Infection with hepatitis B virus (HBV) can cause severe diseases such as liver cirrhosis and hepatocellular carcinoma (HCC), with approximately one million deaths annually around the world.1 Nucleos(t)ide analogues (NAs), including lamivudine (LAM), adefovir (ADV), telbivudine (LdT), entecavir (ETV), and tenofovir (TDF) were recommended by international guidelines for suppressing HBV replication, and have been shown to decrease the rate of complications.2–4

Despite high drug efficacy for inhibiting HBV, one of the largest obstacles with long-term NAs therapy is the development of viral resistance. LAM, used widely as the first NA, has the highest viral resistance rate of 70% after five years of therapy.5 ADV and LdT are superior to LAM for their lower prevalence of viral resistance, while ETV and TDF effectively suppress HBV DNA replication with minimum drug resistance.6 NAs treatment failure has been linked to mutations at the reverse transcriptase (rt) region of the HBV DNA polymerase gene, and mutations at individual codons in the HBV polymerase gene conferring resistance to different NAs. For example, rtM204V/I pathway is responsible for the resistance to L-NAs, such as LAM, LdT, clevudine (CLD), and ETV. Also, rtN236T pathway could confer ADV and TDF resistance, and rtA181T/V pathway could affect LAM, ADV, LdT, and TDF efficacy.5,7,8

It has been found that natural HBV reverse transcriptase mutations exist even in treatment-naive patients with CHB from Europe, Asia, and the Middle East, but the prevalence vary from 0% to 57%.9–13 This wide range might be caused by different study designs, regions, ethnicities, mutation detection methods, sample sizes, and so on. Using a systematic review and meta-analysis, Zhang et al.13 found an 8.0% prevalence of total and primary mutations among untreated CHB patients in China, and that Southern China had a little higher pooled total mutation rate of 8.22% than Northern China 7.55%. Therefore, it is important to understand the true prevalence of HBV DNA polymerase mutation analysis among treatment-naive patients with CHB in different areas of China, because most of them will be treated with NAs at first. To our knowledge, there has been no study in patients of Central China. The purpose of this study is to determine the prevalence and clinical characteristics of spontaneous mutations in Central Chinese cohort, and focus on clinically significant HBV DNA polymerase associated with NAs resistance mutations (rtM250V, rtN236T, rtM204V/I, rtS202I, rtA181T/V, rtT184G, rtL180M, and rtM/V207I) by using the sensitive INNO-LiPA assay.

The occurrence of viral resistance to NAs is closely related with amino acid substitutions in HBV reverse transcriptase, defined as primary and secondary drug resistance mutations. There are currently eight codons associated with primary drug resistance to NAs: rtI169T, rtA181T/V, rtT184S/C/G/A, rtA194T, rtS202G/C/I, rtM204V/I, rtN236T, and rtM250V; and three codons associated with secondary resistance to NAs: rtL80I, rtV173L and rt180M.7 The primary mutations refer to the amino acid changes that result in reduced susceptibility to an antiviral agent. The secondary mutations cause amino acid substitutions that restore replication function defects resulting from primary resistance and may give rise to a low level of reduced susceptibility.14,15

Currently, mutations associated with HBV drug resistance are reported to occur in treatment-naive patients with CHB for any oral anti-HBV therapy.16–29 This might due to both HBV viral factors and host factors. First, these mutations occur naturally, because HBV has a high replication rate, with an estimated production of about 1012virions/day, and a high mutational rate of 10−5 to 10−4 substitutions/base/cycle. Moreover, the HBV reverse transcriptase does not have a proofreading function to repair incorrectly incorporated nucleotides.30 Second, a treatment-naive patient might be infected by a mutant virus derived from a NAs treated patient. Third, the patient could be unaware of receiving some treatment with direct anti-HBV activity.

This study has confirmed the existence of HBV DNA polymerase resistance mutations for NAs in 24 (8.9%) of 269 treatment-naive patients, composed of five primary drug resistance mutations (rtT184G, rtS202I, rtM204I/V, rtN236T, and rtM250V), one secondary resistance mutation (rtL180M), and one codon mutation that might be baseline polymorphisms without any clinical significance (rtV207I). Among the 24 mutations, rtM204V/I was the prevailing type with a higher prevalence of 4.5% (12/269), compared with other codons with prevalence as low as 2.6% (7/269) of rtL180M, 1.9% (5/269) of rtN236T, 1.1% (3/269) of rtM250V, 0.4% (1/269) of rtT184G, 0.4% (1/269) of rtS202I, and 0.4% (1/269) of rtV207I. This means up to 6.5% treatment-naive patients with CHB would encounter LAM or LdT or ADV or ETV primary treatment failure initially, because rtM204V/I (YMDD/YIDD) and rtL180M linked to LAM and ETV resistance, rtM204I accounts for LdT resistance, and rtN236T could confer ADV resistance. Meanwhile, 2.2% (6/269) are multiple resistance mutations including 1.9% (5/269) rtM204V/I & rtL180M, and 0.4% (1/269) rtM204V/I & rtM250V. These treatment-naive patients with CHB would encounter ETV primary treatment failure initially, because the combination of rtM204V/I, rtL180M and one of the following mutations – rtM184S/C/G/A, rtS202G/I/C, or rtM250V – would lead to ETV resistance.7,8,16

To date, the actual prevalence of HBV DNA polymerase mutations in treatment-naive patients with CHB is unknown, and the reported prevalence varied from 0% to 57%.9,12,16–29 Such a wide range is likely due to differences in study design, study population and geography, study method, study sample size, and whether study patients were treatment-naive. Previous studies on the prevalence of drug resistance mutations in treatment-naive patients often used less sensitive methods or included rather small series of patients.9,19 Now a new sensitive ultra-deep pyrosequencing assay is able to detect minor viral quasispecies at a level under 1%.21 A systematic review and meta-analysis from Zhang et al.13 studied a total of 70 articles including 8156 treatment-naive patients with CHB in China, 21 articles of which were from Northern China and 49 from Southern China. They found that Southern China had a little higher pooled total mutation rate of 8.22% than Northern China (7.55%), as well as higher primary mutation rate of 8.12% versus 7.13%. The HBV variants known to confer resistance against NAs were detected and have a frequency of mutations quite comparable to what we found. In contrast, Nishijima et al.21 reported a relatively high mutation rate of 35.7% in 14 untreated patients with CHB in Japan using ultra-deep pyrosequencing. The difference between their studies and our results could be explained by a smaller sample size as well as the different method they used since ultra-deep pyrosequencing analysis is much more sensitive than INNO LiPA. Contrary to the above reports, several studies from Europe,9 North America19 and Latin America,28 reported that less than 5% primary mutation rate associated to NAs were found in treatment-naive patients. They used direct sequencing assay that could potentially underestimate the true prevalence because of its limited ability to reliably detect viral subspecies below 20%–25% of the circulating viral load.21 Meanwhile, the cohort of patients was predominantly infected with HBV genotypes A, D and C which were not common genotypes in China, especially Central China.

This study has the following advantages. Firstly, to our knowledge, this is the first comprehensive research about resistance mutations of the HBV polymerase in Central China, while all prior studies only focused on a few provinces and municipalities of Northern, Southern, and Eastern China such as Beijing, Shanghai, and Jiangsu.10,17,25,26,29 Secondly, all 269 patients have been required to finish a detailed questionnaire including questions about whether they accepted prior NA therapy, while the majority of other studies collected and analyzed serum for DNA polymerase mutations but did not validate whether the patients received treatment for NAs. Thirdly, we focused on mutations at the reverse transcriptase region of the HBV DNA polymerase gene that has been linked to different NAs treatment failure, while most researches merely focused on rtM204V/I, since LAM had the highest prevalence of resistance and YMDD mutation was extensively studied. Other primary mutations and secondary mutations associated with resistance to NAs did not receive enough attention. Lastly, the INNO-LiPA assay in our study is a sensitive and specific assay that can detect subspecies at 5% of the circulating viral population, while the direct sequencing assay is limited by its lower sensitivity that detects mutants only if they consist of greater than 20%–25% of the circulating viral load.12

In conclusion, the prevalence of HBV DNA polymerase mutations associated with NAs resistance was found to be up to 8.9% of our 269 treatment-naive patients with CHB in Central China. The LAM-associated rtM204V/I mutation occurred more frequently than other primary mutations. In China, LAM, ADV, and LdT are commonly used alone or together, ETV and TDF are seldom used. These findings suggest that HBV DNA polymerase associated with NAs resistance mutation should be analyzed before initiating antiviral therapy such as LAM, ADV, and LdT for the patients with CHB.

The authors declare no conflicts of interest.