A

BoM(wRChrCd©

K

1

spbs2i

CRT

1d

Journal of Clinical Virology 40 (2007) 312–317

Frequency and significance of hepatitis B virus surface gene variantcirculating among ‘antiHBc only’ individuals in Eastern India

Arup Banerjee a,1, Partha K. Chandra a,1, Sibnarayan Datta a, Avik Biswas a, PrasunBhattacharya b, Subhasis Chakraborty b, Sekhar Chakrabarti a,c,

Sujit Kumar Bhattacharya a,c, Runu Chakravarty a,∗a ICMR Virus Unit, Kolkata, ID & BG Hospital Campus, Kolkata, India

b Institute of Blood Transfusion Medicine and Immunohematology, Kolkata, Indiac National Institute of Cholera and Enteric Diseases, Kolkata, India

Received 14 February 2007; received in revised form 9 August 2007; accepted 21 August 2007

bstract

ackground: Genetic mutation might account for the presence of hepatitis B virus (HBV) DNA among antiHBc only individuals. The aimf the study was to assess the prevalence and significance of surface gene mutations among antiHBc only cases in our population.ethods: Three hundred and three antiHBc(+) sera of adults (mean age, 33.7 ± 11.0; range 18–65 years) as well as HBsAg(+)/HBV DNA(+)

n = 19) controls were included in this study. Surface gene and basal core promoter (BCP)-precore region were amplified and surface geneas analyzed after direct sequencing.esults: One hundred and seventy-eight out of 303 (58.8%) was antiHBc only, 39/171 (22.8%) of them was HBV DNA(+). Genotypes A,, D were found among both HBsAg(+) and antiHBc(+) samples. Single or multiple amino acids substitutions were found in 82% samples,owever, G145R vaccine escape mutation was rare. Individuals having substitutions within as well as outside major hydrophilic loop (MHL)

egion were detected; some of these mutations were in overlapping RT domain of polymerase (Pol) gene.onclusions: The existence of occult HBV infection among antiHBc only individuals could not be explained fully by mutations in the ‘a’eterminant region of surface gene in our population.

2007 Elsevier B.V. All rights reserved.

rface ge

2i1ee

c

eywords: Occult HBV infection; AntiHBc only in general population; Su

. Introduction

Occult hepatitis B virus (HBV) infection, defined by per-istent HBV viremia in surface antigen negative, HBsAg(−)atients with or without markers of previous infection (anti-ody to core antigen; antiHBc), is an emerging problem inafety of blood transfusion in Asian countries (Liu et al.,

006). Recently, there has been concern about antiHBc onlyndividuals (Grob et al., 2000; Jilg et al., 2001; Weber et al.,

∗ Corresponding author at: ICMR Virus Unit, Kolkata, ID & BG Hospitalampus, GB-4, 1st Floor (East Wing), 57 Dr. Suresh Chandra Banerjeeoad, P.O. Beliaghata, 700010 Kolkata, West Bengal, India.el.: +91 33 2353 7424; fax: +91 33 2353 7425.

E-mail address: runugc@hotmail.com (R. Chakravarty).1 These authors contributed equally to this work.

io‘bf(els

386-6532/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.jcv.2007.08.009

ne variant

001) in which antiHBc is the only detectable HBV markern the absence of HBsAg or antibody to HBsAg (antiHBs),0–40% of these cases have persistent HBV DNA (Alhababit al., 2003; Jilg et al., 1995; Kroes et al., 1991; Weinbergert al., 1997).

Our recent study from Eastern India documented signifi-ant prevalence of antiHBc marker (Chowdhury et al., 2005)n our population, thus leaving open the important questionn the prevalence of occult HBV infection, especially amongantiHBc only’ cases in this population. Previous studiesy our group provided a potential molecular explanationor the presence of HBV DNA in HBsAg(−) individuals

Chakravarty et al., 2002; Datta et al., 2006a,b). Alhababit al. (2003) suggested that in communities where the circu-ation of antigenic determinant (‘a’) mutants are increasing,urveillance of antiHBc only may be a valuable tool to iden-

Clinical Virology 40 (2007) 312–317 313

tmlHt

2

2

iivwr

2

aNFbH(tostiaawp

2a

ael1fsPUl(d

2

r

Table 1Demographic, serological and virological characteristics of 303 HBsAgnegative and antiHBc positive subjects

Features Total

N 303Age (years) [mean ± S.D.] 33.7 ± 11.0Male/female 239/64

Serological testHBsAg(−)/antiHBc(+) 303HBsAg(−)/antiHBs(+) and antiHBc(+) 125AntiHBc only 178

PCR positive (%) 60/289 (20.7)Surface region only (%) 26/60 (43.4)Surface and BCP-precore (%) 34/60 (56.6)

Viral load measured 25/60

HBV genotype by sequencing (%)A 4/50 (8.0)

1ta

3

3

iiHaaHWfi(iiplp(

3

cpredominant followed by HBV/C (18.0%) and HBV/A

A. Banerjee et al. / Journal of

ify the mutants. Reports on prevalence of antiHBc only as aarker of HBV infection in general population in India are

imited. We, therefore, determined the prevalence of occultBV infection among antiHBc only population and assessed

he significance of surface gene mutations in our population.

. Materials and methods

.1. Study subjects

From January 2001 to December 2005, 1294 HBsAg(−)ndividuals, registered in our lab for antiHBc testing werencluded in this study. The majority of the samples were fromoluntary blood donors from both urban and rural areas asell as from the family members of HBsAg positive donors

eferred to our unit for screening for HBV infection.

.2. Serological analysis

HBsAg (Organon Teknika, Boxtel, The Netherlands),ntiHBs (Hepanostika anti-HBs, Biomerieux, Boxtel, Theetherlands) and antiHBc (Hepanostika anti-HBc Uni-orm, Biomerieux, Boxtel, The Netherlands) were testedy commercial ELISA. Samples positive for either anti-CV (Ortho-Clinical Diagnostics, NJ, USA) and/or antiHIV

Biomerieux, Boxtel, The Netherlands) were excluded fromhe study. Using a stringent sample to cutoff (S/CO) valuef <0.3 as the criterion of a positive antiHBc result, 303era of adults (mean age, 33.7 ± 11.0; range 18–65 years)ested positive for antiHBc in two different occasions, werencluded in this study. Nineteen HBsAg(+) individuals (meange, 31.0 ± 9.5; range 18–60 years) were randomly selecteds control to obtain a consensus sequence. Informed consentas obtained from participants. The study was a part of arogram approved by Institutional ethical committee.

.3. Serum HBV DNA isolation, detection, sequencingnd quantification

HBV DNA isolation and detection were done from surfacend basal core promoter (BCP)-precore region as describedarlier (Banerjee et al., 2005; Datta et al., 2006a). Guide-ines for avoiding false positive results (Kwok and Higuchi,991) were followed strictly. Samples repeatedly positiveor HBV DNA by surface gene amplification, were directlyequenced using BigDye Terminator mix and a model ABIRISM 3100 sequencer (Applied Biosystems, Foster City,SA) as described previously (Datta et al., 2006a). Viral

oad was measured by using the real time Taqman PCR assayApplied Biosystems SDS7000, Foster City, USA) with loweretection limit 102 copies by the method of Abe et al. (1999).

.4. Sequence analysis

The amino acids (aa 49–172) of the immunodominantegion of the 50 HBsAg(−) isolates were compared with the

(A(s

C 9/50 (18.0)D 37/50 (74.0)

9 HBsAg(+) subjects. Genotypes/subgenotypes were ascer-ained as described by Norder et al. (2004) and Banerjee etl. (2006).

. Results

.1. Prevalence of occult HBV infection

The demographic, serological and virological character-stics of 303 HBsAg(−) and antiHBc(+) subjects are shownn Table 1. Out of total 303 samples, 125 (41.2%) were anti-Bs(+) and 178 (58.8%) were antiHBc only. Two hundred

nd eighty-nine out of 303 (95.0%) samples were avail-ble for HBV DNA analysis of which 60/289 (20.7%) wereBV DNA(+) by nested PCR from surface gene region.hen these samples were further subjected to PCR ampli-

cation using primers from BCP-precore region, only 34/6056.7%) was successfully amplified. However, no serolog-cal as well as virological differences were found amongndividuals who were positive for both surface and BCP-recore genes or were positive only for surface gene. Viraload could be measured in 25/60 (41.7%) HBV DNA(+) sam-les and only five of them had viral load above 104 copies/mlTable 2).

.2. Genotype/subgenotype distribution

Three genotypes (HBV/A, HBV/C, HBV/D) were cir-ulating in our study population. HBV/D (74.0%) was

8.0%; Table 1). Although one subgenotype of genotype(HBV/Aa) and genotype C (HBV/Cs) was mainly found

Table 2), HBV/D was highly divergent and distributed in fourubgenotypes, HBV/D1, HBV/D2, HBV/D3, and HBV/D5.

314 A. Banerjee et al. / Journal of Clinical Virology 40 (2007) 312–317

Table 2Characteristics of 60 hepatitis B virus DNA positive HBsAg(−) and 19 HBsAg(+) subjects

HBsAg(−)/antiHBc(+) (n = 289)a Total HBsAg(+) (n = 19)

AntiHBs(+) (n = 118) AntiHBc only (n = 171)

HBV DNA positive (%) 21 (17.8) 39 (22.8) 60 (20.7) 19 (100)

Age (years)Mean ± S.D. 32.1 ± 8.9 33.9 ± 9.6 33.3 ± 9.3 31.0 ± 9.5

GenderMale 16 30 46 16Female 5 9 14 3

Surface gene sequencing (n) 16/21 34/39 50/60 19/19

Viral load (n) 10/16 15/34 25/50 ND<104 copies/ml 7 13 20≥104 copies/ml 3 2 5

Genotype/subgenotype (%)A/Aa 1 (6.2) 3 (8.8) 4 (8.0) 2 (10.5)C/Cs 2 (12.5) 7 (20.6) 9 (18.0) 5 (26.3)D/D1 (0) 1 (3.0) 1 (2.0) 4 (21.0)D/D2 3 (18.7) 2 (5.8) 5 (10.0) 5 (26.3)D/D3 7 (43.7) 9 (26.4) 16 (32.0) 1 (5.3)D/D5 3 (18.7) 12 (16.6) 15 (30.0) 2 (10.5)

Nfor HB

3

(itsi

TV

S

H

1111111111

mt(

D: not done.a Out of 303 HBsAg(−)/antiHBc(+) subjects 289 samples were available

.3. Amino acids substitutions in surface gene

Tables 3 and 4 represent the variability of surface geneaa 49–168) in 19 HBsAg(+) and 50 HBsAg(−)/antiHBc(+)

ndividuals. Among HBsAg(−) samples, single or mul-iple amino acids substitutions were found in 41 of 50amples (82.0%). No significant differences were observedn the mean age of anti-HBc only subjects with single

able 3irological characteristics of 19 HBsAg(+) subjects

I. no. Age Sex Genotype/subgenotype

Subtype Mutations in Sgene

BsAg(+) (n = 19)1 31 M A/Aa adw2 Wild type2 33 M A/Aa adw2 P62L3 44 M C/Cs ND R122Q4 24 M C/Cs adr F80S5 38 M D/D1 ayw2 Wild type6 24 M D/D1 ayr L88Q7 28 M D/D1 ayw2 T118V, A128V8 27 F D/D2 ayw3 T131A9 28 M D/D2 ayw3 T118V, A128V,

T131A0 30 M D/D5 ayw3 T125M1 18 M C/Cs adw2 Wild type2 28 M D/D5 ayw3 T68A, T125M3 50 F D/D1 ayw2 Wild type4 22 M D/D3 ayw3 T125M5 14 F D/D2 ayw3 T118V6 35 F D/D2 ayw3 T118V7 27 M D/D2 ayw3 T118A8 24 M C/Cs adr Wild type9 53 F C/Cs adr Wild type

ds7tCfachpHAHc(

Haia

snosig

V DNA testing.

utation (34.7 ± 8.9), with more than one amino acid substi-utions (31.9 ± 10.3) or even with HBsAg positive subjects31.0 ± 9.5).

All together 26 different amino acid substitutions wereetected in 41 samples. The most common amino acid sub-titution found within MHL region was T125M (30/41;3.0%), common in HBV/D3 and HBV/D5. Other substitu-ions at codon Q129N/R, G130R, T131A, M133T, Y134H/F,137G/Y, and S143T were also found (Table 4). Moreover,

ew novel mutations, not reported previously, were detectedmong HBsAg(−) samples. Although amino acid T wasommon at codon 118 in HBV/A and HBV/C, HBV/D waseterogeneous at this position. Most of the HBsAg(−) sam-les (33/37; 89.1%) had T at 118, whereas only 7/12 (58.3%)BsAg(+) samples had this mutation and the rest had eitheror V at codon 118. T125M mutation was found in both

BsAg(+) and HBsAg(−) HBV/D samples but was signifi-antly higher in HBsAg(−) samples [3/12 (25%) versus 30/3781.0%), P = 0.02] along with T118.

Subtype ayw3 was most frequent than ayw2 amongBV/D samples. However, HBV/A and HBV/C had subtype

dw2 and adr, respectively. Subtypes could not be ascertainedn one sequence (BB04623) due to substitution present at 122mino acid position (Table 3).

The most frequent amino acid substitution (T68A) outide the MHL was found in 9/41 (22.0%) due to A356Gucleotide substitution in surface gene, among antiHBc

nly (9/37; 24.3%) HBV/D samples (Table 4). This sub-titution also caused an amino acid change at N422Sn the overlapping RT domain of the polymerase (Pol)ene.

A. Banerjee et al. / Journal of Clinical Virology 40 (2007) 312–317 315

Table 4Virological characteristics of 50 HBsAg (−) subjects

SI. no. Age Sex S gene BCP-precore Genotype/subgenotype

Subtype Mutations in S gene HBV DNA(copies/ml)

HBsAg (−)/AntiHBs(+)/antiHBc(+) (n = 13)1 22 M + + C/Cs adr Wild type ND2 30 M + + D/D2 adw3 T118V, A128V 103 to 104

3 25 F + + D/D5 ayw3 T125M <102

4 46 M + + D/D3 ayw3 T125M 104 to 105

5 29 M + + D/D3 ayw3 T125M 103 to 104

6 34 M + + D/D3 ayw3 T125M 103

7 35 M + + D/D3 ayw3 T125M 104 to 105

8 32 F + + D/D3 ayw3 T125M 104 to 105

9 25 M + − D/D5 ayw3 T125M <102

10 40 M + − D/D2 ayw3 T118V, A128V ND11 27 F + − D/D3 ayw3 T125M ND12 38 F + − D/D3 ayw3 T125M <102

13 43 M + − A/Aa adw2 P62L <102

AntiHBc only (n = 37)14 28 M + + D/D2 ayw3 N59S, T118V,

Q129R, A157TND

15 45 M + + D/D3 ayw3 T125M ND16 30 F + + D/D3 ayw3 T125M ND17 40 M + + D/D3 ayw3 T125M ND18 35 M + + A/Aa adw2 Wild type ND19 20 M + + C/Cs adr Wild type ND20 34 M + + C/Cs adr Wild type ND21 30 M + + C/Cs adr Wild type ND22 30 M + + C/Cs adr T131A ND23 37 M + + D/D5 ayw3 T125M ND24 35 M + + D/D5 ayw3 T125M <102

25 22 M + + D/D5 ayw3 T125M <102

26 38 M + + D/D5 ayw3 T125M, Y134F ND27 30 F + + D/D5 ayw3 M133T, T125M <102

28 25 F + + C/Cs adr C69*, P70H, L88P <102

29 55 F + + C/Cs adr Wild type ND30 20 M + + D/D5 ayw3 T68A, T125M,

S143T103 to 104

31 23 F + + D/D5 adw3 T68A, T125M 107 to 108

32 30 M + + D/D5 ayw3 T68A, T125M 102 to 103

33 31 M + + D/D5 ayw3 T68A, T125M 102 to 103

34 24 M + + A/Aa adw2 A159T 105 to 106

35 28 M + − C/Cs adr Q129N ND36 48 M + − C/Cs adw2 P67Q, ND37 20 M + − D/D2 ayw3 P11Q, T118V,

G130R, Y134H,C137Y

<102

38 28 F + − A/Aa adw2 Wild type ND39 50 F + − D/D3 ayw3 T125M <102

40 32 M + − D/D3 ayw3 T125M <102

41 28 M + − D/D2 ayw3 Wild type ND42 30 M + − D/D3 ayw3 T125M ND43 28 M + − D/D3 ayw2 D99Y, S117T ND44 28 M + − D/D5 ayw3 T68A, T125M,

C137GND

45 50 F + − D/D1 ayw2 Wild type ND46 32 M + − D/D3 ayw3 T125M ND47 60 F + − D/D5 ayw3 T68A, S114T,

T125M<102

48 28 M + − D/D5 ayw3 T68A, T125M ND49 30 M + − D/D5 ayw3 T68A, T125M <102

50 30 M + − D/D5 ayw3 T68A, T125M <102

ND: not done.

3 Clinica

4

HsDoDSHbp

ctmHitsterwda

aHH(cIeH

etitstMecitwMbeaCoto

na(tttHRs

o2tl<tvCosompsdd

mHemnnst

A

oBTa

R

A

A

16 A. Banerjee et al. / Journal of

. Discussion

The study suggested that serological pattern of anti-Bc only was not infrequent in our population. In this

tudy, among 58% antiHBc only subjects, 22.8% were HBVNA(+). Although nested PCR was employed for detectionf occult HBV infection, results of amplification of HBVNA from more than one region were not always concordant.imilarly, failure to detect amplification from both regions ofBV genome of antiHBc only samples, were also reportedy Alhababi et al. (2003). Differences in the sensitivity of therimers might be one of the reasons in these cases.

Reactivity for antiHBc was an incidental finding in theseases. They were healthy and not aware of liver inflamma-ion, although 20.7% of them were HBV DNA(+). Viral load

easured in 20 samples (out of 25) was low (<104 copies/ml).owever, little is known about the long-term outcome of

ndividuals with antiHBc only; most seem to remain asymp-omatic or ‘healthy’ carriers. Indirect evidence strongly hasuggested that in these individuals, the risk of progressiono cirrhosis and hepatocellular carcinoma does exist (Huot al., 1998; Paterlini et al., 1990). On the other hand, in aecent report from low HBV prevalence area, no evidenceas found to indicate occult HBV alone causes severe liveramage in individuals with antiHBc only serology (Knoll etl., 2006).

Phylogenetic analysis revealed (tree not shown) thatlthough HBV/D with subgenotype HBV/D1, HBV/D2,BV/D3 and HBV/D5 was predominant, HBV/Cs andBV/Aa were also present, supporting our previous reports

Banerjee et al., 2006; Datta et al., 2006a). Amino acid atodon 118 was reported to be variable for genotype D fromndia (Gandhe et al., 2003). In our study, T118 was presentxclusively in occult infection while A/V was common inBsAg(+) subjects.Very few surface gene sequences are available from gen-

ral population through out the world. Our study revealedhat relatively high frequency of amino acid substitutionsncluding few novel mutations were present in occult infec-ion. To the best of our knowledge, so far only two extensivetudies on general population reported surface gene muta-ional data (Weinberger et al., 2000; on German population,

inuk et al., 2005; on North American population). How-ver, considerable difference was found in the prevalence ofommon immune escape variants such as D144A or G145Rn these studies. In our study, in spite of the fact that morehan 80.0% was mutated sequences; none of the mutationas G145R or D144A. Only three samples showed either133T, Q129R, G130R mutation, previously described to

e associated with diagnostic and vaccine failure (Carmant al., 1997; Hou et al., 2001; Jongerius et al., 1998; Lee etl., 1997; Oon et al., 1995). In addition, the substitution of

137G/Y may affect the conformational epitope recognitionf the protein by impeding the correct formation of essen-ial disulphide bridges. In the outside MHL region, 24.3%f antiHBc only sequences had T68A mutation which has

B

l Virology 40 (2007) 312–317

ot been reported previously. This substitution also causedmino acid mutation (N422S) in the reverse transcriptaseRT) domain in the pol gene. Since genetic alteration inhe surface gene may have an effect on the RT domain ofhe pol gene, therefore, the possibility cannot be ruled outhat the typically low level of HBV DNA observed for anti-Bc only individuals, may be related to the mutation inT domain which in turn resulted in low level of proteinynthesis.

The present study as well as our previous pilot studyn antiHBc positive blood donors (Bhattacharya et al.,007) showed significant prevalence of occult HBV infec-ion among antiHBc positive individuals. In our study,evel of viremia tested for antiHBc only samples are low,104 copies/ml. However, risk of HBV transmission through

hese donated blood samples may still exist due to the largeolume of blood received during transfusions as suggested byonjeevaram and Lok (2001). Therefore, individuals withccult HBV infection need to be monitored by follow-uptudies to assess its significance in our population. In termsf safety of blood transfusion, some countries have imple-ented the antiHBc/NAT screening of blood donors and

ositive blood units are deferred (Roth et al., 2002). In India,creening of large number of samples should be done fromifferent parts of the country to assess the possibility of intro-ucing these tests for donor screening.

In conclusion, serological pattern of antiHBc only is com-on in our population; however, the existence of occultBV infection among antiHBc only individuals is not fully

xplained by surface gene mutations among them. Pol geneutations might be one of the factors that may cause HBsAg

egativity at least in some of the cases, but their clinical sig-ificance is yet to be determined. Therefore, antiHBc onlyubjects should be carefully monitored by follow-up studieso assess its significance in our population.

cknowledgements

The study was supported by the grant from Indian Councilf Medical Research (ICMR), Government of India and Westengal Aids Control and Prevention Society, India. We thankapan Chakraborty and Sreekanta Deb for excellent technicalssistance.

eferences

be A, Inoue K, Tanaka T, Kato J, Kajiyama N, Kawaguchi R, et al. Quan-titation of hepatitis B virus genomic DNA by real-time detection PCR.J Clin Microbiol 1999;37:2899–903.

lhababi F, Sallam TA, Tong CYW. The significance of ‘anti-HBc only’ inthe clinical virology laboratory. J Clin Virol 2003;27:162–9.

anerjee A, Banerjee S, Chowdhury A, Santra A, Chowdhury S, Roy-chowdhury S, et al. Nucleic acid sequence analysis of basal corepromoter/precore/core region of hepatitis B virus isolated from chroniccarriers of the virus from Kolkata, eastern India: low frequency of muta-tion in the precore region. Intervirology 2005;48:389–99.

Clinica

B

B

C

C

C

C

D

D

G

G

H

H

J

J

J

K

K

K

L

L

M

N

O

P

R

W

W

Weinberger KM, Bauer T, Bohm S, Jilg W. High genetic variability of

A. Banerjee et al. / Journal of

anerjee A, Datta S, Chandra PK, Chowdhury A, Santra A, Roychowd-hury S, et al. Distribution of hepatitis B virus genotypes: phylogeneticanalysis and virological characteristics of genotype C circulatingamong HBV carriers in Kolkata, Eastern India. World J Gastroenterol2006;12:5964–71.

hattacharya P, Chandra PK, Datta S, Banerjee A, Chakraborty S, RajendranK, et al. Significant increase in HIV, HBV, HCV and syphilis infec-tions among blood donors in West Bengal, Eastern India 2004–2005:exploratory screening reveals high frequency of occult HBV infection.World J Gastroenterol 2007;13:3730–3.

arman WF, Van Deursen FJ, Mimms LT, Hardie D, Coppola R, Decker R, etal. The prevalence of surface antigen variants of hepatitis B virus in PapuaNew Guinea, South Africa, and Sardinia. Hepatology 1997;26:1658–66.

hakravarty R, Neogi M, Roychowdhury S, Panda CK. Presence of hepatitisB surface antigen mutant G145R DNA in the peripheral blood leukocytesof the family members of an asymptomatic carrier and evidence of itshorizontal transmission. Virus Res 2002;90:133–41.

howdhury A, Santra A, Chakravorty R, Banerjee A, Pal S, Dhali GK, et al.Community-based epidemiology of hepatitis B virus infection in WestBengal, India: prevalence of hepatitis B and antigen-negative infectionand associated viral variants. J Gastroenterol Hepatol 2005;20:1712–20.

onjeevaram HS, Lok AS. Occult hepatitis B virus infection: a hiddenmenace? Hepatology 2001;34:204–6.

atta S, Banerjee A, Chandra PK, Chowdhury A, Chakravarty R. Genotype,phylogenetic analysis, and transmission pattern of occult hepatitis Bvirus (HBV) infection in families of asymptomatic HBsAg carriers. JMed Virol 2006a;78:53–9.

atta S, Banerjee A, Chandra PK, Mahapatra PK, Chakraborti S,Chakravarty R. Hepatitis B virus genotypes among injecting drug userswith occult HBV infection of Manipur India. Emerg Infect Dis 2006b;12:1990–3.

andhe SS, Chadha MS, Arankalle VA. Hepatitis B virus genotypes andserotypes in western India: lack of clinical significance. J Med Virol2003;69:324–30.

rob P, Jilg W, Bornhak H, Gerken G, Gerlich W, Gunther S, et al. Sero-logical pattern ‘anti-HBc alone’: report on a workshop. J Med Virol2000;62:450–5.

ou J, Wang Z, Cheng J, Lin Y, Lau GK, Sun J, et al. Prevalence of naturallyoccurring surface gene variants of hepatitis B virus in nonimmunized sur-face antigen-negative Chinese carriers. Hepatology 2001;34:1027–34.

uo TI, Wu JC, Lee PC, Chau GY, Lui WY, Tsay SH, et al. Sero-clearanceof hepatitis B surface antigen in chronic carriers does not necessarilyimply a good prognosis. Hepatology 1998;28:231–6.

ilg W, Sieger E, Zachoval R, Schatzl H. Individuals with antibodies againsthepatitis B core antigen as the only serological marker for hepatitis B

infection: high percentage of carriers of hepatitis B and C virus. J Hepatol1995;23:14–20.

ilg W, Hottentrager B, Weinberger K, Schlottmann K, Frick E, Holstege A,et al. Prevalence of markers of hepatitis B in the adult German population.J Med Virol 2001;63:96–102.

l Virology 40 (2007) 312–317 317

ongerius JM, Wester M, Cuypers HT, van Oostendorp WR, Lelie PN, vander Poel CL, et al. New hepatitis B virus mutant form in a blood donor thatis undetectable in several hepatitis B surface antigen screening assays.Transfusion 1998;38:56–9.

noll A, Hamoshi H, Weislmaier K, Jilg W. Serological pattern “anti-HBcalone”: characterization of 552 individuals and clinical significance.World J Gastroenterol 2006;28:1255–60.

roes ACM, Quint WGV, Heijtink RA. Significance of isolated hepati-tis B core antibodies detected by enzyme immunoassay in a high riskpopulation. J Med Virol 1991;35:96–100.

wok S, Higuchi R. Avoiding false positives with PCR. Nature1991;339:237–8.

ee PI, Chang LY, Lee CY, Huang LM, Chang MH. Detection of hepatitisB surface gene mutation in carrier children with or without immunopro-phylaxis at birth. J Infect Dis 1997;176:427–30.

iu C-J, Chen D-S, Chen P-J. Epidemiology of HBV infection in Asianblood donors: emphasis on occult HBV infection and the role of NAT. JClin Virol 2006;36:S33–44.

inuk GY, Sun DF, Uhanova J, Zhang M, Caouette S, Nicolle LE, et al.Occult hepatitis B virus infection in a North American community-basedpopulation. J Hepatol 2005;42:480–5.

order H, Courouce AM, Coursaget P, Echevarria JM, Lee SD, MushahwarIK, et al. Genetic diversity of hepatitis B virus strains derived world-wide: genotypes, subgenotypes, and HBsAg subtypes. Intervirology2004;47:289–309.

on CJ, Lim GK, Ye Z, Goh KT, Tan KL, Yo SL, et al. Molecular epi-demiology of hepatitis B virus vaccine variants in Singapore. Vaccine1995;13:699–702.

aterlini P, Gerken G, Nakajima E, Terre S, D’Errico A, Grigioni W, et al.Polymerase chain reaction to detect hepatitis B virus DNA and RNAsequences in primary liver cancers from patients negative for hepatitis Bsurface antigen. N Engl J Med 1990;323:80–5.

oth WK, Weber M, Petersen D, Drosten C, Buhr S, Sireis W, et al.NAT for HBV and anti-HBc testing increase blood safety. Transfusion2002;42:869–75.

eber B, Melchior W, Gehrke R, Doerr HW, Berger A, Rabenau H. Hep-atitis B virus markers in anti-HBc only positive individuals. J Med Virol2001;64:312–9.

einberger KM, Kreuzpaintner EA, Hottentrager B, Neifer S, Jilg W. Muta-tions in the S-gene of hepatitis B virus (HBV) isolates from chroniccarriers with anti-HBc as the only serological marker of HBV infection.In: Rizzetto M, Purcell RH, Gerin JL, Verme G, editors. In Viral Hep-atitis and Liver Disease. Turin: Edizioni Minerva Medica; 1997. p. 138–43.

the group-specific a determinant of hepatitis B virus surface antigen(HBsAg) and the corresponding fragment of the viral polymerase inchronic virus carriers lacking detectable HBsAg in serum. J Gen Virol2000;81:1165–74.