Many studies have established that T-lymphocyte antigen-4 (CTLA4) is a susceptible gene for Graves’ disease (GD). Also many studies showed the association between the CTLA4 exon-1 49A/G polymorphism and the risk of developing Graves’ ophthalmopathy (GO) in GD patients. But those results were inconsistent. In recent years many new studies were published which helped to shed light on the relationship of CTLA4 SNP49 with GO. So we performed the meta-analysis to explore the association between the SNP49 and GO susceptibility in GD patients. Studies up to February 29, 2012, were searched by using PubMed. The odds ratio was used to evaluate the strength of the association. Altogether 12 case-control studies involving 2,505 participants were included in the meta-analysis. Results showed that the G allele was related to the increased risk of GO compared with the A allele under allelic genetic model (OR = 1.14, 95% CI: 1.14–1.72,
Graves’ disease is a thyroid autoimmune disorder with 25–50% individuals having ophthalmopathy which is called thyroid-associated ophthalmopathy (TAO) or GO. The clinical manifestations of GO can mostly be explained by the increased volume of extraocular muscles and orbital connective tissues [
Human CTLA-4 gene, located on 2q33, encodes a molecule which plays an important role in the downregulation of CD28 interaction with the ligands on the surface of antigen-presenting cells (APCs). The important inhibitory role of CTLA-4 in T-cell function has made it be one candidate gene when exploring autoimmune diseases.
Many reports have established the association of CTLA-4 with GD [
One meta-analysis exploring the association of CTLA-4 gene polymorphism with GO was performed in 2006 which found no definite result [
We searched PubMed up to February 29, 2012, for all possible publications on the association between the CTLA-4 49A/G polymorphism and GO. Search terms were used as follows: “Cytotoxic T-lymphocyte associated antigen-4 or CTLA4,” “Graves’ disease,” and “ophthalmopathy.” We also manually searched all reference lists of the relevant articles for additional papers. Language was limited to English. All eligible studies met the following criteria: (1) case-control design investigating the associations of CTLA4 49A/G polymorphism and GO; (2) providing genotype distribution information in both cases and controls or provided odds ratio (OR) with 95% confidence interval (95% CI) (or sufficient data that allowed us to calculate these). When a study reported results on different subpopulations, we treated each subpopulation as a separate comparison in the meta-analysis. For overlapping data, the most complete or recent study was included.
Two authors (Pengfei Du and Changjiang Wang) independently extracted data and reached consensus on all items. The following information was sought from each report: authors, journal and year of publication, country of origin, selection and characteristics of GO cases and controls, demographics, ethnic group of the study population, eligible and genotyped cases and controls, and number of cases and controls for the genotype. For studies including subjects of different ethnic groups, data were extracted separately for each ethnicity whenever possible.
The primary analysis compared GO cases with controls for the frequency of G versus A alleles. This analysis aims to detect overall differences. Pooled OR with 95% CI was used to evaluate the strength of the associations of CTLA-4 49A/G polymorphism and GO. When zero events occurred, we treated this problem by adding 0.5 to all the 2
The primary literature search yielded 17 papers [
General characteristics of the selected studies in the meta-analysis.
Ethnic | Country | SNP genotyping | Grouping method | Eligible subjects | First author (year) [reference] | ||
---|---|---|---|---|---|---|---|
GO | Non-GO | GO | Non-GO | ||||
European | USA | RFLP | NOSPEC3–6 | No clinical feature | 85 | 52 |
Villanueva (2000) [ |
UK | RFLP | NOSPEC3–6 | NOSPEC < 3 | 161 | 323 | Allahabadia (2001) [ | |
Slovenia | RFLP | Clinical features | No clinical feature | 33 | 34 | Zaletel (2002) [ | |
Iran | RFLP | NOSPEC3–6 | NOSPEC < 3 | 105 | 100 | Esteghamati (2009) [ | |
Poland | RFLP | NOSPEC3–6 | No clinical feature | 95 | 169 | Bednarczuk (2003) [ | |
UK | RFLP | NOSPEC3–6 | NOSPEC < 3 | 124 | 168 | Vaidya (2003) [ | |
Poland | Minisequencing | NOSPEC3–6 | NOSPEC < 3 | 50 | 49 | Frydecka (2004) [ | |
Italy | RFLP | NOSPEC2–6 | NOSPEC < 3 | 90 | 60 | Petrone (2005) [ | |
|
|||||||
Asian | China | DCFH | NOSPEC3–6 | NOSPEC < 3 | 142 | 119 | Han (2006) [ |
China | RFLP | NOSPEC3–6 | No clinical feature | 33 | 56 | Zhang (2006) [ | |
China | RFLP | NOSPEC2 | NOSPEC3, 4 | 34 | 22 | Chong (2008) [ | |
Japan | RFLP | NOSPEC3–6 | No clinical feature | 99 | 220 | Bednarczuk (2003) [ | |
Korea | RFLP | Clinical feature | No clinical feature | 26 | 47 | Park (2000) [ |
Distribution of CTLA4 49A/G alleles among GO and controls in the included studies.
First author | Country | A/A ( |
A/G ( |
G/G ( |
|||
---|---|---|---|---|---|---|---|
GO | Non-GO | GO | Non-GO | GO | Non-GO | ||
Villanueva (2000) [ |
USA | 26 | 16 | 42 | 25 | 17 | 11 |
Allahabadia (2001) [ |
UK | 53 | 83 | 82 | 180 | 26 | 60 |
Zaletel (2002) [ |
Slovenia | 11 | 14 | 17 | 17 | 5 | 3 |
Esteghamati (2009) [ |
Iran | 48 | 66 | 43 | 28 | 14 | 6 |
Bednarczuk (2003) [ |
Poland | 27 | 48 | 41 | 82 | 27 | 39 |
Vaidya (2003) [ |
UK | 27 | 61 | 62 | 77 | 40 | 34 |
Frydecka (2004) [ |
Poland | 11 | 21 | 29 | 21 | 10 | 7 |
Petrone (2005) [ |
Italy | NA | NA | NA | NA | NA | NA |
Han (2006) [ |
China | 18 | 14 | 55 | 39 | 69 | 66 |
Zhang (2006) [ |
China | 1 | 1 | 11 | 18 | 21 | 37 |
Chong (2008) [ |
China | 1 | 0 | 9 | 8 | 24 | 14 |
Bednarczuk (2003) [ |
Japan | 12 | 16 | 41 | 99 | 46 | 105 |
Park (2000) [ |
Korea | 1 | 3 | 10 | 15 | 15 | 29 |
NA: not available.
A total of 12 studies concerning the 49A/G polymorphism included 2,505 individuals (1,082 cases and 1423 controls). The pooled frequency of G allele was 68.1% and 51.8% for cases and controls, respectively. Two studies suggested an at-risk effect of G allele [
Meta-analysis for the effect of the G allele versus the A allele on the risk of GO in GD patients. Each comparison is presented by the name of the first author and the year of publication. The point estimate of the odds ratio and the accompanying 95% confidence interval (CI) are shown. “Total” represents the summary random effects estimation for the comparison along with the respective 95% confidence interval. Values above 1 denote an increased risk for GO with the G allele.
Meta-analysis for the effect of the GG versus the AA genotype on the risk of GO in GD patients in European subgroup. Each comparison is presented by the name of the first author and the year of publication. The point estimate of the odds ratio and the accompanying 95% confidence interval (CI) are shown. “Total” represents the summary random effects estimation for the comparison along with the respective 95% confidence interval. Values above 1 denote an increased risk for GO with the GG genotype.
Meta-analysis for the effect of the CTLA4 49A/G allele in G carriers and in A carriers on the risk of GO in GD patients. Each comparison is presented by the name of the first author and the year of publication. The point estimate of the odds ratio and the accompanying 95% confidence interval (CI) are shown. “Total” represents the summary random effects estimation for the comparison along with the respective 95% confidence interval. Values above 1 denote an increased risk for GO with the G allele.
The current meta-analysis including 12 case-control studies was in an effort to clarify the relationship between CTLA-4 gene polymorphism and GO susceptibility. The overall results indicated that the CTLA-4 49A/G polymorphism was associated with susceptibility of GO which was more significant in European population. Although obvious heterogeneity was detected for 49A/G associations, sensitivity analyses did not materially alter the overall and subgroup results under different genetic models, indicating that the results were stable and reliable.
The polymorphism of CTLA-4 had been suggested in many diseases which included Addison’s disease, autoimmune hypothyroidism, and rheumatoid arthritis. The CTLA-4 49A/G SNP in exon-1 leads to the substitution of Ala with Thr in the signal peptide part which was reported to cause misprocessing of CTLA-4 in the ER resulting in less efficient glycosylation and diminished surface expression of CTLA-4 protein [
The association of CTLA-4 49A/G SNP with GO was more significant in European population but not in Asian population. The CTLA4 49A/G SNP shows the same effect on the GO development with different significance. The results are confusing but not conflicting. This can be explained by the following reasons. Firstly, both GD and GO are multigenic conditions. It is unlikely to have one major susceptibility gene contributing to the development of GO. In the studies of Xu et al., they transiently transfected a T-cell line with a CTLA-4 construct harboring either the G or the A allele of the 49A/G SNP and found no difference in CTLA-4 expression and/or function harboring the A or the G allele [
In conclusion, our analysis first intensified the association between 49A/G and GO susceptibility, which showed that the G allele is a risk factor for GO. More risk factors should be considered in future association studies of the CTLA-4 SNP with GO. The gene-gene and gene-environment interaction should also be taken into consideration in future research. The identification of the candidate gene and understanding of the mechanisms by which they cause disease will identify those individuals at risk for developing GO in the future. Treatment will evolve towards more preventive strategies. The treatment will be personalized by targeting the specific pathways that are most contributory to GO development. We believe that the future will witness the engagement of novel molecules, such as altered peptide ligands [
The authors declare that they have no financial and personal relationships with other people or organizations that can inappropriately influence their work.