increasing pattern. Although all patients who showed PVR at week 48 and VR at week 96 (n=10) had decreasing pattern of HBV DNA from week 48 to week 72, the patterns and differences of HBV DNA level from week 48 to week 72 were not selected as a predictor for VR at week 96 (all P >0.05). It is a great challenge for physicians to manage PVR which forewarns treatment failure in the use of third-generation NUCs such as ETV. Pan et al. reported that among 236 CHB patients receiving ETV therapy for 68 weeks, 3% had a suboptimal response (<1 log copies/ml of HBV DNA reduction at week 24). After switching to tenofovir (TDF) monotherapy, all patients achieved HBV DNA <160 copies/ml within 24 weeks. 37 However, there is risk 20

28 of resistance with a switch-to strategy when maintaining an alternative monotherapy for a long period. On the other hand, in add on strategy, not only uncertain safety data for the concurrent use of ETV and TDF but also increased cost can be a problem. So far, there are conflicting opinions among experts on this issue, and more investigations should be conducted. To the best of our knowledge, this is the first attempt to establish optimal PVR criteria for treatment of CHB using the scientific method. We propose a practical management guideline based on our PVR criteria (Fig. 3). An initial response assessment should be performed at week 12. Responders should continue ETV and primary non-responders should be checked for compliance. The treatment response should be assessed again at week 48. Patients with a favorable response (<35 IU/mL of HBV DNA at week 48) should continue ETV, whereas patients with PVR ( 35 IU/mL of HBV DNA at week 48) should be managed differently according to their baseline HBV DNA level, if they are compliant. Our study still leaves several issues unresolved. First, the PVR criteria in our study should be validated in a cohort with different ethnicities and settings to test its extended use and corresponding management flow. Second, we could not predict the long-term VR beyond 2 years of ETV therapy because there were few patients who used ETV for more than 96 weeks, and even fewer for more than 144 weeks. Besides, 195 patients were initially excluded only because they received ETV therapy for less than 96 weeks. Therefore, long-term large population based studies should be followed to deduce comprehensive conclusions. Lastly, although we demonstrated that measuring HBV DNA level at week 48 is the best way to predict VR at week 96 among available variables, this criterion could not perfectly discriminate patients who 21

29 would benefit from continued use of ETV for up to 2 years (23.4% of patients with PVR). Thus, if future studies identify new factors that are better predictors than currently available ones, such as differences in viral sequence, our PVR criteria can be modified using these newly identified predictors. 22

30 V. CONCLUSION In conclusion, an HBV DNA level exceeding 35 IU/mL (2.24 log10 copies/ml, 174 copies/ml) at week 48 was the optimal criterion for treatment-naïve patients with CHB receiving ETV. The results of this study support the PVR criteria suggested by the EASL guidelines for treatment-naïve patients with CHB receiving ETV. 23

31 REFERENCES 1. Hepatitis B vaccines: WHO position paper--recommendations. Vaccine 2010;28: Lee WM. Hepatitis B virus infection. N Engl J Med 1997;337: Chen C, Yang HI, Su J, Jen C, You S, Lu S, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA 2006;295: Iloeje UH, Yang HI, Su J, Jen C, You S, Chen C. Predicting cirrhosis risk based on the level of circulating hepatitis B viral load. Gastroenterology 2006;130: Keeffe EB, Dieterich DT, Han SB, Jacobson IM, Martin P, Schiff ER, et al. A treatment algorithm for the management of chronic hepatitis B virus infection in the United States: 2008 update. Clin Gastroenterol Hepatol 2008;6: Liaw Y, Leung N, Kao J, Piratvisuth T, Gane E, Han K, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2008 update. Hepatol Int 2008;2: Lok AS, McMahon BJ. Chronic hepatitis B. Hepatology 2007;45: Ahn SH, Chan HL, Chen P, Cheng J, Goenka MK, Hou J, et al. Chronic hepatitis B: whom to treat and for how long? Propositions, challenges, and future directions. Hepatol Int 2010;4: European Association For The Study Of The Liver UH. EASL Clinical Practice Guidelines: management of chronic hepatitis B. J Hepatol 2009;50: Lampertico P. Partial virological response to nucleos(t)ide analogues in naïve patients with chronic hepatitis B: From guidelines to field practice. J Hepatol 2009;50:

32 11. Lai C, Gane E, Liaw Y, Hsu C, Thongsawat S, Wang Y, et al. Telbivudine versus lamivudine in patients with chronic hepatitis B. N Engl J Med 2007;357: Safadi R, Xie Q, Chen Y, Yin Y-K, Weis Y, Hwang S. A randomized trial of switching to telbivudine versus continued lamivudine in adults with chronic hepatitis B: results of the primary analysis at Week 24. J Hepatol 2007;46:S Marcellin P, Heathcote EJ, Buti M, Gane E, de Man RA, Krastev Z, et al. Tenofovir disoproxil fumarate versus adefovir dipivoxil for chronic hepatitis B. N Engl J Med 2008;359: Hadziyannis SJ, Tassopoulos NC, Heathcote EJ, Chang T, Kitis G, Rizzetto M, et al. Long-term therapy with adefovir dipivoxil for HBeAg-negative chronic hepatitis B for up to 5 years. Gastroenterology 2006;131: Lai C, Shouval D, Lok AS, Chang T, Cheinquer H, Goodman Z, et al. Entecavir versus lamivudine for patients with HBeAg-negative chronic hepatitis B. N Engl J Med 2006;354: Chang T, Gish RG, de Man R, Gadano A, Sollano J, Chao Y, et al. A comparison of entecavir and lamivudine for HBeAg-positive chronic hepatitis B. N Engl J Med 2006;354: Leung N, Peng C, Hann H, Sollano J, Lao-Tan J, Hsu C, et al. Early hepatitis B virus DNA reduction in hepatitis B e antigen-positive patients with chronic hepatitis B: A randomized international study of entecavir versus adefovir. Hepatology 2009;49: Tenney DJ, Pokornowski KA, Rose RE. Entecavir maintains a high genetic barrier to HBV resistance through 6 years of naive patients. J Hepatol 2009;50:S Guo J, Li Q, Shi X, Zhang D, Zeng A, Feng T, et al. Dynamics of hepatitis B virus resistance to entecavir in a nucleoside/nucleotide-naïve 25

33 patient. Antiviral Res 2009;81: Suzuki F, Akuta N, Suzuki Y, Yatsuji H, Sezaki H, Arase Y, et al. Selection of a virus strain resistant to entecavir in a nucleoside-naive patient with hepatitis B of genotype H. J Clin Virol 2007;39: Cho J, Yoon K, Lee K, Park D, Lee Y, Moon H, et al. Distribution of hepatitis B virus genotypes in Korea. Korean J Hepatol 2009;15: Bae SH, Yoon SK, Jang JW, Kim CW, Nam SW, Choi JY, et al. Hepatitis B virus genotype C prevails among chronic carriers of the virus in Korea. J Korean Med Sci 2005;20: Song B, Cui XJ, Kim H. Hepatitis B virus genotypes in Korea: an endemic area of hepatitis B virus infection. Intervirology 2005;48: Cho J, Park D, Kim T, Yeom J, Kim H, Moon J, et al. Genotypes of hepatitis B virus on Jeju island. Korean J Hepatol 2004;10: Lee JM, Kim HJ, Park JY, Lee CK, Kim DY, Kim JK, et al. Rescue monotherapy in lamivudine-resistant hepatitis B e antigen-positive chronic hepatitis B: adefovir versus entecavir. Antivir Ther 2009;14: Martinot-Peignoux M, Boyer N, Cazals-Hatem D, Pham BN, Gervais A, Le Breton V, et al. Prospective study on anti-hepatitis C virus-positive patients with persistently normal serum alanine transaminase with or without detectable serum hepatitis C virus RNA. Hepatology 2001;34: Persico M, Persico E, Suozzo R, Conte S, De Seta M, Coppola L, et al. Natural history of hepatitis C virus carriers with persistently normal aminotransferase levels. Gastroenterology 2000;118: Kim DY, Ahn SH, Lee HW, Park JY, Kim SU, Paik YH, et al. Clinical course of virologic breakthrough after emergence of YMDD mutations in HBeAg-positive chronic hepatitis B. Intervirology 2008;51: Lok AS, Zoulim F, Locarnini S, Bartholomeusz A, Ghany MG, 26

34 Pawlotsky J, et al. Antiviral drug-resistant HBV: standardization of nomenclature and assays and recommendations for management. Hepatology 2007;46: Ahn SH, Park YN, Park JY, Chang H, Lee JM, Shin JE, et al. Long-term clinical and histological outcomes in patients with spontaneous hepatitis B surface antigen seroclearance. J Hepatol 2005;42: Yuen M, Fong DY, Wong DK, Yuen JC, Fung J, Lai C. Hepatitis B virus DNA levels at week 4 of lamivudine treatment predict the 5-year ideal response. Hepatology 2007;46: Kim DY, Kim SU, Ahn SH, Park JY, Lee JM, Park YN, et al. Usefulness of FibroScan for detection of early compensated liver cirrhosis in chronic hepatitis B. Dig Dis Sci 2009;54: Kim SU, Han K, Nam CM, Park JY, Kim DY, Chon CY, et al. Natural history of hepatitis B virus-related cirrhotic patients hospitalized to control ascites. J Gastroenterol Hepatol 2008;23: Youden WJ. Index for rating diagnostic tests. Cancer 1950;3: Perrillo RP, Marcellin P. Effect of newer oral antiviral agents on future therapy of chronic hepatitis B. Antivir Ther 2010;15: Yao G, Ren H, Xu D, Zhou X, Jia J, Wang Y, et al. Results of 3 years of continuous entecavir treatment in nucleos(t)ide-naive chronic hepatitis B patients. Zhonghua Gan Zang Bing Za Zhi 2009;17: Pan C, Hu K. Excellent treatment response to tenofovir (TDF) monotherapy in chronic hepatitis B (CHB) patients with prior suboptimal response to entecavir (ETV) monotherapy. Hepatology 2008;48:725A. 27

35 ABSTRACT(IN KOREAN) 만성 B 형간염환자에서엔테카비어를이용한초치료시 부분적바이러스반응의임상적의의 < 지도교수안상훈 > 연세대학교대학원의학과 전영은 서론 : 2009년유럽간학회 (European Association for the Study of the Liver) 에서제시한만성 B형간염에관한임상지침에따르면, 부분적바이러스반응의정의는 B형간염바이러스 DNA (HBV DNA) 측정값의최소수준이나측정시점을정의하는데있어과학적근거가미약하다. 따라서, 본연구는엔테카비어초치료를받은만성 B형간염환자에있어 96주째의바이러스반응을가장잘예측할수있는이상적인부분적바이러스반응의정의가무엇이지조사해보고자하였다. 재료및방법 : 96주동안엔테카비어초치료를받은 175명 ( 남자 126명, 여자 49명 ) 의환자를대상으로전향적으로연구하였다. 96주째의바이러스반응을예측하는데있어, receiver operating characteristics (ROC) 곡선과 area under the receiver operating characteristics curve (AUC) 방법을이용하여 HBV DNA 의가장적합한측정시점을정하였으며, Youden index를이용하여가장적합한 HBV DNA의최소수준을정하였다. 결과 : 96주간의엔테카비어치료후에, 139(79.4%) 명의환자가바이러스반응을달성하였다. 96주째의바이러스반응을예측하는 AUC 값은 24주째보다 48주때통계학적으로 28

36 유의하게더크게나타났다 (AUC=0.806 vs , P=0.023). 초치료 48주째부분적바이러스반응에해당하는가장이상적인 HBV DNA 최소수준은 35 IU/mL 이었다. 41(23.4%) 명의환자가엔테카비어치료시상기부분적바이러스정의 (48주째의 HBV DNA 수준 35IU/mL) 에해당하였다. 결론 : 48주째의 HBV DNA 수준 35 IU/mL 초과는엔테카비어를초치료로한환자에있어 96주째의치료실패를예측할수있는이상적인부분적바이러스반응의정의이다. 본연구는유럽간학회에서제시한부분적바이러스반응의정의가과학적근거가있음을지지해주는것이다 핵심되는말 : 만성 B형간염, 엔테카비어, 부분적바이러스반응, 바이러스반응 29