Breast Cancer Research and Treatment

, Volume 131, Issue 1, pp 27–31

The associations between two polymorphisms in the interleukin-10 gene promoter and breast cancer risk

  • Ke-Da Yu
  • Ao-Xiang Chen
  • Chen Yang
  • Lei Fan
  • A-Ji Huang
  • Zhi-Ming Shao
Preclinical study

DOI: 10.1007/s10549-010-1133-3

Cite this article as:
Yu, KD., Chen, AX., Yang, C. et al. Breast Cancer Res Treat (2012) 131: 27. doi:10.1007/s10549-010-1133-3

Abstract

The association between single-nucleotide polymorphisms (SNPs) in the interleukin-10 (IL-10) gene promoter and breast cancer risk is still ambiguous. We here performed a meta-analysis based on the evidence currently available from the literature to make a more precise estimation of the relationship between two genetic variants in the IL-10 gene promoter, −1082A > G (rs1800896) and −592C > A (rs1800872), and breast cancer. Odds ratios (ORs) with 95% confidence intervals (CIs) were used to assess the corresponding strengths of association under the codominant, dominant, and recessive models. A total of ten studies (4,181 cases and 4,384 controls) were eligible for meta-analysis. There were six studies with 3,032 cases and 3,190 controls for rs1800872, and eight studies with 1,636 cases and 1,670 controls for rs1800896. Meta-analysis showed that neither of the two polymorphisms had any association with increased breast cancer risk (for rs1800896: OR = 1.060, 95% CI = 0.785–1.432 in the dominant model, and OR = 1.152, 95% CI = 0.958–1.386 in the recessive model; and for rs1800872: OR = 0.952, 95% CI = 0.859–1.056 in the dominant model, and OR = 0.892, 95% CI = 0.741–1.072 in the recessive model). The results did not change when the analyses were restricted in Caucasians, or in the studies fulfilling Hardy–Weinberg equilibrium, or according to source of controls. In outlier analysis, no individual study affected the overall OR dominantly, since omission of any single study made no material huge difference. In conclusion, the present meta-analysis suggests a lack of association between the two SNPs (rs1800896 and rs1800872) in the IL-10 gene promoter and breast cancer risk. Further studies, either with larger sample size or regarding other SNPs/haplotypes within the IL-10 gene, are needed to clarify the role of IL-10 in breast carcinogenesis.

Keywords

Interleukin-10 rs1800896 rs1800872 Breast cancer Meta-analysis 

Introduction

For most sporadic breast cancers, a substantial component of risk may be determined by multiple low-penetrance susceptibility genes [1]. It has also been widely accepted that common variants within genes involving breast carcinogenesis-related pathways are candidate loci for breast cancer susceptibility [2]. Multifunctional cytokines are related to the development of inflammatory and immunological responses which play a crucial role in the pathogenesis of autoimmune and malignant diseases [3, 4, 5]. Some single nucleotide polymorphisms (SNPs) within specific cytokine genes might influence the expression and/or activity of encoding proteins, and thereby might make the host predispose to certain cancer [3, 6, 7]. Among cytokines, interleukin-10 (IL-10), as an immune response modulator, is a multifunctional cytokine with both immunosuppressive and anti-angiogenic functions, and consequently has both tumor-promoting and tumor-inhibiting properties under different circumstances [7]. Some clinical researches, though not all, had indicated that serum IL-10 levels might be a useful marker for monitoring the prognosis of breast cancer patients [7, 8].

The IL-10 gene comprises five exons and is located on chromosome 1q31–32. The promoter region contains at least 40 polymorphic sites according to dbSNP (http://www.ncbi.nlm.nih.gov/sites/snp), seven of which have been validated by HapMap project (http://www.hapamap.org). Polymorphisms in the promoter region have been subjected to the greatest scrutiny, particularly with regard to possible influences on gene transcription [9]. The SNPs within the IL-10 gene promoter include rs1800896 (−1082A > G), rs1800871 (−819C > T), and rs1800872 (−592C > A). It was reported that SNP rs1800896 (−1082A > G) and haplotype rs1800896_rs1800871_rs1800872 (−1082_−819_−592) were associated with differential IL-10 expression in vitro, with the −1082A_−819T_−592A haplotype associated with decreased IL-10 expression, compared with the −1082G_−819C_−592C haplotype [10, 11]. The homozygous AA genotype of rs1800896 (−1082A > G) which was related to a lower IL-10 expression, was implied to be associated with several cancers, including melanoma, prostate cancer, and breast cancer [7].

To date, several studies have shown the possible involvement of IL-10 in the pathogenesis of breast cancer, but the results are yet undetermined. To our best knowledge, rs1800896 and rs1800872 are the most extensively studied polymorphisms in the IL-10 gene in breast cancer susceptibility. To conclude, we performed the present meta-analysis to evaluate the association between these two polymorphisms and breast cancer risk.

Methods

Publication search

We searched the Medline, PubMed, and Web of Science databases (updated to June 20, 2010) using the following search terms: (“interleukin-10” or “IL-10’’ or “IL10”) and “breast’’. Eligible studies were retrieved and examined carefully. The references were checked as well for other relevant publications. Review articles were also inspected to find additional eligible studies. Only studies published in English were included in this study; we did not define any minimum number of subjects to be included. For overlapping studies, only the ones with the largest sample number were included.

Eligible studies and data extraction

The articles thus identified were assessed independently by two of the authors (K.D.Yu and L.Fan), and any discrepancy in studies’ eligibility were adjudicated by Professor Z.M.Shao. The inclusion criteria were as follows: (i) evaluation of the association between rs1800896 and/or rs1800872 and breast cancer risk (cancer patients versus cancer-free controls), (ii) retrospective case–control studies or prospective cohort studies, and (iii) having available odds ratio (OR) with its 95% confidence interval (95% CI) of studied polymorphisms, or with sufficient available genotyping data to estimate corresponding parameters. Studies included in meta-analysis should meet all the above criteria. Any study with wrong data or inconsistent data was excluded. The following variables were extracted from each study if available: first author’s surname, publication year, source of controls (hospital-based or population-based), ethnicity, and number of cases and controls in different genotypes of studied polymorphisms whenever possible. Information was carefully extracted from all the eligible publications, independently by two of the authors (K.D.Yu and L.Fan). Disagreement was resolved by discussion between the authors. If they could not reach a consensus, then another investigator (Z.M.Shao) adjudicated over the disagreement.

Statistical methods

Meta-analysis was mainly performed as described previously [12, 13, 14, 15, 16, 17]. In brief, for each study, the OR with its 95% CI was calculated to assess the association strength. The pooled OR was calculated by a fixed-effects model (using the Mantel–Haenszel method) or a random-effect model (using the DerSimonian and Laird method) according to the heterogeneity among studies. If P value of heterogeneity <0.10, then the between-studies heterogeneity was considered to be significant, and we chose the random-effects model to calculate the pooled OR; otherwise, the fixed-effects model was employed. Four different types of OR were calculated: (i) Aa genotype versus AA genotype, (ii) aa genotype versus AA genotype, (iii) aa + Aa genotypes versus AA genotype (the dominant model), and (iv) aa genotype versus Aa + AA genotypes (the recessive model). For each study, Hardy–Weinberg equilibrium (HWE) was also evaluated using the goodness-of-fit chi-square test. Departure from HWE was evaluated in the control population with the same ethnicity (P < 0.05 indicating a departure from HWE), but a deviation from HWE in a mixed control population was allowed [18, 19]. We performed subgroup analysis in subpopulations according to their ethnicity, or in studies fulfilling HWE, or by source of control populations. The potential publication bias was examined visually in a funnel plot of log (OR) against its standard error (SE), and the degree of asymmetry was tested using Egger’s test (P < 0.05 considered to be statistically significant). We also performed sensitivity analysis by omitting each study to find potential outliers. All of the statistical analyses were performed using Stata/SE version 10.0 (Stata Corporation, College Station, TX).

Results

A total of ten studies (4,181 cases and 4,384 controls) met the inclusion criteria (Table 1). Of those, eight studies investigated the association between rs1800896 and breast cancer risk (1,636 cases and 1,670 controls), and six studies were for rs1800872 (3,032 cases and 3,190 controls). Four studies investigated these two polymorphisms simultaneously.
Table 1

Characteristics of studies included for meta-analysis of two SNPs in the IL-10 gene promoter region

SNP

Study number

Author

Year

Control source

Ethnicity

Cases

Controls

HWE

MAF

aaa

Aaa

AAa

aa

Aa

AA

rs1800896

1

Giordani [21]

2003

HB

Caucasian (Italy)

11

54

60

16

51

33

0.615

0.415

(−1082A > G)

2

Smith [22]

2004

PB

Caucasian (UK)

39

58

32

57

120

46

0.239

0.525

 

3

Guzowski [23]

2005

HB

Mixed (USA)

12

28

10

4

12

9

1.000

0.400

 

4

Balasubramanian [24]

2006

PB

Caucasian

121

253

123

117

260

121

0.323

0.496

 

5

Onay [25]

2006

PB

Caucasian

103

205

90

71

194

107

0.308

0.452

 

6

Scola [26]

2005

HB

Caucasian (Italy)

16

40

28

21

45

40

0.206

0.410

 

7

Gonullu [27]

2007

HB

Turkey

3

22

13

1

7

16

0.834

0.188

 

8

Kong [28]

2010

HB

Chinese

1

29

285

2

35

285

0.422

0.061

     

Total

1,636

1,670

  

rs1800872

3

Guzowski [23]

2005

HB

Mixed (USA)

3

17

30

2

10

12

0.967

0.292

(−592C > A)

9

Langsenlehner [29]

2005

PB

Caucasian (Austrian)

21

210

269

36

199

261

0.818

0.273

 

6

Scola [26]

2005

HB

Caucasian (Italy)

5

30

49

12

35

59

0.067

0.278

 

7

Gonullu [27]

2007

HB

Turkey

5

17

16

4

10

10

0.586

0.375

 

10

Pharoah [30]

2007

PB

Caucasian (> 98% white)

115

679

1,251

116

764

1,338

0.610

0.225

 

8

Kong [28]

2010

HB

Chinese

119

135

61

134

131

57

0.014

0.620

     

Total

3,032

3,190

  

HB hospital-based, PB population-based, SNP single-nucleotide polymorphisms, HWE Hardy–Weinberg equilibrium, MAF minor allele frequency

a“A” denotes a major allele; “a” denotes a minor allele

Table 2 presents in detail the results of meta-analysis. For rs1800896, the women harboring the rare allele (G allele) were not associated with altered breast cancer risk either in the dominant model (pooled OR = 1.060, 95% CI: 0.785–1.432) or in the recessive model (OR = 1.152, 95% CI: 0.958–1.386). Furthermore, neither heterozygous carriers (OR = 1.026, 95% CI: 0.773–1.363) nor rare homozygous carriers (OR = 1.147, 95% CI: 0.920–1.431) had significantly different risk compared with common homozygous carriers. These results were not changed when the analysis was restricted in Caucasians. Similar results were observed in terms of rs1800872, which did not seem to be associated with breast cancer risk in any genetic model [in the codominant models: ORs (95% CIs) were 0.964 (0.865–1.074) and 0.887 (0.723–1.088); in the dominant model: OR (95% CI) was 0.952 (0.859–1.056); and in the recessive model, OR (95% CI) was 0.892 (0.741–1.072)]. We also performed subgroup analysis in studies fulfilling HWE (Table 2) as well as according to the source of controls (data not shown). The results displayed that neither rs1800896 nor rs1800872 had any association with breast cancer in sub-populations.
Table 2

Pooled ORs of the two SNPs in different models

Analysis model

rs1800896 OR (95% CI)

rs1800872 OR (95% CI)

Overall (n = 8; 1,636:1,670a)

Caucasian (n = 5; 1,283:1,324a)

Fulfilling HWE (n = 8; 1,636:1,670a)

Overall (n = 6; 3,032:3,190a)

Caucasian (n = 3; 2,667:2,844a)

Fulfilling HWE (n = 5; 2,717:2,868a)

Codominant (Het. vs. Common Hom.)

1.026 (0.773–1.363)

0.972 (0.805–1.173)

1.026 (0.773–1.363)

0.964 (0.865–1.074)

0.967 (0.863–1.083)

0.964 (0.862–1.078)

P/Phb

0.859/0.034

0.766/0.119

0.859/0.034

0.506/0.979

0.561/0.861

0.520/0.942

Codominant (Rare Hom. vs. Common Hom.)

1.147 (0.920–1.431)

1.024 (0.681–1.538)

1.147 (0.920–1.431)

0.887 (0.723–1.088)

0.912 (0.721–1.154)

0.904 (0.717–1.138)

P/Ph

0.222/0.064

0.908/0.032

0.222/0.064

0.250/0.368

0.443/0.080

0.390/0.259

Dominant (Rare Hom. + Het. vs. Common Hom.)

1.060 (0.785–1.432)

0.958 (0.706–1.300)

1.060 (0.785–1.432)

0.952 (0.859–1.056)

0.960 (0.862–1.070)

0.956 (0.859–1.065)

P/Ph

0.703/0.009

0.785/0.038

0.703/0.009

0.354/0.984

0.465/0.969

0.415/0.964

Recessive (Rare Hom. vs. Het. + Common Hom.)

1.152 (0.958–1.386)

1.144 (0.948–1.381)

1.152 (0.958–1.386)

0.892 (0.741–1.072)

0.921 (0.730–1.160)

0.913 (0.727–1.145)

P/Ph

0.132/0.358

0.160/0.150

0.132/0.358

0.223/0.303

0.484/0.057

0.430/0.208

Het. heterozygous, Hom. homozygous, OR odds ratio, CI confidence interval, HWE Hardy–Weinberg equilibrium

a Study number, number of cases and controls

b Ph denotes P value of Q test for heterogeneity test

In addition, we evaluated the publication bias using Begg’s funnel plot and Egger’s test. The funnel plots did not reveal an evident asymmetry, and the Egger’s test suggested the absence of publication bias in the dominant model (P = 0.219 for 1800896; P = 0.093 for rs1800872) and in other genetic models. Moreover, we evaluated the influence of any individual study on the overall OR of 1800896 and rs1800872, respectively. No individual study affected the overall OR dominantly, since omission of any single study made no material huge difference.

Discussion

The present meta-analysis including ten studies systematically evaluates the associations between two polymorphisms in the IL-10 gene promoter, rs1800896 and rs1800872, and breast cancer risk. The results indicate that neither rs1800896 nor rs1800872 is a conspicuous low-penetrant risk factor for developing breast cancer. In subgroup analysis, no significant association was found in Caucasian population. Our findings are consistent with most of the related studies summarized in the meta-analysis.

Although it is biologically plausible that rs1800896 which affects IL-10 levels could have influenced the susceptibility to breast cancer, the current evidence provides a null outcome. While, rs1800872, which is located in the −1082_−819_−592 (rs1800896_rs1800871_rs1800872) haplotype (which is associated with decreased IL-10 expression [10]), also has no association with breast cancer according to our meta-analysis. [AQ: The sentence, “While, rs1800872, which is located in the….” needs reconstruction as it suffers from lack of clarity. Please revert with your advice.] The potential explanation is that the effect of a single polymorphism might have a limited impact on breast cancer susceptibility than that has been anticipated. Actually, other polymorphisms at −3575, −1349, −819, +19 and two microsatellite variants (CA)n have been identified in the IL-10 gene promoter region in some populations [7]. These findings indicate a complex regulatory network governing the IL-10 gene expression. More comprehensive haplotype-based or multiple SNP-based approaches rather than a single polymorphism-based strategy may provide more precise information on genetic contribution of IL-10 to breast cancer etiology. It is regretted that the current data could make a haplotype-based meta-analysis. Thus, we could not rule out the possibility of potential risk contribution of the rs1800896_rs1800872 haplotype or other haplotypes in the IL-10 gene to breast cancer. In subgroup analysis, we consistently showed no association between breast cancer and the two SNPs. Such results, however, should be treated with caution. Relatively small samples of subpopulation and multiple hypothesis tests render the interpretation of negative results difficult. Further studies with larger sample size should be undertaken to elucidate those effects.

Some limitations of this meta-analysis should be acknowledged. First, some control populations are hospital based; such women might have benign breast disease and correspond to a potentially incremental risk of breast cancer [20]. Second, the overall outcomes are based on individual unadjusted ORs, while a more precise evaluation should be adjusted by other potentially suspected factors. Third, although the available data on rs1800872 comprise at least 3,000 cases and 3,000 controls, the total subject number of rs1800896 studies seems to be insufficient (approximately 1,600 cases and 1,600 controls) to reach a reliable conclusion. Relatively speaking, the limited number of studies also makes the results from subgroup analysis (e.g., by ethnicity, fulfilling HWE, or source of controls) less reliable. Last, but not the least, many other factors such as family history of breast cancer, menopausal status, and other SNPs might also affect the association between the two SNPs (rs1800896 and rs1800872) and susceptibility to breast cancer. However, among the ten studies for meta-analysis, few studies had provided sufficient data of the fore-mentioned factors, and we therefore had very limited information for subgroup analysis.

Despite these limitations, the current meta-analysis suggests a lack of association between two SNPs (rs1800896, rs1800872) in the IL-10 gene promoter and breast cancer risk. Further case–control studies, either with larger sample size or regarding other SNPs and haplotypes in the IL-10 gene, are needed to clarify the role of IL-10 in breast carcinogenesis.

Acknowledgments

This research is supported by grants from the National Basic Research Program of China (2006CB910501), 2009 Youth Foundation of Shanghai Public Health Bureau, 2009 Youth Foundation of Shanghai Medical College, and the National Natural Science Foundation of China (30971143, 30972936).

Conflict of interest

All authors declared no potential conflicts of interest.

Copyright information

© Springer Science+Business Media, LLC. 2010

Authors and Affiliations

  • Ke-Da Yu
    • 1
    • 2
  • Ao-Xiang Chen
    • 1
    • 2
  • Chen Yang
    • 1
    • 2
  • Lei Fan
    • 1
    • 2
  • A-Ji Huang
    • 1
    • 2
  • Zhi-Ming Shao
    • 1
    • 2
    • 3
  1. 1.Department of Breast Surgery, Cancer Center/Cancer InstituteFudan UniversityShanghaiPeople’s Republic of China
  2. 2.Department of Oncology, Shanghai Medical CollegeFudan UniversityShanghaiPeople’s Republic of China
  3. 3.Institutes of Biomedical ScienceFudan UniversityShanghaiPeople’s Republic of China

Personalised recommendations