Archives of Medical Research 45 (2014) 270e279

ORIGINAL ARTICLE

HLA-DR Polymorphism and Primary Biliary Cirrhosis: EvidenceFrom a Meta-analysis

Man Li,a,* Hao Zheng,b,* Qing-bao Tian,a Mei-na Rui,a and Dian-wu Liua

aDepartment of Epidemiology and Statistics, School of Public Health, Hebei Medical University, Shijiazhuang, ChinabDepartment of Ultrasonography, Hebei Chest Hospital, Shijiazhuang, China

Received for publication September 22, 2013; accepted February 26, 2014 (ARCMED-D-13-00518).

*These authors con

Address reprint req

versity, #361 Zhong-sh

311-86265531; FAX: 3

0188-4409/$ - see frohttp://dx.doi.org/10

Background and Aims. We undertook this study to review and quantitatively analyze theassociation between human leukocyte antigen (HLA) DR polymorphisms and susceptibil-ity of primary biliary cirrhosis (PBC).

Methods. All relevant publications on the association between HLA-DR polymorphismsand PBC were searched through June 2013. Odds ratios (OR) and confidence intervals(CI) for the comparisons between case and control group were calculated. Statistical anal-ysis was performed using Stata 11.0 software.

Results. Nineteen articles (or 20 studies including the substudies) were identified. ForDR*7 allele, the ORs (95% CIs) were 1.530 (1.310, 1.788), 1.757 (1.285, 2.403) and1.495 (1.211, 1.845) in overall, Asian and European populations, respectively. ForDR*8 alleles, the ORs (95% CIs) were 3.158 (1.822, 5.475), 2.803 (2.420, 3.247) and3.056 (2.573, 3.629) in Asian, American and European subgroups, respectively. The sub-group analysis for DR*11 and DR*13 showed a significant association in Asian and Eu-ropean population. For DR*12 and *15 alleles, the overall ORs (95% CIs) were 0.551(0.404, 0.753) and 0.721 (0.607, 0.857). However, in subgroup analysis for DR*12 allele,the association was only found in Asian population. In addition, statistical significanceexists in American and European populations in the subgroup analysis for DR*15 allele.

Conclusion. Our meta-analysis suggested that HLA-DR *7 and *8 allele polymorphismscontributed to the susceptibility of PBC, whereas DR*11, *12, *13 and *15 allele poly-morphisms are protective factors in certain population. � 2014 IMSS. Published byElsevier Inc.

Key Words: HLA, Meta-analysis, Primary biliary cirrhosis, Polymorphism.

Introduction

Primary biliary cirrhosis (PBC) is an autoimmune chroniccholestatic liver disease characterized by portal inflamma-tion and immune-mediated destruction of the intrahepaticbile ducts (1). Loss of bile ducts leads to decreased bilesecretion and the retention of toxic substances within theliver, resulting in further hepatic damage, fibrosis, cirrhosisand, eventually, liver failure (2).

The prevalence of PBC seems to vary in different parts ofthe world with prevalence rates of up to 492 cases per million

tributed equally to this work.

uests to: Dian-wu Liu, Professor, Hebei Medical Uni-

an East Road, Shijiazhuang, 050017, PRChina; Phone:

11-86265531; E-mail: dwliu1956@hotmail.com

nt matter. Copyright � 2014 IMSS. Published by Elsevier.1016/j.arcmed.2014.03.002

population levels reported during the last decade by studiesperformed in Newcastle (UK), North America, Canada,southern Israel, and southern China (3e7). Many genetic var-iants including CTLA (8), TNF (9), VDR (10) and severalhuman leukocyte antigen (HLA) loci (11,12) have beenshown to be associated with PBC. However, few of the asso-ciations have been proven to be conclusive. The mechanismof susceptibility to PBC has yet to be elucidated. Severallines of evidence suggest that immune-mediated mechanismsplay a crucial role. The disease has many similarities withother autoimmune disorders including female prevalence,common in subjects with a family history of PBC, andfrequent coexistence with other autoimmune diseases (13).

The human major histocompatibility complex (MHC)class I and class II regions include the genes HLA A,B,

Inc.

271HLA-DR Polymorphism and Primary Biliary Cirrhosis

Cw (class I); and DR, DQ and DP (class II), respectively (14).The class I and II antigens are critical components of the T-cell immune response. HLA class II molecules are expressedon specialized antigen-presenting cells, present peptides of13e21 amino acid residues and are restricted by CD4 core-cognition, which mediates many key immunological eventssuch as definition of self-tolerance or cellular immune re-sponses to tumors and pathogens (15,16). The bound and pre-sented peptide is partly determined by the structure of theMHC binding site and up to 90% of the inherited variationin the HLA genes encodes amino acid residues at this site.Thus, HLA allele determines the structure of the antigen-binding site and the specific antigen-binding characteristicsof each expressed HLA molecule. Therefore, the HLA allelecan act as a determinant of both susceptibility and resistance(17). Similar to other genetically complex diseases, HLA hasbeen extensively studied in PBC, but for decades data havecumulatively suggested only a weak association with theclass II HLA DRB1. Recent genome-wide studies of PBC(18,19) have shown the strongest association signals in theHLA region. Association of HLA class II alleles with PBChas been widely studied in Caucasian and Asian patient.

However, there is still a lack of conclusive evidence ofthe association between polymorphisms and the suscepti-bility of PBC. The relationship between them is not uni-versal for all the investigated populations. For example,a number of associations with HLA II molecules havebeen found in Caucasian as well as in Japanese popula-tions (20e22), but they were not confirmed by studiesfrom Italy, Brazil and the UK (23e25). In this meta-analysis, all available data were pooled to examine the as-sociation between common HLA-DR alleles and PBCsusceptibility.

Subjects and Methods

Study Selection

Several databases (Medline/PubMed and Chinese ChinaNational Knowledge Infrastructure CNKI, EMBASE, Webof Science, Cochrane databases) were searched throughJune 2013 for all publications on the association betweenHLA DR polymorphism and PBC. The search terms wereas follows: (‘‘primary biliary cirrhosis’’ or ‘‘PBC’’) and(‘‘HLA’’ or ‘‘human leukocyte antigen’’ or ‘‘DR’’) and(‘‘polymorphism’’ or ‘‘variant’’ or ‘‘genotype’’). No lan-guage limitations were used. In addition, we also searchedreferences of retrieved articles. Studies needed to meet thefollowing criteria: (1) case-control study; (2) primarybiliary cirrhosis as outcome; (3) sufficient published datafor estimating an odds ratio (OR) with 95% confidence in-terval (CI), and (4) at least two comparison groups (PBCpatient group vs. control group) involved in a single study.Exclusion criteria were no reporting about the genotype fre-quency or insufficient information for data extraction.

Data Extraction

Data extraction was performed independently by two inves-tigators according to the inclusion criteria listed above.The third participant was consulted for discussion toreach agreement concerning discrepancies. The followingitems were extracted from each study: first author’s lastname, publication date, country of origin, ethnicity, theNewcastle-Ottawa Scale (NOS), numbers of cases and con-trols, genotyping method, and association allele.

Statistical Analysis

Stata11.0 software was used for statistical analysis toperform meta-analysis. Heterogeneity among studies wasassessed with Q-test with I2 statistic interpreted as the pro-portion of total variation contributed by between-studyvariation (26). Mantel-Hansel fixed-effect model (27) orDersimonian Laird random-effect model (28) was then usedto calculate the pooled OR and 95% CI. If there was a sta-tistical difference in terms of heterogeneity ( p !0.1), arandom-effect model was selected to pool the data. Other-wise, a fixed-effect model was used. Begg’s funnel plotwas used to examine small study effects. To validate thecredibility of outcomes in this meta-analysis, a sensitivityanalysis was performed by sequential omission of individ-ual studies or by omitting studies plotted by the Galbraithplot method as the possible major source of heterogeneity.The method of Begg and Mazumdar was used to assesspublication bias with a p !0.1 for considering statisticalsignificance. In addition, subgroup analysis by geographywas also performed.

Results

Ninety-two articles were retrieved using the search strat-egy. After the first screening, 64 articles were excluded(14,17,29e91). The remaining 28 articles then underwentfull text assessing and three articles without sufficientdata (18,92,93) and one study (94) about anticentromereantibody-positive patients were excluded. In view ofthe low resolution of RFLP-PCR method, the studies(20,24,95e97) using RFLP-PCR method were alsoexcluded from this study. Finally, this meta-analysis,which included a total number of 6057 cases and 16107controls, identified 19 articles (or 20 studies includingthe substudies) on HLA-DR polymorphisms (11,12,21-23,25,83,98e109) (Figure 1). The most commonly inves-tigated polymorphisms were HLA-DR1, DR3, DR4, DR8,DR11, DR13, DR15, which were reported in 14, 10, 14,19, 16, 17 and 14 studies, respectively. Quality of theincluded articles was evaluated using NOS scale. TheNOS contains eight items categorized into three dimen-sions including selection, comparability, and exposure.For each item a series of response options is provided.

Figure 1. Flow diagram of literature search. (A color figure can be found in the online version of this article.)

272 Li et al./ Archives of Medical Research 45 (2014) 270e279

A star system is used to allow a semiquantitative assess-ment of study quality, such that the highest quality studiesare awarded a maximum of one star for each item with theexception of the item related to comparability that allowsthe assignment of two stars. The NOS ranges between zeroand nine stars (110). The characteristics of each study arepresented in Table 1.

Association Between Polymorphisms of HLA-DR andPBC

The results of this meta-analysis are listed in Table 2. ForHLA-DR*1, DR*3, DR*4, *DR9, *DR10, *DR14 and*DR16 alleles, no evidence of associations betweenHLA-DR polymorphism and susceptibility of PBC werefound. However, in subgroup analysis by geography,DR*3 polymorphism was found in association with PBC.ORs and 95% CIs were 1.919 (1.236, 2.977) in Asian and1.414 (1.091, 1.831) in European population, respectively.At the same time, DR*4 polymorphism was found in asso-ciation with PBC in Asian and European population. ORsand 95% CIs were 1.217 (1.005, 1.474) and 1.360 (1.178,1.571), respectively.

For the HLA-DR*7 polymorphism, there was no evi-dence of heterogeneity (I2 5 0, p 5 0.725). A fixed-effectmodel was used in the calculation of the ORs. Overall OR(95% CI) was 1.530 (1.310, 1.788) with p !0.001. Sub-group analysis showed summary ORs (95% CI) were1.757 (1.285, 2.403) for Asian and 1.495 (1.211, 1.845)for European, which showed that the HLA-DR*7 polymor-phism significantly increased the risk of PBC ( p!0.001 foreach) (Figure 2).

For the HLA-DR*8 polymorphism, there was a moder-ate level of heterogeneity (I2 5 47.2, p 5 0.012). A randomeffect model was used in the OR calculation. Subgroupanalysis by ethnicity showed summary OR (95% CI) were3.158 (1.822, 5.475) for Asian, 2.803 (2.420, 3.247) forAmerican and 3.056 (2.573, 3.629) for European, whichshowed that the HLA-DR*8 polymorphism also contri-buted to susceptibility of PBC ( p !0.001 for each)(Supplemental Figure 1).

For the HLA-DR*11 polymorphism, there was a moder-ate level of heterogeneity (I2 5 38.4, p 5 0.060). A randomeffect model was used in the OR calculation (SupplementalFigure 2). Subgroup analysis by ethnicity showed summaryORs (95% CI) were 0.320 (0.197, 0.520) for Asian, 0.892

Table 1. Main characteristic of included studies

No Study Year Country Genotyping method NOS score

Numbers

Association allelesCases Controls

1 Zhao 2013 China SBT 7 145 500 08, 07, 12, 11

2 Umemura 2012 Japan SSO-PCR 6 229 523 0405, 0803, 1101, 1302

3 Liu 2012 UK Immunochip 8 2861 8514 0801, 1501, 1101, 1104, 0404

4 Nakamura 2010 Japan SBT 6 334 258 0405, 0803, 1101, 1302, 1501

5 Zepeda-Gomez 2009 Mexico SSO-PCR 6 16 99 01, 04

6 Vazquez-Elizondo 2009 Mexico PCR-SSP 6 18 762 None

7 Niro 2009 Italy PCR-SSP 6 107 141 08, 13, 11

8 Invernizzi 2008 Italy PCR-SSP 8 664 1992 02, 08, 11, 13

9 Liu 2006 China PCR-SSP 7 65 431 07

10 Donaldson 2006 UK SSO-PCR 8 412 236 08, 13

Italy SSO-PCR 8 80 95 08, 11, 13

11 Jiang 2004 China PCR-SSP 6 52 43 08

12 Mullarkey 2005 USA SSO-PCR 7 72 381 08

13 Invernizzi 2003 Italy SSO-PCR 8 186 558 11

14 Bittoncourt 2003 Brazil PCR-SSP 7 61 83 None

15 Washmuth 2002 Sweden SSO-PCR 8 99 158 11, 14, 08

16 Stone 2002 Multicenter SSO-PCR 8 154 216 08

17 Donaldson 2001 UK SSO-PCR 7 164 102 08

18 Onishi 1994 Japan SSO-PCR 6 31 215 08

19 Begovich 1994 USA SSO-PCR 7 51 240 15, 13, 08, 09

273HLA-DR Polymorphism and Primary Biliary Cirrhosis

(0.550, 1.447) for American and 0.447 (0.350, 0.572) forEuropean, which showed that the HLA-DR*11 polymor-phism decreased the risk of PBC in Asian and Europeanpopulations ( p !0.001 for both).

For the HLA-DR*12 polymorphism, there was no het-erogeneity (I2 5 4.1, p 5 0.400). A fixed-effect modelwas used in the OR calculation. Overall OR (95% CI)was 0.551 (0.404, 0.753) with p !0.001 (Figure 3). Sub-group analysis by ethnicity showed summary ORs (95%CI) were 0.485 (0.338, 0.697) for the Asian population,which showed a protective role in PBC ( p !0.001). How-ever, no association was found in the American subgroup( p 5 0.148) and European subgroup ( p 5 0.829).

For the HLA-DR*13 polymorphism, there was a moder-ate level of heterogeneity (I2 5 34.4, p 5 0.081). A randomeffect model was used in the OR calculation (SupplementalFigure 3). Subgroup analysis by ethnicity showed summaryORs (95% CI) were 0.468 (0.326, 0.671) for Asian popula-tion and 0.671 (0.470, 0.957) for European population,which showed that the HLA-DR*13 polymorphismdecreased the risk of PBC in Asian and European popula-tions ( p !0.001 and 0.028, respectively).

For the HLA-DR*15 polymorphism, there was a moder-ate level of heterogeneity (I2 5 35.2, p 5 0.094). A randomeffect model was used in the OR calculation (Figure 4).Subgroup analysis by ethnicity showed summary ORs(95% CI) were 0.824 (0.610, 1.113) for Asian population,which showed no association with the risk of PBC ( p 50.206). ORs (95% CIs) were 0.592 (0.374, 0.937) for Amer-ican and 0.671 (0.591, 0.761) for European populations,which indicate the protective role ( p !0.001 for both).

Sensitivity Analysis

A sensitivity analysis for each HLA-DR allele interactionanalysis was performed by sequential omission of individ-ual studies. A single study included in this meta-analysiswas deleted each time to observe the influence of the indi-vidual result on the pooled ORs. The corresponding pooledORs were not materially altered, indicating that our resultswere statistically robust.

Publication Bias

In this meta-analysis, funnel plot and Egger’s test were per-formed to evaluate the publication bias. All p values of Eg-ger’s tests ( p Egger’s) are shown in Table 2. p Egger’s withO0.05 was considered as statistical evidence of the funnelplots’ asymmetry. The Egger’s test results showed that pub-lication bias in our meta-analyses was not remarkable inmost of the subgroup analysis.

Discussion

Worldwide prevalence rates for PBC seem to show signifi-cant differences across geographic or ethnic boundaries.For example, although PBC is found in all races, it is rarelyreported in Africa and the Indian subcontinent (111). Re-ports of clustering of cases (e.g., within families) (112) sug-gested a possible underlying genetic influence. BecausePBC displays features of autoimmunity, many publicationshave attempted to identify genes with a role in disease sus-ceptibility and progression by evaluating small numbers of

Table 2. Overall and group-specific statistics for association between HLA-DR and PBC

Allele Subgroup Studies No.

Heterogeneity

Ma

OR

Publication biasbI2 (%) p OR (95% CI) p

DR1 All PBC 14 33 0.126 F 1.097 (0.921, 1.308) 0.378 0.669

Asian 4 0.0 0.860 F 1.051 (0.649, 1.591) 0.814 0.262

American 5 71.2 0.008 R 1.325 (0.570, 3.079) 0.514 0.705

European 5 0.0 0.413 F 1.126 (0.862, 1.472) 0.383 0.187

DR3 All PBC 10 40.6 0.087 R 1.250 (0.936, 1.669) 0.130 0.552

Asian 4 0 0.589 F 1.919 (1.236, 2.977) 0.004 0.969

American 4 27.5 0.247 F 0.851 (0.597, 1.212) 0.370 0.455

European 2 0 0.698 F 1.414 (1.091, 1.831) 0.009 —

DR4 All PBC 14 78.5 !0.001 R 1.177 (0.900, 1.539) 0.235 0.670

Asian 6 0 0.870 F 1.217 (1.005, 1.474) 0.044 0.435

American 5 91.9 !0.001 R 0.988 (0.350, 2.792) 0.982 0.754

European 3 0 0.483 F 1.360 (1.178, 1.571) !0.001 0.985

DR7 All PBC 10 0 0.725 F 1.530 (1.310, 1.788) !0.001 0.782

Asian 4 35 0.202 F 1.757 (1.285, 2.403) !0.001 0.261

American 3 0 0.991 F 1.381 (0.978, 1.951) 0.067 0.028

European 3 0 0.928 F 1.495 (1.211, 1.845) !0.001 0.351

DR8 All PBC 19 47.2 0.012 R 2.818 (2.294, 3.463) !0.001 0.782

Asian 6 79.2 !0.001 R 3.158 (1.822, 5.475) !0.001 0.278

American 5 0 0.788 F 2.803 (2.420, 3.247) !0.001 0.397

European 8 0 0.454 F 3.056 (2.573, 3.629) !0.001 0.860

DR9 All PBC 9 29.4 0.184 F 1.017 (0.814, 1.270) 0.882 0.138

Asian 4 0 0.601 F 0.942 (0.745, 1.190) 0.616 0.994

American 2 67.1 0.081 R 1.959 (0.806, 1.532) 0.493 —

European 3 0 0.679 F 2.535 (0.940, 6.837) 0.066 0.342

DR10 All PBC 9 0 0.601 F 1.363 (0.885, 2.098) 0.159 0.297

Asian 4 0 0.834 F 1.573 (0.755, 3.279) 0.227 0.347

American 2 78.7 0.030 R 1.120 (0.106, 1.843) 0.925 —

European 3 0 0.709 F 1.455 (0.774, 2.734) 0.244 0.023

DR11 All PBC 16 38.4 0.060 R 0.467 (0.377, 0.577) !0.001 0.630

Asian 5 0 0.951 F 0.320 (0.197, 0.520) !0.001 0.508

American 3 0 0.949 F 0.892 (0.550, 1.447) 0.644 0.738

European 8 47.7 0.063 R 0.447 (0.350, 0.572) !0.001 0.634

DR12 All PBC 9 4.1 0.400 F 0.551 (0.404, 0.753) !0.001 0.551

Asian 4 0 0.969 F 0.485 (0.338, 0.697) !0.001 0.743

American 2 0 0.414 F 0.335 (0.076, 1.472) 0.148 —

European 3 41.7 0.180 F 1.079 (0.541, 2.150) 0.829 0.886

DR13 All PBC 17 34.4 0.081 R 0.663 (0.538, 0.817) !0.001 0.562

Asian 6 0 0.780 F 0.468 (0.326, 0.671) !0.001 0.193

American 4 0 0.932 F 0.859 (0.609, 1.212) 0.386 0.157

European 7 61.1 0.017 R 0.671 (0.470, 0.957) 0.028 0.580

DR14 All PBC 13 56.4 0.007 R 0.971 (0.687, 1.372) 0.868 0.391

Asian 5 37.6 0.171 F 0.945 (0.685, 1.303) 0.729 0.361

American 2 0 0.630 F 0.740 (0.325, 1.687) 0.474 —

European 6 74.1 0.002 R 1.150 (0.618, 2.140) 0.660 0.936

DR15 All PBC 14 35.2 0.094 R 0.721 (0.607, 0.857) !0.001 0.898

Asian 4 46.9 0.130 F 0.824 (0.610, 1.113) 0.206 0.470

American 5 56.4 0.057 R 0.592 (0.374, 0.937) !0.001 0.207

European 5 0 0.512 F 0.671 (0.591, 0.761) !0.001 0.384

DR16 All PBC 6 50.8 0.071 R 0.709 (0.334, 1.504) 0.370 0.639

Asian 3 79.3 0.008 R 0.811 (0.204, 3.229) 0.767 0.905

American 2 0 0.889 F 0.661 (0.231, 1.888) 0.439 —

European 1 — — — 0.435 (0.085, 2.233) 0.319 —

aModel of meta-analysis; R, random-effects model; F, fixed-effects model.bp value for Egger’s test.

274 Li et al./ Archives of Medical Research 45 (2014) 270e279

variants in one or a few specific candidate genes by meansof case-control study designs. They focused on genetic var-iations in genes related to immunity, such as HLA-DR.However, such approaches have led to little insight into

the genetic basis of PBC, mainly due to lack of robust repli-cation. To our knowledge, no systematic review has beenpreviously published on the association between HLA-DRand PBC.

Figure 2. OR and 95% CI for HLA-DR*7 polymorphisms in fixed-effects model.

275HLA-DR Polymorphism and Primary Biliary Cirrhosis

HLA is one of the most widely studied regions in the hu-man genome, although the true role of the various HLA al-leles in inducing autoimmune reactions remains unclear tosome extent (113). However, it is possible that specific

Figure 3. ORs and 95% CIs for HLA-DR *12

HLA alleles enhance autoimmune activation by enhancingimmunogenicity and influencing the expressed repertoire ofT cells. Many studies have reported associations of HLAclass II alleles and PBC in populations of Caucasian and

polymorphisms in fixed-effects model.

Figure 4. ORs and 95% CIs for HLA-DR*15 polymorphisms in random-effects model.

276 Li et al./ Archives of Medical Research 45 (2014) 270e279

Asian ethnicity (83,98,99). The association with HLADRB1*08 allele has been found most frequently amongstudies from the UK (83), U.S. (114) and Sweden (22), thusindicating that this allele may act as a risk factor for PBCamong Caucasians. However, it is notable that several Eu-ropean studies have failed to confirm an association withDRB1*08 (97,101).

In our meta-analysis, HLA-DR*7 and *8 allele polymor-phisms were found to play the role of risk factors for PBC,whereas HLA-DR*11, *12, *13 and *15 were protectivefactors for PBC. In addition, because the genetic back-ground may greatly vary among different geographicgroups, our analysis evaluated the association of HLA-DRin PBC by comparing different geographic cases with con-trol populations. For DR*8 allele, the subgroup analysisshowed statistical significance. For the HLA-DR*7, *11and *13 polymorphism, subgroup analysis by geographyindicated the association with PBC in Asian and Europeansubgroups but not in the American subgroup. Statistical sig-nificance between DR*15 and PBC exists in American andEuropean subgroups but not in the Asian subgroup. Forthose populations carrying risk alleles and at risk of PBC,they should pay more attention to their lifestyle, diet, andavoid some adverse factors.

Meta-analyses are currently increasingly used to helpmake clinical decisions because pooling data from a num-ber of studies should reduce both bias and uncertainty.However, the conclusion may be affected by a type of selectbias, which was caused by those rapidly published studieswith statistically significant results. In this meta-analysis,

funnel plots, the graph of estimates of the effect of eachtrial vs. sample size are used to detect publication bias byevaluating an asymmetry. Publication bias is consideredto exist when funnel plots are asymmetrical. In addition,we also used other statistical methods such as Egger’smethod to detect publication bias for the reason that Eg-ger’s method had stronger statistical power than Begg’smethod or Mac skill’s method (115,116). The resultsshowed that for most of the DR alleles, p values for publi-cation bias are not significant except for DR*7 subgroupanalysis in which the p value was 0.028 in the Americansubgroup. We then used another method, the trim-and-fillmethod, to correct estimates for publication bias. However,no publication bias was found using this method. Beyondthat, the sensitivity analysis showed that the correspondingpooled ORs were not materially altered when omitting onestudy at one time, which indicated that our results werestatistically robust. Some limitations exist in this meta-analysis: First, the sample size for some of the subgroup an-alyses is small. It is necessary to collect studies for furtheranalysis in the future. Second, the gene-gene or gene-environment interaction may influence the association be-tween HLA-DR polymorphism and PBC. However, noappropriate data are available from the reported studies.

In conclusion, our meta-analysis suggested that HLA-DR*7 and *8 are risk factors for PBC, whereas DR*11and *13 are protective factor for Asian and European pop-ulations. The significant associations between DR*12 and*15 polymorphisms and PBC were only found in certainpopulations.

277HLA-DR Polymorphism and Primary Biliary Cirrhosis

Conflict of Interest

The authors declare no conflict of interest.

Financial Support

This work was supported by grants from the Fostering Tal-ents project of Hebei Medical University (No. 0620970001)and Hebei Key Medical Guidance Topics of Health Depart-ment (20130457).

Supplementary data

Supplementary data associated with this article can befound in the online version at http://dx.doi.org/10.1016/j.arcmed.2014.03.002.

References1. Kaplan MM. Primary biliary cirrhosis. N Engl J Med 1996;335:

1570e1580.

2. Invernizzi P, Selmi C, Gershwin ME. Update on primary biliary

cirrhosis. Dig Liver Dis 2010;42:401e408.

3. James OF, Bhopal R, Howel D, et al. Primary biliary cirrhosis once

rare, now common in the United Kingdom? Hepatology 1999;30:

390e394.

4. Boberg KM, Aadland E, Jahnsen J, et al. Incidence and prevalence of

primary biliary cirrhosis, primary sclerosing cholangitis, and autoim-

mune hepatitis in a Norwegian population. Scand J Gastroenterol

1998;33:99e103.

5. Kim WR, Lindor KD, Locke GR 3rd, et al. Epidemiology and natural

history of primary biliary cirrhosis in a US community. Gastroenter-

ology 2000;119:1631e1636.

6. Liu H, Liu Y, Wang L, et al. Prevalence of primary biliary cirrhosis in

adult referring hospital for annual health check-up in Southern China.

BMC Gastroenterol 2010;10:100.

7. Myers RP, Shaheen AA, Fong A, et al. Epidemiology and natural his-

tory of primary biliary cirrhosis in a Canadian health region: a

population-based study. Hepatology 2009;50:1884e1892.

8. Li M, Zheng H, Li T, et al. Cytotoxic T-lymphocyte associated

antigen-4 gene polymorphisms and primary biliary cirrhosis: a sys-

tematic review. J Gastroenterol Hepatol 2012;27:1159e1166.

9. Chen R, Cai X, Han Z, et al. Meta-analysis: the relationship between

TNF-alpha polymorphisms and susceptibility to primary biliary

cirrhosis. Hepatogastroenterology 2012;60:568e576.

10. Tanaka A, Nezu S, Uegaki S, et al. Vitamin D receptor polymor-

phisms are associated with increased susceptibility to primary biliary

cirrhosis in Japanese and Italian populations. J Hepatol 2009;50:

1202e1209.

11. Zhao DT, Liao HY, Zhang X, et al. Human leucocyte antigen alleles

and haplotypes and their associations with antinuclear antibodies fea-

tures in Chinese patients with primary biliary cirrhosis. Liver Int;

2013.

12. Invernizzi P, Selmi C, Poli F, et al. Human leukocyte antigen poly-

morphisms in Italian primary biliary cirrhosis: a multicenter study

of 664 patients and 1992 healthy controls. Hepatology 2008;48:

1906e1912.

13. Selmi C, Torok NJ, Affronti A, et al. Genomic variants associated

with primary biliary cirrhosis. Genome Med 2010;2:5.

14. Invernizzi P. Human leukocyte antigen in primary biliary cirrhosis:

an old story now reviving. Hepatology 2011;54:714e723.

15. Klein J, Sato A. The HLA system. First of two parts. N Engl J Med

2000;343:702e709.16. Klein J, Sato A. The HLA system. Second of two parts. N Engl J

Med 2000;343:782e786.

17. Jones DE, Donaldson PT. Genetic factors in the pathogenesis of pri-

mary biliary cirrhosis. Clin Liver Dis 2003;7:841e864.18. Hirschfield GM, Liu X, Xu C, et al. Primary biliary cirrhosis associ-

ated with HLA, IL12A, and IL12RB2 variants. N Engl J Med 2009;

360:2544e2555.19. Liu X, Invernizzi P, Lu Y, et al. Genome-wide meta-analyses identify

three loci associated with primary biliary cirrhosis. Nat Genet 2010;

42:658e660.

20. Underhill J, Donaldson P, Bray G, et al. Susceptibility to primary

biliary cirrhosis is associated with the HLA-DR8-DQB1*0402 haplo-

type. Hepatology 1992;16:1404e1408.

21. Donaldson P, Agarwal K, Craggs A, et al. HLA and interleukin 1

gene polymorphisms in primary biliary cirrhosis: associations with

disease progression and disease susceptibility. Gut 2001;48:

397e402.

22. Wassmuth R, Depner F, Danielsson A, et al. HLA class II markers

and clinical heterogeneity in Swedish patients with primary biliary

cirrhosis. Tissue Antigens 2002;59:381e387.

23. Bittencourt PL, Palacios SA, Farias AQ, et al. Analysis of major his-

tocompatibility complex and CTLA-4 alleles in Brazilian patients

with primary biliary cirrhosis. J Gastroenterol Hepatol 2003;18:

1061e1066.

24. Zhang L, Weetman AP, Bassendine M, et al. Major histocompatibil-

ity complex class-II alleles in primary biliary cirrhosis. Scand J Im-

munol 1994;39:104e106.

25. Begovich AB, Klitz W, Moonsamy PV, et al. Genes within the HLA

class II region confer both predisposition and resistance to primary

biliary cirrhosis. Tissue Antigens 1994;43:71e77.

26. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency

in meta-analyses. BMJ 2003;327:557e560.

27. Mantel N, Haenszel W. Statistical aspects of the analysis of data from

retrospective studies of disease. J Natl Cancer Inst 1959;22:719e748.

28. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control

Clin Trials 1986;7:177e188.

29. Zhou Ye, Li Hai-ying, Chen Yan, et al. Association between HLA

polymorphism and primary biliary cirrhosis. Zhong Hua Xiao Hua

Za Zhi 2004;24:616e617.

30. Liu Hai-ying, Deng An-mei, Zhang Jian, et al. Analysis of HLA al-

leles polymorphism in Chinese patients with primary biliary

cirrhosis. Zhong Hua Gan Zang Bing Za Zhi 2005;13:410e413.

31. Jiang Xiao-hua, Zhong Ren-qian, Fang Xiao-yun, et al. Association

between HLA eDRB1, DQB1 allele polymorphism and primary

biliary cirrhosis. Zhong Hua Gan Zang Bing Za Zhi 2004;12:436.

32. Liberal R, Grant CR, Mieli-Vergani G, et al. Autoimmune hepatitis: a

comprehensive review. J Autoimmun 2013;41:126e139.

33. Nakamura M, Nishida N, Kawashima M, et al. Genome-wide associ-

ation study identifies TNFSF15 and POU2AF1 as susceptibility loci

for primary biliary cirrhosis in the Japanese population. Am J Hum

Genet 2012;91:721e728.

34. Srivastava B, Mells GF, Cordell HJ, et al. Fine mapping and replica-

tion of genetic risk loci in primary sclerosing cholangitis. Scand J

Gastroenterol 2012;47:820e826.

35. Coss Adame E, Granados J, Uribe M, et al. Does HLA-DR7 differ-

entiate the overlap syndrome of auto-immune hepatitis-primary

biliary cirrhosis (AIH-PBC) from those with auto-immune hepatitis

type 1? Ann Hepatol 2011;10:28e32.

36. Ishibashi H, Komori A, Shimoda S, et al. Risk factors and prediction

of long-term outcome in primary biliary cirrhosis. Intern Med 2011;

50:1e10.

37. Chong VH, Telisinghe PU, Jalihal A. Primary biliary cirrhosis in

Brunei Darussalam. Hepatobiliary Pancreat Dis Int 2010;9:622e628.

278 Li et al./ Archives of Medical Research 45 (2014) 270e279

38. Yanagisawa M, Takagi H, Takahashi H, et al. Familial clustering and

genetic background of primary biliary cirrhosis in Japan. Dig Dis Sci

2010;55:2651e2658.

39. Cassinotti A, Birindelli S, Clerici M, et al. HLA and autoimmune

digestive disease: a clinically oriented review for gastroenterologists.

Am J Gastroenterol 2009;104:195e217. quiz 194, 218.

40. Etxagibel A, Julia MR, Brotons A, et al. Drug-induced hepatitis

superimposed on the presence of anti-SLA antibody: a case report.

J Med Case Rep 2008;2:25.

41. Jakab SS, Navarro VJ, Colombe BW, et al. Human leukocyte antigen

and adult living-donor liver transplantation outcomes: an analysis of

the organ procurement and transplantation network database. Liver

Transpl 2007;13:1405e1413.42. Czaja AJ. Autoimmune liver disease. Curr Opin Gastroenterol 2000;

16:262e270.

43. Farias AQ, Goncalves LL, Bittencourt PL, et al. Applicability of the

IAIHG scoring system to the diagnosis of antimitochondrial/anti-M2

seropositive variant form of autoimmune hepatitis. J Gastroenterol

Hepatol 2006;21:887e893.

44. Chapman R, Cullen S. Etiopathogenesis of primary sclerosing chol-

angitis. World J Gastroenterol 2008;14:3350e3359.45. Hagymasi K, Tulassay Z. [Genetics of primary biliary cirrhosis]. Orv

Hetil 2005;146:2151e2155.

46. Liu HY, Deng AM, Zhang J, et al. [Analysis of HLA allele polymor-

phism in Chinese patients with primary biliary cirrhosis]. Zhonghua

Gan Zang Bing Za Zhi 2005;13:410e413.

47. Invernizzi P, Selmi C, Mackay IR, et al. From bases to basis: linking

genetics to causation in primary biliary cirrhosis. Clin Gastroenterol

Hepatol 2005;3:401e410.

48. Selmi C, Mayo MJ, Bach N, et al. Primary biliary cirrhosis in mono-

zygotic and dizygotic twins: genetics, epigenetics, and environment.

Gastroenterology 2004;127:485e492.49. Ichiki Y, Shimoda S, Ishibashi H, et al. Is primary biliary cirrhosis a

model autoimmune disease? Autoimmun Rev 2004;3:331e336.

50. Feld JJ, Heathcote EJ. Epidemiology of autoimmune liver disease. J

Gastroenterol Hepatol 2003;18:1118e1128.

51. Nemoto T, Kubota K, Kita J, et al. Unusual onset of chronic

graft-versus-host disease after adult living-related liver trans-

plantation from a homozygous donor. Transplantation 2003;75:

733e736.

52. Heathcote J. Variant syndromes of autoimmune hepatitis. Clin Liver

Dis 2002;6:669e684.

53. Lohr HF. [Autoimmune hepatitis and overlap syndrome: therapy].

Praxis (Bern 1994) 2002;91:1347e1351.

54. Tanaka A, Igarashi M, Kakinuma M, et al. The occurrence of various

collagen diseases in one family: a sister with ISSc, PBC, APS, and

SS and a brother with systemic lupus erythematosus. J Dermatol

2001;28:547e553.

55. Gershwin ME. The natural history of primary biliary cirrhosis: of

genes and cooperation. J Hepatol 2001;35:412e415.56. Toda G. Revised criteria for diagnosis of autoimmune hepatitis.

Intern Med 2000;39:999e1000.

57. O’Donohue J, Oien KA, Donaldson P, et al. Co-amoxiclav jaundice:

clinical and histological features and HLA class II association. Gut

2000;47:717e720.

58. Czaja AJ, Carpenter HA, Santrach PJ, et al. Autoimmune cholangitis

within the spectrum of autoimmune liver disease. Hepatology 2000;

31:1231e1238.59. Gershwin ME, Ansari AA, Mackay IR, et al. Primary biliary

cirrhosis: an orchestrated immune response against epithelial cells.

Immunol Rev 2000;174:210e225.60. Uibo R, Salupere V. The epidemiology of primary biliary cirrhosis:

immunological problems.Hepatogastroenterology1999;46:3048e3052.

61. Springer JE, Cole DE, Rubin LA, Cauch-Dudek K, et al. Vitamin

D-receptor genotypes as independent genetic predictors of decreased

bone mineral density in primary biliary cirrhosis. Gastroenterology

2000;118:145e151.62. Agarwal K, Jones DE, Bassendine MF. Genetic susceptibility to pri-

mary biliary cirrhosis. Eur J Gastroenterol Hepatol 1999;11:603e606.

63. Lohse AW, zum Buschenfelde KH, Franz B, et al. Characterization of

the overlap syndrome of primary biliary cirrhosis (PBC) and autoim-

mune hepatitis: evidence for it being a hepatitic form of PBC in

genetically susceptible individuals. Hepatology 1999;29:1078e1084.

64. Jones DE, Watt FE, Grove J, et al. Tumour necrosis factor-alpha pro-

moter polymorphisms in primary biliary cirrhosis. J Hepatol 1999;

30:232e236.

65. Banner BF, Karamitsios N, Smith L, et al. Enhanced phenotypic

expression of alpha-1-antitrypsin deficiency in an MZ heterozygote

with chronic hepatitis C. Am J Gastroenterol 1998;93:1541e1545.

66. Vuoristo M, F€arkkil€a M, Gylling H, et al. Expression and therapeutic

response related to apolipoprotein E polymorphism in primary biliary

cirrhosis. J Hepatol 1997;27:136e142.67. Niveloni S, Dezi R, Pedreira S, et al. Gluten sensitivity in patients

with primary biliary cirrhosis. Am J Gastroenterol 1998;93:404e408.

68. Broome U, Grunewald J, Scheynius A, et al. Preferential V beta3 us-

age by hepatic T lymphocytes in patients with primary sclerosing

cholangitis. J Hepatol 1997;26:527e534.

69. Uring-Lambert B, Valette C, Hartmann D, et al. [Changes in the

polymorphism and concentrations of components of the class II com-

plement after orthotopic transplantation of the liver]. Gastroenterol

Clin Biol 1995;19:976e979.

70. Underhill JA, Donaldson PT, Doherty DG, et al. HLA DPB polymor-

phism in primary sclerosing cholangitis and primary biliary cirrhosis.

Hepatology 1995;21:959e962.

71. Donaldson P, Doherty D, Underhill J, et al. The molecular genetics of

autoimmune liver disease. Hepatology 1994;20:225e239.

72. Gregory WL, Bassendine MF. Genetic factors in primary biliary

cirrhosis. J Hepatol 1994;20:689e692.

73. Gregory WL, Daly AK, Dunn AN, et al. Analysis of HLA-class-II-

encoded antigen-processing genes TAP1 and TAP2 in primary biliary

cirrhosis. Q J Med 1994;87:237e244.

74. Seki T, Kiyosawa K, Ota M, et al. Association of primary biliary

cirrhosis with human leukocyte antigen DPB1*0501 in Japanese pa-

tients. Hepatology 1993;18:73e78.75. Hietarinta M, Koskimies S, Lassila O, et al. Familial scleroderma:

HLA antigens and autoantibodies. Br J Rheumatol 1993;32:

336e338.

76. Fugger L, Morling N, Ryder LP, et al. Restriction fragment length

polymorphism of two HLA-B-associated transcripts genes in five

autoimmune diseases. Hum Immunol 1991;30:27e31.

77. Shibasaki T, Shimada T, Morita T, et al. Immunogenetical penetrance

in patients with primary biliary cirrhosis. Jpn J Med 1990;29:

533e536.

78. Prochazka EJ, Terasaki PI, Park MS, et al. Association of primary

sclerosing cholangitis with HLA-DRw52a. N Engl J Med 1990;

322:1842e1844.

79. Fugger L, Morling N, Ryder LP, et al. NcoI restriction fragment

length polymorphism (RFLP) of the tumour necrosis factor (TNF

alpha) region in primary biliary cirrhosis and in healthy Danes. Scand

J Immunol 1989;30:185e189.

80. Morita T, Okuda S, Okada K, et al. [A case of primary biliary

cirrhosis with seroimmunological abnormalities]. Nihon Naika Gak-

kai Zasshi 1987;76:1067e1072.81. Tsuji H, Nomiyama K, Watanabe H, et al. [Primary biliary cirrhosis

in two sisters]. Nihon Naika Gakkai Zasshi 1987;76:522e527.

82. Ercilla G, Pares A, Arriaga F, et al. Primary biliary cirrhosis associ-

ated with HLA-DRw3. Tissue Antigens 1979;14:449e452.

83. Liu JZ, Almarri MA, Gaffney DJ, et al. Dense fine-mapping study

identifies new susceptibility loci for primary biliary cirrhosis. Nat

Genet 2012;44:1137e1141.

279HLA-DR Polymorphism and Primary Biliary Cirrhosis

84. Czaja AJ, Carpenter HA, Santrach PJ, et al. Genetic predispositions

for immunological features in chronic liver diseases other than auto-

immune hepatitis. J Hepatol 1996;24:52e59.

85. Yokosawa S, Yoshizawa K, Ota M, et al. A genomewide DNA micro-

satellite association study of Japanese patients with autoimmune hep-

atitis type 1. Hepatology 2007;45:384e390.86. BoX, BroomeU, RembergerM, et al. Tumour necrosis factor alpha im-

pairs function of liver derived T lymphocytes and natural killer cells in

patients with primary sclerosing cholangitis. Gut 2001;49:131e141.87. Strnad P, Kucukoglu O, Lunova M, et al. Non-coding keratin variants

associate with liver fibrosis progression in patients with hemochro-

matosis. PLoS One 2012;7:e32669.

88. Yoshizawa K, Umemura T, Ota M. Genetic background of autoim-

mune hepatitis in Japan. J Gastroenterol 2011;(46 Suppl 1):42e47.

89. Zappala F, Grove J, Watt FE, et al. No evidence for involvement of

the interleukin-10-592 promoter polymorphism in genetic suscepti-

bility to primary biliary cirrhosis. J Hepatol 1998;28:820e823.90. David E, Jones J, Fiona E, et al. Tumour necrosis factor—a promoter

polymorphisms in primary biliary cirrhosis. J Hepato 1999;30:

232e236.

91. Gordon MA, Oppenheim E, Camp NJ, et al. Primary biliary cirrhosis

shows association with genetic polymorphism of tumour necrosis

factor alpha promoter region. J Hepatol 1999;31:242e247.

92. Juran BD, Hirschfield GM, Invernizzi P, et al. Immunochip analyses

identify a novel risk locus for primary biliary cirrhosis at 13q14, mul-

tiple independent associations at four established risk loci and epis-

tasis between 1p31 and 7q32 risk variants. Hum Mol Genet 2012;

21:5209e5221.93. Mells GF, Floyd JA, Morley KI, et al. Genome-wide association

study identifies 12 new susceptibility loci for primary biliary

cirrhosis. Nat Genet 2011;43:329e332.

94. Akimoto S, Abe M, Ishikawa O, et al. HLA-DRB1 and DQB1 genes

in anticentromere antibody positive patients with SSc and primary

biliary cirrhosis. Ann Rheum Dis 2001;60:639e640.

95. Morling N, Sandberg-Wollheim M, Fugger L, et al. Immunogenetics

of multiple sclerosis and optic neuritis: DNA polymorphism of HLA

class II genes. Immunogenetics 1992;35:391e394.

96. Gregory WL, Mehal W, Dunn AN, et al. Primary biliary cirrhosis:

contribution of HLA class II allele DR8. Q J Med 1993;86:393e399.97. MehalWZ,GregoryWL, LoYM, et al. Defining the immunogenetic sus-

ceptibility to primary biliary cirrhosis. Hepatology 1994;20:1213e1219.

98. Umemura T, Joshita S, Ichijo T, et al. Human leukocyte antigen class II

molecules confer both susceptibility and progression in Japanese pa-

tients with primary biliary cirrhosis. Hepatology 2012;55:506e511.

99. Nakamura M, Yasunami M, Kondo H, et al. Analysis of HLA-DRB1

polymorphisms in Japanese patients with primary biliary cirrhosis

(PBC): The HLA-DRB1polymorphism determines the relative risk

of antinuclear antibodies for disease progression in PBC. Hepatol

Res 2010;40:494e504.

100. Zepeda-Gomez S, Montano-Loza A, Zapata-Colindres JC, et al.

HLA-DR allele frequencies in Mexican mestizos with autoimmune

liver diseases including overlap syndromes. Immunol Invest 2009;

38:276e283.

101. Vazquez-Elizondo G, Ponciano-Rodriguez G, Uribe M, et al. Human

leukocyte antigens among primary biliary cirrhosis patients born in

Mexico. Ann Hepatol 2009;8:32e37.

102. Niro GA, Poli F, Andriulli A, et al. TNF-alpha polymorphisms in pri-

mary biliary cirrhosis: a northern and southern Italian experience.

Ann NY Acad Sci 2009;1173:557e563.

103. Liu HY, Deng AM, Zhou Y, et al. Analysis of HLA alleles polymor-

phism in Chinese patients with primary biliary cirrhosis. Hepatobili-

ary Pancreat Dis Int 2006;5:129e132.104. Donaldson PT, Baragiotta A, Heneghan MA, et al. HLA class II al-

leles, genotypes, haplotypes, and amino acids in primary biliary

cirrhosis: a large-scale study. Hepatology 2006;44:667e674.

105. Mullarkey ME, Stevens AM, McDonnell WM, et al. Human leuko-

cyte antigen class II alleles in Caucasian women with primary biliary

cirrhosis. Tissue Antigens 2005;65:199e205.

106. Invernizzi P, Battezzati PM, Crosignani A, et al. Peculiar HLA poly-

morphisms in Italian patients with primary biliary cirrhosis. J Hepa-

tol 2003;38:401e406.

107. Stone J, Wade JA, Cauch-Dudek K, et al. Human leukocyte antigen

Class II associations in serum antimitochondrial antibodies (AMA)-

positive and AMA-negative primary biliary cirrhosis. J Hepatol 2002;

36:8e13.

108. Onishi S, Sakamaki T, Maeda T, et al. DNA typing of HLA class II

genes; DRB1*0803 increases the susceptibility of Japanese to pri-

mary biliary cirrhosis. J Hepatol 1994;21:1053e1060.

109. Jiang XH, Zong RQ, Fang XY, et al. [Relationship between alleles of

HLA-DRB and HLA-DQB1 and Chinese patients with primary

biliary cirrhosis]. Zhonghua Gan Zang Bing Za Zhi 2004;12:436.

110. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the

assessment of the quality of nonrandomized studies in meta-analyses.

Eur J Epidemiol 2010;25:603e605.111. Neuberger J. Primary biliary cirrhosis. Lancet 1997;350:875e879.

112. Bach N, Schaffner F. Familial primary biliary cirrhosis. J Hepatol

1994;20:698e701.

113. Carson DA. Genetic factors in the etiology and pathogenesis of auto-

immunity. FASEB J 1992;6:2800e2805.

114. Invernizzi P, Ransom M, Raychaudhuri S, et al. Classical HLA-

DRB1 and DPB1 alleles account for HLA associations with primary

biliary cirrhosis. Genes Immun 2012;13:461e468.115. Hayashino Y, Noguchi Y, Fukui T. Systematic evaluation and com-

parison of statistical tests for publication bias. J Epidemiol 2005;

15:235e243.116. Sterne JA, Gavaghan D, Egger M. Publication and related bias in

meta-analysis: power of statistical tests and prevalence in the litera-

ture. J Clin Epidemiol 2000;53:1119e1129.