The subjects of this study are hepatitis patients whose ALT is greater than or equal to 80 U/L but the ALT is less than or equal to 10 times the upper limit of normals and who did not receive the antiviral drug therapy. Specifically, HBV >10⁵ copies/mL in case of HBeAg positive and HBV >10⁴ copies/mL in case of HBeAg negative were included in this study. Acute hepatitis A patients and chronic hepatitis C patients were excluded. Among liver cirrhosis patients, those who had hepatic coma, esophageal variceal hemorrhage, spontaneous bacterial peritonitis, and uncompensated liver cirrhosis of Child-Pugh score ≥10 were excluded. Also, patients with hepatocellular carcinoma and malignant tumors in other organs, patients with creatinine of greater than 1.5 times ULN, patients given immunosuppressive drugs in 6 months prior to the enrollment in this study, and patients with serious diseases believed to affect the test other than liver disease (e.g. uncontrollable diabetes, alcoholics, and malignant tumor, etc.) were excluded.

Subjects are randomized at an one to one ratio to the entecavir monotherapy and the entecavir plus carnitine complex (Godex®, Seoul, Celltrion pharm) combination therapy. After the central institution grants randomized groups to the study organization, randomization is carried out by stratifying the groups according to the presence of HBeAg by checking if they were HBeAg positive or negative at the time of screening. Computer program generates a list of random numbers. First, with the date of random number generation set as a seed number, two random numbers are created and then used as seed numbers for creating random numbers for two HBeAg positive and negative groups. The block sizes of four and six are both used. Sufficient random numbers are generated in each layer. In each block, entecavier plus carnitine complex combination therapy (test group) was allocated with small numbers and entecavir monotherapy (control group) was allocated to big numbers depending to the size of random numbers. The randomization table is randomly allocated to subjects depending on HBeAg type in the order of the subject patients' enrollment in the study organization. They are randomly selected to be divided into a monotherapy group and a combination therapy group. Entecavir (0.5 mg) and entecavir plus carnitine complex (2 cap./time, 3 times/day, totally 2,472 mg/day, oral administration) were given to the monotherapy group and the combination therapy group, respectively for 12 months.

Before and 3, 6, 9, 12 months after medication, ALT, AST, bilirubin, HBV-DNA PCR, HBsAg, HBsAb, HBeAg, and HBeAb were measured. Then abdominal ultrasonography and AFP examination were carried out every six months. HBV DNA was measured by real-time PCR method on a central lab using a COBAS TaqMan analyser (Roche Diagnostics, Mannheim, Germany) with a detection limit of 20 IU/mL. Quantitative HBsAg was tested on a central lab using the ARCHITECT HBsAg QT immunoassay (Abbott Diagnostics, Wiesbaden, Germany). Quantitative HBsAg was examined every 3 months during the follow-up period.

Interferon gamma enzyme-linked immunosorbent spot (ELISPOT) examination was performed using the ELISPOT test to evaluate the ability of leucocytic interferon-γ secretion caused by HBsAg stimulation. Blood specimen was transferred to the central laboratory within 2 days after blood-collecting. Peripheral blood mononuclear cells (PBMC) were separated here to be stored in a nitrogen tank. Seed 100 µL of cell suspension in 5-well plates and put 1uL of dilution of core and surface protein 1 to medium 9. For a positive control group, mAb CD3-2 was added until its maximum concentration reached 100 ng/mL and cultured in 5% CO2 incubator at 37℃ for 12-48 hours. The cell suspension was removed and washed five times using PBS. The detection reagent (mAb 7-B6-1) was added in the ratio of 200 to 1 in PBS where 10% FBS is contained and another 100 µL of it was added into each well. Incubate it at room temperature for 2 hours and wash five time using PBS. Filter it at 0.45 µm and put 100 µL of substrate solution (BCIP/NBT-plus) to be developed for 15 minutes. Water-wash and dry it at room temperature to be counted using ELISpot reader (AELVIS, Germany).

We collected 5 mL of peripheral venous blood before and after administration of each drug an put it in EDTA tube to be kept in a freezer at -80℃. Process it in a centrifuge at room temperature and extract DNA from it, then measure peripheral blood mitochondria DNA using real-time polymerase chain reaction (PCR) using COBAS Ampliprep (Roche, Indianapolis, IN, USA). Develop a probe for GAPDH and mtDNA and mark 5' terminal and 3' terminal each with 5-carboxyfluorescein(FAM) reporter and 6-carboxy-tetramethyl-rhodamine (TAMRA) quencher. PCR was performed for mtDNA and PCR mixture was 15 µL which consists of TaqMan Universal PCR Master Mix (ABI, USA) 7.5 µL, primer forward (5 pmol/mL) 1.5 µL, primer reverse (5 pmol/mL) 1.5 µL, probe (1 pmol/µL) 1.5 µL, and distilled water 0.5 µL, DNA 2.5 µL. PCR used ABI 7500 Real-Time PCR System and amplification was repeated forty times on conditions of 2 minutes at 50℃, once for 10 minutes, 15 seconds at 95°, and 1 minute at 60°. For a qualitative analysis of mitochondrial DNA copy number, the comparative analysis using the amplification of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was implemented.

Since the primary end point is ALT normalization rate when 12-month treatment is completed, Fisher's exact test or Chi-square test needs to be used to verify the difference between two groups. But, if there are significant differences between demographic data and clinical history data, logistic regression analysis is used for correction.

Continuous variables (HBsAg titer, HBV DNA titer, ELISPOT and the peripheral blood mitochondrial DNA copy number) were analyzed by t -test or Wilcoxon Rank test. Categorical variables (rate of ALT normalization and HBV DNA <300 copies/mL) were analyzed by the Fisher's exact test of Chi-square test.

A total of 130 patients were randomized in this clinical trial. Out of those randomized patients, six were excluded from analysis due to withdrawal, two due to follow-up failure, one due to adverse reactions, and two due to inclusion/exclusion criteria, while a total of 119 patients were included in this study (Fig. 1). The average ages of the monotherapy group and the combination therapy group were 44.9 and 43.0 years old, respectively. There was no difference in ALT, AST, and virus concentration in serum between two groups (Table 1).

Normalization rates of subjects' serum ALT were 58.9% and 87.5% in 3-month dosing and 6-month dosing for the monotherapy group, and 95.2% and 100% for the combination therapy group, which showed that the figure was statistically higher in the combination therapy group (P<0.0001 vs. P=0.0038). After 9-month dosing, the figures showed no significant difference between the monotherapy group and the combination therapy group (P=0.1011). At the time of 12-month dosing, the figures showed statistically significant difference between two groups with 85.7% and 100% (P=0.0019) (Fig. 2). On condition of the ALT threshold value below than 30 U/L for males and 19 U/L for females, the serum ALT normalization rates after 3, 6, 9-month dosing were statistically significantly higher in the combination therapy. At the time of 12-month dosing when the study is completed, the normalization rates were 69.6% in the monotherapy group and 93.7% in the combination therapy group, which showed statistically significant difference between two groups (P=0.0006) (Table 2). At the time of 12-month dosing, the average values of serum ALT were 25.2±15.0 in the monotherapy group and 13.6±5.1 in the combination therapy group, which showed lower in the combination therapy group (P<0.0001). When pretreatment ALT was over 200 U/L, the serum ALT normalization rates after 3-month dosing were 53.9% and 88.2% in the monotherapy group and the combination therapy group, respectively (P=0.0348). At the time of 12-month dosing when the study is completed, the ALT normalization rates were 69.2% and 100% in the monotherapy group and the combination therapy group, respectively, which showed statistically significant difference between two groups (P=0.014). Normalization rates of subjects' serum AST were 73.2% and 79.4% in 3-month dosing and 12-month dosing for the monotherapy group, and 92.6% and 88.9% for the combination therapy group, which showed no significant difference between the monotherapy group and the combination therapy group.

The rates of ALT normalization and HBV DNA of less than 300 copies/mL after 3, 6, 9, 12-month dosing were 43.6%, 50.9%, 69.1%, 75.9% in the monotherapy group and 54.8%, 50.0%, 64.4%, 70.7% in the combination therapy group, respectively, which showed the increasing trend of serum virus negative conversion rate in both groups according to treatment period, but no difference in HBV DNA negative conversion rate between two groups.

HBeAg seroclearance rates at the time of therapeutic termination were 14.3% and 5.6% in the monotherapy group and the combination therapy group, respectively while HBeAg seroconversion rates were 10.2% and 5.6% in the monotherapy group and the combination therapy group, respectively, which showed no difference between two groups. In the monotherapy group, quantitative HBsAg level declined from 42,810 IU/mL to 10,433 IU/mL at the time of 12-month dosing on average, while in the combination therapy group, quantitative HBsAg level reduced from 44,037 IU/mL to 8,621 IU/mL at the time of 12-month dosing on average, which showed no difference between two groups.

Before treatment, the abilities of leucocytic interferon-γ secretion caused by hepatitis B surface antigen stimulation were 56.5 SFC/10⁶ PBMCs and 50.5 SFC/10⁶ PBMCs in the monotherapy group and the combination therapy group, respectively. But after 12-month treatment, the ELISPOT values increased to 81.3 SFC/10⁶ PBMCs and 99.2 SFC/10⁶ PBMCs, respectively. The monotherapy group showed no statistical difference in the ability of leucocytic interferon-γ secretion between pretreatment and 12-month treatment (P=0.131), but the combination therapy group showed a significant increase in the 12-month treatment compared to pretreatment (P=0.030) (Fig. 3).

At the baseline and end of treatment, peripheral blood mitochondria copy numbers were 1,298.3 copies/mL and 2,944.1 copies/mL in the monotherapy group and 1,437.2 copies/mL and 2,320.8 copies/mL in the combination therapy group, respectively, but no statistical significance was found in both groups (P=0.7351).

The most frequently reported adverse effects occurred during the study period were fatigue, skin rash, and indigestion. During 12-months, the numbers of adverse events in monotherapy group and combination therapy group were 22 and 35, respectively. Among them only six cases were highly suggested having causal relationship with medication. Indigestion, skin rash, and drowsiness were the frequent adverse events of the medication. However, the prevalence of each symptom was not statistically significant (Table 3). Rest of symptoms did not seem to have causal relation with medication. The prevalence of fatigue, general weakness, chronic gastritis, and upper respiratory infection, has not related with treatment (P=0.2573).