Associations of Interleukin-4 Receptor Gene Polymorphisms (Q551R,I50V) with Rheumatoid Arthritis: Evidence from a Meta-Analysis

Abstract

Background and Aims: Published data on the associations between interleukin-4 receptor (IL-4R) gene polymorphisms (Q551R, I50V) and rheumatoid arthritis (RA) risk are controversial. To quantitatively evaluate the relationships, a meta-analysis was performed. Methods: Studies were identified from the databases of PubMed, MEDLINE, Chinese Biomedical Literature Database, and Chinese National Knowledge Infrastructure, with the last report up to June 2012. The effect summary odds ratio (OR) and 95% confidence interval (CI) were obtained. Results: A total of six separate comparisons involving 2173 patients and 1892 controls were included to assess the association of IL-4R gene Q551R polymorphism and RA susceptibility. Overall, no significantly elevated RA risk was found in the meta-analysis. The pooled OR for the minor R allele was 0.942 (95% CI: 0.848-1.047, p=0.268) in patients with RA. After stratification by ethnicity, there was still no significant association detected in the European population (OR=0.979, 95% CI: 0.875-1.094). As for I50V polymorphism, there were four comparisons involving 1653 patients and 1584 controls in this meta-analysis. The pooled OR for the V allele was 1.104 (95% CI: 1.001-1.217) in RA, the V allele of the IL-4R gene I50V variant might be a risk factor for RA. However, the relationship between the V allele of IL-4R gene I50V polymorphism and rheumatoid factor positive in patients with RA was not identified through a minor meta-analysis, including four independent relevant comparisons. Conclusions: This meta-analysis indicates that the I50V polymorphism of IL-4R gene may confer susceptibility to RA; up to now, there is still not enough evidence to reveal the association of the IL-4R gene Q551R polymorphism with RA risk.

Introduction

Rheumatoid arthritis (RA) is a chronic, systemic, inflammatory autoimmune disease characterized by chronic joint inflammation and destruction that affects close to 1% of the population worldwide (Brennan and Beech, 2007; Bronson et al., 2008). Although the clear etiology of RA remains to be elucidated, it is widely accepted that RA is a complex disease and both environmental and genetic factors can contribute to disease initiation as well as its evolution (Perdigones et al., 2009). The human leukocyte antigen (HLA) class II molecules are the most powerful recognized genetic factors for RA. However, with the development of genetic technology, such as genome-wide association studies and other candidate gene approaches for identifying further susceptibility variants of RA, it has been proved that this association only accounts for one-third of genetic susceptibility and that non-HLA genes are also involved (MacKay et al., 2002; Wellcome Trust Case Control Consortium, 2007). Thus, to better understand the pathogenesis of RA, genes other than HLA should also be explored.

As small proteins secreted by cells mostly after activation, cytokines can jointly play key roles in the regulation of systemic inflammation, local tissue damage, and immunomodulation. A central feature of RA is a relative imbalance in proinflammatory and anti-inflammatory cytokines. Cytokines also play a critical role in the pathogenesis of RA (John et al., 1998). Interleukin-4 (IL-4) and its receptor (IL-4R) have been proved to be associated with the development of RA, since diminished production of IL-4 is believed to contribute to the characteristic Th1-mediated autoimmune rheumatoid inflammation (Skapenko et al., 2003). Furthermore, a variety of immune disorders, including RA disease, can be induced by an imbalance in the Th1:Th2 ratio in the adaptive immune response. As a B-cell stimulating factor, IL-4 is the first B-cell growth factor discovered, and it mainly promotes proliferation of T cells and contributes to production of antibodies by B cells. Pleiotropic activities of IL-4 are mediated by its specific receptor IL-4R at the surface of the target cell to play a role in the immune system. IL-4R represents a complex transmembrane receptor composed of two subunits, the common γ chain shared by several interleukin receptors, can amplify the signal transduced by the IL-4Rα, a 140-kDa high-affinity binding chain. The α chain binds IL-4 and mediates its effect through kinases of the Janus family attached to the intracellular domain (Nelms et al., 1999).

The IL4R gene is located at the short arm of chromosome 16p12.1. To date, seven single-nucleotide polymorphism (SNP) sites have been identified in the coding region of the gene (Kruse et al., 1999). Two of these SNPs, Q551R (rs1801275) and I50V (rs1805010), are nonsynonymous (Mitsuyasu et al., 1999) and are suggested to be associated with various autoimmune diseases, such as asthma and atopy (Sandford et al., 2000; Hytönen et al., 2004), connective tissue disease (Youn et al., 2000), and Crohn's disease (Aithal et al., 2001).

Recently, studies have evaluated the associations of the IL-4R gene polymorphisms (Q551R, I50V) with RA in different ethnic groups (Genevay et al., 2002; Prots et al., 2006; Suppiah et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012). However, the results are inconsistent. Therefore, in the present study, we performed an exhaustive meta-analysis to comprehensively assess the association of IL-4R gene Q551R, I50V polymorphisms and RA risk. Additionally, several studies on the association between the IL-4R gene I50V variant and rheumatoid factor (RF) in RA patients have been performed (Prots et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012), but the results are still disputable. Thus, a minor meta-analysis was also conducted to draw a valuable conclusion on this relationship.

Materials and Methods

Literature search

All studies examining the association of IL-4R gene Q551R, I50V polymorphisms (rs1801275, rs1805010) with RA were fully considered and carefully selected. Studies were identified from the databases of PubMed, MEDLINE, Chinese Biomedical Literature Database, and Chinese National Knowledge Infrastructure, with the last report up to June 2012. The following keywords were used for searching: “interleukin-4 receptor,” “IL-4R,” “polymorphism,” “rheumatoid arthritis,” and “RA.” Studies published in English and Chinese were all considered to be included and had to be in accordance with the following criteria: (1) study must be used an unrelated case-control design, familial studies were excluded as their linkage considerations; (2) available allele frequency or genotype distribution data for comparison; (3) sufficient published data for estimating an odds ratio (OR) with 95% confidence interval (CI); (4) genotype distribution of the control group must be in Hardy-Weinberg equilibrium (HWE).

Data extraction

Two investigators (H.P. and M.Z.) independently assessed all potentially relevant studies with the inclusion criteria. The collected data comprised the name of the first author, journal, and publication year, country of origin, ethnicity of the study population, numbers of cases and controls, demographics, matching criteria, allelic frequencies, and genotype distribution of IL-4R gene polymorphisms (Q551R, I50V).

Statistical analyses

Allelic frequencies of IL-4R polymorphisms (Q551R, I50V) from the respective study were determined by the allele counting method. The chi-square test was used to assess the HWE of genotypes in the control group of each study.

We examined the contrast of the allelic effect of R (minor allele) versus Q (common allele) and V versus I; subsequently, the contrast of RR+QR versus QQ, VV+IV versus II (dominant effect) genotypes as well as the contrast of RR versus QR+QQ, VV versus IV+II (recessive effect) genotypes were also conducted. The presence of heterogeneity between studies was evaluated with the Q- and I²-statistics (Petitti, 1994). Heterogeneity was considered significant for p<0.10. I²=100%×(Q−df)/Q (Higgins and Thompson, 2002), I² values of 25%, 50%, and 75% were defined as low, moderate, and high estimates, respectively. If p-value of the heterogeneity Q-statistic was less than 0.10, the random effects model was used for meta-analysis, otherwise, we used a fixed effects model.

The pooled estimate of risk OR was obtained using the Mantel-Haenszel method in the fixed effects model (Mantel and Haenszel, 1959), while using the DerSimonian and Laid method in the random effects model (DerSimonian and Laird, 1986). And the Z-test was used to determine the pooled OR with the significance set at p<0.05. In addition, to clearly assess the ethnicity-specific effect, analyses by ethnic subgroups were subsequently performed.

Evaluation of publication bias

Publication bias was investigated with the funnel plot, in which, the standard error of log (OR) of each study was plotted against its OR. An asymmetric plot suggested possible publication bias. Funnel plot asymmetry was assessed by the method of the Egger's linear regression test, which used a linear regression approach to measure the funnel plot asymmetry on the natural logarithm scale of the OR (Egger et al., 1997).

Statistical analyses were carried out using the Stata software package version 11.0 (Stata Corporation, College Station, TX). All the p-values were two sided. A p-value of<0.05 was taken as statistically significant.

Results

Literature search and characteristics of eligible studies

A total of 107 potentially relevant articles were identified after an initial search from the selected electronic databases, of which, six reports (Genevay et al., 2002; Prots et al., 2006; Suppiah et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012) were eligible for inclusion. Figure 1 shows the literature selection process. Characteristics (first author's name, publication year, country of origin, demographic characteristics of case and control, etc.) of each study are described in Table 1. The distributions of the genotype and allele of IL-4R gene Q551R, I50V polymorphisms in each case-control study are respectively shown in Tables 2 and 3, and the p-values of HWE testing for the genotypic distribution of the control groups are also showed.

FIG. 1.

Flow diagram of the study selection process. IL-4R, interleukin-4 receptor; RA, rheumatoid arthritis.

Table 1.

Characteristics of the Case-Control Studies Included in the Meta-Analysis

		Demographic characteristics
First author, year	Country	Case	Control	Gene polymorphism	Type of study	Genotyping method
Hussein, 2012	Egypt	172 patients with RA (172 females; age 27-65 years, mean age 47.75±9.3 years)	172 healthy controls (172 females; age 29-62 years, mean age 49.35±8.7 years)	IL-4R Q551R	Hospital-based	PCR
				IL-4R I50V
Marinou, 2008	UK	965 patients with RA	988 healthy controls	IL-4R Q551R	Hospital-based	TaqMan
				IL-4R I50V
Moreno, 2007	Spain	102 patients with RA (mean age 51±12.8 years)	102 health controls (gender-, age- matched, mean age 50.3±12.5 years)	IL-4R Q551R	Hospital-based	PCR
				IL-4R I50V
Prots, 2006	Germany	471 patients with RA (78% female, mean age 56.9±13.4 years)	371 age-, gender-, and ethnic-matched health controls	IL-4R Q551R	Hospital-based	PCR
				IL-4R I50V
Suppiah, 2006	UK	294 patients with RA (170 females)	165 healthy controls (79 females)	IL-4R Q551R	Hospital-based	TaqMan
Genevay, 2002	Switzerland	233 patients with RA (174 females)	148 healthy matched controls	IL-4R Q551R	Hospital-based	TaqMan

IL-4R, interleukin-4 receptor; RA, rheumatoid arthritis; PCR, polymerase chain reaction.

Table 2.

Distribution of IL-4R Q551R Genotypes, and Allelic Frequency in RA Patients and Controls

	Genotype of IL-4R Q551R						Allele
	Case			Control			Case		Control
First author	QQ	QR	RR	QQ	QR	RR	Q	R	Q	R	HWE for controls (p-value)
Hussein	84	72	16	66	80	26	240	104	212	132	0.828
Marinou	558	316	38	601	302	41	1432	392	1504	384	0.695
Moreno	50	45	7	41	52	9	145	59	134	70	0.186
Prots	304	145	22	227	131	13	753	189	585	157	0.261
Suppiah	175	102	17	90	58	17	452	136	238	92	0.106
Genevay	150	66	6	97	38	3	366	78	232	44	0.747

HWE, Hardy-Weinberg equilibrium.

Table 3.

Distribution of IL-4R I50V Genotypes, and Allelic Frequency in RA Patients and Controls

	Genotype of IL-4R I50V						Allele
	Case			Control			Case		Control
First author	II	IV	VV	II	IV	VV	I	V	I	V	HWE for controls (p-value)
Hussein	44	92	36	68	72	32	180	164	208	136	0.103
Marinou	250	460	198	301	438	200	960	856	1040	838	0.085
Moreno	20	60	22	28	52	22	100	104	108	96	0.815
Prots	134	242	95	111	174	86	510	432	396	346	0.266

Association between IL-4R gene Q551R polymorphism and RA risk

Six studies (Genevay et al., 2002; Prots et al., 2006; Suppiah et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012) with a total of 2173 cases and 1892 controls explored the association between the IL-4R gene Q551R polymorphism and RA susceptibility. Meta-analysis showed that there was no significant association of the R allele with RA risk in all subjects (OR=0.942, 95% CI: 0.848-1.047) (Table 4). After stratification by ethnicity, there was still no significant association detected in the European population (OR=0.979, 95% CI: 0.875-1.094) (Table 4).

Table 4.

Meta-Analysis of the IL-4R Q551R, I50V Polymorphisms, and RA Susceptibility Associations

		Sample size			Test of associations				Test of heterogeneity
Polymorphisms	Study	Case	Control	No. of studies	OR (95% CI)	Z	p-Value	Model	p-Value	I²(%)
IL-4R Q551R
R vs. Q	Overall	2173	1892	6	0.942 (0.848-1.047)	1.11	0.268	F	0.106	45.0
	European	2001	1720	5	0.979 (0.875-1.094)	0.38	0.705	F	0.274	22.1
RR+QR vs. QQ	Overall	2173	1892	6	0.951 (0.837-1.080)	0.78	0.438	F	0.135	40.5
	European	2001	1720	5	0.987 (0.863-1.128)	0.20	0.843	F	0.272	22.3
RR vs. QQ+QR	Overall	2173	1892	6	0.839 (0.636-1.108)	1.24	0.216	F	0.383	5.2
	European	2001	1720	5	0.912 (0.671-1.240)	0.59	0.558	F	0.434	0.0
IL-4R I50V
V vs. I	Overall	1653	1584	4	1.104 (1.001-1.217)	1.98	0.048	F	0.292	19.6
	European	1481	1412	3	1.074 (0.968-1.191)	1.35	0.177	F	0.554	0.0
VV+IV vs. II	Overall	1653	1584	4	1.269 (1.091-1.477)	3.08	0.002	F	0.204	34.7
	European	1481	1412	3	1.207 (1.027-1.418)	2.29	0.022	F	0.540	0.0
VV vs. II+IV	Overall	1653	1584	4	0.986 (0.833-1.167)	0.16	0.870	F	0.694	0.0
	European	1481	1412	3	0.968 (0.811-1.157)	0.35	0.723	F	0.590	0.0

OR, odds ratio; vs., versus; CI, confidence interval; F, fixed effects model.

Meta-analysis of the RR+QR versus QQ and RR versus QR+QQ genotypes indicated similar results to the R allele analysis (Table 4). The overall ORs for RR+QR and RR were 0.951 (95% CI: 0.837-1.080) and 0.839 (95% CI: 0.636-1.108), respectively (Table 4). Subgroup analysis of Europeans still failed to find any marked association of the IL-4R gene Q551R polymorphism and RA susceptibility in all genetic models.

Association between IL-4R gene I50V polymorphism and RA risk

A total of four studies (Prots et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012), including 1653 RA patients and 1584 controls, were available for meta-analysis on the IL-4R gene I50V polymorphism and RA susceptibility. Meta-analysis of the V allele and VV+IV genotypes revealed significant associations with RA. And the overall ORs were 1.104 (95% CI: 1.001-1.217) and 1.269 (95% CI: 1.091-1.477), respectively (Fig. 2, Table 4). Although no association of V allele and RA was detected in a subgroup of European Caucasians, in a dominant model, contrast of VV+IV versus II genotypes, the relationship was marked (OR=1.207, 95% CI: 1.027-1.418) (Fig. 2, Table 4).

FIG. 2.

Forest plots for meta-analyses of European and overall populations on the association of IL-4R gene I50V polymorphisms and RA susceptibility. Significant associations were detected in the overall allelic contrast and the genotypic dominant model (VV+IV vs. II) in overall and European populations. (a) allele V versus I; (b) genotype VV+IV versus II (dominant model); (c) genotype VV versus II+IV (recessive model).

Association between IL-4R gene I50V polymorphism and RF positive in RA patients

Totally, four studies (Prots et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012) that assessed the relationship between the IL-4R gene I50V polymorphism and RF were included in the meta-analysis. Due to the insufficient data of genotypes in two studies (Moreno et al., 2007; Hussein et al., 2012), contrast of genotypes (dominant or recessive model) was not performed. Since significant heterogeneity was observed across studies, we conducted analyses using the random effects model. Regretfully, we did not detect any associations between the V allele of IL-4R gene I50V polymorphism and RF positive in RA patients of the overall or European population. The overall ORs were 1.129 (95% CI: 0.657-1.941) and 0.880 (95% CI: 0.530-1.462), respectively (Table 5).

Table 5.

Distribution of IL-4R I50V Genotypes, and Allelic Frequency in RA Patients with RF Positive and RF-Negative Patients, and the Results of Meta-Analysis

	Genotype of IL-4R I50V						Allele
	RF-positive RA patients			RF-negative RA patients			RF-positive RA		RF-negative RA
First author, ethnicity	II	IV	VV	II	IV	VV	I	V	I	V
Hussein, non-European	NA	NA	NA	NA	NA	NA	106	130	74	34
Marinou, European	122	288	165	59	136	64	532	618	254	264
Moreno, European	NA	NA	NA	NA	NA	NA	74	58	24	48
Prots, European	65	98	57	27	40	15	228	212	94	70
Total	NA	NA	NA	NA	NA	NA	940	1018	446	416
Pooled OR for V allele (Overall)							1.129 (95% CI: 0.657-1.941)
Pooled OR for V allele (European)							0.880 (95% CI: 0.530-1.462)

NA, not available; RF, rheumatoid factor.

Heterogeneity and publication bias

The p-values of the Q-test and I²-statistics for heterogeneity are all described in Table 4. No significant between-study heterogeneity was observed in the meta-analysis of overall or subgroups when we explored the associations between IL-4R gene polymorphisms (Q551R, I50V) and RA risk. However, there was high heterogeneity (p<0.001, I²=87.7%) between studies in analysis for the V allele of IL-4R gene I50V variant and RF positive. When studies of an Egyptian and Spanish population by Hussein et al. (2012) and Moreno et al. (2007) were excluded, the heterogeneity was not identified (p=0.602, I²=0.0%), indicating that these two studies may contribute to the heterogeneity. Nevertheless, excluding these studies did not materially affect our result.

The funnel plot and Egger's test were performed to access publication bias. No obvious publication bias was shown in the results (Table 6).

Table 6.

Tests for Publication Bias (Egger's Test) in Overall and European Populations

	Overall			European
Comparisons	No. of studies	t	p	No. of studies	t	p
IL-4R Q551R
R vs. Q	6	−1.57	0.192	5	−1.27	0.295
RR+QR vs. QQ	6	−1.98	0.119	5	−1.48	0.235
RR vs. QQ+QR	6	−0.05	0.966	5	−0.02	0.987
IL-4R I50V
V vs. I	4	0.71	0.550	3	−0.00	0.997
VV+IV vs. II	4	1.06	0.399	3	0.38	0.770
VV vs. II+IV	4	0.08	0.941	3	−0.35	0.788
IL-4R I50V and RF
V vs. I	4	−0.10	0.930	3	−1.11	0.466

Discussion

RA is one of the most heterogeneous complex autoimmune diseases. It is well recognized that there is individual susceptibility to autoimmune diseases even with the same environmental exposure. Host factors, including polymorphisms of genes involved in the pathogenesis of autoimmune diseases, may contribute to this difference. Therefore, genetic susceptibility to RA has been a research focus in the scientific community.

The IL-4Rα chain is responsible for IL-4 binding, transduction of the signal into the cell, and cellular response. Previous studies have suggested that the I variant of IL-4R gene I50V polymorphism of the extracellular domain could significantly upregulate the receptor response to IL-4, increase cell proliferation, and IgE production (Mitsuyasu et al., 1998; Bukhari et al., 2002). It has been proved that the anti-inflammatory effect of IL-4 is more pronounced in the presence of mutant variants of the gene of IL-4Rα chain (IL-4R gene Q551R variant) (Sandford et al., 2000; Shirakawa et al., 2000). These results indicate that I and R alleles are protective for RA. Since the first case-control study for the association between IL-4R gene Q551R polymorphisms and RA was reported in a Swiss population (Genevay et al., 2002), five studies have been undertaken to replicate the association (Prots et al., 2006; Suppiah et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012), what is more, the relation of IL-4R gene I50V polymorphisms and RA have been explored by four studies (Prots et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012). However, these results are inconclusive.

Meta-analysis is a means of augmenting the effective sample size through the pooling of data from individual association studies and enhancing the statistical power of the analysis for the estimation of genetic effects (Peng et al., 2012). In the current study, we did a meta-analysis to estimate the relationships between IL-4R gene Q551R, I50V polymorphisms and the susceptibility to RA. In the meta-analysis, the data of the overall and European subgroups showed that the IL-4R gene Q551R polymorphism may not be an RA susceptibility gene across populations. The pooled ORs for the R allele in the overall and European populations were 0.942 (95% CI: 0.848-1.047) and 0.979 (95% CI: 0.875-1.094). A marked association between IL-4R gene I50V polymorphisms and RA risk was finally detected in the overall and European populations, which was consistent with the previous studies (Mitsuyasu et al., 1998; Bukhari et al., 2002).

In addition, there are several studies investigating the association of IL-4R gene Q551R, I50V variants and clinical features in RA patients, but in different populations, these reveal enormous variability. For example, Prots et al. (2006) reported an association of these genetic polymorphisms with the rapid development of erosions and a diminished responsiveness to IL-4 by CD₄⁺ T cells in German patients with RA, whereas Marinou et al. (2008) failed to support such associations in an English population. In Egyptian RA patients, Hussein et al. (2012) found that subjects with the VV genotype of IL-4R gene I50V were significantly more likely to have erosive arthropathy (OR=2.6, 95% CI: 1.1-6.1). The distinction between the effects of the IL-4R gene Q551R, I50V polymorphisms on the risk for several manifestations of RA in different studies may result from genetic heterogeneity, size of the studied groups, and different exposure of the populations to environmental factors. What is more, since the diagnostic standard of some clinical characteristics for RA in different places are not all the same, due to the high heterogeneity, it is hard to use a method of meta-analysis to assess these associations. We only conducted a meta-analysis based on the data from the previous four studies (Prots et al., 2006; Moreno et al., 2007; Marinou et al., 2008; Hussein et al., 2012) to evaluate the association of the V allele of IL-4R gene I50V variant and RF positive in RA patients. Finally, no marked relationship was detected in our current study. Whereas the result may be possibly affected by the high heterogeneity between studies (p<0.001, I²=87.7%), it can be explained by differences in patient populations (e.g., age and year from onset, female proportion, and disease severity).

This study is the first meta-analysis to assess the associations between IL-4R gene Q551R, I50V polymorphisms with the risk of RA. Nevertheless, several limitations of this study should be discussed. First, the number of total samples in the meta-analysis is not large, especially in the analysis performed to evaluate the relations of I50V variant and RA or RF positive. Furthermore, we only conducted a meta-analysis for the European population, because of the data limitation of the studies in other ethnic groups. Therefore, more studies stratified by ethnicity are still needed in future research. Second, significant between-study heterogeneity was identified in the meta-analysis on the relationship of IL-4R gene I50V polymorphisms with RF in RA patients. However, the heterogeneity had no substantial influence on the overall effect according to the results of the sensitivity analysis. Third, as the limitation of the genotypes data in two studies, genetic models (dominant and recessive model) in the meta-analysis for IL-4R I50V and RF were not performed. Fourth, our literature searching depended on English and Chinese, language bias might be considered.

In conclusion, the present study suggests that the IL-4R gene I50V polymorphism might contribute to the susceptibility to RA. However, there is still not enough evidence to indicate the association of the Q551R variant and the development of RA or V allele of IL-4R gene I50V polymorphisms and RF. To reach a definitive conclusion, further gene-gene and gene-environment interaction studies based on a larger sample size and stratified by ethnicity are still needed.

Footnotes

Acknowledgments

This work was partly supported by grants from the key program of the National Natural Science Foundation of China (30830089) and the Anhui Provincial Natural Science Foundation (11040606M183).

Author Disclosure Statement

The authors declare that there are no conflicts of interest.

References

Aithal

, Day

, Leathart

et al. 2001. Association of single nucleotide polymorphisms in the interleukin-4 gene and interleukin-4 receptor gene with Crohn's disease in a British population. Genes Immun, 2:44-47.

Brennan

, Beech

. 2007. Update on cytokines in rheumatoid arthritis. Curr Opin Rheumatol, 19:296-301.

Bronson

, Criswell

, Barcellos

. 2008. The MHC2TA-168A/G polymorphism and risk for rheumatoid arthritis: a meta-analysis of 6861 patients and 9270 controls reveals no evidence for association. Ann Rheum Dis, 67:933-936.

Bukhari

, Lunt

, Harrison

et al. 2002. Rheumatoid factor is the major predictor of increasing severity of radiographic erosions in rheumatoid arthritis. Arthritis Rheum, 46:906-912.

DerSimonian

, Laird

. 1986. Meta-analysis in clinical trials. Control Clin Trials, 7:177-188.

Egger

, Davey Smith

, Schneider

et al. 1997. Bias in meta-analysis detected by a simple, graphical test. BMJ, 315:629-634.

Genevay

, Di Giovine

, Perneger

et al. 2002. Association of interleukin-4 and interleukin-1B gene variants with Larsen score progression in rheumatoid arthritis. Arthritis Rheum, 47:303-309.

Higgins

, Thompson

. 2002. Quantifying heterogeneity in a meta-analysis. Stat Med, 21:1539-1558.

Hussein

, El-Tarhouny

, Mohamed

et al. 2012. Association of interleukin-4 receptor gene polymorphisms with rheumatoid arthritis in Egyptian female patients. Joint Bone Spine, 79:38-42.

10.

Hytönen

, Löwhagen

, Arvidsson

et al. 2004. Haplotypes of the interleukin-4 receptor alpha chain gene associate with susceptibility to and severity of atopic asthma. Clin Exp Allergy, 34:1570-1575.

11.

John

, Myerscough

, Marlow

et al. 1998. Linkage of cytokine genes to rheumatoid arthritis. Evidence of genetic heterogeneity. Ann Rheum Dis, 57:361-365.

12.

Kruse

, Japha

, Tedner

et al. 1999. The polymorphisms S503P and Q576R in the interleukin-4 receptor alpha gene are associated with atopy and influence the signal transduction. Immunology, 96:365-371.

13.

MacKay

, Eyre

, Myerscough

et al. 2002. Whole-genome linkage analysis of rheumatoid arthritis susceptibility loci in 252 affected sibling pairs in the United Kingdom. Arthritis Rheum, 46:632-639.

14.

Mantel

, Haenszel

. 1959. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst, 22:719-748.

15.

Marinou

, Till

, Moore

et al. 2008. Lack of association or interactions between the IL-4, IL-4R alpha and IL-13 genes, and rheumatoid arthritis. Arthritis Res Ther, 10:R80.

16.

Mitsuyasu

, Izuhara

, Mao

et al. 1998. Ile50Val variant of IL-4R upregulates IgE synthesis and associates with atopic asthma. Nat Genet, 19:119-120.

17.

Mitsuyasu

, Yanagihara

, Mao

et al. 1999. Cutting edge: dominant effect of Ile50Val variant of the human IL-4 receptor alpha-chain in IgE synthesis. J Immunol, 162:1227-1231.

18.

Moreno

, González

, Saaibi

et al. 2007. Polymorphisms in the IL4 and IL4RA genes in Colombian patients with rheumatoid arthritis. J Rheumatol, 34:36-42.

19.

Nelms

, Keegan

, Zamorano

et al. 1999. The IL-4-receptor: signaling mechanisms and biologic functions. Annu Rev Immunol, 17:701-738.

20.

Peng

, Zhou

, Xu

et al. 2012. Association of PTPN22 C1858T Polymorphism and Type 1 Diabetes: a Meta-analysis. Immunol Invest, 41:484-496.

21.

Perdigones

, Lamas

, Vigo

et al. 2009. 6q23 polymorphisms in rheumatoid arthritis Spanish patients. Rheumatology (Oxford), 48:618-621.

22.

Petitti

. 1994. Meta-Analysis, Decision Analysis, and Cost-Effectiveness Analysis. Oxford University Press: New York.

23.

Prots

, Skapenko

, Wendler

et al. 2006. Association of the IL4R single-nucleotide polymorphism I50V with rapidly erosive rheumatoid arthritis. Arthritis Rheum, 54:1491-1500.

24.

Sandford

, Chagani

, Zhu

et al. 2000. Polymorphisms in the IL4, IL4RA and FCERIB genes and asthma severity. J Allergy Clin Immunol, 106:135-140.

25.

Shirakawa

, Deichmann

, Izuhara

et al. 2000. Atopy and asthma: genetic variants of IL-4 and IL-13 signalling. Immunol Today, 21:60-64.

26.

Skapenko

, Niedobitek

, Kalden

et al. 2003. The Th2 cytokines IL-4 and IL-10 are internal controllers of human Th1-biased immunity in vivo. Arthritis Res Ther, 5:88.

27.

Suppiah

, Rooney

, Vandenbroeck

. 2006. Polymorphisms in the interleukin-4 and IL-4 receptor genes modify risk for chronic inflammatory arthropathies in women. Exp Mol Pathol, 81:239-244.

28.

Wellcome Trust Case Control Consortium. 2007. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature, 447:661-678.

29.

Youn

, Hwang

, Cho

et al. 2000. Association of the interleukin-4 receptor alpha variant Q576R with Th1/Th2 imbalance in connective tissue disease. Immunogenetics, 51:743-746.