INTRODUCTION
Nucleos(t)ide analog (NA) antivirals effectively control chronic hepatitis B (CHB) to prevent or retard the progression of chronic liver disease, portal hypertensive complications and hepatocellular carcinoma (HCC) [
1,
2]. The long-term maintenance of hepatitis B virus (HBV) DNA suppression by NAs is an important factor that can slow the progression of liver diseases [
3-
6]. The long-term use of NAs, particularly less potent and lower genetic barrier drugs such as lamivudine (LMV) and adefovir (ADV), has increased the likelihood of drug resistance in CHB patients in proportion to the duration of treatment, resulting in treatment failure [
1,
2]. CHB is highly prevalent in Korea, and the newer and more potent NAs, entecavir (ETV) and tenofovir (TDF), were unavailable until 2007 and 2012, respectively. Therefore, the occurrence of genotypic resistance due to use of less-potent NAs has been a primary issue in the management of CHB patients. Furthermore, sequential rescue therapies for CHB patients with genotypic resistance can lead to multidrug resistant HBV strains due to sequential selection of mutations that cannot be effectively suppressed by antivirals; this also contributes to the progression of chronic liver diseases [
7,
8].
TDF has the most potent antiviral activity and safety profile among NAs in CHB-naïve patients [
9]. However, few studies describe the efficacy of TDF-based rescue therapy against partial virologic response (PVR) with genotypic resistance developing after long-term LMV, ADV, ETV or combination therapies. It is also unclear whether any form of genotypic resistance is associated with the efficacy of TDF. Thus, we investigated the antiviral efficacy of TDF-based rescue therapy in CHB patients who had newly developed resistance to prior NAs or PVR caused by genotypic resistance, and attempted to identify the viral factors associated with a favorable response.
MATERIALS AND METHODS
Patients and study design
A total of 54 CHB patients with newly developed genotypic resistance to NAs, or PVR to previous antiviral therapies, were enrolled retrospectively at Hallym University Medical Center, Korea, from December 2012 to September 2013. The criteria for enrollment were 1) adherence to previous NA treatment, 2) measurable baseline serum HBV DNA at the time of TDF-based rescue therapy, and 3) genotypic resistance with newly developed genotypic resistance or PVR to the previous therapies. None of the patients were co-infected with hepatitis C virus, human immunodeficiency virus, or hepatitis D virus and they did not have other concomitant liver diseases such as alcoholic liver cirrhosis, autoimmune hepatitis, or HCC. The study protocol was approved by the institutional review boards of each institution and followed the ethical guidelines of the 1975 Declaration of Helsinki.
Definition of partial virologic response
PVR was defined as a decrease in serum HBV DNA > 1 log
10 IU/mL or copies/mL with detectable serum HBV DNA after 6 months of prior antiviral therapy maintenance [
1]. Viral breakthrough was defined as an increase > 1 log
10 IU/mL or copies/mL in HBV DNA from baseline [
2].
Clinical and laboratory assessments
Serum alanine aminotransferase (ALT), albumin, bilirubin, creatinine, creatinine clearance, hepatitis B surface antigen (HBsAg), antibody to HBsAg (anti-HBs), hepatitis B e antigen (HBeAg), antibody to HBeAg (anti-HBe), and serum HBV DNA (lower limit of detection of 20 IU/mL) were measured every 3 months by the investigators. Laboratory data were also obtained from the medical records. Genotypic resistance to antivirals such as LMV, ADV, and ETV was determined using restriction fragment mass polymorphism at initiation of TDF-based rescue therapy.
Study objectives and endpoints
The primary study objective was to investigate the efficacy, measured as viral response (VR), of TDF-based rescue therapy in NA-experienced patients with genotypic resistance. Therefore, the primary end point was defined as serum HBV DNA negativity (< 20 IU/mL) after 3, 6, and 12 months of TDF-based rescue therapy. Secondary endpoints were ALT normalization, HBeAg loss or seroconversion and emergence of genotypic resistance to TDF.
Statistical analyses
Serum HBV DNA levels were logarithmically transformed for analyses. Categorical variables were analyzed using the χ2-test, and continuous variables were analyzed using the Mann-Whitney U-test, as appropriate. The Kaplan-Meier methodology was used to evaluate the cumulative rate of HBV DNA negativity, and subgroup comparisons were performed using the log-rank test. The Cox proportional hazards model was used to evaluate the factors independently associated with VR. A P-value < 0.05 was considered to indicate statistical significance. Statistical analyses were performed using SPSS for Windows software (ver. 21.0; SPSS Inc., Chicago, IL, USA).
DISCUSSION
The present study revealed that TDF-based rescue therapy effectively suppressed HBV, even in CHB patients with PVR caused by genotypic resistance to previously used antiviral drugs. In practice, the cumulative rate of undetectable serum HBV DNA at 12 months reached ~76%, indicating that TDF has similar antiviral efficacy in CHB-naïve patients. Recent studies also demonstrated that TDF-based rescue therapy may effectively suppress viral replication, even in patients with multidrug-resistant HBV or PVR to previous antiviral therapies. Even TDF monotherapy could provide VR comparable to that seen with TDF with ETV combination therapy in those patients [
10-
16].
NAs with lesser potency and a lower genetic barrier may give rise to genotypic resistance, especially with long-term use. Moreover, sequential NAs monotherapy or combination therapies promote multidrug resistance, and consequently demonstrate insufficient suppression of HBV replication. In this study, LMV was administered as a first-line therapy in 52 patients (96%) in whom viral breakthrough and LMV-resistance developed, as expected. In 34 patients (65.4%) treated with ADV alone or ADV combined with LMV as the second-line rescue therapy, HBV suppression was ineffective. However, the replication of LMV-resistant HBVs was suppressed effectively when the rescue therapy was designed on the basis of TDF. Favorable VR was observed in ETV-resistant HBVs even when they were not mutants of rtS202 or rtM250.
A previous in vitro study showed that rtA181T/V or rtN236T may decrease susceptibility to TDF fourfold, but had little effect on ETV [
17]. However, in this study, the patients with rtA181T/V or rtN236T showed a similar VR to those with other genotypic resistance treated with TDF rescue regimens (
p = 0.292). In addition, five of nine patients with rtA181T/V and rtN236T in our study achieved VR, although a previous study reported that CHB patients with these mutations were refractory to TDF monotherapy or TDF with LMV combination therapy [
10]. In particular, two patients with these mutations achieved VR within 9 months of TDF monotherapy, and another four of seven patients achieved VR within 12 months of using TDF and ETV combination therapy. Lim et al. reported that the patients who had a single ADV-resistance mutation or double ADV-resistance mutations (rtA181T/V or/and rtN236T) showed similar VR to each treatment regimen, TDF monotherapy, or TDF with ETV combination therapy [
14]. Therefore, any TDF-based rescue therapy may be effective in patients with HBV at rtA181T/V and/or rtN236T strains, although further evaluation with a large number of cases is warranted.
As expected, the baseline serum HBV DNA level was an independent predictor of VR. Several studies have shown that a lower baseline serum HBV DNA prior to therapy is strongly associated with antiviral efficacy measured by viral suppression, HBeAg loss or seroconversion, serum ALT normalization and histological findings [
18-
22]. Poor VR to antiviral agents is observed primarily in CHB patients with high baseline serum HBV-DNA levels and HBeAg positivity [
23,
24]. In the present study, a serum HBV DNA level < 20,000 IU/mL at pre-treatment could be used as a cut-off for predicting achievement of VR with TDF-based rescue therapy. Negative HBeAg was also an independent factor predicting favorable VR in CHB patients with genotypic resistance. In fact, all HBeAg-negative CHB patients showed undetectable serum HBV-DNA within 6 months. These findings suggest that patients negative for HBeAg, or who had lower baseline serum HBV-DNA levels, may achieve VR more easily than those positive for HBeAg or with higher baseline serum HBV-DNA levels.
One of the most interesting findings of this study is that the cumulative probabilities of achieving VR were significantly lower in patients with the rtS202 mutation than in those without the rtS202 mutation (29% vs. 84% at 12 months,
p = 0.005). Also, two patients with the rtM250 mutation and one patient with mutations of both rtT184 and rtS202G did not achieve VR within 12 months. However, other ETV-resistance strains of HBV without rtS202 or rtM250 did not show a difference in VR for TDF-based rescue therapy. According to a previous report, the rtT184, rtS202, and rtM250 mutations may affect VR for antiviral therapy by impacting the sequence of the overlapping HBV surface antigen gene [
25], although the association of these mutations with VR in TDF-based rescue therapy requires clarification. Interestingly, the patients with mutations of both rtT184 and rtS202 or rtM250 alone did not achieve VR during follow-up. The changes at rtT184 or rtS202 act by repositioning the YMDD loop and affecting the size of the ETV-triphosphate binding pocket; these substitutions then change the stabilizing H-bonding network between rtT184 and rtS202. A previous study found that the isolate containing both rtT184 and rtS202 substitutions displays the highest level of ETV resistance [
25,
26]. Our study also showed that one patient with the HBV mutation at rtT184 and rtS202 did not achieve VR. Substitution of rtM250 is unique in that resistance appears to reside primarily in minus-strand DNA synthesis, suggesting that contacts with the primer-template may be involved in resistance. The experiments showed that mutations of rtM250 and both rtT184 and rtS202 significantly more decreased nearly equivalent manner for ETV susceptibility compared with a single mutation at rtT184 or S202 [
25]. Genotypic analysis of the ETV-resistant HBV revealed that 89% of ETV-resistant HBV had mutations at rtT184, rtS202, or both, with 17% having a combination of the two. Only 11% had mutations at rtM250, alone (5.5%) or in combination with those at rtT184 or rtS202 (5.5%). These findings suggest that a mutation at rtM250 is rarer than that at rtT184 and rtS202 [
27]. Interestingly, a mutation at rtM250 results in phenotypic resistance to ETV [
28]. Although the mutations at rtT184 or rtS202 may affect the YMDD loop position, our results suggest that the patients with a mutation at rtT184 showed favorable VR compared with other patients but the patients with a mutation at rtS202 showed unfavorable VR. This suggests that the existence of mutations at both rtT184 and rtS202, rtS202 alone or rtM250 alone should be considered when VR is predicted in a clinical setting, although the number of patients with these mutations was too small to draw a definite conclusion. Therefore, further long-term follow-up may determine a method of suppressing HBV with rtT184, rtS202, and rtM250 mutations in a large population.
Other prospective clinical trials of TDF and TDF with ETV rescue therapy in ADV-resistant or ETV-resistant patients have shown that 48 weeks of TDF monotherapy provided a comparable rate of VR compared with TDF and ETV combination therapy [
14,
15]. Our study showed similar results to the above trials. TDF monotherapy in particular may be a reasonable treatment option for multidrug resistant CHB patients without mutations at rtS202, rtM250, or both rtS202 and rtT184.
There were several limitations to our study. First, the sample size was too small to conclude that TDF-based rescue therapy is the only option for CHB with PVR with genotypic resistance to prior therapies. However, it should be considered that the inclusion of a large number of such cases is impractical. It was also difficult to enroll more patients due to the shortage of multidrug-resistant CHB patients using TDF as a first-line therapy. Second, baseline genotypic resistance was heterogeneous and TDF-based rescue regimens varied according to the clinicians’ preference. Third, the follow-up period was relatively short; therefore, a study of prolonged duration would be required to investigate long-term efficacy. Finally, the study used a retrospective observational design. Thus, a large-scale prospective study should be conducted in future, although patients who develop new resistance to NAs and have partial response to sequential rescue therapies are relatively rare.
In conclusion, TDF-based rescue therapy appeared to be effective in patients who had newly developed genotypic resistance to prior NAs or PVR to sequential rescue therapies. A serum HBV-DNA level of < 20,000 IU/mL, HBeAg negativity, and S202G non-mutations at baseline may be predictive of VR with TDF-based rescue therapy. TDF monotherapy in particular may be a reasonable treatment option for multidrug resistant CHB patients without mutations at rtS202, rtM250, or at both rtS202 and rtT184.