Introduction

Hepatitis B virus (HBV) is an important public health problem worldwide, which could be life-threatening by causing liver infection. According to the World Health Organization, approximately 257 million people were infected with HBV, and in 2015, 887,000 people died mostly from complications (WHO 2017). The prevalence of HBV in the world varies according to the geographical regions (from 0.1 to 20%), and Turkey is among the countries with intermediate endemicity (2–8%) for hepatitis B infection (Tozun et al. 2015).

HBV is an enveloped, incomplete double-stranded circular DNA virus, a retrovirus-like replication strategy which depends on reverse transcription, from Hepadnaviridae family (Ezzikouri et al. 2014). After entering the hepatocytes, incomplete double-stranded circular DNA is converted into covalently closed circular DNA (cccDNA) form in the nucleus. Hence, cccDNA, which is not directly targeted by current anti-HBV therapeutics, represents the viral persistence reservoir (Schreiner and Nassal 2017).

Despite improvements in HBV vaccination programs and treatment, it has been reported that HBV-associated liver cirrhosis or deaths due to hepatocellular carcinoma increased worldwide between 1990 and 2013. For this reason, the primary treatment goal for chronic hepatitis B (CHB) is to suppress viral replication and prevent disease progression to prevent HBV-related complications (Shirvani-Dastgerdi et al. 2017). Currently, there are two basic treatment options for patients with CHB, which are nucleos(t)id analogs (NAs) and interferon alfa. The approved NAs for HBV therapy in Europe are lamivudine (LAM), adefovir dipivoxil (ADV), entecavir (ETV), telbivudine (LdT), and tenofovir disoproxil fumarate (TDF). In terms of resistance development, these NAs can be classified as drugs with low (LAM, ADV, LdT) and high genetic barrier (ETV, TDF) (Tacke and Kroy 2016). These analogs inhibit HBV replication by targeting viral polymerase, which catalyzes the reverse transcription of a pregenomic RNA intermediate to form the DNA genome (Ohno et al. 2015). However, mutations in the viral polymerase gene during antiviral therapy make the virus partly or completely resistant to certain antiviral drugs (Hermans et al. 2016).

It is not possible to completely eliminate cccDNA with existing NA, and this is often associated with the development of drug resistance mutations, since it may require long-term treatment (Ohno et al. 2015). In addition, lack of proofreading activity of the viral polymerase results in the appearance of viral mutants (Sayan et al. 2010). HBV variants may be transmitted to patients susceptible to HBV from patients receiving NA therapy (Lok et al. 2017). However, these HBV polymerase gene variants that mediate resistance in HBV are known to cause cross-resistance to other NAs (Hermans et al. 2016).

Early detection of drug resistance mutations during antiviral therapy in patients with CHB will reduce the risk of developing life-threatening complications by ensuring that the most appropriate treatment protocol occurs. It will also help to reduce the risk of toxicity and the economic burden by preventing the use of unnecessary medicines. In this study, it was aimed to evaluate the drug resistance mutations and the rates of drug resistance mutations in patients followed up by the diagnosis of CHB.

Materials and methods

Study population

This retrospective study included 318 patients with CHB who were examined for HBV drug resistance mutations between January 2013 and March 2017 at the Division of Virology and Fundamental Immunology, Department of Medical Microbiology, Istanbul University, Istanbul Faculty of Medicine, Istanbul, Turkey. Patients whose DNA amount was below the detection limit were not evaluated for mutation analysis. Ethical approval for this study was obtained by Ethics Committee of Istanbul University, Istanbul Faculty of Medicine (reference number: 2017/652/11).

HBV DNA extraction

Extraction of HBV DNA was performed using the QIAamp DNA Blood Mini Kit (Qiagen, Germany) according to the manufacturer’s recommended protocol, with 200 μl of serum.

Determination of drug resistance mutations

The A-F domains of the HBV polymerase gene were amplified using the HotStarTaq DNA Polymerase Kit (Qiagen, Germany) containing biotin-labeled primers prior to the hybridization step. The presence of PCR products was confirmed by examining on 2% agarose gel electrophoresis. An INNO-LiPA commercial kit (INNO-LiPA HBV Multi-DR, Fujirebio Europe NV, Belgium) based on the reverse hybridization principle to detect HBV drug resistance mutations was used. The biotinylated DNA material generated from the HBV polymerase gene was hybridized with specific oligonucleotide probes immobilized as parallel lines on the membrane-based strips. After washing, streptavidin labeled with alkaline phosphatase was added and bound to any biotinylated hybrid. The bands formed after incubation with a specific chromogen were evaluated according to the manufacturer’s recommendation.

Statistical analysis

Statistical analysis was performed using the SPSS 21 software (SPSS Inc., Chicago, IL, USA). The normal distribution suitability of the variables was examined by visual methods (histogram and probability plots) and the Kolmogorov-Smirnov test. The mean age between groups was compared by the Student t test. Pearson’s chi-square or Fisher’s tests were used determining whether there were any differences in frequency between groups. p values of less than 0.05 were considered statistically significant.

Results

Out of the total 318 patients with HBV in this study, it consisted of 195 (61.32%) male and 123 (38.68%) female patients. The mean age of the patients was 44.18 ± 16.07 years and ranged from 4 to 90 years. Drug resistance mutations were detected in 46.86% (149/318) of patients with CHB. Of the 149 patients with NA resistance, 96 (64.43%) were males and 53 (35.57%) were females. The mean age of the patients with drug resistance mutations was significantly higher than those with drug sensitive (48.00 ± 16.85; 40.81 ± 14.58, respectively, p < 0.001). Of the 149 patients with drug resistance mutations, 15 (10.07%) were treatment-naive patients and 134 (89.93%) were patients receiving NA therapy. In addition, 82.09% (110/134) of 134 patients received LAM or combined therapy with LAM. The rates of the drug resistance mutations were detected in 36.79% (117/318) for LAM, in 12.58% (40/318) for ETV, and in 7.86% (25/318) for ADV. In 10 patients, the probable TDF resistance (3.14%) was found. There was no significant difference in the distribution ratios according to the gender of the patients with LAM, ADV, ETV, and TDF resistance (p = 0.764, p = 0.475, p = 0.609, and p = 0.567, respectively).

The patients with probable TDF resistance were three females and seven males. Four patients with probable TDF resistance had cirrhosis and one of them was treatment-naive and three patients had received LAM therapy during 1–3 years. Of six non-cirrhotic patients with probable TDF resistance, five patients were treatment-naive and one had received LAM therapy for 6 months. Among patients with probable TDF resistance, single-drug resistance was detected in four patients, double-drug resistance (LAM + TDF, ADV + TDF, ETV + TDF) in three patients, and triple-drug resistance (LAM + ADV + TDF in two patients, LAM + ETV + TDF in one patient) in three patients (Table 1).

Table 1 Distribution of single, double, and triple drug resistance in chronic hepatitis B patients
Full size table

Among all patients with drug resistance mutations, single-drug resistance was detected in 110 patients (34.59%), double-drug resistance in 35 patients (11.01%), and triple-drug resistance in four patients (1.26%) and the distribution of drug resistance mutations rates is shown in Table 1. The most common drug resistance was LAM (25.79%) among single-drug resistance, the LAM + ETV (7.86%) among double-drug resistance, and the LAM + ADV + TDF (0.63%) among triple-drug resistance.

The distribution of drug resistance mutations detected in 149 patients is shown in Table 2. When the distributions of mutations associated with LAM resistance were examined, it was found the M204I mutation in 22.01% (70/318), L180M in 21.07% (67/318), M204V in 17.30% (55/318), and L80V/I in 16.67% (53/318) of the cases. The single mutation was detected in 11.32% (36/318) of the cases, and the most frequent mutation was M204I. The most common mutations associated with LAM resistance were the combination of M204I and L80V/I (16.98%, 54/318), and M204V and L180M (15.09%, 48/318) mutations. In 24 of 25 patients with ADV resistance, A181T (7.55%, 24/318) was a dominant mutation. The N236T mutation was present in only three patients (0.94%, 3/318). ETV-associated mutations found in 40 patients were S202G/C/I mutation in 5.97% (19/318), T184G/S/C/I/L mutation in 5.66% (18/318), and M250V/I/L mutation in 2.83% (9/318) of the cases. Of the 40 patients with ETV resistance, 31 (77.50%) had received LAM therapy in the past. The A194T mutation determining the probable TDF resistance was 3.14% (10/318). Four patients in whom this mutation was found, the A194T mutation was identified alone.

Table 2 Drug resistance mutations
Full size table

Discussion

NAs prevent HBV replication by inhibiting the HBV reverse-transcriptase. Unfortunately, mutations in the HBV reverse-transcriptase gene make the HBV strain resistant to NAs (Chantratita et al. 2016). In general, drug resistance is a major issue associated with the long-term use of NAs, particularly those with low potency or a low genetic barrier (Ahn et al. 2017). Although there is no mutation pattern determining TDF resistance, insufficient responses to TDF in cases have been reported. In vitro, the A194T polymerase mutation is associated with a partial drug resistance to TDF and reduces susceptibility to TDF (Tacke and Kroy 2016).

In a study conducted with 44 CHB patients in Afyon, 34.1% (15/44) of the patients had mutations. Resistance rates were determined as 27.3% for LAM, 4.5% for ADV, and 2.3% for the combination of LAM + ADV (Kalaycı et al. 2010). In another study in Ankara, the resistance-associated mutation was found in 16% (9/56) of the samples by LIPA test. LAM resistance in nine samples and ETV resistance in two samples were detected; ADV and TDF resistances were not detected (Aydogan et al. 2013). In one study in Aydın, LAM resistance was found in 24% (12/50) and, in four patients, the combination of LAM + ADV resistance was detected (Kırdar et al. 2016). The study conducted in Kocaeli included 194 patients who received long-term NA treatment. While drug resistance was detected in 63% (122/194) of the patients, resistance rates of 43% (85/194) in LAM, 8% (16/194) in ADV, and 5% (10/194) in ETV were reported. TDF resistance was not detected (Sayan et al. 2011). In the literature, no study was found to determine the distribution of HBV resistance mutations in CHB patients in Istanbul. This study is the first study conducted in a large patient population in Istanbul, including 318 CHB patients. In a multicenter study involving 1568 patients in Europe, drug resistance mutations in 52.7% of the cases were detected. Resistance rates were 60.1% (584/972) for LAM, 18.6% (11/59) for ADV, and 10% (5/50) for ETV (Hermans et al. 2016). In another study conducted in China between 2010 and 2013, drug resistance rates were 46.5% (569/1223) for LAM, 11.4% for ADV, and 1.7% for ETV (Meng et al. 2017).

Drug resistance mutations were detected in 46.86% (149/318) of patients with CHB in this study. Of the 149 patients in whom drug resistance mutations were detected, 15 (10.07%) were treatment-naive patients and 134 (89.93%) were patients receiving NA therapy. Additionally, 82.09% (110/134) of 134 patients had received LAM or combined therapy with LAM. Resistance rates were 36.79% for LAM, 12.58% for ETV, and 7.86% for ADV, and the probable TDF resistance was 3.14%. Although the second most common antiviral resistance in various studies in Turkey and other countries was ADV, the second highest antiviral resistance in this study was ETV. Cross-resistance among some NA may occur, and the studies have shown that LAM resistance mutations (L180M/M240V) can lead to ETV resistance (Yamada et al. 2017). The high rate of ETV resistance in this study may be explained by the fact that most of these patients (77.50%) had received LAM therapy in the past.

In limited studies in Turkey and other countries, the probable TDF resistance was reported. In France, in each of the two different case studies published in 2009 (Pastor et al. 2009) and 2012 (Dupouey et al. 2012), the probable TDF resistance was found. In addition, the following figures were reported: A194T mutation associated with the probable TDF resistance in one patient in Italy in 2015 (De Francesco et al. 2015), one patient in India in 2016 (Sharma 2016), two patients in China (Jiang et al. 2016), and 13 (20.3%) of 64 patients with CHB in a study carried out in Iran in 2016 (Motahar et al. 2016). In Turkey, Sayan et al. (2010) in 2010 and Ergunay et al. (2013) in 2013 reported the probable TDF resistance mutations in each one treatment-naive patient. In this study, the probable TDF resistance rate was 3.14% (10/318). Four patients with probable TDF resistance had cirrhosis and one of them was treatment-naive and three patients received LAM therapy during 1–3 years. Of six non-cirrhotic patients with probable TDF resistance, five patients were treatment-naive and one received LAM therapy for 6 months. To our knowledge, this was the highest rate detected in Turkey. This rate was higher than the probable TDF resistance rates reported from various countries, but it was lower than the resistance rate reported from our neighboring Iran.

The most common mutation in patients which is not giving response to LAM in hepatitis B treatment is the M204V/I mutation at codon 204 of the reverse transcriptase. Compensatory mutations in LAM-resistant strains can also be seen in V173L and L80V/I (Kırdar et al. 2016). Mutations associated with ADV resistance are N236T and A181V (Kalaycı et al. 2010). The S202G/I, T184S/C, and M250V/I mutations cause ETV resistance and usually require coexistence with LAM resistance mutations (M204V/I ± L180M) (Shirvani-Dastgerdi et al. 2017). The A194T is the mutation associated with the probable TDF resistance (Sayan et al. 2010). In a study conducted in Izmir, resistance mutations were detected in 43 of 172 patients treated with LAM. LAM resistance was detected in all patients in whom the mutation was determined. The rates of the mutations were 8.13% for M204V + L180M, 4.65% for M204I + L180M + L80V, and 4.06% for M204I + L80V (Kose et al. 2015). In Kocaeli, the LAM resistance rate was 43% (84/194), and M204V/I mutation was detected in 26 patients. In addition, 16 L80I/V and 24 L180M mutations accompanying the M204I mutation were detected. Six A181T/V, eight N236T, and two A181T + N236T mutations associated with ADV were detected. Eight T184A/I/S and two S202C mutations associated with ETV were detected (Sayan et al. 2011).

In this study, drug resistance mutations were detected in 46.86% of 318 CHB patients. LAM resistance mutations were identified in 117 patients (36.79%). The most frequent mutations associated with LAM resistance were M204I (22.01%), L180M (21.07%), M204V (17.30%), and L80V/I (16.67%). The mutations in 40 patients (12.58%) with ETV resistance were S202G/C/I, T184G/S/C/I/L, and M250V/I/L. The rates of ADV resistance mutations were 7.55% (24/318) for A181T/V and 0.94% (3/318) for N236T. The A194T mutation associated with the probable TDF resistance was also detected in 10 (3.14%) patients. The mutations rates in this study were different than other studies. The differences in these resistance ratios may be related to the inadequate number of cases in the studies, regional differences, as well as the nature of the drug, the duration of treatment, treatment options in patients with drug resistance, and combination therapies.

The limitation of this study was that sequence analysis could not be performed to confirm the probable TDF resistance because the study was retrospective.

In conclusion, in this study, resistance rates of LAM, ETV, ADV, and TDF were found different than those of the other studies. Although ADV was the second most frequent resistance in various studies, the second most frequent resistance in this study was ETV. Unlike studies in Turkey and other countries, in this study, remarkable resistance for ETV and TDF was found, which have high genetic barriers to resistance. The high rate of the probable TDF resistance was striking, with 3.14%. For the control of hepatitis B, which is still a serious problem in our country, it is of great importance to carry out extensive studies to determine drug resistance mutations in patients with CHB.