Tenofovir disoproxil fumarate monotherapy for nucleos(t)ide analogue-naïve and nucleos(t)ide analogue-experienced chronic hepatitis B patients

Patients

Patients with chronic HBV infection who were treated with TDF monotherapy (300 mg/day) for at least 12 weeks between December 2012 and July 2014 at a single center were consecutively enrolled. Patients with coinfection with hepatitis C virus or HIV and patients who have experienced multi-drug resistance and were not indicated for TDF monotherapy were excluded. The study protocol was approved by the institutional review board of Ilsan Paik Hospital, Inje University College of Medicine, Goyang, Korea.

Data collection

History of previous antiviral treatment, antiviral drug resistance, presence of cirrhosis and adherence to medication were collected. HBeAg, HBeAb, serum alanine aminotransferase (ALT), and serum HBV DNA were also collected at baseline and every 12 weeks. Viral mutation tests were performed at baseline in NA-experienced patients and in patients with virological breakthrough (VBT).

Laboratory assay

Serologic markers of HBV were tested using commercially available electrochemiluminescence immunoassay (ECLIA, Roche Diagnostics, Branchburg, NJ). Serum HBV DNA levels were measured by real-time PCR using The COBAS® Amplicor assay (Roche Diagnostics, limit of detection 20 IU/mL). Genotypic analysis was done by the restriction fragment mass polymorphism (RFMP) method.

Definition

A complete virological response (CVR) is defined as a decrease in serum HBV DNA to an undetectable level as determined by real-time PCR assay (≤20 IU/mL). A partial virological response (PVR) is defined as a decrease in HBV DNA of more than 2 log IU/mL but with detectable serum HBV DNA at week 48 for entecavir (ETV), ADV and TDF, and at week 24 for LAM and telbivudine (LdT).15 VBT is defined as a confirmed increase in HBV DNA level of more than 1 log IU/mL compared to the nadir HBV level achieved during therapy. Biochemical response is defined as a normalization of ALT (≤40 IU/L). Diagnosis of liver cirrhosis was based on the histological findings or radiological findings together with clinical features indicative of portal hypertension, such as thrombocytopenia (platelet count <100,000/mm³), gastroesophageal varices, or ascites.

Statistical analysis

Continuous variables were summarized as the median (interquartile range) or mean±SD. Serum HBV DNA levels were logarithmically transformed for analysis. Continuous variables were compared using the Student's t-test, ANOVA, nonparametric Mann-Whitney or Kruskal-Wallis test as appropriate and categorical variables were compared using chi-square test. CVR rate was calculated with Kaplan-Meier's curve and compared with the log rank test. A Cox proportional hazard model including variables with P-values ≤0.1 in univariate analysis and clinically relevant factors was used to identify predictive factors independently associated with the time to a CVR. P-value <0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 21.0 (SPSS Inc., Chicago, IL, USA).

Baseline characteristics of patients

From November 2012 to July 2014, a total of 212 patients were treated with TDF monotherapy. Among those, 18 patients were excluded for multidrug resistance, and 58 patients were excluded for follow-up duration < 3 months. A total of 136 patients (male 96 (70.6%), median age 49) were finally included in this study. Baseline characteristics of patients were summarized in Table 1.

Table 1.

Baseline characteristics of the patients

Ninety-four patients (69.1%) were HBeAg-positive and 51 patients (37.5%) had liver cirrhosis. Median follow-up duration was 56 weeks. The mean baseline serum HBV DNA level was 4.9±2.3 log IU/mL and the mean ALT level was 155±252 IU/L.

Seventy-four patients had experienced prior NA therapy (NA-exp group). Among those, 31 patients had documented antiviral resistance to lamivudine (LAM-R group). Twenty-eight patients have not been tested for genotypic mutation. The reasons for medication change to TDF in NA-experienced patients were PVR (n=45), withdrawal hepatitis after discontinuation of prior NA therapy (n=13), virological breakthrough (n=10), emergence of antiviral resistance (n=2), myopathy (n=2) and pregnancy (n=2). Mean duration of washout period in patients with withdrawal hepatitis was 49.7 weeks. Prior history of NA treatment and types of antiviral resistance were summarized in Table 2. Baseline serum HBV DNA was significantly higher in NA-naïve than NA-exp group (5.9±2.0 log IU/mL vs. 3.9±2. log IU/mL, P<0.01). Baseline serum ALT was also significantly higher in NA-naive patients than in NA-experienced patients.

Table 2.

Patterns of prior antiviral treatment and genotypic resistance analysis in NA-exp patients

Virological and biochemical responses to TDF monotherapy

Mean reduction of serum HBV DNA was 3.03 log IU/mL at week 24 and 3.33 log IU/mL at week 48 respectively. Reduction of serum HBV DNA was significantly higher in NA-naïve group than NA-Exp group both at week 24 (4.03 vs. 2.30 log IU/mL, P<0.01) and at week 48 (4.28 vs. 2.46 log IU/mL, P<0.01) (Fig. 1). However, there was no significant difference of the serum HBV DNA reduction between NA-naive and NA-Exp group who had washout periods for previous NA therapy and similar baseline serum HBV DNA with NA-naive patients at week 24 (4.04 vs. 4.59 log IU/mL, P=0.61) and week 48 (4.30 vs. 4.64 log IU/mL, P=0.82).

Figure 1

Mean changes in serum HBV DNA level during TDF treatment.

HBV, hepatitis B virus; NA, nucleos(t)ide analogue; LAM-R, lamivudine resistance.

CVR was achieved in 47.1% at week 24 and 71.5% at week 48 respectively. CVR rate at week 48 in NA-naïve group (71.7%) was not significantly different from NA-exp group (n=74, 71.3%) (Fig. 2A). When NA-naïve group was compared with and LMV-R (n=31, 66.1%), there was no significant difference of CVR between two groups. There was no difference of CVR according to types of NA (NA naive 71.7% vs. patients with exposure to LMV only 74.1% vs. patients with exposure to ETV 68.9%, patients with exposure to ADV 72.9%) either. When VR was defined as <60 IU/mL, VR at week 48 was 88.1% in NA-naïve patients, 88.1% in NA-exp group and 87.9% in LAM-R group respectively. CVR rate at week 48 was significantly higher in HBeAg-negative (87.6%) than HBeAg positive patients (75%) (Fig. 2B). Patients with baseline serum HBV DNA <4.3 log IU/mL achieved significantly higher CVR (87.8%) than those with baseline serum HBV DNA ≥4.3 log IU/mL (59.5%) (Fig. 2C). To compensate the difference of baseline serum HBV DNA in comparison of CVR between NA-naïve and NA-exp group, we analyzed CVR between two groups according to baseline serum HBV DNA level. For patients with baseline serum HBV DNA <4.3 log IU/mL, CVR in NA-naïve patients (n=13, 92.3%) was not significantly different compared with NA-exp group (n=43, 82.7%), or LMV-R group (n=15, 92%) (Fig. 3A). For patients baseline serum HBV DNA ≥4.3 log IU/mL, CVR in NA-naïve group (n=49, 65.2%) had tendency to be higher compared with NA-exp group (n=31, 50.3%, P=0.117), or LMV-R group (n=16, 43%, P=0.063) (Fig. 3B).

Figure 2

Cumulative probability of CVR to TDF according to variables. (A) omparison of CVR rate according to NA-experience. CVR rate in NA-naïve group was not significantly different from that in NA-exp group or LAM-R group. (B) Comparison of CVR rate according to HBeAg status. CVR rate was significantly higher in HBeAg negative patients than HBeAg positive patients. (C) Comparison of CVR rate according to baseline serum HBV DNA. CVR rate was significantly higher in patients with serum HBV DNA <4.3 log IU/mL than those with serum HBV DNA ≥4.3 log IU/mL. CVR, complete virological response; TDF, tenofovir disoproxil fumarate; NA, nucleos(t)ide analogue; LAM-R, lamivudine resistance.

Figure 3

Cumulative probability of a CVR to TDF according to baseline serum HBV DNA. (A) CVR in patents with baseline serum HBV DNA <4.3 log IU/mL. CVR in NA-naïve patients (n=13, 92.3%) was not significantly different from that in NA-exp group (n=43, 85.8%), or LMV-R group (n=15, 92%). (B) CVR in patents with baseline serum HBV DNA ≥4.3 log IU/mL, CVR in NA-naïve patients (n=49, 65.2%) had tendency to be higher as compared with NA-exp group (n=31, 50.3%. P=0.117), or LMV-R group (n=16, 43%, P=0.063). CVR, complete virological response; TDF, tenofovir disoproxil fumarate; NA, nucleos(t)ide analogue; LAM-R, lamivudine resistance.

ALT normalization at week 48 was achieved in 82.6% (Fig. 4). Proportion of ALT normalization were also not significantly different compared NA-naïve (79.3%) with NA-Exp (88.9%) or LAM-R (89.7%) group.

Figure 4

Cumulative probability of a biochemical response to TDF monotherapy. The biochemical response rate did not differ significantly between the NA-naïve group and the NA-exp group or the LAM-R group. TDF, tenofovir disoproxil fumarate; NA, nucleos(t)ide analogue; LAM-R, lamivudine resistance.

HBeAg loss occurred in 9.6% (9/94). HBeAg loss was 5.0% in NA-naïve patients, 13.0% in NA-exp group and 12.0% in LAM-R group respectively. Virological breakthrough developed in 5 patients but no genotypic resistance was identified. Serum HBV DNA declined without medication change in four patients except one patient who was lost to follow up.

Predictive factors for a complete virological response to TDF monotherapy

Univariate Cox regression analyses of the total patients revealed that baseline serum HBV DNA level (HR, 0.830; 95% CI 0.888 to 0.911) and HBeAg negativity (HR, 1.685;95% CI 1.096 to 2.590) were significantly associated with a CVR. Age, gender, baseline ALT, liver cirrhosis, NA-experience and resistance to lamivudine did not show any association with a CVR. In multivariate Cox proportional hazard model including variables with P-value of ≤0.1 and NA-experience which might be a clinically significant factor, baseline serum HBV DNA was the only predictive factor for a CVR (HR, 0.809; 95% CI 0.729-0.898).

Safety

There were no case who discontinued TDF for significant adverse events during the TDF treatment. Renal function was maintained well in all patients. Also there was no patient with hypophosphatemia during the TDF treatment.