The association between methylenetetrahydrofolate reductase gene C677T polymorphisms and breast cancer risk in Chinese population

Tumor Biol. (2015) 36:9153 9158 DOI 10.1007/s13277-015-3321-6 EDITORIAL The association between methylenetetrahydrofolate reductase gene C677T polymorphisms and breast cancer risk in Chinese population Yadong Wang 1 & Haiyan Yang 2 & Huiyan Gao 1 & Haiyu Wang 1 Received: 5 March 2015 /Accepted: 12 March 2015 /Published online: 4 November 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015 Abstract With great interest, we read the recent article entitled Methylenetetrahydrofolate reductase polymorphisms and breast cancer risk in Chinese population: a meta-analysis of 22 case-control studies published online in Tumor Biology, 2014, 35: 1695 1701. This article suggests that methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism was significantly associated with breast cancer risk in Chinese population. The result is encouraging. Nevertheless, three key issues in this meta-analysis are worth noticing. Keywords MTHFR. Polymorphism. Breast cancer. Risk. Meta-analysis. Chinese We read with great interest the recent paper entitled Methylenetetrahydrofolate reductase polymorphisms and breast cancer risk in Chinese population: a meta-analysis of 22 case-control studies published online in Tumor Biology, 2014, 35: 1695 1701 [1]. Liang et al. conducted a metaanalysis to investigate the association between methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphisms and breast cancer risk in the Chinese population on the basis of 13 case-control studies with 3272 cases and 4419 controls. The authors found that MTHFR C677T * Yadong Wang wangyd76@yahoo.com; wangyd76@126.com; wangyd76@163.com 1 2 Department of Toxicology, Henan Center for Disease Control and Prevention, No. 105 of South Nongye Road, Zhengzhou 450016, China Department of Epidemiology, School of Public Health, Zhengzhou University, Zhengzhou 450052, China polymorphism was significantly associated with breast cancer risk in the Chinese population under three genetic models [T vs. C: (OR)=1.12, 95 % confidence interval (95 % CI)=1.02 1.23, P=0.015; TT vs. CC: OR=1.35, 95 % CI= 1.10 1.67, P=0.005; TT vs. CC+CT: OR=1.37, 95 % CI= 1.11 1.70, P=0.004]. It is an extremely valuable study. Nevertheless, after carefully examining the data provided by Liang et al. (shown in Table 1 in their original text) [1], we found that three key issues should be worth noticing. Firstly, one overlapping paper [2] was not properly excluded from the study of Liang et al. [1]. Secondly, the metaanalysis of Liang et al. failed in including two eligible articles [3, 4] that were available at the time their meta-analysis was performed. Thirdly, the data reported by Liang et al. [1] for the study of Gao et al. [5] did not seem in line with the data provided by Gao et al. [5] in their original publication. The numbers reported by Gao et al. [5] (shownintable2 of the original paper of Gao et al.) for CC, CT, and TT, in cases and controls, are 202-305-117 and 235-301-88, respectively. Interestingly enough, after carefully examining the data reported by Liang et al. [1], the numbers are 217-327-125 in cases and 257-329-96 in controls, respectively. Therefore, the conclusions by Liang et al. are not entirely reliable. It is required to clarify the association between MTHFR C677T polymorphisms and the risk of breast cancer in Chinese population comprehensively and objectively. We reassessed this association by performing an updated meta-analysis on the basis of a total of 18 studies with 6808 cases and 7872 controls. Further subgroup analysis was conducted in this study stratified by source of control and Hardy-Weinberg equilibrium (HWE) in controls. In addition, cumulative meta-analysis was carried out to examine the tendency of results by accumulating single study year by year, which could be used to determine whether new relevant studies are needed or

9154 Tumor Biol. (2015) 36:9153 9158 Table 1 Main characteristics of selected studies in this metaanalysis Author Year Region Source of control Number of case Number of control P value of HWE Cheng C [6] 2008 Taiwan HB 349 530 0.623636 Gao C [5] 2009 Nanjing PB 624 624 0.592439 Han H [3] 2011 Shandong HB 47 52 0.073532 He J [7] 2014 Henan HB 310 381 0.000019 Hua Z [8] 2011 Yunnan PB 95 90 0.021217 Huang C [9] 2014 Taiwan HB 1232 1232 0.288652 Inoue M [10] 2008 Singapore PB 380 608 0.527225 Kan X [11] 2007 Yunnan PB 125 103 0.041775 Li W [12] 2009 Guangdong PB 65 143 0.187377 Liu Y [13] 2013 Guangdong HB 435 435 0.222884 Qi J [14] 2004 Beijing PB 217 218 0.593105 Qiao J [15] 2014 Henan HB 535 673 0.000002 Shrubsole M [16] 2004 Shanghai PB 1112 1160 0.442466 Wu X [17] 2012 Yunnan HB 75 75 0.800632 Wu Y [4] 2012 Shandong HB 200 200 0.023998 Xi J [18] 2014 Guangdong HB 818 848 0.355550 Yu C [19] 2007 Taiwan PB 109 420 0.336255 Yuan H [20] 2009 Heilongjiang HB 80 80 0.515565 HB hospital-based control, PB population-based control, HWE Hardy-Weinberg equilibrium Table 2 The summary for the association of MTHFR C677T polymorphisms with breast cancer risk in Chinese population Genetic model Cases/controls Heterogeneity test Summary OR (95 % CI) Hypothesis test df Begg s test Egger s test Q P Z P Z P t P Total CT vs. CC 5810/7045 18.88 0.34 1.03 (0.95 1.10) 0.69 0.49 17 0.08 0.940 0.06 0.954 TT vs. CC 4069/4635 56.68 <0.00001 1.57 (1.26 1.97) 3.95 <0.0001 17 1.59 0.112 1.53 0.145 CT+TT vs. CC 6808/7872 36.45 0.004 1.12 (1.00 1.25) 2.05 0.04 17 0.68 0.495 0.91 0.377 T vs. C 13,616/15,744 64.46 <0.00001 1.20 (1.08 1.34) 3.30 0.001 17 1.36 0.173 1.31 0.208 Stratification by HWE Yes CT vs. CC 4785/5765 13.15 0.36 1.01 (0.93 1.09) 0.27 0.79 12 0.43 0.669 0.62 0.545 TT vs. CC 3229/3696 44.10 <0.0001 1.51 (1.15 1.98) 2.95 0.003 12 1.65 0.100 1.71 0.115 CT+TT vs. CC 5543/6425 25.56 0.01 1.09 (0.96 1.23) 1.36 0.17 12 1.40 0.161 1.39 0.192 T vs. C 11,086/12,850 45.40 <0.00001 1.16 (1.03 1.31) 2.44 0.01 12 1.65 0.100 1.58 0.143 Stratification by source of control Population-based control CT vs. CC 2285/2937 9.97 0.19 0.98 (0.88 1.10) 0.32 0.75 7 1.11 0.266 0.55 0.601 TT vs. CC 1532/1881 22.71 0.002 1.34 (0.94 1.92) 1.62 0.11 7 0.62 0.536 1.14 0.298 CT+TT vs. CC 2727/3366 15.30 0.03 1.04 (0.87 1.25) 0.41 0.68 7 0.12 1.000 0.27 0.796 T vs. C 5454/6732 23.60 0.001 1.09 (0.93 1.29) 1.06 0.29 7 0.12 0.902 0.58 0.584 Hospital-based control CT vs. CC 3525/4108 8.00 0.53 1.05 (0.96 1.16) 1.14 0.25 9 0.36 0.721 0.70 0.506 TT vs. CC 2537/2754 24.98 0.003 1.75 (1.33 2.29) 4.03 <0.0001 9 1.07 0.283 0.97 0.359 CT+TT vs. CC 4081/4506 18.08 0.03 1.18 (1.03 1.36) 2.40 0.02 9 1.07 0.283 1.24 0.250 T vs. C 8162/9012 32.80 0.0001 1.29 (1.12 1.48) 3.50 0.0005 9 1.43 0.152 1.36 0.211 HWE Hardy-Weinberg equilibrium, OR, 95 % CI 95 % confidence interval

Study or Subgroup Tumor Biol. (2015) 36:9153 9158 9155 Fig. 1 Forest plots of the odds ratio for MTHFR C677T polymorphisms associated with breast cancer risk (a TT vs. CC, b CT+TT vs. CC, c T-allele vs. C- allele, and d CT vs. CC) A Case Control Odds Ratio Odds Ratio Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI 31 216 41 309 6.4% 1.10 [0.66, 1.81] 117 319 88 323 7.8% 1.55 [1.11, 2.16] 17 27 5 21 2.3% 5.44 [1.52, 19.41] 54 213 44 264 6.9% 1.70 [1.09, 2.66] 9 74 11 63 3.5% 0.65 [0.25, 1.70] 175 713 103 699 8.3% 1.88 [1.44, 2.47] 21 260 43 382 6.1% 0.69 [0.40, 1.20] 22 96 9 74 4.0% 2.15 [0.92, 4.99] 10 48 3 93 2.1% 7.89 [2.06, 30.29] 32 282 30 285 6.2% 1.09 [0.64, 1.84] 71 113 54 113 6.2% 1.85 [1.09, 3.14] 105 346 82 447 7.8% 1.94 [1.39, 2.70] 183 557 196 583 8.5% 0.97 [0.76, 1.24] 13 45 6 43 3.0% 2.51 [0.85, 7.35] 50 111 21 91 5.5% 2.73 [1.48, 5.05] 50 539 53 550 7.2% 0.96 [0.64, 1.44] 9 65 25 250 4.2% 1.45 [0.64, 3.27] 29 45 13 45 3.8% 4.46 [1.84, 10.84] Total (95% CI) 4069 4635 100.0% 1.57 [1.26, 1.97] Total events 998 827 Heterogeneity: Tau² = 0.14; Chi² = 56.68, df = 17 (P < 0.00001); I² = 70% Test for overall effect: Z = 3.95 (P < 0.0001) 0.01 0.1 1 10 100 Favours experimental Favours control B Study or Subgroup Case Control Odds Ratio Odds Ratio Events Total Events Total Weight M-H, Random, 95% CI M-H, Random, 95% CI 164 349 262 530 7.1% 0.91 [0.69, 1.19] 422 624 389 624 8.0% 1.26 [1.00, 1.59] 37 47 36 52 1.3% 1.64 [0.66, 4.10] 151 310 161 381 6.5% 1.30 [0.96, 1.75] 30 95 38 90 2.7% 0.63 [0.35, 1.15] 694 1232 636 1232 9.9% 1.21 [1.03, 1.42] 141 380 269 608 7.3% 0.74 [0.57, 0.97] 51 125 38 103 3.2% 1.18 [0.69, 2.02] 27 65 53 143 2.7% 1.21 [0.66, 2.20] 185 435 180 435 7.2% 1.05 [0.80, 1.37] 175 217 159 218 4.1% 1.55 [0.99, 2.43] 294 535 308 673 8.2% 1.45 [1.15, 1.82] 738 1112 773 1160 9.5% 0.99 [0.83, 1.18] 43 75 38 75 2.4% 1.31 [0.69, 2.49] 139 200 130 200 4.5% 1.23 [0.81, 1.86] 329 818 351 848 9.0% 0.95 [0.78, 1.16] 53 109 195 420 4.4% 1.09 [0.72, 1.66] 64 80 48 80 2.0% 2.67 [1.31, 5.41] Total (95% CI) 6808 7872 100.0% 1.12 [1.00, 1.25] Total events 3737 4064 Heterogeneity: Tau² = 0.03; Chi² = 36.45, df = 17 (P = 0.004); I² = 53% Test for overall effect: Z = 2.05 (P = 0.04) 0.01 0.1 1 10 100 Favours experimental Favours control C D Study or Subgroup Case Control Odds Ratio Odds Ratio Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI 133 318 221 489 6.9% 0.87 [0.66, 1.16] 305 507 301 536 7.9% 1.18 [0.92, 1.51] 20 30 31 47 0.5% 1.03 [0.39, 2.72] 97 256 117 337 4.2% 1.15 [0.82, 1.61] 21 86 27 79 1.4% 0.62 [0.32, 1.22] 519 1057 533 1129 17.8% 1.08 [0.91, 1.28] 120 359 226 565 7.9% 0.75 [0.57, 0.99] 29 103 29 94 1.5% 0.88 [0.48, 1.62] 17 55 50 140 1.3% 0.81 [0.41, 1.57] 153 403 150 405 6.3% 1.04 [0.78, 1.38] 104 146 105 164 1.9% 1.39 [0.86, 2.25] 189 430 226 591 7.2% 1.27 [0.98, 1.63] 555 929 577 964 15.4% 1.00 [0.83, 1.20] 30 62 32 69 1.1% 1.08 [0.55, 2.15] 89 150 109 179 2.7% 0.94 [0.60, 1.46] 279 768 298 795 12.6% 0.95 [0.77, 1.17] 44 100 170 395 2.6% 1.04 [0.67, 1.62] 35 51 35 67 0.6% 2.00 [0.93, 4.28] Total (95% CI) 5810 7045 100.0% 1.03 [0.95, 1.10] Total events 2739 3237 Heterogeneity: Chi² = 18.88, df = 17 (P = 0.34); I² = 10% Test for overall effect: Z = 0.69 (P = 0.49) 0.01 0.1 1 10 100 Favours experimental Favours control

9156 Tumor Biol. (2015) 36:9153 9158 A B.5 1 5 C.5 1 5 D.5 1 5.1 1 10 Fig. 2 Forest plots for cumulative meta-analysis of for MTHFR C677T polymorphisms associated with breast cancer risk (a TT vs. CC, b CT+TT vs. CC, c T-allele vs. C-allele, and d CT vs. CC) Fig. 3 Funnel plot analysis to detect publication bias for MTHFR C677T polymorphisms associated with breast cancer risk (a TT vs. CC, b CT+ TT vs. CC, c T-allele vs. C-allele, and d CT vs. CC)

Tumor Biol. (2015) 36:9153 9158 9157 Fig. 4 Sensitivity analysis on the association between MTHFR C677T polymorphisms and breast cancer risk (T-allele vs. C-allele) Meta-analysis fixed-effects estimates (exponential form) Study ommited 1.07 1.10 1.15 1.21 1.26 not. We hope that our results are able to provide comprehensive and objective evidence for the association of MTHFR C677T polymorphisms with breast cancer risk in Chinese population. Table 1 listed the main characteristics of selected studies in this meta-analysis. Table 2 listed the summary s of the association between MTHFR C677T polymorphisms and the risk of breast cancer in Chinese population. Overall, we observed an increased risk of breast cancer among the subjects carrying MTHFR C677T TT and CT+TT genotypes (OR= 1.57, 95 % CI 1.26 1.97 and OR=1.12, 95 % CI 1.00 1.25, respectively) compared with those carrying CC genotype in total Chinese population (Fig. 1a, b). We also observed an increased risk of breast cancer among the subjects carrying MTHFR C677T T-allele (OR=1.20, 95 % CI 1.08 1.34) compared with those carrying C-allele in total Chinese population (Fig. 1c). We did not observe the association of MTHFR C677T polymorphism with breast cancer risk under the genetic model of CT vs. CC (OR=1.03, 95 % CI 0.95 1.10) (Fig. 1d). The cumulative meta-analysis accumulated the studies in the light of the year of publications, and the results showed that there were still significant associations between MTHFR C677T polymorphisms and breast cancer risk under the genetic models of TT vs. CC, CT+TT vs. CC, and T-allele vs. C-allele; the cumulative ORs were 1.57 with 95 % CI 1.26 1.97, 1.12 with 95 % CI 1.00 1.25, and 1.20 with 95 % CI 1.08 1.34, respectively (Fig. 2a c). When limiting the analysis to the studies with controls in agreement with HWE, we observed an increased risk of breast cancer among the subjects carrying TT genotype compared with those carrying CC genotype (OR=1.51, 95 % CI 1.15 1.98) and among the subjects carrying T-allele compared with those carrying C-allele (OR=1.16, 95 % CI 1.03 1.31) (Table 2). In a subgroup analysis by source of control, we observed an increased risk of MTHFR C677T polymorphisms for breast cancer under the genetic models of TT vs. CC, CT+TT vs. CC and T-allele vs. C-allele on the basis of hospital-based controls (OR=1.75 with 95 % CI 1.33 2.29, OR=1.18 with 95 % CI 1.03 1.36 and OR=1.29 with 95 % CI 1.12 1.48, respectively) (Table 2). We did not observe any association between MTHFR C677T polymorphisms and breast cancer risk on the basis of population-based controls (Table 2). The shape of funnel plots seemed to be approximately symmetrical among total population (Fig. 3a d). Both Egger s test and Begg s test results suggested that no evidence of publication bias existed in this meta-analysis (Table 2). To evaluate the stability of the results of the meta-analysis, we performed a sensitivity analysis through sequentially removing each individual study. The sensitivity analysis showed that the results were robust and were not influenced by any single study (Fig. 4). In summary, the results of the study by Liang et al. [1] should be expounded with caution. To reach a definitive conclusion, further well-designed and larger sample size studies are still needed to evaluate the association between MTHFR C677T polymorphisms and breast cancer risk in Chinese population. We hope that our remarks will contribute to more accurate elaboration and substantiation of the results provided by Liang et al. [1]. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (No. U1404815). Conflicts of interest None

9158 Tumor Biol. (2015) 36:9153 9158 References 1. Liang H, Yan Y, Li T, Li R, Li M, Li S, et al. Methylenetetrahydrofolate reductase polymorphisms and breast cancer risk in Chinese population: a meta-analysis of 22 case-control studies. Tumour Biol. 2014;35: 1695 701. 2. Lin WY, Chou YC, Wu MH, Huang HB, Jeng YL, Wu CC, et al. The MTHFR C677T polymorphism, estrogen exposure and breast cancer risk: a nested case-control study in Taiwan. Anticancer Res. 2004;24:3863 8. 3. Han H, Han L, Gao H, Hou L. Genetic polymorphisms of methylenetetrahydrofolate reductase, methionine synthase, methylation of NF2, and their association with breast cancer morbidity. Chin J Curr Adv Gen Surg. 2011;14:846 50. 4. Wu Y, Wu L, Wang Y, Cao W, Hou L. Relation between the SNPs in methylenetetrahydrofolate reductase gene C677T and G1793A and the susceptibility of sporadic breast cancer. Prog Mod Biomed. 2012;12:2609 14. 5. Gao CM, Tang JH, Cao HX, Ding JH, Wu JZ, Wang J, et al. MTHFR polymorphisms, dietary folate intake and breast cancer risk in Chinese women. J Hum Genet. 2009;54:414 8. 6. Cheng CW, Yu JC, Huang CS, Shieh JC, Fu YP, Wang HW, et al. Polymorphism of cytosolic serine hydroxymethyltransferase, estrogen and breast cancer risk among Chinese women in Taiwan. Breast Cancer Res Treat. 2008;111:145 55. 7. He JM, Pu YD, Wu YJ, Qin R, Zhang QJ, Sun YS, et al. Association between dietary intake of folate and MTHFR and MTR genotype with risk of breast cancer. Genet Mol Res. 2014;13:8925 31. 8. Hua Z, Wang Y, Ni J, Ge F, Zou T. Folic acid, vitamin B 12, MTHF R, MS gene polymorphisms associate with the risk of breast cancer. Mod Oncol. 2011;19:428 31. 9. Huang CY, Chang WS, Shui HA, Hsieh YH, Loh CH, Wang HC, et al. Evaluation of the contribution of methylenetetrahydrofolate reductase genotypes to Taiwan breast cancer. Anticancer Res. 2014;34:4109 15. 10. Inoue M, Robien K, Wang R, Van Den Berg DJ, Koh WP, Yu MC. Green tea intake, MTHFR/TYMS genotype and breast cancer risk: the Singapore Chinese Health Study. Carcinogenesis. 2008;29: 1967 72. 11. Kan X, Zou T, Wu X, Wang X. Association between MTHFR genotype polymorphism and breast cancer susceptibility in human population from Yunnan. Cancer Res Prev Treat. 2007;34:716 8. 12. Li W, Chen S. MTHFR C677T gene polymorphism associated with breast cancer. Pract J Med. 2009;25:2031 3. 13. Liu Y, Zhou LS, Xu XM, Deng LQ, Xiao QK. Association of dietary intake of folate, vitamin B 6 and B 12 and MTHFR genotype with breast cancer risk. Asian Pac J Cancer Prev. 2013;14: 5189 92. 14. Qi J, Miao XP, Tan W, Yu CY, Liang G, Lu WF, et al. Association between genetic polymorphisms in methylenetetrahydrofolate reductase and risk of breast cancer. Zhonghua Zhong Liu Za Zhi. 2004;26:287 9. 15. Qiao J, Jiao D, Lu Z, Cui S, Liu Z. Association of methylenetetrahydrofolate reductase and methionine synthase polymorphisms with breast cancer risk and interaction with folate, vitamin B 6,and vitamin B 12 intakes. Tumor Biol. 2014;35:11895 901. 16. Shrubsole MJ, Gao YT, Cai Q, Shu XO, Dai Q, Hebert JR, et al. MTHFR polymorphisms, dietary folate intake, and breast cancer risk: results from the Shanghai Breast Cancer Study. Cancer Epidemiol Biomarkers Prev. 2004;13:190 6. 17. Wu XY, Ni J, Xu WJ, Zhou T, Wang X. Interactions between MTHFR C677T-A1298C variants and folic acid deficiency affect breast cancer risk in a Chinese population. Asian Pac J Cancer Prev. 2012;13:2199 206. 18. Xi J, Su Y, Fadiel A, Lin Y, Su F, Jia W, et al. Association of physical activity and polymorphisms in FGFR2 and DNA methylation related genes with breast cancer risk. Cancer Epidemiol. 2014;38:708 14. 19. Yu CP, Wu MH, Chou YC, Yang T, You SL, Chen CJ, et al. Breast cancer risk associated with multigenotypic polymorphisms in folate-metabolizing genes: a nested case-control study in Taiwan. Anticancer Res. 2007;27:1727 32. 20. Yuan H, Xu X, Wang Z. The relation between polymorphism of methylenetetrahydrofolate reductase C677T and the risk of breast cancer. J Mudanjiang Med Univ. 2009;30:2 4.