Clinical Research On Gene Variations Of HBV RT/S Region Based On Next-generation Sequencing

Hepatitis B Virus(HBV)infection is a global health problem,which affecting approximately 400 million people worldwide and result in almost 1 million deaths each year due to decompensated cirrhosis or hepatocellular carcinoma(HCC).HCC is the sixth most common cancers as well as the third most common cause of cancer-related death worldwide.HBV infection is particularly prevalent in China and Southeast Asia,where almost 85% of the HCC patients with positive hepatitis B virus surface antigen(HBsAg).It is estimated that there are about 930 million hepatitis B carriers in China,including about 20 million cases of chronic hepatitis B(CHB)patients.The HBV gene sequence is about 3.2kb in total length which is partially doublestranded circular DNA.It is composed of four overlapping open reading frames(ORF): P,S,C,X,which encoding virus reverse transcriptase,virus surface antigen(L-HBsAg,M-HBsAg,S-HBsAg),regulatory protein(Hepatitis B X antigen,HBx)and viral core antigen(Hepatitis B core antige,HBcAg),respectively.The P region mainly contains terminal protein,revers transcriptase(reverse transcriptase,RT),noncoding spacer region and H region of RNA enzyme.Due to the lack of proofreading activity of the polymerase in the process of HBV replication,the sequence of HBV gene varies greatly.Therefore,HBV genome has a high rate of variation and exists in complex quasispecies in patients.Previous studies were mainly based on the Sanger sequencing which is limited by the number of sequence results and can only analysis HBV quasispecies limitedly.This making it difficult to delineate and explore the gene mutation and quasispecies more comprehensively.However,next generation sequencing(NGS)has be used in the clinical application.Compared with the Sanger sequencing,it has the advantages of high-throughput and high-sensitivity.By using it we can maximize the sampling all quasispecies of HBV,so as to obtain the closest real HBV quasispecies and mutation sequences within the patients.This project uses the intrinsic advantages of nextgeneration sequencing to perform ultra-deep sequencing of HBV gene sequences,combining with bioinformatics methods to explore the impact of HBV mutation levels on antiviral drugs,on hepatitis B surface antigen-antibody binding reactions and to assess its use as a differentiation from CHB and HCC patients.Part Ⅰ: Characteristic of HBV-RT region in hepatocellular carcinoma patients based on next generation sequencingThe aim of this project was to characterize the mutation in HBV RT gene of HCC patients.According to the changes of HBV DNA level in patients after a long followup,we divided the patients into thress groups: virology breakthrough group,partial virology response group and virology response group,respectively.There are statistically significant difference in six mutations: L180 M,T184A,S202 G,M204V/I,N238 H amongt the three groups.Among them,N238 H is potential new drug resistance mutation,their average mutation percentage in virology breakthrough group is 23.801% respectively.Besides,their average mutation percentage in partial virology response group are 0.021% and 1.041%.By statistical analyzing the drug resistance mutation of 77 HBV related HCC patients,we observed that the percentage of M204V/I and L180M+M204V/I substitution account for the highest,followed by A181T/V,their mutation rates are 12.99%,16.88% and 11.69%,respectively.Furthermore,we divided 55 reported drug resistance mutation into four categories based on literature,namely,primary drug resistance,compensatory mutation,putative NA resistatance mutation,and pretreatment mutation.The mutation rates on these three groups were calculated and the differences were compared respectively.We also found that the next generation sequencing is effective and could provide doctors with guidance for making antiviral therapy decision.Part Ⅱ: Mutations in the RT/S gene of hepatitis B virus-related hepatocellular carcinoma patients with coexistence of HBsAg and anti-HBs antibodiesThis study mainly analyzed the characteristics of clinical and HBV S sequence of HCC patients with with coexistence of HBsAg and anti-HBs antibodies.We also explored whether the different mutations in amino acids of the RT and S region between the HBsAg+/HBsAb+ and HBsAg+/HBsAb-group could be used as a distinguishing feature.By clustering the mutation characteristics of all the amino acid positions in the RT region and the S region of the two groups of patients,we found that the characteristic of amino acid mutations in the RT region could not be used to distinguish these two group absolutely.However,the amino acid mutations in the S region could distinguish the two groups completely.This shows that the mutation characteristics of amino acids in the S regions are quite different in two groups.In the end,we found that there were 25 mutation of S region are significantly different in the two groups of patients,of which the top 10 are: S204 R,F220C,F219 S,V224A,I195 M,W196L,T131 N,M133S,T47 K,V190A.In this study,we get the mutation map of the HBV S region by next generation sequencing and revealed the association between it and HBsAg immune escape.Part Ⅲ: Detection of hepatocellular carcinoma based on next generation sequencing of hepatitis B virus RT/S geneIn this study,we used the next generation sequencing technology to sequence the HBV RT region of the HBV sequence of 107 HCC patients and 120 chronic HBV(CHB)infected individuals.Besides,there are 77 HCC and 76 CHB patients belong to HBV C genotype.The differences between CHB and HCC patients in HBV RT gene were compared based on their nucleotide and RT/S region amino acid substitutions.Further more,the mutation pattern were also used to build prediction models for HCC status using both K-nearest geighbors(KNN)and support vector machine(SVM).It was observed that the mutation of HCC are higher than CHB on many nucleotide sites.By clustering of all the nucleotide mutations in the two groups,it was founthat the two groups of patients could be separated perfectly.In addition,in the RT and S region amino acids,45 and 49 mutation with statistical differences were found between the two groups.We found that KNN are suitable for the characteristic of entropy to distinguish HCC and CHB patients better.The AUC of KNN model based on the entropy of HBV RT1 can reach 0.907 in the independent validation cohort.The AUC of KNN model based on the entropy of HBV RT2 can reach 0.915 in the independent validation cohort.The combination of HBV RT1 and HBV RT2 is 0.926.Our study shows clearly that mutation frequencies of HBV sequences contain much information about the composition of different HBV genotypes.Quantitative NGS data combined with a machine learning method could be a powerful approach for prediction of the status of different diseases.Furthermore,the characteristic of mutations in HBV RT region based on the next sequencing may be used as a new "marker" to distinguish HCC and CHB patients,which opens a new venue for the study of early diagnosis of liver cancer.Part Ⅳ: Correlation between hepatitis B core-related antigen and HBV viral load in chronic hepatitis B and hepatocellular carcinoma patientsHepatitis B core-related antigen has been revealed as an important marker of HBV infection recently.We aimed to evaluate the HBcrAg assay for indication of HBV loads in CHB and HCC patients.HBcrAg correlated positively with HBV DNA irrespective of HBeAg status.Both HBcrAg and HBV DNA were associated with cccDNA in patients with elevated serum HBV DNA(>4 log IU/ml).In patients with non-elevated HBV DNA(≤4 log IU/ml),no relationship between HBV DNA and cccDNA was observed,but we still documented a modest correlation between HBcrAg and cccDNA.In addition,a decreased correlation between HBcrAg and cccDNA in HCC tumorigenesis was documented.Finally,a moderate correlation occurred between HBcrAg and albumin,alanine transaminase,aspartate transaminase in both CHB and HCC patients.Our data show that HBcrAg can serve as a biomarker to assess HBV loads in patients and provide a better method for monitoring cccDNA in HCC patients than HBV DNA.