Association between the basal core promoter mutation and the development of hepatocelluar carcinoma in patients with chronic hepatitis B virus infection

----------------Abstract ] Jt has been suggested that Ihe A to T mulation at nucleotide 1762 and/or lhe G 10 A mutation al nucleotide 1764 (A1762T/G1764A) in Ihe basal core promoler (BCP) in hepalills B virus (HB 이 might be associated 에 Ih Ihe developmenl 01 hepalocellular carcinoma (HCCl. This sludy invesligaled Ihe associa1ion belween Ihe A 1762T/G1764A mulalions 01 HBV and Ihe development 01 HCC by adjusling age ‘ sex. and genolypes 01 HBV. A lolal 01 106 HCC palienls were sludied. Age ， sex ， and genolype-matched palients were assigned in a 2:1:1 ratio (HCC: inactive HBsAg carrier: liver cirrhosisl. The prevalence of A 1762T/G1764A mulalions was higher in LC (94.3%) (P=0.004) and HCC palienls (96.2%) (P (0.001> Ihan inaclive HBsAg carrlers (73.6%). There was no dil1erence in Ihe prevalence 01 A 1762T/G1764A mulalions belween LC and HCC palienls (P=0.69l. In mullivariale analysis ， palienls wilh cirrhosis (odds ra1io rOR1 ， 6.0; 95% c 。깨 idence inlerval rCIl ， 1.6-22.3) and HCC (OR. 9.2; 95% CI ， 2.8-29.5) had a grealer likelihood 01 A 1762T1G 1764A mulalions Ihan in inaclive HBsAg carriers. There was no increased likelihood of A1762T/G1764A mu!ations in HCC patients compared with LC patienls (OR ， 1.7: 95% CI ， 0.4-8.4: P=0.5). These dala suggesl Ihal A 1762T/G1764A mulalions Ihemselves mighl nol be associaled wilh Ihe developmenl 01 HCC ， especially in palienls wilh genolype C. (J Med Life SCi 2009;6:113-118)


Hepatitis B virus (HBV) infection is associated
HBsAg HBeAg and anti-HBe were tested by third generation microparticle enzyme immunoassays using commerciaJ kits (Abbo!t. North Chicago n" USA) HBV DNA levels were measured by U1eDigene Hybrid Capture assay (Digene Corporation Gaithersburg MD USA) The limit of detection of this assay was O 5pg/mL Genotyping 01 H8V and Sequencing Nucleic acids were extl'acted from 200 /.Ù .. J serum that had been s1.oredat 8O't using a Hi 1 Pure Viral Nucleic Acid Kit (Roche Penzberg Gennany). HBV genotype was determined by genotype-specific primers 25 ) To detect the mutations in the BCP and the precore region of the HBV we did bi-directional sequencing with ABl PRISM Bi )Ye Terminator v3.1 Cycle Sequencing Kit (Aplied Biosystems Foster. USA) by automatic sequencing machine (ABI PRISM 3100 Genetic An zer Tokyo Japan) Polymerase chain reaction and direct sequencing were perfonned as described previously We undertook all necessary precautions 10 prevent cross-contamination and distilled water was used as a negative ntrol at each PCR step

Slatislical Analyses
The results were expressed as mean SD (range) or percentage. The differences between categorical variables were 8llalyzed by Fisher s exact test or chi-square test. For the continuous variables. Sωdent's t-test was used. Logistic regresslon 1alysis was used 10 assess the independent predictive factors for A1762T/G1764A mutations. A P value f less than O.05(two-tailed) was considered to be statistically significant

Results
During the study period. 113 consecutive patients were diagnosed having HCC Among them. l06patients (88men and 18 women: the mean age 53.8yr) .vi.th HCC were studied Seven patients with HCC were excluded because of the exclusion criteria. Demographic features of the study patients are presented in Table 1 AIl patients were infected with HBV genotype C. π1e details of the mutation pattern in the BCP region are presented in TabJe 1. The prevalence of A1762T/G1764A mutations was 90.1%. Among these. 1 patient showed deletion from nucleotide 1749 to nucleotide 1771 e A ω T mutation at nucleotide 1762 and the G 1.0 A mutation at nucleotide 1764 in the BCP accompanied in 189 of 192 (98.4%) palients The prevalence of A1762T/G 1764A mutations was higher in patients with LC (94.3%) (vs inactive HBsAg carriers P=O OO4) and HCC (96 2%) (vs inactive HBsAg calTÎer. P(O.OOl) than in inactive HBsAg cameπ (73.6%). However there was no difference in the prevalence of A1762T/G1764A mutations between the patients with LC and HCC (p=O69) In patients with HCC c1inical1iver cirrhosis was accompanied in 86 patienfs (81%) There was no difference in the prevalence of A1762T IG 1764A mutations between the HCC patients with LC (83/86 96 5%) and without LC (19/20 95%) (P=O8) η1e G ω A mutation at nucleotide 1896 (GI896A> was more frequent in patients with LC and HCC than inactive HBsAg caπers. however they did not differ between the paüents with LC and HCC (Table 1)   P1'eviously. a significant correlalion has been found between Iiver damage and the presence of A176'Zr/G 1764A mutations l9l . Live1'cirrhosis surrogate of live1'd nage IS a majo1'risk fac101' in ilie development of HCCll. 1n this study. patie s with c1inical liver cirrhosis ld HCC had a greater likelihood of A1762'l'/G1764A rnutations than in inactive HBsAg carrie1's. However there was no increased likelihood of A1762TIG1764A mutations in patients with HCC compared with LC. suggesting that these mutations mi 1t have no direct 1'ole in the development of HCC. 10stead.
A1762T/G 1764A mutations were associated with the progres! n ω Iiver cirrhosis. indicating thal A1762T/G 1764A mutations may be. in part. indirectly associated with the development of HCC. However A1762T/G1764A mutations we1'e also high even in patients with inactive HBsAg camers. It is speculated that the c1inical outcomes of HBV infection may be deterrnined by multiple mutations aJong tl1e HBV genomes29)01'other viral fac10rs in the development of HCC The correlation between the G1896A mutation and ilie development of HCC was also analyzed. However. there was 00 co1'relation between G1896A mutation and HCC as previously reportecFl Recently. HBV genotypes may cont1'ibute to the development of HCC10. 111 In addition. mutation in lhe surface gene was also 1'eported to be associaled with hepa1ocarcinogenesis 3Cll .
1be1'efore the possibility that other viral mutants contJibuted 10 the development of HCC could not be excluded. The1'efo1'e. to address lhese issues prospective fo!low-up full-length viral genomic st.udies f1'om the clinically silent stage to the development of HCC needed in the futu1'e 1n conclusion although this data showed that A1762T/G1764A mul8tions in the BCP were hi lly prevalent in patients with HCC it was also commonly found in patients with LC and inact.ive HBsAg carriers. Therefore. the BCP mutations themselves might not be directly associated with the development of HCC. especially in patients infected with HBV genotwe C.

Lon
In lhis study AI762T/GI764A mul8üons were detected in most (90.1%) of the study patients. In addition. the prevalence of A1762T/G1764A mutations even in inactive HBsAg c ers was exσ'aordinarily high (73.6%) compared with previous 1'epo1'ts5-7) in which lhe prevalence of A1762T/G1764A mul8tions was observed in 20%-40% in inactive HBsAg camers. However. our data are in agreement with a study from Korea by Yoo et al. 26 >. In U1eir study 78.3% of the study patients were conside1'ed to be inactive HBsAg carriers and 21.7% were c1inical liver cirrhosis A1762T/G1764A mutations were found in 89.9% using direct sequencing in patients Nith HBeAg-negative. HBV DNAnegative by hybridizalion. and normal ALT. In addition. Shindo et a1.27) also repo1'ted that the prevalence of A1762T/G1764A mutations was 97% in patients with HBeAg-negative HBV DNA-negative by hybridization. and nonnal ALT. even though U1ep1'oportionof the patients with liver cirrhosis and inactive HBsAg carriers were not assessed in their study. Taking those in10 account. it is unlikely that the discrepancy between the previous stu ieg5-7 28)and ou1' study is caused by methodologicaJ p1'oblem. A possible explanation fo1' the discrepancy in the p1'evalence of A1762T/G1764A mutations in inactive HBsAg caniers is the diffe1'ent distribution of HBV genotype. because this is an import.ant predictive factor fo1' A1762T/A1764G mutations in U1eBCpI8.19) Previously. some case-cont1'ol studies suggested that A1762T/G1764A mutations had a 1'olein the development of HCC' l 'evious studies are potentially su ected 10 some sampling bias and U1epotenlial confounding effects of age and HBV genotype. In lhe study by Fang et a1. 5 ) as control group HBeAg-positive asymptomatic caniers. who we1'e young and considered to be in a state of immune t.olerant phase in ilienatural course of chronic hepatitis B. were more selected than in HCC patients. 1n the study of Baptista et al.6). U1ere was no mention of the genotype of HBV e age wac; difTerent among the study groups in ilie study of Kao et a1. 7 ) ln addition. Chan et 1I) reported that A1762T/GI764A mutations were not associated with the development of HCC in a prospective longitudinal cohort study. It has been also 1'eported that A1762T/G.l764A mutations were nol associated with the development of HCC in patients with HBV genot;ype C8l. Therefore these observations suggest that A1762T