Lack of Effects of Dietary Folate Intake on Risk of Breast Cancer : An Updated Meta-analysis of Prospective Studies

Breast cancer is one of the most common malignancies that cause death among females worldwide (Ferlay et al., 2010; Shakeel et al., 2013). As a multifactorial disease, breast cancer is associated not only with environmental and hereditary factors but also dietary factors including folate, vitamin B6, vitamin B12 and alcohol (Ma et al., 2009; Gou et al., 2013; Liu et al., 2013). Folate, as a methyl donor in one-carbon metabolism, has been shown to mediate carcinogenesis by participating in DNA synthesis, repair, and methylation (Eichholzer et al., 2001; Stevens et al., 2010; Harris et al., 2012). Methionoine and vitamin B12 are also involved in onecarbon metabolism, and any change in the levels of these nutrients could affect one-carbon metabolism, effecting folate availability on this pathway (Zhang et al., 2011). Studies have investigated the association between dietary folate intake and the risk of breast cancer, but the findings are inconsistent. Although a meta-analysis (Larsson et al., 2007) published in 2007 suggested no association of dietary folate intake with the risk of breast cancer, subsequently seven publications (Cho et al., 2007; Kabat et al., 2008; Larsson et al., 2008; Maruti et al., 2009; Stevens et al., 2010; Shrubsole et al., 2011; Bassett et al., 2013) from February 2007 to 2013 presented different conclusions. Stevens et al. (2010) observed an association of folate dietary intake with decreased risk of breast cancer in postmenopausal women. A negative


Introduction
Breast cancer is one of the most common malignancies that cause death among females worldwide (Ferlay et al., 2010;Shakeel et al., 2013).As a multifactorial disease, breast cancer is associated not only with environmental and hereditary factors but also dietary factors including folate, vitamin B6, vitamin B12 and alcohol (Ma et al., 2009;Gou et al., 2013;Liu et al., 2013).
Folate, as a methyl donor in one-carbon metabolism, has been shown to mediate carcinogenesis by participating in DNA synthesis, repair, and methylation (Eichholzer et al., 2001;Stevens et al., 2010;Harris et al., 2012).Methionoine and vitamin B12 are also involved in onecarbon metabolism, and any change in the levels of these nutrients could affect one-carbon metabolism, effecting

Lack of Effects of Dietary Folate Intake on Risk of Breast Cancer: An Updated Meta-analysis of Prospective Studies
Meng Liu, Lian-Hua Cui*, Ai-Guo Ma, Na Li, Jin-Mei Piao correlation was seen in one study (Shrubsole et al., 2011) and the remaining (Cho et al., 2007;Kabat et al., 2008;Larsson et al., 2008;Maruti et al., 2009;Bassett et al., 2013) showed no association.Therefore, we performed an updated meta-analysis on a total of 1,854,013 participants and 24,620 breast cancer cases from sixteen prospective studies, to assess further the association of dietary folate intake with the risk of breast cancer.

Literature search and selection criteria
We conducted a comprehensive English literature search up to August 2013 by two independent researchers on PUBMED, MEDLINE and EMBASE databases.Search terms "dietary folate intake" or "dietary folic acid consumption" in combination with "breast cancer" or "breast neoplasm" were used.Eligible studies have to meet the following inclusion criteria: 1) prospective study design; 2) the exposure of interest was dietary folate intake; 3) number of incident breast cancer cases and total participants; 4) provided relative risk (RR) or hazard ratio(HR) with 95% confidence interval (CI).

Data extraction and quality assessment
Two independently investigators abstracted data from each eligible publication: the last name of first author, publication year, area where the study carried out, simple size and breast cancer cases, a baseline age of participants, follow-up period, RR or HR combination with 95%CI, detailed categories and quantiles of dietary folate intake, methods of estimating dietary intake, adjustments of variables in the analysis.The risk estimates also should be extracted, and which reflecting the greatest degree of control for potential confounders.
The quality of the studies was assessed by two researchers (LM and LN) using the 9-star Newcastle-Ottawa Quality Assessment Scale (NOS) (Wells et al., 2012).Detailed grading standards of the NOS for casecontrol or cohort study were listed as follows: selection (maximum score=4), comparability (maximum score=2), and exposure (case-control)/outcome (cohort) assessment (maximum score=3).A high score(≥7) out of a total of nine points indicated high study quality.

Statistical analysis
Since the incidence and mortality rate of breast cancer is relatively low, RR and HR will be approximately equal, and the measure of effect-estimates is referred to as RR in our meta-analysis (Lin et al., 2013).Cochrane Q test and Higgins I-square (I 2 ) statistics were used to assess heterogeneity among studies.The p-value of 0.1 was used for the Cochrane Q test on testing the heterogeneity, and the values of 25, 50 and 75% of I 2 statistic were used as low, moderate and high heterogeneity, respectively.Based on the test on heterogeneity, the fixed-effects model (Mantel et al., 1959) or random effects mode (DerSimonian et al., 1986) was used to obtain pooled estimates.In addition, we performed meta-regression, subgroup and sensitivity.
For the dose-response analysis, we used the generalized least-squares trend estimation (GLST) method developed by Greenland and Orsini (Greenland et al., 1992;Orsini et al., 2006).The method requires that the average categories of dietary folate dose, number of breast cancer cases, person-years or noncases, and adjusted logarithm of the RR with its SE (Greenland et al., 1992;Berlin et al., 1993).The median value of dietary folate intake in each category was assigned to the corresponding RR for each study when provided in the paper.For studies that reported the range of dietary folate, the midpoint of the interval was chosen.For the lowest category was open ended, the lowest boundary was considered to be zero.
For the open-ended upper interval, the value arbitrarily assigned was 20% higher than the low end of the interval (Berlin et al., 1993;Aune et al., 2012) .The does-response results are presented for a 220ug/d increment.
The publication bias among studies was examined with Funnel plots, Egger´s liner regression test (Egger et al., 1997) and Begg´s test (Begg et al., 1994) with a significance level of 0.1.Sensitivity analyses were conducted to assess the stability of individual studies, by excluding any single study each time.In addition, we also performed subgroup analysis based on menstrual status, hormonal status and the consumption of alcohol, methionine and vitamin B12.All statistical analyses were conducted with STATA software (version 12.0; College Station, TX).All statistical tests were two sided and considered statistically significant when <0.05.

Eligible studies and studies characteristics
Our search strategy identified fifteen prospective cohort studies and one nested case-control study, including 1854013 participants and 24620 breast cancer patients.Of those, fifteen studies contained results about dietary intake of folate.Eight studies (Zhang et al., 1999;Sellers et al., 2001;Cho et al., 2003;2007;Feigelson et al., 2003;Maruti et al., 2009;Stevens et al., 2010;Basset et al., 2013) were conducted in the United States, two articles each in Canada (Rohan et al., 2000;Kabat et al., 2008;) and Australia (Baglietto et al., 2005;Stolzenberg-Solomon et al., 2006) and the rest from China, Swedish, France and Denmark (Lajous et al., 2006;Tjønneland et al., 2006;Larsson et al., 2008;Shrubsole et al., 2011), separately.To estimate dietary folate intake, all studies used the Food Frequency Questionnaire (FFQ).The baseline characteristics of all selected studies were summarized in Table 1.The study quality score ranged from 5 to 8 according to the 9-star Newcastle-Ottawa Scale, and was ≥7 (indicating high quality) for the majority (10/16) of studies.

Dietary folate intake and breast cancer risk
All studies provided detail results on dietary folate intake.The RR or HR for the highest versus lowest   (Lajous et al., 2006;Shrubsol et al., 2011) showed remarkably inverse relationship, and eleven studies (Zhang et al., 1999;Rohan et al., 2000;Sellers et al., 2001;Cho et al., 2003;2007;n Feigelson et al., 2003;Tjønneland et al., 2006;Kabat et al., 2008;Larsson et al., 2008;Maruti et al., 2009;Bassett et al., 2013) showed no association.The summary RR for breast cancer for the highest versus lowest categories of the folate dietary intake was 0.98 (95%CI 0.90-1.05, Figure 1), indicating that no association was found between dietary folate intake and breast cancer risk.

Sensitivity analysis and heterogeneity assessment
Sensitivity analyses were performed to test the stability of the pooled results.The effect of each study on the overall meta-analysis estimate was assessed by omitting one study at a time, but the pooled RRs were always persistent, demonstrating that our results were robust.
Heterogeneity test showed there was significantly different among the including studies (I²=53.8%,p=0.007,Q=30.29), therefore, a randomized-effects model was employed to pool them to obtain the overall RR.

Publication bias
There was no significant publication bias based on funnel plot (Figure 2).Egger's and Begg's test indicated that there was not a possibility of publication bias for the relationship of dietary folate intake with breast cancer risk (p=0.568 and p=0.488, Figure 2, Table 2).

Meta-analysis results of dietary folate intake by stratification of alcohol intake
Six studies investigated combined effects of dietary folate and alcohol intake on breast cancer.Three of these studies (Zhang et al., 1999;Rohan et al., 2000;Baglietto ett al., 2005) had found marked reductions in breast cancer risk among those who consuming higher alcohol.Three other studies (Cho et al., 2003;Feigelson et al., 2003;Stevens et al., 2010) reported that the relationship between folate intake and breast cancer was not modified by alcohol intake.In our meta-analysis, there was no significant association in breast cancer risk for high versus low folate intake by alcohol stratification (Table 2).

Meta-analysis results of dietary folate intake by stratification of menstrual or hormonal status
Meta-analysis results of dietary folate intake and menstrual status, hormonal status are illustrated in Table 2. Subgroup analysis of different menopausal statuses showed the relationship of dietary folate intake with cancer risk did not differ in postmenopausal and premenopausal breast cancer patients (postmenopausal vs premenopausal RR=0.98, 95%CI 0.89 to 1.07).The same results were also observed when the stratified analyses were carried out by estrogen receptor (ER) and progesterone receptor (PR) status (Table 2).

Meta-analysis results of dietary folate by stratification of methionine or vitamin B12 intake
Three prospective studies (Stolzenberg-Solomon et al., 2006;Stevens et al., 2010;Shrubsole et al., 2011) have examined whether the association between folate intake and risk of breast cancer is modified by methionine intake.Two prospective studies (Lajous et al., 2006;Shrubsole et al., 2011) have evaluated the association between dietary folate intake and risk of breast cancer by strata of intakes of vitamin B12.In our meta-analysis, there were no significant interactions between dietary folate, methionine and vitamin B12 intake (Table 2).

Discussion
Our meta-analysis was based on prospective cohort studies evaluating the relationship of dietary folate intake

56
and the risk of breast cancer.We found no evidence to support the association of dietary folate exposure and the risk of breast cancer by using the random-effects model, where the pooled estimate for the highest versus the lowest exposure level was 0.98 (95%CI: 0.90-1.05).We further observed that there was no association in subgroup analysis of menstrual status, hormonal status, the consumption of alcohol, methionine or vitamin B12.In addition, the result from dose-response analysis showed that dietary folate intake in increments of 220 μg/day was not associated with breast cancer risk Folate has a critical role in DNA methylation (Kim et al., 2004;Nazki et al., 2014).Low folate intake might alter DNA methylation and thereby affect gene expression, DNA integritty and stability (Ma et al., 2009).Folate may also mediate carcinogenesis by an alternative pathway.A form of folate, 5,10-methylene tetrahydrofolate, is a methyl donor that plays an important role in the conversion of dUMP to dTMP.Failure to synthesis dTMP will lead to nucleotide deficiency, and in turn result in inappropriate incorporation of uracil into DNA in place of thymidine, resulting in DNA strand breaks.It has been hypothesized that low dietary folate intake might be associated with breast cancer, affecting the methylation of the ER receptor, which might have an influence on silencing genes (Zhu et al., 1998;Zhang et al., 2005;Gou et al., 2013).However, our meta-analysis result showed that dietary folate intake was not significantly associated with the risk of breast cancer.In fact, a similar result was also seen in metaanalysis studies on the association between dietary folate intake and other malignancies risk, such as ovarian cancer (Li et al., 2013), pancreatic cancer (Bao et al., 2011), and lung cancer (Cho et al., 1999).
Vitamin B12, as cofactors, and methionine may affect carcinogenesis due to their critical roles in the one-carbon metabolism pathway, which plays an important role in DNA synthesis, methylation, and repair.They may also influence folate metabolism and its physiologic effects (Harris et al., 2012).In addition, menstrual status and hormonal status are known risk factors for breast cancer.Further stratified analyses were conducted by menstrual status, hormonal status and the consumption of alcohol, methionine and vitamin B12.We found that those stratified factors didn't change the association of dietary folate intake with breast cancer risk, Alcohol is likely to affect folate methylation pathways by promoting the degradation, inhibiting the absorption, and increasing the excretion of folate (Kato et al., 1999).Thus, we conducted alcohol stratification analysis based on six prospective studies, and the results indicated that no significant association between high versus low dietary folate intake and breast cancer risk.Similarly, Flood et al. (2002) reported that alcohol consumption couldn't modify the relationship between dietary folate intake and the risk of colorectal cancer.However, a previous metaanalysis (Larsson et al., 2007) in 2007, only including two prospective studies, indicated that high folate intake was associated with a statistically significant decreased risk of breast cancer among women with moderate or high alcohol consumption, but not among women with low or no alcohol consumption.As these results are inconsistent, large prospective studies are warranted to clarify further the interaction of alcohol and dietary folate consumption and the risk of breast cancer.
Begg's and Egger's tests were used to detect the potential publication bias, and no significant discrepancy was seen from the meta-analysis.Our study, consisting of 1,854,013 participants and 24,620 breast cancer patients, included studies that were based on a prospective cohort; therefore, the conclusions are highly credible.
In conclusion, the current meta-analysis demonstrated that there was no association between dietary folate intake and the risk of breast cancer.Also, no differences were observed in the interactions between dietary folate intake and menstrual status, hormonal status and the consumption of alcohol, methionine or vitamin B12 on the risk of breast cancer.Further prospective studies are essential to confirm the observed results.

Figure 1 .
Figure 1.Forest Plot for the Meta-Analysis of the Association Between Dietary Folate Intake and the Risk Breast Cancer(Highest Category vs Lowest Category)

Table 1 . Characteristics of Prospective Studies Included in the Meta-Analysis of Dietary Foate Intake and Breast Cancer Risk
DOI:http://dx.doi.org/10.7314/APJCP.2014.15.5.2323Lack of Dietary Folate Intake Effects on Risk of Breast Cancer: An Updated Meta-analysis of Prospective Studies , length of follow-up, alcohol intake, total energy intake, age at first birth, history of breast cancer in mother or sister, history of begin breast cancer, BMI at age 18 years, height in cm, age at menopause, postmenopausal hormone, use , weight change from age 18 years.age, practice of breast self-examination, number of live births, energy intake, age at menarche, menopausal status, randomization group, study center alcohol consumption, family history of breast cancer in a first-degree relative., family history of breast cancer, age at menarche, age at menopause, oral contraceptive use, age at first birth, parity, hormone replacement therapy, waist-to-hip ratio, BMI, BMI at age 18, smoking, physical activity, other B vitamins, alcohol, height., body mass index , animal fat, parity, age at menarche, alcohol intake, energy, family history of breast cancer, contraceptive use, age at first birth ,history of benign breast disease., dietary folate, methionine, race, age at first birth, age at menopause,history of breast lump, education, multivitamin use, parity, energy, first degree family history of breast cancer, physical activity, HRT use, age at menarche mammographic history, adult weight gain, BMI., mammography screening history , birth control pill use history of benign breast disease , age at menarche, age at first birth ,number of live births, family history of breast cancer, age at menopause, hormonere placement therapy., history of benign breast disease, family breast cancer, age at menarche, parity, age at first birth, vitamin supplement use, breastfeeding, years since last use of oral contraceptives, alcohol intake, BMI, years of hormone replacement therapy use, age at menopause, regular mammographic evaluation., parity , age at first birth, BMI, history of benign breast disease, family history of breast cancer, age at menarche, animal fat, oral contraceptive use, intakes of alcohol ,energy.BMI, smoking, education, oral contraceptive use, menopausal status, intake of calories, family history of breast cancer, history of breast biopsy, age at menarche, parity, hormone replacement therapy, alcohol intake., body mass index, height, parity, age at first birth, age at menarche, age at menopause, use of oral contraceptives, use of postmenopausal hormones, family history of breast cancer, history of benign breast disease, intakes of alcohol and total energy.,family history of breast cancer, mammography within 2 year, preceding baseline, history of breast biopsy, age at menarche , age at first birth, age at menopause, years of combined estrogen and progestin PMH, BMI , total physical activity , alcohol intake in the past year .

Funnel Plot Analysis and Begg's Test for the Publication Bias
DOI:http://dx.doi.org/10.7314/APJCP.2014.15.5.2323Lack of Dietary Folate Intake Effects on Risk of Breast Cancer: An Updated Meta-analysis of Prospective Studies