Secondary Logo

Folate intake, serum folate, and risk of esophageal cancer

a systematic review and dose–response meta-analysis

Ni, Yingchuna,b; Du, Jingea,b; Yin, Xiaolina,b; Lu, Minga,b

European Journal of Cancer Prevention: May 2019 - Volume 28 - Issue 3 - p 173–180
doi: 10.1097/CEJ.0000000000000441
Review Article: Gastrointestinal Cancer
Free
SDC

The dose–response relationship between folate and the risk of esophageal cancer (EC) is not clear. To further elucidate their relationships, we carried out a dose–response meta-analysis of folate intake, serum folate, and the risk of EC. PubMed, Embase, Web of Science, and China National Knowledge Infrastructure were searched for observational studies until September 2016. Then, we carried out a systematic review and dose–response meta-analysis using Stata 14.0 software. Subgroup analyses were further carried out according to study characteristics and adjustment confounders. A total of 23 studies with a total of 3886 patients were enrolled in this study. The pooled odds ratios for EC in the highest versus the lowest levels of folate intake and serum folate were 0.64 (0.54–0.76, P<0.001) and 0.45 (0.19–1.07, P=0.071), respectively. Dose–response meta-analyses were carried out to assess associations between folate intake, serum folate, and EC risk. When serum folate is 10 μg/l higher than the lowest reference dosage (3.44 μg/l), EC decreased risk with an increase in serum folate levels. When folate intake is 50 μg/day higher than the lowest reference dosage (125.21 μg/day), the EC risk is decreased with an increase in folate intake. Finally, the results support that folate can promote public health through decreasing EC risk in a certain dosage range; otherwise, the protective effects might be reduced.

aClinical Epidemiology Unit, Qilu Hospital of Shandong University

bDepartment of Epidemiology and Health Statistics, School of Public Health, Shandong University, Jinan, China

Correspondence to Ming Lu, PhD, Clinical Epidemiology Unit, Qilu Hospital of Shandong University, Jinan 250000, China Tel: +86 531 82169034; fax: +86 531 86927544; e-mail: lvming@sdu.edu.cn

Received July 31, 2017

Accepted September 5, 2017

Back to Top | Article Outline

Introduction

Folate is considered an established protective factor for the onset of esophageal cancer (EC) (Tio et al., 2014). Although several reviews (Larsson et al., 2006; Kubo, 2010; Liu et al., 2011) have investigated the association between folate and EC risk, the study of quantitative relationships between folate levels and EC risk is still rare and unclear.

To date, only one article (Zhao et al., 2017) has reported the dose–response relationship between dietary folate intake and the risk of EC. However, there was no linear or nonlinear relationship between the serum folate levels and EC risk. Also, Xiao et al. (2014) have pointed out that folate intake was not associated with esophageal adenocarcinoma (EAC), whereas the low folate intake was associated with an increased risk of esophageal squamous cell carcinoma (ESCC). Several articles (Hong et al., 2012; Wu et al., 2012) have also described the biologic mechanisms of folate that reduce the risk of cancer. They pointed out that folate deficiency may lead to genetic mutation, chromosomal damage, and altered epigenetic modification.

The aim of this study is to further illustrate the link between folate and EC risk. Therefore, we carried out a quantitative meta-analysis to systematically identify their associations.

Back to Top | Article Outline

Methods

Databases, source, and searches

We systematically searched PubMed, Embase, Web of Science, and China National Knowledge Infrastructure databases for relative articles published from the database inception to the end of September 2016 (using following search keywords: ‘esophageal’ OR ‘esophagus’ OR ‘oesophagus’ OR ‘oesophageal’, ‘folate’ OR ‘vitamin B’ OR ‘folic acid’ OR ‘folacin’, ‘cancer’ OR ‘tumor’ OR ‘neoplasm’ OR ‘carcinoma’ with restrictions of title or abstract) and screened carefully. Only relevant articles published in English or in Chinese and studies carried out in humans were included. References of all included studies were screening carefully to identify any other additional studies. The article search was performed by two independent authors.

Back to Top | Article Outline

Study inclusion and exclusion

The inclusion criteria were as follows: (i) original studies on the association between folate and the risk of EC; (ii) case–control study or cohort study; (iii) reported odds ratio (OR) or relative risk (RR) estimates with 95% confidence intervals (CIs) or data for calculation; and (iv) presented at least three folate dosage levels. The exclusion criteria were as follows: (i) reviews without original data, comments, ecological studies, case reports, and editorials, and (ii) reported the outcome not as EC. If several publications involved overlapped individuals, we included the latest published study.

Back to Top | Article Outline

Data extraction and quality assessment

From each eligible study, we extracted the following information: first author’s last name, year of publication, country of the population studied, study period, age, source of controls, outcome of cases, folate type, OR with their 95% CIs, adjusted variables adjusted in the multivariable model, and the Newcastle–Ottawa Quality Assessment Scale (NOQAS) (Table 1). To reduce the risk of possible unmeasured confounders, we extracted OR and RR estimates from the maximally adjusted model. NOQAS (Stang, 2010) was used to assess the quality of the included studies based on the selection of the study groups, the comparability of the groups, and the ascertainment of exposure and outcome for case–control study and cohort study. The score ranged from 0 (as poor) to 9 (as excellent). The present work followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement (Knobloch et al., 2011).

Table 1

Table 1

Table 1

Table 1

Back to Top | Article Outline

Statistical analysis

Stata MP 14.0 software (StataCorp., College Station, Texas, USA) was used for statistical analysis. We quantified the relationship between folate and EC risk by pooling the ORs for the highest compared with the lowest category. Subgroup analyses were further carried out, if feasible, according to area, study design, outcome of cases, study quality, and adjustment for confounders (smoking, alcohol, BMI, energy intake, and education). Sensibility analysis was carried out to validate whether the result is robust. Heterogeneity was assessed using the Higgins I2 statistics. If heterogeneity was significantly small (I2<50%), the fixed-effect model was adopted as the pooling method; otherwise, the random-effect model was used. Publication bias was tested using Begg’s test (Stata METABIAS command). All tests were two sided, with a significance level of 0.05.

Moreover, we carried out a dose–response meta-analysis using the median or mean folate levels and the adjusted natural log of the RRs or ORs with their SE. If folate intake was reported by range, we assigned the midpoint of the upper and lower boundaries in each category as the average intake. If the highest category was open-ended, we considered the width of the category to be the same as that of the adjacent category. If the lowest category was open-ended, the lowest boundary was set to 0 (Aune and Norat, 2011; Norat et al., 2002; Wu et al., 2013) To derive the dose–response curve, we modeled folate using restricted cubic splines with four knots at the 5th, 35th, 65th, and 95th percentiles of the distribution (Mei-Ling et al., 2014). We included studies for this dose–response analysis only if they reported the distributions of cases and persons or person-years, as well as the ORs (RRs) and 95% CI with the variance estimates for at least three quantitative exposure categories.

Back to Top | Article Outline

Results

Figure 1 shows the literature research and selection. After exclusion of duplicates and studies that did not fulfill the inclusion criteria, 20 articles (Brown et al., 1988; Zhang et al., 1997; Franceschi et al., 2000; Mayne et al., 2001; Chen et al., 2002; Yang et al., 2005; de Stefani et al., 2006; Galeone et al., 2006; Aune et al., 2011; Ibiebele et al., 2011; Jessri et al., 2011; Ting et al., 2011; Zhao et al., 2011; Tavani et al., 2012; Bao et al., 2013; Huang et al., 2013; Sharp et al., 2013; Fanidi et al., 2014; Xiao et al., 2014; Chang et al., 2015) including 23 studies were included finally, with a total of 3886 EC patients. Among these articles, two articles (Ibiebele et al., 2011; Xiao et al., 2014) reported the associations between folate and both two pathological types (EAC and ESCC). The basic characteristics of these studies are summarized in Table 1. Among the 23 studies, two cohort studies (Xiao et al., 2014) reported the correlations between folate intake and EAC risk and ESCC risk. 16 case–control studies reported the association between folate intake and EC risk and five case–control studies reported the relationship between serum folate and EC risk. The quality scores of all the articles ranged from 6 to 8 points (Supplementary Table 1, http://links.lww.com/EJCP/A187 and Supplementary Table 2, Supplemental digital content 2, http://links.lww.com/EJCP/A188).

Fig. 1

Fig. 1

After analyzing 16 case–control studies and two cohort studies that reported the link between folate intake and EC risk, the result showed significant heterogeneity (I2=57.9%) and the pooled OR of EC for the highest versus the lowest level of folate intake was 0.64 (0.54–0.76, P<0.001) (Fig. 2). The results of subgroup meta-analyses for the association between folate intake and the risk of EC in area, study design, outcome of cases, study quality, and adjustment for confounders (smoking, alcohol, BMI, energy intake, and education) showed consistency with the primary findings (Table 2). There was almost no publication bias (folate intake: Egger’s test, P=0.088; Begg’s test, P=0.081). During sensitivity analysis (Supplementary Fig. 1, Supplemental digital content 3, http://links.lww.com/EJCP/A189), we found that the study by Tavani et al. (2012) had an obvious impact on the statistical robustness of the result. After its exclusion, the pooled OR of EC for the highest versus the lowest level of folate intake was 0.67 (0.57–0.79) (I2=48.4%; publication bias: Egger’s test P=0.149, Begg’s test P=0.174). We found that none of the selected studies had a huge impact on the pooled results.

Fig. 2

Fig. 2

Table 2

Table 2

Five case–control studies (Ting et al., 2011; Huang et al., 2013; Fanidi et al., 2014; Chang et al., 2015) were analyzed on the association between serum folate and the risk of EC, showing significant heterogeneity (I2=88.7%) and the pooled OR of EC for the highest versus the lowest level of serum folate was 0.45 (0.19–1.07, P=0.071) (Fig. 2). Because there were only five studies, we did not carry out subgroup analyses. There was almost no publication bias (serum folate: Egger’s test, P=0.558; Begg’s test, P=0.142) between these five enrolled studies.

The other three studies (Zhang et al., 1997; Bollschweiler et al., 2002) were not included because the numbers of the dosage groups in both case and control groups were unavailable. Finally, nine folate intake studies and four serum folate studies were included in the dose–response meta-analysis. Figure 3 shows an inverse relationship between serum folate levels and EC risk. When serum folate was 10 μg/l higher than the lowest reference dosage [3.44 μg/l (Ting et al., 2011)], there was an obvious negative correlation between serum folate levels and EC risk. With an increase in serum folate levels, the EC risk decreases. Figure 4 described a nonlinear trend between folate intake and EC risk. We can see that when the level of folate intake is 50 μg/day higher than the lowest reference dosage [125.21 μg/day (Aune et al., 2011)]; the EC risk is decreased with the increase of folate intake. Also, when the level of folate intake is higher than the lowest reference dosage 100 μg/day, there is a short platform period. Then, the risk of EC increases slightly, but its OR value is also below 1.00.

Fig. 3

Fig. 3

Fig. 4

Fig. 4

Back to Top | Article Outline

Discussion

In one latest study (Zhao et al., 2017) , the pooled OR of EC for the highest versus the lowest level of dietary folate intake was 0.63 (95% CI: 0.56–0.71), whereas in ours, the pooled OR was 0.64 (95% CI: 0.54–0.76). Our findings are consistent with Zhao’s. The results of the pooled OR of EC for the highest versus the lowest level of serum folate are also similar. However, Chang et al. (2015) did not find a clear relationship between serum folate and EC risk. We have described the relationships between folate intake, serum folate, and EC risk using a dose–response meta-analysis method. The result showed that EC risk decreased obviously with an increase in folate levels only in a certain dosage range. In other words, the protective effects of folate might be reduced when folate levels are too high or too low. This information provides insights into the prevention of EC. Our results are different from those of a previous study (Zhao et al., 2017). The results of that study showed that dietary folate intake was associated linearly with the risk of EC, whereas in our study, folate intake and EC risk were nonlinearly related. Because our study is the most recent one and we updated the results by incorporating recently published studies into the meta-analysis, we believe that our results are more reliable.

Folate, as a one-carbon source, is involved in many biological regulatory processes, such as DNA methylation and repair. An animal experiment found that folate sensing by mechanistic target of rapamycin represents a novel pathway by which folate can modulate cell functions (Silva et al., 2017). Another study (Bach et al., 2017) pointed out that DNA double-strand breaks in cells occurred at low folate levels, and repair activity and chromatin re-organization were folate dependent. The role of folate in tumorigenesis and cancer prevention depends on the stage of cell transformation at the time of folate intervention and the dose of the folate supplement (Song et al., 2000; Lawrance et al., 2007). There are two prominent mechanisms by which folate deficiency may influence the risk of cancer: (i) by inducing misincorporation of uracil into DNA, which could lead to chromosomal breaks and mutations (Blount et al., 1997) and/or (ii) by causing folate appears to enhance carcinogenesis aberrant DNA methylation, resulting in altered expression of critical proto-oncogenes and tumor-suppressor genes (Choi and Mason, 2000). Folate also plays a fundamental role in purine and thymidylate synthesis, and a folate-deficient diet can lead to an over-representation of uracil in the nucleotide pool, which can result in reiterative repair of DNA strand breaks and abasic sites (Melnyk et al., 1999). Folate may prevent tumor development before the existence of preneoplastic lesions, but increase tumorigenesis once preneoplastic lesions are established (Kim, 2004, 2007). Ulrich et al. (2010) pointed out that folate supplementation may promote the development and progression of already existing, undiagnosed, premalignant lesions, whereas folate supplementation does not alter the development or progression of the disease.

As with most meta-analyses, our study also had some limitations. First, we may have failed to control for potential confounders. Although most studies controlled for some lifestyle factors, it is difficult to completely eliminate the residual confounding factors. Second, recall bias may have been present in original studies when food-frequency and recall questionnaires were used to assess diet. This would have inevitably led to some degree of misclassification of folate intake (Larsson et al., 2006). Third, publication bias may be present because of language restrictions. In the subgroup analyses based on area, design, outcome of cases, NOQAS score, and five adjusted confounders, we observed an inverse association between folate intake and the risk of EC in almost all groups, but not in the cohort study design (OR=1.01, 95% CI: 0.79–1.30) and in the group with a NOQAS score of 6 (OR=0.96, 95% CI: 0.59–1.57). Only two cohort studies exist to date. The limited data from cohort studies might lead to bias in this subgroup result.

Through these analyses, we can conclude that folate is a protective factor for EC within a certain dosage range; otherwise, the protective effects might be reduced.

Back to Top | Article Outline

Acknowledgements

All work was completed at Shandong University, China. I thank my study team during the writing of this thesis for their support.

Back to Top | Article Outline

Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline

References

Aune D, Norat T (2011). Dietary fibre, whole grains, and risk of colorectal cancer: systematic review and dose–response meta-analysis of prospective studies. BMJ 343:1756–1833.
Aune D, Deneo-Pellegrini H, Ronco AL, Boffetta P, Acosta G, Mendilaharsu M, et al (2011). Dietary folate intake and the risk of 11 types of cancer: a case–control study in Uruguay. Ann Oncol 22:444–451.
Bach M, Savini C, Krufczik M, Cremer C, Roesl F, Hausmann M (2017). Super-resolution Localization microscopy of gamma-H2AX and heterochromatin after folate deficiency. Int J Mol Sci 18:8.
Bao L, Peng J, Huang G, Wang S, Yin H, Wang T, et al (2013). The study on the relationship between serum folic acid and vitamin B2 levels and esophageal cancer. Wei Sheng Yan Jiu 42:451–454.
Blount BC, Mack MM, Wehr CM, MacGregor JT, Hiatt RA, Wang G, et al (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.
Bollschweiler E, Wolfgarten E, Nowroth T, Rosendahl U, Monig SP, Holscher AH (2002). Vitamin intake and risk of subtypes of esophageal cancer in Germany. J Cancer Res Clin Oncol 128:575–580.
Brown LM, Blot WJ, Schuman SH, Smith VM, Ershow AG, Marks RD, et al (1988). Environmental factors and high risk of esophageal cancer among men in coastal South Carolina. J Natl Cancer Inst 80:1620–1625.
Chang SC, Goldstein BY, Mu L, Cai L, You NC, He N, et al (2015). Plasma folate, vitamin B12, and homocysteine and cancers of the esophagus, stomach, and liver in a Chinese population. Nutr Cancer 67:212–223.
Chen H, Tucker KL, Graubard BI, Heineman EF, Markin RS, Potischman NA, et al (2002). Nutrient intakes and adenocarcinoma of the esophagus and distal stomach. Nutr Cancer 42:33–40.
Choi SW, Mason JB (2000). Folate and carcinogenesis: an integrated scheme. J Nutr 130:129–132.
De Stefani E, Ronco AL, Boffetta P, Deneo-Pellegrini H, Acosta G, Correa P, et al (2006). Nutrient intake and risk of squamous cell carcinoma of the esophagus: a case–control study in Uruguay. Nutr Cancer 56:149–157.
Fanidi A, Relton C, Ueland PM, Midttun O, Vollset SE, Travis RC, et al (2014). A prospective study of one-carbon metabolism biomarkers and cancer of the head and neck and esophagus. Int J Cancer 136:915–927.
Franceschi S, Bidoli E, Negri E, Zambon P, Talamini R, Ruol A, et al (2000). Role of macronutrients, vitamins and minerals in the aetiology of squamous-cell carcinoma of the oesophagus. Int J Cancer 86:626–631.
Galeone C, Pelucchi C, Levi F, Negri E, Talamini R, Franceschi S, et al (2006). Folate intake and squamous-cell carcinoma of the oesophagus in Italian and Swiss men. Ann Oncol 17:521–525.
Hao T, Maimaitmng T, Yan-Ping W (2011). The association between nutritional status, folic acid levels and esophageal cancer among 234 Kazakh people in Xinyuan Xinjiang. J Xinjiang Med Uni 34:429–431.
Hong Z, Tian C, Zhang X (2012). Dietary calcium intake, vitamin D levels, and breast cancer risk: a dose–response analysis of observational studies. Breast Cancer Res Treat 136:309–312.
Huang GL, Wang SK, Su M, Wang TT, Cai HZ, Yin H, et al (2013). Serum folate, MTHFR C677T polymorphism and esophageal squamous cell carcinoma risk. Biomed Environ Sci 26:1008–1012.
Ibiebele TI, Hughes MC, Pandeya N, Zhao Z, Montgomery G, Hayward N, et al (2011). High intake of folate from food sources is associated with reduced risk of esophageal cancer in an Australian population. J Nutr 141:274–283.
Jessri M, Rashidkhani B, Hajizadeh B, Gotay C (2011). Macronutrients, vitamins and minerals intake and risk of esophageal squamous cell carcinoma: a case–control study in Iran. Nutr J 10:137.
Kim YI (2004). Folate, colorectal carcinogenesis, and DNA methylation: lessons from animal studies. Environ Mol Mutagen 44:10–25.
Kim YI (2007). Folate and colorectal cancer: an evidence-based critical review. Mol Nutr Food Res 51:267–292.
Knobloch K, Yoon U, Vogt P (2011). Preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement and publication bias. J Craniomaxillofac Surg 39:91–92.
Kubo A (2010). Dietary factors and the risks of oesophageal denocarcinoma and Barrett’s oesophagus. Nutr Res Rev 23:230–246.
Larsson SC, Giovannucci E, Wolk A (2006). Folate intake, MTHFR polymorphisms, and risk of esophageal, gastric, and pancreatic cancer: a meta-analysis. Gastroenterology 131:1271–1283.
Lawrance AK, Deng L, Brody LC, Finnell RH, Shane B, Rozen R (2007). Genetic and nutritional deficiencies in folate metabolism influence tumorigenicity in Apcmin/+ mice. J Nutr Biochem 18:305–312.
Liu YX, Wang B, Wan MH, Tang WF, Huang FK, Li C (2011). Meta-analysis of the relationship between the Metholenetetrahydrofolate reductase C677T genetic polymorphism, folate intake and esophageal cancer. Asian Pac J Cancer Prev 12:247–252.
Mayne ST, Risch HA, Dubrow R, Chow WH, Gammon MD, Vaughan TL, et al (2001). Nutrient intake and risk of subtypes of esophageal and gastric cancer. Cancer Epidemiol Biomarkers Prev 10:1055–1062.
Mei-Ling L, Xi-Jian L, Ru-Chun L, Hao L, Tian-Mu C, Qiang H, et al (2014). Realizing the dose–response meta analysis in Stata Software. J Evid Based Med 14:182–187.
Melnyk S, Pogribna M, Miller BJ, Basnakian AG, Pogribny IP, James SJ (1999). Uracil misincorporation, DNA strand breaks, and gene amplification are associated with tumorigenic cell transformation in folate deficient/repleted Chinese hamster ovary cells. Cancer Lett 146:35–44.
Norat T, Lukanova A, Ferrari P, Riboli E (2002). Meat consumption and colorectal cancer risk: dose–response meta-analysis of epidemiological studies. Int J Cancer 98:241–256.
Sharp L, Carsin AE, Cantwell MM, Anderson LA, Murray LJ (2013). Intakes of dietary folate and other B vitamins are associated with risks of esophageal adenocarcinoma, Barrett’s esophagus, and reflux esophagitis. J Nutr 143:1966–1973.
Silva E, Rosario FJ, Powell TL, Jansson T (2017). Mechanistic target of rapamycin is a novel molecular mechanism linking folate availability and cell function. J Nutr 147:1237–1242.
Song J, Medline A, Mason JB, Gallinger S, Kim YI (2000). Effects of dietary folate on intestinal tumorigenesis in the apcMin mouse. Cancer Res 60:5434–5440.
Stang A (2010). Critical evaluation of the Newcastle–Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 25:603–605.
Tavani A, Malerba S, Pelucchi C, Dal Maso L, Zucchetto A, Serraino D, et al (2012). Dietary folates and cancer risk in a network of case–control studies. Ann Oncol 23:2737–2742.
Tio M, Andrici J, Cox MR, Eslick GD (2014). Folate intake and the risk of upper gastrointestinal cancers: a systematic review and meta-analysis. J Gastroenterol Hepatol 29:250–258.
Ulrich CM, Xu X, Liu A, Chen J (2010). Folate. Nutrition and Health: Bioactive Compounds and Cancer. In: Milner JA, Romagnolo DF, editors. Springer Science Business Media. LLC: New York. pp.387–410.
Wu W, Zheng Y, Wang R, Huang W, Liu L, Hu X, et al (2012). Antitumor activity of folate-targeted, paclitaxel-loaded polymeric micelles on a human esophageal EC9706 cancer cell line. Int J Nanomedicine 7:3487–3502.
Wu W, Kang S, Zhang D (2013). Association of vitamin B6, vitamin B12 and methionine with risk of breast cancer: a dose–response meta-analysis. Br J Cancer 109:1926–1944.
Xiao Q, Freedman ND, Ren J, Hollenbeck AR, Abnet CC, Park Y (2014). Intakes of folate, methionine, vitamin B6, and vitamin B12 with risk of esophageal and gastric cancer in a large cohort study. Br J Cancer 110:1328–1333.
Yang CX, Matsuo K, Ito H, Shinoda M, Hatooka S, Hirose K, et al (2005). Gene–environment interactions between alcohol drinking and the MTHFR C677T polymorphism impact on esophageal cancer risk: results of a case–control study in Japan. Carcinogenesis 26:1285–1290.
Zhang ZF, Kurtz RC, Yu GP, Sun M, Gargon N, Karpeh M Jr, et al (1997). Adenocarcinomas of the esophagus and gastric cardia: the role of diet. Nutr Cancer 27:298–309.
Zhao P, Lin F, Li Z, Lin B, Lin J, Luo R (2011). Folate intake, methylenetetrahydrofolate reductase polymorphisms, and risk of esophageal cancer. Asian Pac J Cancer Prev 12:2019–2023.
Zhao Y, Guo C, Hu H, Zheng L, Ma J, Jiang L, et al (2017). Folate intake, serum folate levels and esophageal cancer risk: an overall and dose–response meta-analysis. Oncotarget 8:10458–10469.
Keywords:

esophageal cancer; folate; meta-analysis

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.