| BackgroundHepatitis B virus(HBV) infection is a global disease, about 2 billion people worldwide has been infected with HBV with more than 350 million chronic Hepatitis B virus infection around the world. China is an endemic area for HBV infection. The seroepidemiological survey on HBV infection conducted in 2006 showed hepatitis B surface antigen (HBsAg) carrier rate in our country is about 7.18%, and there are about 20 million chronic hepatitis B infection (CHB) patients, these patients progressed to decompensated liver disease, liver cirrhosis and primary liver cancer (Hepatocellular carcinoma, HCC) that may endanger the patients’lives, so HBV is still an important public health problem in our country.For the ability to inhibit HBV replication, Lamivudine was the first nucleoside analogues used for chronic hepatitis B treatment. Long-term LAM treatment can significantly reduce serum HBV DNA level and promote HBeAg seroconversion, reduce liver inflammation activity, delay the progress of liver disease, reduce the occurrence of decompensated liver cirrhosis and liver cancer, reduce the mortality of CHB patients. Due to the safety and tolerability, Lamivudine was widely used for treating chronic hepatitis B in many economic developing counties. However, drug resistance was the main concern during long-term treatment with LAM. Under the treatment of LAM, the incidence of drug resistance in 1 year,2 years,3 years,4 years and 5 years were 22%,38%,53%,66% and 82%, respectively. Drug resistance can result in virological breakthrough or rebound, promote disease progression and increase the risk of liver cirrhosis and liver cancer. Therefore we need to monitor drug resistance during lamivudine treatment and start rescue treatment in time.Rescue treatment after Lamivudine resistance are switching to TDF or add-on ADV suggested by current guidelines. As adding on ADV is more effective than switching to ADV in lamivudine-resistant patients, and adding on ADV can reduce the incidence of ADV resistance, current guideline recommendation for the treatment of lamivudine-resistant patients is switching to or adding on the nucleos(t)ide analogues without cross-resistance. However, with this rescue treatment strategy, patient need to continue treatment for many years and may lead to the occurrence of multiple drug resistance. Pegylated Interferon alfa-2a (Peg-IFNα-2a) can induce a sustained immune response in some patients with limited time, such as continuous HBeAg seroconversion, even HBsAg seroconversion. Thus there are studies suggest Peg-IFNα-2a treatment could be used as a rescue treatment for LAM-resistance patients. To evaluate the efficacy of Peg-IFN a-2a in HBeAg positive patients with lamivudine resistance, we conducted a multicenter, randomized, controlled clinical study in 2005 to 2008. Results showed that the response to peginterferon alfa-2a among patients with lamivudine resistance was suboptimal, compared to naive patients. After the end of treatment, majority of the patients developed virological rebound. Chinese chronic hepatitis B prevention guidelines suggest that CHB patients with baseline high viral load can initial treatment with combination of two NUCs without cross-resistance or NUCs with high resistance barrier.Treatment efficacy of peginterferon alfa-2afor LAM-resistant patient is not satisfactory. The reason of the suboptimal response is unclear, maybe affected by two main factors, the host’s immune and viruses which interact with each other. Hepatitis B virus exists in patients as a form of quasispecies, HBV quasispecies evolution with different viral replication capacities and under host immune pressure. Under treatment of nucleoside analogues, quasispecies evolution may be mainly decided by the choice of drugs and viral adaptability. However, the treatment of peginterferon alfa-2a does not select the virus resistance mutations, therefore evolution of HBV quasispecies is mainly decided by the host immune pressure and viral adaptability. Therefore dynamic change and evolution of LAM-resistance under the treatment of Pegylated Interferon alfa-2a may reveal the relation between host’s immune and viral factors. The dynamic LAM-resistance change with treatment of peginterferon has not been reported before.Pyrosequencing has been developed as a method to detect HBV drug resistance mutation due to the characteristics of high sensitivity, accuracy and high throughput in recent years. Our research use commercial kit designing based on Pyrosequencing, which can detect ten common HBV drug resistance mutations related sites (rt169 rt173 rt180, rt181, rtl84, rt194, rt202, rt204 rt236 and rt250).Therefore, first we evaluate the quality and clinical applicability of Pyrosequencing assay kit for detecting Hepatitis B virus resistance. Then we use Pyrosequencing to quantitatively detect dynamic change of LAM-resistance mutations under the treatments of peginterferon alfa-2a to investigate the relationship between the dynamic change of LAM-resistance mutation and the response.Subjects and methods1. Subjects 1.1 Samples for pyrosequencing kit validationOne hundred and two CHB outpatients or hospitalized patients were selected from Hepatology Unit, Nanfang Hospital, Guangzhou, from January 2011 to December 2011. CHB diagnosis standard was based on 2005 Chinese chronic hepatitis B viral prevention guidelines. Patients should receive NA treatment for more than 6 months. Besides, they should suffer virological breakthrough (increase of HBV DNA over one log from nadir) or biochemical breakthrough during treatment, or after one year of antiviral treatment their HBV DNA levels were still higher than 1000 copies/ml or even higher. All samples had finished HBV drug resistance detection by Sanger sequencing and HBV RT gene seuencing.Pyrosequencing kit was using negative reference from standard samples for quantitative HBV DNA from National Institutes for Food and Drug control. PCR lower limit of detection was validated by the world health organization (WHO) HBV DNA quantitative reference standard.Plasmids containing RT gene of the hepatitis B virus (one wild type and 10 with different mutant sites) were used for preparing series of samples with different mutation ratios.1.2 Samples for investigating relationship between the dynamics ofLAM-resistance mutation and responsePatients in the present study were from a randomized, open-label study, which was carried out in 13 centers across China (including Hong Kong) from 2005 to 2008. A total of 255 patients were randomized to receive peginterferon alfa-2a treatment (171 patients) or adefovir treatment (84 patients). At last 235 patients met the inclusion criteria in that study. Because of the volume of serum, finally our study enrolled 123 peginterferon alfa-2a treatment patients.2. Experiment tests2.1 HBV DNA ExtractionHBV DNA was extracted from 200μL serum samples by QIAamp UltraSens Virus Kit following manufacturer’s protocol with necessary measures to prevent from cross-contamination.2.2 PyrosequencingPyrosequencing commercial kit was provided by Qiagen, which allowed us to detect 10 common HBV mutations (rt169, rt173,rt180, rt181,rt184, rt194, rt202, rt204, rt236 and rt250). Pyrosequencing workstation was PyroMark(?) Q24 MDx (QIAGEN, Germany); all procedure was following manufacturer’s protocol.3. Analysis of sequencing dataAnalysis software was supported by pyrosequencing workstation. First, single nucleotide polymorphism (SNP) analysis was performed to determine whether the mutation existed. If the mutation-site was heterozygous mutations, the proportion of mutant virus would be analyzed, and the proportion of real mutation ratio was calculated through linear relationship.4. Validation of pyrosequecing commercial kit4.1 limit of detection of PCRGradient dilution series sample of WHO hepatitis B virus nucleic acid quantitative standard reference were used for validation of limit of detection of PCR in the kit.4.2 Linear range and accuracy evaluationPlasmids containing RT gene of the hepatitis B virus (one wild type and 10 with different mutantion sites) were used for preparing series of samples with different mutation ratios. Every mutation ratio mixes were tested at least 10 times for evaluating the accuracy of the kit and establishing a mutation rate linear relationship between detection values and theoretical values.4.3 Comparison between pyrosequencing and Sanger sequencingOne hundred and two clinical samples which have been sequenced and analyzed by Sanger sequencing were retested by pyrosequencing to determine the concordance of these two methods.5. Laboratory tests5.1 Serum HBV markersHBsAg, antibody against HBsAg (HBsAb), HBeAg, antibody against HBeAg (HBeAb) were determined using a commercially available radioimmunoassay (Abbott Laboratories).5.2 Serum HBV DNA levelsSerum HBV DNA levels were measured using a real-time PCR quantification, the Cobas Ampliprep/Cobas TaqMan, version 2.0 (CAP/CTM, Roche Molecular Systems, Inc, Pleasanton, CA, USA).5.3 ALTSerum ALT levels were measured by automated techniques. Upper limit of normal (ULN) was 40 IU/ml6. Statistical analysisData analyses were performed by SPSS for Windows, version 19.0. Continuous variables were expressed as mean and standard deviation (SD) or median (range); categorical variables were expressed as percentages. HBV DNA levels and HBsAg levels were expressed in logarithmic units (log IU/mL).χ2 test and t test were applied when appropriate to determine whether the results were statistically different. Area under the receiver operator characteristic curve (AUROC), sensitivity, specificity positive predictive value (PPV), negative predictive value (NPV), and several cut-off value in predicting treatment outcome were calculated. The statistical significance of all tests was set as P<0.05 by 2-tailed tests.Results1. Validation of pyrosequecing commercial kitThe lower detection limit of HBV DRT was 50 IU/mL. Except for RT236, correlation between true mutation rate and detection mutation rate in the other 9 resistance-related mutation sites were very high with R2>0.98. In detection of 102 clinical samples,4 samples were not amplified successfully by PCR. Concordance between HBV DRT and Sanger sequencing results was observed for 897/969(92.6%) amino acid positions. Concordant results were achieved in 46/98(46.9%) samples at all 10 mutation sites. For a single mutation site, concordance rates differed from 71.5% to 100% at 10 mutation sites respectively. Analysis of discordant samples showed that in 87.5%(63/72) samples, Sanger sequencing detected wild (WT) while HBV DRT detected wild/mutation (WT/MT). In 5.6%(4/72) samples, Sanger sequencing detected wild/mutation while HBV DRT detected wild. In the remaining 6.9%(5/72) samples, Sanger sequencing detected WT but HBV DRT detected MT.2. Relationship between the dynamic of LAM-resistance mutation and response2.1 Result of pyrosequencingA total of 123 peginterferon alfa-2a treatment patients were included in the analysis. Pyrosequencing were performed for the sample of all the 123 patients at week 24. Success rate of pyrosequencing was 91%(112/123). The success rate in 40 patients for dynamic investigation at baseline, week 24, week 48 and week 72 were 100%(40/40),82.5%(33/40),70%(28/40),87.5%(35/40), respectively.2.2 Baseline clinic characteristicA total of 123 peginterferon alfa-2a treatment patients were included in our study. Among them,101 were male and 22 female, whose median age was 33 years (range, 19 to 63 years). The median serum HBV DNA level was 7.15±1.2 log10 IU/mL, and the median ALT level was 72 U/L (range,9 to 1776 U/L). Thirty four patients were genotype B and others were genotype C.2.3 Baseline mutation characteristicsPyrosequencing were performed for the sample of 40 patients at baseline. The rates of patients who developed mutation of M204V/I, L180M, A181V or V173L were 100%(40/40),82.5%(33/40),60%(24/40) and 10%(4/40), respectively. The median mutation ratio of these four mutation sites were 95.8%(range,29.6% to 100%), 29.6%(range,0% to 97.1%),0%(range,0% to 16.8%) and 0%(range,0% to 97%).2.4 Dynamic of mutation virusIn complete response group (CR), the mean mutation rates of rtM204V/I at baseline, week 24, week 48, week72 were 91.3±19.4%,19.9±27.1%,9.1±26.4% and 15.8±27.4%, respectively. In non-response group (NR), the mean mutation rates of rtM204V/I at baseline, week 24, week 48, week72 were 90±11.3%,47.2±32.2%, 31.2±29.4% and 21.3±29.9%, respectively.In complete response group(CR), the mean mutation rates of rtL180Mat baseline, week 24, week 48, week 72 were 50±44.6%,11.2±27.1%,13.6±23.9% and 13.4±25.5%, respectively. In non-response group (NR), the mean mutation rates of rtL180M at baseline, week 24, week 48, week 72 were 69.4±32.4%,31.2±29.6%, 11.66±29.7% and 9.2±18.9%, respectively.In complete response group (CR), the mean mutation rates of rtL181M at baseline, week 24, week 48, and week72 were 5.8±5.9%,1.3±1.3%,1.0±2.1% and 2.3±2.2%, respectively. In non-response group (NR), the mean mutation rates of rtA181M at baseline, week 24, week 48, week 72 were 6.0±5.2%,4.6±2.8%, 4.5±3.7% and 5.3±3.2%, respectively.2.5 Relationship between baseline mutation rates and treatment outcomeThere were no significant differences between two groups in the mutation rate of rtM204V/I (91.3±19.4% vs.90±11.3%, P=0.132), rtL180M (50±44.6% vs69.4±32.4%,P=0.307), rtA181V (5.8±5.9% vs.4.6±2.8%, P=0.311) at baseline.2.6 Relationship between week 24 mutation rates and treatment outcomeThere were significant differences between two groups in the mutation rate of rtM204V/I, rtL 180M and rtA 181V at week 24.In the 40 patients, the AUROC of HBV DNA, HBsAg, M204V/I, L180M and A181V at week 24 were 0.866,0.872,0.771,0.781 and 0.862, respectively.In the 123 patients, the AUROC of HBV DNA, HBsAg, M204V/I, L180M and Al 81V at week 24 were 0.809,0.833,0.672,0.759 and 0.764, respectively.2.7 Best cut-off value for predictionAt week 24, the best cut-off value for predicting the treatment outcome of DNA, HBsAg, and the mutation rates of M204V, L180M, A181V were 6 log10 IU/mL,2 log10 IU/mL,20%,5% and 5%, respectively.Conclusion1. Pyrosequecing showed high sensitivity and accuracy in detecting antiviral drug-resistant mutations. The method is superior to Sanger sequencing in detecting minor mutation and can be used for early detection of resistance mutation.2. Based on the data from the study, there were no differences between CR and NR groups in virus mutation rates of M204V/I, L180M, A181V at baseline. Hence, the baseline virus resistant mutation rates cannot predict treatment outcome.3. At week 24, lower mutation rates of M204V/I, L180M, A181V can predict treatment outcome of Peg-IFN as rescue therapy for LAM-resistant patients. In all 123 patients, the AUROC of M204V/I, L180M and A181V for predicting treatment outcome was 0.672,0.759 and 0.764, respectively. However, the strongest predictors were HBV DNA levels and HBsAg levels.4. At week 24, the mutation rates combined with HBV DNA levels and HBsAg levels can increase the power of prediction. When the HBV DNA< 6 log10 IU/mL, HBsAg< 2 log10 IU/mL and rtA181V< 5%, the sensitivity, specificity, PPV and NPV for predicting complete response at week 72 were 84.62%,81.48%,68.75%,91.67%, respectively. It suggests quantitative detection of mutation rates of LAM-resistant virus can be used for predicting treatment outcome of Peg-IFN in LAM-resistant patients. |
PDF Full Text Request