Evidence of Association Between Methylenetetrahydrofolate Reductase Gene and Susceptibility to Breast Cancer: A Candidate-Gene Association Study in a South-Eastern European Population

Abstract

The methylenetetrahydrofolate reductase (MTHFR) gene has been proposed as a candidate gene for breast cancer (BC). However, the specific role of MTHFR polymorphisms and haplotypes has not been fully clarified and replicated. We examined the association of two common MTHFR polymorphisms (C677T and A1298C) and their haplotypes in a candidate-gene association study, involving 300 female patients with BC and 283 healthy women. The single locus analysis for the two polymorphisms revealed an association only for the C677T polymorphism [odds ratio (95% confidence interval), OR=2.05 (1.21–3.48)], but adjustment for age diminished this association [OR=1.76 (0.92–3.42)]. The menopausal status showed no significant effect in the association between the MTHFR polymorphisms and BC. The analysis of haplotypes showed an association for the C677-A1298 haplotypes (p=0.04). The available evidence from our study may support a contributory role of MTHFR polymorphisms in BC development. Future larger studies may help in elucidating the genetics of BC further.

Introduction

B reast cancer (BC) is the most common cancer among women, with an estimated incidence rate of 1 million cases per year worldwide. Although many environmental etiologic factors have been established by epidemiological studies (Madigan et al., 1995), there is evidence that genetic susceptibility can play a substantial role in developing BC (Zintzaras, 2006). High-penetrance BC susceptibility genes, such as BRCA1 and BRCA2, account for the majority of familial cancers; however, these genes contribute only a small fraction (10%) of BC cases (Milne and Benitez, 2008). The pathogenesis of sporadic BC is poorly understood and the search for low-penetrance genes by hypothesis-driven candidate-gene studies (genetic association study [GAS]) and hypothesis-free genome-wide association studies (GWAS) is ongoing (Hindorff et al., 2009).

Epidemiologic studies have indicated that folate deficiency may be related to development of several cancers, including BC (Blount et al., 1997; Kim, 1999; Sharp et al., 2002). Methylenetetrahydrofolate reductase (MTHFR) is a regulatory enzyme in folate metabolism that catalyzes the irreversible conversion of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate and directs the flux of intracellular folate toward the conversion of homocysteine to methionine at the expense of nucleotide synthesis (Frosst et al., 1995; Blount et al., 1997; Weisberg et al., 1998). The MTHFR gene is located at 1p36.3 and C677T (Ala222Val) and A1298C (Glu429Ala) are two common gene polymorphisms that have been described for this enzyme (Frosst et al., 1995; Weisberg et al., 1998). Both polymorphisms influence the activity of the enzyme, the homocysteine levels, and the plasma folate concentration with C677T having a higher effect than A1298C (Forsti et al., 2004; Chen et al., 2005).

Previous GAS that investigated the association between BC and the C677T and A1298C polymorphisms have produced controversial or inconclusive results and the replication record of these studies is relatively poor (Zintzaras, 2006). Typically, these studies involved small numbers of subjects, and they recruited different populations and used different sampling strategies (Zintzaras, 2006). Further, the implication of MTHFR polymorphism in BC development was not proven by GWAS (Easton et al., 2007; Murabito et al., 2007; Gold et al., 2008; Antoniou et al., 2010; Turnbull et al., 2010). Although meta-analysis of GAS may play a role in testing for replication and estimating the effect size efficiently, these studies did not provide strong evidence of the association status so far (Lewis et al., 2006; Zintzaras, 2006; Lissowska et al., 2007; Macis et al., 2007; Qi et al., 2010; Zhang et al., 2010). Therefore, the status of association for the two MTHFR polymorphisms remains ambiguous.

The aim of this article was to test the hypothesis of association between the two commonly investigated MTHFR polymorphisms (C677T and A1298C) in a GAS for BC in a South-Eastern European population (Greeks). Analysis of gene haplotypes was also considered (Zintzaras and Lau, 2008a).

Materials and Methods

Study population

A total of 300 cases with BC and 283 controls were recruited from the Department of Medical Oncology, University Hospital of Larissa. The same cohort of cases was also used in previous studies (Zintzaras et al., 2010). The mean age (±SD) was 59.86±11.90 and 43.55±17.73 for cases and controls respectively; all subjects were women. The postmenopausal women in cases and controls were 87% and 48%, respectively. The clinical characteristics of the cases are shown in Table 1. The cases were women treated with chemotherapy (combinations of 5-fluorouracil, anthracyclines, taxanes, and cyclophosphamide) for BC. The study was approved by University Hospital of Larissa ethics committee and all subjects provided an informed consent. Control subjects were free of chronic illness (renal, cardiovascular, mental, hepatic, endocrine disorders, or cancer) and without any history of malignancy. Controls were not blood-related to the patients. A blood sample for biochemical measurements and DNA extraction was taken from each individual.

Table 1.

Characteristics of Patients with Breast Cancer

Stage	%
I	11
II	32
III	24
IV	33
ER status
Positive	84
Negative	16
Histology
Ductal	84
Lobular	9
Mixed	3
Other	4
Grade
I	9
II	56
III	35
PR status
Positive	60
Negative	40
LNs involvement
Yes (n>0)	62
No (N0)	38

ER, estrogen receptors; DCIS, ductal carcinoma in situ; PR, progesterone receptors; LNs, lymph nodes.

Laboratory assays

Genomic DNA was extracted from whole blood using the QIAamp DNA blood kit (QIAGEN, Valencia, CA) following the manufacturer's instructions. Genotyping of each polymorphism was performed by amplification of 50 to 100 ng of genomic DNA. The primer sequences used and the laboratory conditions for genotyping (polymerase chain reaction, restriction enzymes, and agarose electrophoresis) for each MTHFR polymorphism have been previously described (Frosst et al., 1995; Weisberg et al., 1998; Zintzaras et al., 2010). The genotyping was performed by laboratory personnel blinded to clinical status.

Data analysis

The genotypic distribution in the control group was tested for conforming with the Hardy–Weinberg equilibrium (HWE) and the existence of linkage disequilibrium (LD) between the two polymorphisms in both cases and control was tested by using exact tests according to Weir (1996). The genotype distribution and dominant and recessive models of the cases were compared to the control group using a logistic model. The associations were expressed as odds ratios (ORs), unadjusted and adjusted for age, with the corresponding 95% confidence interval. The logistic modeling was performed using SPSS v11.5. The haplotype frequencies were estimated and compared using SHEsis (Shi and He, 2005). A result was considered statistically significant when p<0.05.

Results

The distributions of MTHFR C677T and A1298C genotypes according to clinical status (overall, in postmenopausal and premenopausal subjects) are presented in Table 2. Overall, significant differences were detected in the distribution of genotypes of C677T polymorphism between cases and controls (p=0.03). In contrast, there were no differences for the A1298C polymorphisms (p=0.47). Homozygous for the mutant allele had the smallest proportion in both cases and controls in each polymorphism. In both pre- and postmenopausal women, genotype was not associated with BC susceptibility for C677T (p=0.18 and p=0.76, respectively) and A1298C (p=0.70 and p=0.92, respectively).

Table 2.

Distribution (%) of the C677T and A1298C Genotypes Among Cases and Control Subjects

Overall subjects
SNP	Genotype	Cases, n=300	Controls, n=283	p-value
C677T	CC	35%	35%	0.03
	CT	50%	57%
	TT	15%	8%
A1298C	AA	43%	48%	0.47
	AC	45%	41%
	CC	12%	11%

Postmenopausal subjects
SNP	Genotype	Cases, n=261	Controls, n=135	p-value
C677T	CC	34%	34%	0.18
	CT	50%	56%
	TT	16%	10%
A1298C	AA	42%	47%	0.70
	AC	46%	42%
	CC	12%	11%

Premenopausal subjects
SNP	Genotype	Cases, n=39	Controls, n=148	p-value
C677T	CC	36%	36%	0.76
	CT	54%	57%
	TT	10%	7%
A1298C	AA	45%	49%	0.92
	AC	42%	40%
	CC	13%	11%

The p-values for testing the association between genotype distribution of each polymorphism and susceptibility to breast cancer are shown.

Table 3 shows the association results for the recessive and dominant models of the mutant alleles. For the C677T polymorphism, the recessive model showed significant association [OR=2.05 (1.21–3.48)], indicating that 677 TT homozygous have twofold risk for BC in relation to C677 carriers. However, the age-adjusted model diminished the significance [OR=1.76 (0.92–3.42)]. Moreover, the dominant model for 677 T allele produced nonsignificant association. Regarding the A1298C polymorphism, the recessive and dominant models produced nonsignificant associations in both the unadjusted and age-adjusted analysis. In the subgroup analysis by menopausal status (Table 3), both polymorphisms had nonsignificant associations for both examined models.

Table 3.

Odds Ratio and the Corresponding 95% Confidence Intervals for Testing the Association Between Breast Cancer and Methylenetetrahydrofolate Reductase C677T and A1298C Polymorphisms for the Recessive, Dominant, and Co-Dominant Model

Polymorphism	Genetic contrast	OR (95% CI)	p-Value	OR_adjusted (95% CI)	p-Value
Overall
C677T	Recessive model	2.05 (1.21–3.48)	0.01	1.76 (0.92–3.42)	0.09
	Dominant model	1.05 (0.74–1.47)	0.86	1.16 (0.75–1.81)	0.49
A1298C	Recessive model	1.08 (0.65–1.81)	0.79	0.92 (0.48–1.78)	0.81
	Dominant model	1.23 (0.89–1.70)	0.24	1.01 (0.66–1.56)	0.96
Postmenopausal subjects
C677T	Recessive model	1.80 (0.93–3.48)	0.09	1.73 (0.75–3.98)	0.20
	Dominant model	1.00 (0.64–1.55)	0.98	1.28 (0.73–2.22)	0.39
A1298C	Recessive model	1.09 (0.55–2.17)	0.86	0.97 (0.49–1.97)	0.74
	Dominant model	1.21 (0.78–1.86)	0.44	0.99 (0.55–1.76)	0.96
Premenopausal subjects
C677T	Recessive model	1.58 (0.47–5.33)	0.51	1.21 (0.30–4.80)	0.79
	Dominant model	1.03 (0.49–2.14)	0.97	1.05 (0.42–2.64)	0.91
A1298C	Recessive model	1.15 (0.40–3.31)	0.78	0.91 (0.22–4.04)	0.90
	Dominant model	1.15 (0.57–2.30)	0.73	1.12 (0.48–2.62)	0.82

The ORs adjusted for age are also shown.

OR, odds ratio; CI, confidence interval.

The controls were in HWE for A1298C polymorphism (p=0.68), whereas the controls for the C677T polymorphism did not conform with HWE (p<0.01) due to excess of heterozygotes, indicating the existence of possible unknown stratification; therefore, the results should be interpreted with caution (Zintzaras, 2010b). In both the patient and control populations, the two polymorphisms were in LD (p<0.01): for cases, D=0.82 and r ²=0.24; for controls, D=0.87 and r ²=0.20. The estimated haplotypes of the two MTHFR polymorphisms for BC cases and controls are presented in Table 4. However, the estimated frequencies of haplotype C677-A1298 were different in cases and controls (p=0.04), indicating a protective effect.

Table 4.

Estimated Haplotype Frequencies for the Two Methylenetetrahydrofolate Reductase Polymorphisms (C677T and A1298C)

	Haplotype frequency
C677T-A1298C haplotype	Cases	Controls	p-Value
CA	0.27	0.33	0.04
CC	0.33	0.31	0.53
TA	0.39	0.35	0.28
TC	0.02	0.01	0.28

The p-values for comparing each haplotype between cases and controls, and the p-value for the overall comparison of haplotypes between cases and controls are shown.

Discussion

The present study investigated whether the MTHFR C677T and A1298C polymorphisms were associated with the development of BC. Single locus analysis revealed an association for the C677T polymorphism, but the age-adjusted analysis diminished the association. However, the analysis of haplotypes showed an association for the C677-A1298 haplotype.

The conclusions reached in the present analysis of GAS were based on relatively small number of participants; therefore, any inferences regarding association have to be cautious. Typically, GAS performed by a single center has the tendency to lack the power to detect a significant association (Zintzaras and Lau, 2008b). To detect a significant (p<0.05) modest genetic effect (e.g., OR 1.2) of a polymorphism with minor allele frequency of 10% with a power >80%, a sample size of thousands of subjects would be needed (Zintzaras and Lau, 2008b). Thus, no single institution is able to provide a reasonable number of patients and collaborative studies are needed to provide sufficient power to detect significant associations. Then, the creation of large databases and consortia to facilitate the sharing of data among investigators and to undertake collaborative research would be a step forward in improving power. However, results from GAS originated from individual centers are still needed to incorporate the results in subsequent meta-analyses, providing pooled estimates of the genetic risk effects with decreased level of uncertainty (Zintzaras and Lau, 2008a). Although individual GAS may have limitations, cumulatively with meta-analyses they may point to true positives or negatives and prevent publication bias (Zintzaras and Lau, 2008a). In addition, GAS and meta-analysis may be used as a tool to test the replication validity of possible significant associations (Zintzaras and Lau, 2008a). Nevertheless, the present study may contribute to the current efforts of data sharing and joint analysis, and it may open avenues for future collaborations with other investigators in exploring other candidate genes in BC or performing GWAS (GAIN Collaborative Research Group, 2007; Zintzaras and Kitsios, 2009).

The lack of strong association for the MTHFR C677T polymorphism, at least for the age-adjusted analysis, and the lack of association for A1298C polymorphism might be due to other unidentified functional mutations that exist in the MTHFR gene (Zintzaras, 2010a) that affect the susceptibility to BC. In addition, other polymorphisms involved in folate/homocysteine pathway may affect the risk of BC. Thus, individual MTHFR polymorphisms might not be reliable markers for developing BC. However, the individual polymorphisms are in LD and interaction of the polymorphisms within haplotypes could be a major determinant of disease susceptibility than the individual polymorphism (Zintzaras and Lau, 2008a). The analysis of haplotypes is expected to be more powerful than single-marker analysis since the distribution of haplotypes is likely to be preserved during evolution (Zintzaras et al., 2007). Especially if the haplotypes are composed of markers that define mutations within functional DNA, then these haplotypes could have more of a biological role. The present analysis was focused on haplotypes defined by two commonly investigated and potentially functional MTHFR variants. Other variants and HapMap tagging polymorphisms were not considered since the aim of the study was to test the validity of the association shown by other studies (Ergul et al., 2003) in a South-Eastern European population.

The risk effect of MTHFR polymorphisms may be dependent on effect modifiers like folate intake, life style, exogenous hormones, bone density, and prolactin levels (Bailey, 2003; Zintzaras, 2006). Epidemiological studies have shown association between alcohol intake and development of BC in both pre- and postmenopausal women (Singletary and Gapstur, 2001), but increased folate intake may prevent the development of BC in women who consume alcohol (Zhang, 2004). Although a meta-analysis by Zintzaras (2006) indicated that premenopausal and postmenopausal women have different pathogenesis regarding the genetic effect of MTHFR, the present study did not replicate this finding.

Currently, none of the published GWAS in BC have shown significant association for the MTHFR gene (Easton et al., 2007; Murabito et al., 2007; Gold et al., 2008; Antoniou et al., 2010; Ingle et al., 2010; Turnbull et al., 2010). Further, the meta-analyses performed so far did not prove the role of MTHFR in developing BC (Lewis et al., 2006; Zintzaras, 2006; Lissowska et al., 2007; Macis et al., 2007; Qi et al., 2010; Zhang et al., 2010). However, it may be possible that unknown rare variants or structural variation tagged by the MTHFR polymorphisms account for the inconsistent associations of the common variants investigated by the GWAS (Kitsios and Zintzaras, 2009a, 2009b) since the role of MTHFR gene in BC susceptibility has not been proved. Nevertheless, the available evidence from GAS and GWAS cannot support a major contributory role for MTHFR gene in BC pathogenesis.

In addition to GAS and GWAS (Zintzaras and Lau, 2008a; Kitsios and Zintzaras, 2009a), microarray gene expression analyses (Zintzaras and Ioannidis, 2008) and whole genome linkage scans (Zintzaras and Ioannidis, 2005) may assist in drawing inferences regarding candidate markers in BC by examining the genomic convergence of these different data sources (Kitsios and Zintzaras, 2009b). Convergent findings may identify variants meriting priority in replication studies, limiting the plethora of false-positive results (Kitsios and Zintzaras, 2009b). Although GWAS have significantly advanced our understanding of the genetic basis of complex diseases by uncovering many variants of unprecedented biological implication, only a relative small subset of findings have been supported by independent sources of evidence (Kitsios and Zintzaras, 2009b). Nevertheless, evidence from convergence could provide an additional piece of information in determining the design of replication gene-candidate studies. By comparing the results from different methodologies, we would be able to identify variants or loci with concomitant evidence of implication. Currently, GWAS represent a superior strategy for unraveling genetic complexity, compared to other lines of evidence (Kitsios and Zintzaras, 2009b). However, the findings of GAS may be supportive in replicating existed evidence and in revealing genuine genetic effects that could merit prioritization in future studies. Although replication studies validated the initial findings of a considerable number of GWAS, there are still studies that lack of replication (Zintzaras and Lau, 2008a). Replication from different investigators and different methodologies (e.g., GAS) is essential to make sense of the mass of associations likely to result from GWAS (Storey and Tibshirani, 2003; Thomas, 2006; Zintzaras and Lau, 2008a).

In conclusion, the present GAS detected an association for MTHFR C677T polymorphism in the unadjusted analysis and revealed statistical significance for the C677-A1298 haplotype. The results of the present study should be interpreted with caution since the number of participants was relatively small and the population may have a structure. BC is a complex disease with multifactorial etiology; thus, the minor contributing pathogenetic role of MTHFR polymorphisms in conjunction with gene–gene–environment interaction cannot be totally excluded (Clayton and McKeigue, 2001). The creation of collaborative centers or consortia may help in identifying the contributing role of genetic variants by performing GAS and GWAS with adequate power (Kitsios and Zintzaras, 2009b; Zintzaras and Kitsios, 2009).

Authors' Contribution

Christos N. Papandreou was the principal investigator, he supervised the clinical and laboratory studies, recruited the subjects, and collected the blood samples and the clinical data; Elias Zintzaras, Georgios Bakalos, and Theocharis Koufakis performed the statistical analysis; Chrysa Doxani, Nikos Zdoukopoulos, Dimitris C. Ziogas, Panagiotis J. Vlachostergios, and Eleana Hatzidaki performed the genotyping; Elias Zintzaras, Chrysa Doxani, and Christos N. Papandreou drafted the article.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Antoniou

A.C.

, Wang

, Fredericksen

Z.S.

, McGuffog

, Tarrell

, Sinilnikova

O.M.

, Healey

, Morrison

, Kartsonaki

, Lesnick

et al. 2010. A locus on 19p13 modifies risk of breast cancer in BRCA1 mutation carriers and is associated with hormone receptor-negative breast cancer in the general population. Nat Genet, 42:885–892.

Bailey

L.B.

2003. Folate, methyl-related nutrients, alcohol, and the MTHFR 677C—>T polymorphism affect cancer risk: intake recommendations. J Nutr, 133:3748S–3753S.

Blount

B.C.

, Mack

M.M.

, Wehr

C.M.

, MacGregor

J.T.

, Hiatt

R.A.

, Wang

, Wickramasinghe

S.N.

, Everson

R.B.

, Ames

B.N.

1997. Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: implications for cancer and neuronal damage. Proc Natl Acad Sci USA, 94:3290–3295.

Chen

, Gammon

M.D.

, Chan

, Palomeque

, Wetmur

J.G.

, Kabat

G.C.

, Teitelbaum

S.L.

, Britton

J.A.

, Terry

M.B.

, Neugut

A.I.

, Santella

R.M.

2005. One-carbon metabolism, MTHFR polymorphisms, and risk of breast cancer. Cancer Res, 65:1606–1614.

Clayton

, McKeigue

P.M.

2001. Epidemiological methods for studying genes and environmental factors in complex diseases. Lancet, 358:1356–1360.

Easton

D.F.

, Pooley

K.A.

, Dunning

A.M.

, Pharoah

P.D.

, Thompson

, Ballinger

D.G.

, Struewing

J.P.

, Morrison

, Field

, Luben

et al. 2007. Genome-wide association study identifies novel breast cancer susceptibility loci. Nature, 447:1087–1093.

Ergul

, Sazci

, Utkan

, Canturk

N.Z.

2003. Polymorphisms in the MTHFR gene are associated with breast cancer. Tumour Biol, 24:286–290.

Forsti

, Angelini

, Festa

, Sanyal

, Zhang

, Grzybowska

, Pamula

, Pekala

, Zientek

, Hemminki

, Kumar

2004. Single nucleotide polymorphisms in breast cancer. Oncol Rep, 11:917–922.

Frosst

, Blom

H.J.

, Milos

, Goyette

, Sheppard

C.A.

, Matthews

R.G.

, Boers

G.J.

, den Heijer

, Kluijtmans

L.A.

, van den Heuvel

L.P.

et al. 1995. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet, 10:111–113.

10.

GAIN Collaborative Research Group. Manolio

T.A.

, Rodriguez

L.L.

, Brooks

, Abecasis

Collaborative Association Study of Psoriasis. Ballinger

, Daly

, Donnelly

, Faraone

S.V.

et al. 2007. New models of collaboration in genome-wide association studies: the Genetic Association Information Network. Nat Genet, 39:1045–1051.

11.

Gold

, Kirchhoff

, Stefanov

, Lautenberger

, Viale

, Garber

, Friedman

, Narod

, Olshen

A.B.

, Gregersen

et al. 2008. Genome-wide association study provides evidence for a breast cancer risk locus at 6q22.33. Proc Natl Acad Sci USA, 105:4340–4345.

12.

Hindorff

L.A.

, Sethupathy

, Junkins

H.A.

, Ramos

E.M.

, Mehta

J.P.

, Collins

F.S.

, Manolio

T.A.

2009. Potential etiologic and functional implications of genome-wide association loci for human diseases and traits. Proc Natl Acad Sci USA, 106:9362–9367.

13.

Ingle

J.N.

, Schaid

D.J.

, Goss

P.E.

, Liu

, Mushiroda

, Chapman

J.A.

, Kubo

, Jenkins

G.D.

, Batzler

, Shepherd

et al. 2010. Genome-wide associations and functional genomic studies of musculoskeletal adverse events in women receiving aromatase inhibitors. J Clin Oncol, 28:4674–4682.

14.

Kim

Y.I.

1999. Folate and cancer prevention: a new medical application of folate beyond hyperhomocysteinemia and neural tube defects. Nutr Rev, 57:314–321.

15.

Kitsios

G.D.

, Zintzaras

2009a. Genome-wide association studies: hypothesis-“free” or “engaged”? Transl Res, 154:161–164.

16.

Kitsios

G.D.

, Zintzaras

2009b. Genomic convergence of genome-wide investigations for complex traits. Ann Hum Genet, 73:514–519.

17.

Lewis

S.J.

, Harbord

R.M.

, Harris

, Smith

G.D.

2006. Meta-analyses of observational and genetic association studies of folate intakes or levels and breast cancer risk. J Natl Cancer Inst, 98:1607–1622.

18.

Lissowska

, Gaudet

M.M.

, Brinton

L.A.

, Chanock

S.J.

, Peplonska

, Welch

, Zatonski

, Szeszenia-Dabrowska

, Park

, Sherman

et al. 2007. Genetic polymorphisms in the one-carbon metabolism pathway and breast cancer risk: a population-based case-control study and meta-analyses. Int J Cancer, 120:2696–2703.

19.

Macis

, Maisonneuve

, Johansson

, Bonanni

, Botteri

, Iodice

, Santillo

, Penco

, Gucciardo

, D'Aiuto

et al. 2007. Methylenetetrahydrofolate reductase (MTHFR) and breast cancer risk: a nested-case-control study and a pooled meta-analysis. Breast Cancer Res Treat, 106:263–271.

20.

Madigan

M.P.

, Ziegler

R.G.

, Benichou

, Byrne

, Hoover

R.N.

1995. Proportion of breast cancer cases in the United States explained by well-established risk factors. J Natl Cancer Inst, 87:1681–1685.

21.

Milne

R.L.

, Benitez

2008. Current strategies in the search for low penetrance genes in cancer. Histol Histopathol, 23:507–514.

22.

Murabito

J.M.

, Rosenberg

C.L.

, Finger

, Kreger

B.E.

, Levy

, Splansky

G.L.

, Antman

, Hwang

S.J.

2007. A genome-wide association study of breast and prostate cancer in the NHLBI's Framingham Heart Study. BMC Med Genet, 8,Suppl 1:S6.

23.

, Ma

, Yang

, Fan

, Zhang

, Chen

, Zhou

, Jiang

2010. Methylenetetrahydrofolate reductase polymorphisms and breast cancer risk: a meta-analysis from 41 studies with 16,480 cases and 22,388 controls. Breast Cancer Res Treat, 123:499–506.

24.

Sharp

, Little

, Schofield

A.C.

, Pavlidou

, Cotton

S.C.

, Miedzybrodzka

, Baird

J.O.

, Haites

N.E.

, Heys

S.D.

, Grubb

D.A.

2002. Folate and breast cancer: the role of polymorphisms in methylenetetrahydrofolate reductase (MTHFR) Cancer Lett, 181:65–71.

25.

Shi

Y.Y.

, He

2005. SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res, 15:97–98.

26.

Singletary

K.W.

, Gapstur

S.M.

2001. Alcohol and breast cancer: review of epidemiologic and experimental evidence and potential mechanisms. JAMA, 286:2143–2151.

27.

Storey

J.D.

, Tibshirani

2003. Statistical significance for genomewide studies. Proc Natl Acad Sci USA, 100:9440–9445.

28.

Thomas

D.C.

2006. Are we ready for genome-wide association studies? Cancer Epidemiol Biomarkers Prev, 15:595–598.

29.

Turnbull

, Ahmed

, Morrison

, Pernet

, Renwick

, Maranian

, Seal

, Ghoussaini

, Hines

, Healey

C.S.

et al. 2010. Genome-wide association study identifies five new breast cancer susceptibility loci. Nat Genet, 42:504–507.

30.

Weir

1996. Genetic Data Analysis II: Methods for Discrete Population Genetic Data. Sinauer Associates: Sunderland, MA.

31.

Weisberg

, Tran

, Christensen

, Sibani

, Rozen

1998. A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab, 64:169–172.

32.

Zhang

, Qiu

L.X.

, Wang

Z.H.

, Wu

X.H.

, Liu

X.J.

, Wang

B.Y.

, Hu

X.C.

2010. MTHFR C677T polymorphism associated with breast cancer susceptibility: a meta-analysis involving 15,260 cases and 20,411 controls. Breast Cancer Res Treat, 123:549–555.

33.

Zhang

S.M.

2004. Role of vitamins in the risk, prevention, and treatment of breast cancer. Curr Opin Obstet Gynecol, 16:19–25.

34.

Zintzaras

2006. Methylenetetrahydrofolate reductase gene and susceptibility to breast cancer: a meta-analysis. Clin Genet, 69:327–336.

35.

Zintzaras

2010a. Genetic variants of homocysteine/folate metabolism pathway and risk of inflammatory bowel disease: a synopsis and meta-analysis of genetic association studies. Biomarkers, 15:69–79.

36.

Zintzaras

2010b. Impact of Hardy-Weinberg equilibrium deviation on allele-based risk effect of genetic association studies and meta-analysis. Eur J Epidemiol, 25:553–560.

37.

Zintzaras

, Grammatikou

, Kitsios

G.D.

, Doxani

, Zdoukopoulos

, Papandreou

, Patrikidou

2010. Polymorphisms of the endothelial nitric oxide synthase gene in breast cancer: a genetic association study and meta-analysis. J Hum Genet, 55:743–748.

38.

Zintzaras

, Ioannidis

J.P.

2005. Heterogeneity testing in meta-analysis of genome searches. Genet Epidemiol, 28:123–137.

39.

Zintzaras

, Ioannidis

J.P.

2008. Meta-analysis for ranked discovery datasets: theoretical framework and empirical demonstration for microarrays. Comput Biol Chem, 32:38–46.

40.

Zintzaras

, Kitsios

, Stefanidis

2007. Response to endothelial nitric oxide synthase polymorphisms and susceptibility to hypertension: genotype versus haplotype analysis. Hypertension, 49:E2.

41.

Zintzaras

, Kitsios

G.D.

2009. Synopsis and synthesis of candidate-gene association studies in chronic lymphocytic leukemia: the CUMAGAS-CLL information system. Am J Epidemiol, 170:671–678.

42.

Zintzaras

, Lau

2008a. Synthesis of genetic association studies for pertinent gene-disease associations requires appropriate methodological and statistical approaches. J Clin Epidemiol, 61:634–645.

43.

Zintzaras

, Lau

2008b. Trends in meta-analysis of genetic association studies. J Hum Genet, 53:1–9.