Introduction

Rheumatoid arthritis (RA) is a chronic inflammatory disease of predominantly synovial joints and affects up to 1% of the world’s population [1]. Although the etiology of RA has not been determined, a genetic component of RA susceptibility has been established by twin and family studies, which have estimated that the heritability of RA liability may be as high as 60% [2]. Human leukocyte antigen (HLA) class II molecules are the most powerful genetic factors of RA identified to date. However, family studies suggest that this association accounts for only one-third of genetic susceptibility, and that non-HLA genes are also involved [3].

Signal transducers and activators of transcription (STATs) regulate the entire hematopoietic process by regulating cytokine binding to specific cell-surface receptors [4]. In particular, STAT4 encodes signal transducer and activator of transcription 4, the STAT protein family member activated by interleukin-12 via its receptor that has an essential downstream role in Th1 cell differentiation and proliferation. STAT4 is also involved in the development of a newly discovered subset of Th17 cells, which display a dominant role in autoimmunity-associated inflammation [5].

A recent study by Amos et al. [6] found evidence of linkage at chromosome 2q33 in RA, and a follow-up study revealed that polymorphisms of the STAT4 gene (located at 2q33) are the markers responsible for the linkage signal in 2q33 [7]. Furthermore, the STAT4 rs7574865 polymorphism has been found to be associated with RA in different ethnic groups [7–14]. The allelic frequencies of genes often differ substantially between populations and, thus, ethnic-specific association studies are required to confirm genetic associations in different populations. In the present study, we investigated whether the STAT4 rs7574865 polymorphism contributes to RA susceptibility in different populations using a meta-analysis approach.

Materials and methods

Identification of eligible studies and data extraction

An exhaustive search was conducted for studies that examined associations between STAT4 polymorphisms and RA. The literature was searched using the MEDLINE citation index to identify articles in which the STAT4 polymorphism was determined in RA patients and controls (the most recent article was dated January 2009). In addition, all references mentioned were reviewed to identify additional studies not indexed by MEDLINE. The following key words and subject terms were searched: ‘STAT4’, ‘polymorphism’, ‘rheumatoid arthritis’ and ‘RA.’ The following information was extracted from each study: author, year of publication, ethnicity of the study population, demographics, numbers of cases and controls, and the T allele frequency of the STAT4 rs7574865 polymorphism.

Evaluation of publication bias

A funnel plot and the Egger’s linear regression test were used to detect publication bias [15]. When asymmetry was indicated, we used the ‘trim and fill’ method to adjust summary estimates for observed bias [16]. This method removes small studies until symmetry in the funnel plot is achieved by recalculating the center of the funnel, before the removed studies are replaced with their missing mirror-image counterparts. A revised summary estimate is then calculated using all original studies and hypothetical ‘filled’ studies.

Evaluation of statistical associations

T allele frequencies of the STAT4 rs7574865 polymorphism were determined for each study using the allele counting method. Point estimates of risks, ORs and OR 95% confidence intervals (CI) were determined for each study. Cochran’s Q-statistics were used to assess within- and between-study variations and heterogeneities [17]. The heterogeneity test was used to assess the null hypothesis that all studies were evaluating the same effect. When a significant Q-statistic (P < 0.10) indicated heterogeneity across studies, the random effects model was used for meta-analysis, and if heterogeneity across studies was not indicated, the fixed effects model was used. Fixed effects assume that genetic factors have similar effects on RA susceptibility across all studies, and that observed variations between studies are caused by chance alone [18]. The random effects model assumes that different studies show substantial diversity and assesses both within-study sampling errors and between-study variances [19]. If study groups show no heterogeneity, the fixed and random effects models produce similar results, and, if not, the random effects model usually produces wider CIs than the fixed effects model. The random effects model is used in the presence of significant between-study heterogeneity. We also quantified the effect of heterogeneity by using a recently developed measure, i.e., I 2 = 100% × (Q − df)/Q [17]. I 2 ranges between 0 and 100% and represents the proportion of inter-study variability that can be attributed to heterogeneity rather than chance. I 2 values of 25, 50 and 75% were defined as low, moderate and high estimates, respectively. Statistical manipulations were performed using the Comprehensive Meta-Analysis computer program (Biosta, Englewood, NJ, USA).

Results

Studies included in the meta-analysis

A total of 14 relevant studies, which investigated the relation between the STAT4 polymorphism and RA, were identified using MEDLINE. Six studies were excluded because they were reviews [5, 20] or because they did not concern RA [21–24]. Thus, eight studies met the inclusion criteria [7–14]. Of these, one study contained data on two different groups [12] and three studies contained data on three different groups [7, 10, 11]. In the present study, we analyzed these groups independently. Therefore, a total of 15 separate comparisons were considered in this meta-analysis, which involved 16,088 RA patients and 16,509 controls, and ten European, four Asian, and one Latin American population (Table 1). However, because the sample population in the Latin American group was inadequate, ethnicity-specific meta-analysis was conducted on European and Asian populations.

Table 1 Characteristics of the studies included in the meta-analysis
Full size table

Frequency of the STAT4 rs7574865 T allele in different ethnic groups

The mean frequency of the STAT4 rs7574865 T allele was 23.7% among all normal controls, and the European population had a lower T allele prevalence than the other two ethnic groups. In terms of normal controls, frequencies of the T allele in the European, Latin American and Asian populations were 21.4, 27.9 and 32.0% (Table 2).

Table 2 Prevalence of the STAT4 rs7574865 polymorphism T allele (a) and meta-analysis of the STAT4 rs7574865 polymorphism/RA association (b)
Full size table

Meta-analysis of the association between the STAT4 rs7574865 polymorphism and RA

Meta-analysis was performed on all subjects and by ethnic group. A summary of the meta-analysis findings of the relation between the STAT4 rs7574865 polymorphism and RA is provided in Table 2. Meta-analysis showed an association between the STAT4 rs7574865 polymorphism and RA in all subjects (OR = 1.271, 95% CI = 1.197–1.350, P < 0.001) (Table 2; Fig. 1). After stratification by ethnicity, analysis indicated that the STAT4 rs7574865 T allele was significantly associated with RA in Europeans and Asians (OR = 1.300, 95% CI = 1.195–1.414, P < 0.001; OR = 1.216, 95% CI = 1.135–1.303, P < 0.001). In addition, the original Latin American study also showed a significant association between the STAT4 rs7574865 T allele and RA (Table 2; Fig. 2). Stratification of patients according to the presence of anti-CCP antibody revealed a significant association between the T allele and RA in both anti-CCP-positive and -negative RA patients versus controls. Furthermore, a direct comparison between and anti-CCP-positive and -negative subjects showed no significant difference (Table 2).

Fig. 1
figure 1

ORs and 95% CIs of individual studies and pooled data for the association between the STAT4 rs7574865 polymorphism T allele and RA in all subjects

Full size image
Fig. 2
figure 2

ORs and 95% CI of individual studies and pooled data for the association between the T allele and RA in each ethnic group

Full size image

Heterogeneity and publication bias

Between-study heterogeneity in terms of the ORs of STAT4 rs7574865 T allele was found among all subjects, European populations and anti-CCP-positive subjects and, thus, meta-analyses were performed using a random effects model (Table 2). Publication bias causes a disproportionate number of positive studies and poses a problem for meta-analyses, and evidence of publication bias was found for the meta-analyses of all subjects, European populations and anti-CCP-positive subjects (Egger’s regression test P values <0.1) (Table 2; Fig. 3). Therefore, the ‘trim and fill’ method was used to adjust for publication bias in the three meta-analyses. Adjusted ORs obtained using the ‘trim and fill’ technique remained statistically significant (data not shown).

Fig. 3
figure 3

Funnel plot of studies for the association between the T allele and RA in all subjects (Egger’s regression P value = 0.007)

Full size image

Discussion

In the present study, we subjected previously published data to meta-analysis to evaluate genetic associations between the STAT4 rs7574865 polymorphism and RA susceptibility in different ethnic populations. The prevalence of the rs7574865 T allele was found to vary in ethnic groups from 21.4 to 32.0%. Mean T allele frequency in controls was 23.6%, and was lowest for Europeans and highest for Asians. Furthermore, ethnic-specific meta-analysis showed that an association exists between the STAT4 rs7574865 polymorphism and RA in Europeans and Asians.

Since STAT4 plays a key role in the regulation of Th1 and Th17 cell responses, it may exert an influence on RA via defective signaling. STAT4 is also overexpressed in RA synovium as compared with normal tissues [25]. Furthermore, it has been reported that STAT4-deficient mice are resistant to proteoglycan-induced arthritis and develop significantly less severe collagen-induced arthritis than wild-type controls [26].

Prominent STAT4 expression has been reported in the dendritic cells of RF-positive patients with RA [27]. In view of the fact that these cells are associated with the presence of autoantibodies, such as, RF [10], we investigated whether the association between the STAT4 rs7574865 polymorphism and RA is dependent on anti-CCP antibody positivity. However, stratification by the presence of anti-CCP antibody revealed that the association between the STAT4 rs7574865 polymorphism and RA is not dependant of the presence of this antibody.

The rs7574836 polymorphism examined is located in the third intron of the STAT4 gene, and its functional significance has not been determined. Moreover, since the susceptibility haplotype is located within intron 3 of STAT4, it might be responsible for splice variations or changes in the regulatory effects of STAT4 [7]. Further studies are required to investigate the functional effects of the susceptibility haplotype. Autoimmune diseases are initiated by the breakdown of self-tolerance, and share a common pathogenesis. Furthermore, since STAT4 might play a key role in autoimmune diseases, including RA and systemic lupus erythematosus (SLE), its role in various autoimmune diseases also need to be studied [5].

The present study has some limitations that should be considered. First, publication bias and heterogeneity may have distorted the meta-analysis. However, all studies showed the same directionality with respect to the association between the rs7574865 T allele and RA, and adjustment using the trim and fill method did not change results, which indicates that heterogeneity and publication bias were unlikely to have biased our analysis. Second, our ethnic-specific meta-analysis included data from European and Asian patients and, thus, our results are applicable to these ethnic groups alone. Third, although available genetic data suggest that the STAT4 rs7574865 polymorphism is a determinant of RA susceptibility, the possibilities that the STAT4 haplotypes or other disease-associated polymorphisms of the STAT4 may contribute also need to be examined by meta-analysis. In the present study, we could not conduct meta-analysis on the STAT4 susceptibility haplotype or other STAT4 polymorphisms due to limited data.

In conclusion, this meta-analysis, which included 16,088 RA patients and 16,509 controls, confirms that the STAT4 rs7574865 polymorphism is associated with RA susceptibility in major ethnic groups, and that the prevalence of the rs7574865 T allele depends on ethnicity.