IJC International Journal of Cancer

Transcription

1 IJC International Journal of Cancer Dietary total fat and fatty acids intake, serum fatty acids and risk of breast cancer: A meta-analysis of prospective cohort studies Yi Cao 1, Lin Hou 1 and Weijing Wang 2 1 Department of Biochemistry, Qingdao University Medical College, Qingdao, China 2 Department of Epidemiology and Health Statistics, Qingdao University Medical College, Qingdao, China Results from prospective cohort studies on the association between dietary total fat and fatty acids intake and risk of breast cancer remain controversial. Pertinent prospective cohort studies were identified by a search of Embase and PubMed from inception to September Study-specific relative risks (RRs) with 95% confidence intervals were pooled using a randomeffect model. Between-study heterogeneity and publication bias were assessed, and sensitivity analysis was conducted. Twenty-four independent studies on dietary total fat and fatty acids intake and seven studies on serum fatty acids were included. The pooled RR of breast cancer for the highest vs. lowest category of dietary total fat intake was 1.10 ( ); however, no association was observed in studies adjusting for traditional risk factors of breast cancer. No association was observed between animal fat, vegetable fat, saturated fatty acids (SFA), monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA), n-3 PUFA, n-6 PUFA, eicosapentaenoic acid, docosahexaenoic acid, alpha-linolenic acid, oleic acid, linoleic acid and arachidonic acid and risk of breast cancer. The pooled RRs of breast cancer for the highest vs. lowest category of serum SFA, MUFA, PUFA, n-3 PUFA and n-6 PUFA were 1.00 ( ), 1.41 ( ), 0.59 ( ), 0.81 ( ) and 0.84 ( ), respectively. Results from this meta-analysis suggested that dietary total fat and fatty acids might be not associated with risk of breast cancer. Worldwide, breast cancer is the most frequently diagnosed cancer and the leading cause of cancer death among females, accounting for 25% of all cancer cases and 15% of all cancer deaths among females. 1 In China, breast cancer is now the most common cancer in females and the incidence is increasing, 2 and the current age-standardized 5-year relative survival for female breast cancer is 73.0%. 3 Several modifiable causes of breast cancer have been established, including radiation exposure, alcohol, postmenopausal obesity, lack of physical activity and postmenopausal hormone therapy with estrogen plus progestins. 4 While there continues to be interest in whether fat intake may influence breast cancer risk, 5 the most recent report from the World Cancer Research Fund/ Key words: fat, fatty acids, breast cancer, prospective cohort studies, meta-analysis Additional Supporting Information may be found in the online version of this article. Conflict of interest: Lin Hou receive a grant from Natural Science Foundation of Shandong Province (ZR20120HM89) DOI: /ijc History: Received 12 Sep 2015; Accepted 18 Nov 2015; Online 23 Nov 2015 Correspondence to: Lin Hou, Department of Biochemistry, Qingdao University, No. 38 Dongzhou Road, Qingdao , Shandong, People s Republic of China, Tel.: , Fax: , houlinqdu1963@163.com American Institute for Cancer Research concluded that although a statistically significant positive association between fat intake and breast cancer was found in case control studies, evidence from prospective cohort studies was still not consistent. 6 Furthermore, the varying role that different types of individual fatty acids might have on breast cancer risk deserves further investigation. 6 Relative to case control studies, prospective cohort studies do not suffer from recall bias and are anticipated to be less likely to have selection bias. In addition, prospective cohort studies are also believed to provide better evidence for causality because fat and fatty acids consumption precedes breast cancer incidence. Considering results from recent prospective cohort studies 7 14 have not been quantitatively summarized, we conducted a metaanalysis of prospective cohort studies following the standard methods 15 (Supporting Information Table 1) to (i) first assess the association between dietary total fat and fatty acids intake and breast cancer risk; (ii) then evaluate the modification of key covariates to the associations; (iii) evaluate serum fatty acids and risk of breast cancer (iv) and assess the heterogeneity among studies and publication bias. Methods Literature search and selection We performed a literature search from inception to September 2015 using the databases of PubMed and Embase. Details of the search strategy are shown in Supporting Information

2 Cao et al What s new? Whether or not a woman s risk of breast cancer is increased by a high-fat diet remains unclear. In particular, the results of prospective cohort studies have been inconsistent. In the present meta-analysis, data from relevant prospective studies were pooled and assessed. After adjusting for known risk factors, neither dietary total fat intake nor fatty acid intake were found to be associated with breast cancer risk. Further investigation is needed, however, to clarify findings from studies in Europe, for which measurement error may have affected assessments, notably for postmenopausal women. Table 2. Moreover, we also reviewed the reference lists from retrieved articles to search for further relevant studies. Two investigators (Yi Cao and Weijing Wang) independently reviewed all identified studies, and studies were included if they met the following criteria: (i) a prospective cohort study or nested case control study in which fat and fatty acids consumption precedes breast cancer incidence; (ii) the exposure of interest was dietary total fat, fatty acids intake or serum fatty acids; (iii) the outcome of interest was breast cancer and (iv) relative risk (RR) or hazard ratio and odds ratio (for nested case control studies) with 95% confidence interval (CI) were provided (we presented all results with RR). If data were duplicated in more than one study, we included the study with the most recent one; otherwise, the one with the most number of cases was included. Data extraction We extracted all data using a standardized data collection form (Yi Cao and Weijing Wang). Information was recorded as follows: the first author s last name, publication year, number of cases and participants, measurement of exposure, country where the study was performed, variables adjusted for in the analysis, RR estimates with corresponding 95% CI for the highest versus lowest categories of dietary total fat and fatty acids intake and serum fatty acids. For dose response analysis, the number of cases and participants (person-years) and RR (95% CI) for each category of dietary total fat or fatty acids intake were also extracted. We extracted the RRs that reflected the greatest degree of control for potential confounders. Statistical analysis Pooled measure was calculated as the inverse variance weighted mean of the logarithm of RR with 95% CI to assess the strength of association. A random-effect model was used as the pooling method, which considers both within-study and between-study variation. The I2 was used to assess heterogeneity, and I2 values of 0, 25, 50 and 75% represent no, low, moderate and high heterogeneity, respectively. 16 We performed a sensitivity analysis in which one study at a time was removed and the rest were analyzed to evaluate whether the results could have been affected markedly by a single study. Publication bias was evaluated with Egger regression test. Meta-regression and subgroup analysis were conducted to explore potential sources of heterogeneity and perform comparison between groups, and the p values from meta-regression were calculated with a permutation test of 1000 to control the spurious findings. Study quality was assessed using the 9-star Newcastle-Ottawa Scale ( For dose response analysis, the median or mean level of dietary total fat and fatty acids intake for each category was assigned to the corresponding RR for every study. If the upper boundary of the highest category was not provided, we assumed that the boundary had the same amplitude as the adjacent category. A two-stage random-effects dose response meta-analysis 17 was performed to compute the trend from the correlated log RR estimates across levels of dietary total fat and fatty acids intake, respectively, taking into account the between-study heterogeneity. In the first stage, a restricted cubic spline model with three knots at the 25th, 50th and 75th percentiles of the levels of dietary total fat and fatty acids intake was estimated using generalized least square regression taking into account the correlation within each set of published RRs. Then the study-specific estimates were combined using the restricted maximum likelihood method in a multivariate random-effects meta-analysis. A p value for nonlinearity was calculated by testing the null hypothesis that the coefficient of the second spline is equal to 0. All statistical analyses were performed with STATA version 12.0 (Stata Corporation, College Station, TX). All reported probabilities (p values) were two-sided with p < 0.05 considered statistically significant. Results Literature search and study characteristics The flow chart for study inclusion is shown in Supporting Information Figure 1. A total of 897 articles were identified after excluding duplications in the two databases, and then 73 articles were reviewed in full text after reviewing titles and/or abstracts. Another 43 articles were further excluded for other reasons (Supporting Information Fig. 1). Finally, 24 independent studies from 23 articles 7,8,10 14,18 33 (38,262 cases among 1,387,366 subjects) were included on dietary total fat and fatty acids intake, and seven articles (1,334 cases among 3,511 subjects) were included on serum fatty acids. For dietary total fat and fatty acids intake, 11 studies were conducted in USA, nine in Europe, three in Asia and one in Canada. Dietary information was assessed with self- or interviewer-administrated food frequency questionnaire (FFQ) in 22 studies, while food record and 24-hr recall methods were used in the other two studies. The mean duration of follow-up ranged from 2 to 25 years. Most of the included

3 1896 Fat, fatty acids and breast cancer Figure 1. Forest plot for dietary total fat intake and risk of breast cancer. The size of gray box is positively proportional to the weight assigned to each study, and horizontal lines represent the 95% confidence intervals. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] studies adjusted for energy intake, body mass index and reproductive factors, and about 50% of the included studies adjusted for family history of breast cancer, exogenous female hormones use, alcohol intake and education, while part of studies adjusted for smoking. The included studies met the quality score of 5 7 stars. Detailed information of included studies is shown in Supporting Information Table 3. For serum fatty acids, all included studies were nested case control studies. The mean age of included subjects ranged from 49.0 to 59.6 years. Serum fatty acids (five studies), plasma fatty acids (one study) and erythrocyte membrane fatty acids (one study) were measured by gas chromatography. The included studies met the quality score of 8 9 stars. Detailed information of included studies is shown in Supporting Information Table 4. Quantitative synthesis Dietary total fat and breast cancer. Highest versus lowest levels of dietary total fat were positively associated with the risk of breast cancer [RR (95% CI): 1.10 ( ), I %, Fig. 1]. Publication bias was detected (p ), and small-study effects 24 were found in the funnel plot (Supporting Information Fig. 2). After excluding this single study, no publication bias was found (p ), and the reestimated effect was [1.09 ( ), I %]. Sensitivity analysis showed that no individual study had an excessive influence on the pooled effect (Table 1). In subgroup analysis, the positive association between dietary total fat intake and risk of breast cancer was only observed in postmenopausal women, studies conducted in Europe, studies with a relatively short follow-up duration (<10 years), studies with subjects of mean age >50 years, studies with dietary information assessed with selfadministrated FFQ, studies that did not adjust for education, family history of breast cancer, body mass index, reproductive factors, smoking and alcohol, studies adjusting for less number of covariates (<6) and studies with study quality of <7 stars. Number of covariates adjusted (p ) and study quality (p ) contributed to the between-study heterogeneity significantly. No association was found between animal fat [n 5 5, 0.95 ( ), I %] and vegetable fat [n 5 5, 0.95 ( ), I %] and risk of breast cancer, respectively. Dietary saturated fatty acids and breast cancer. Highest versus lowest levels of dietary total saturated fatty acids (SFA) were not associated with the risk of breast cancer [1.08

4 Cao et al Table 1. Pooled results on dietary total fat and fatty acids with risk of breast cancer Dietary total fat Saturated fatty acids Monounsaturated fatty acids Polyunsaturated fatty acids N RR (95% CI) I2 (%) p N RR (95% CI) I2 (%) p N RR (95% CI) I2 (%) p N RR (95% CI) I2 (%) p Overall ( ) ( ) ( ) ( ) Follow-up duration (years) > ( ) ( ) ( ) ( ) < ( ) ( ) ( ) ( ) Country where the study conducted Asia ( ) ( ) ( ) ( ) Europe ( ) ( ) ( ) ( ) USA ( ) ( ) ( ) ( ) Age at baseline >50 years ( ) ( ) ( ) ( ) years ( ) ( ) ( ) ( ) Dietary assessment Self-administered ( ) ( ) ( ) ( ) Others ( ) ( ) ( ) ( ) Adjust for education Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) Adjust for family history of breast cancer Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) Adjust for body mass index Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) Adjust for reproductive variables Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) 51.40

5 1898 Fat, fatty acids and breast cancer Table 1. Pooled results on dietary total fat and fatty acids with risk of breast cancer (Continued) Dietary total fat Saturated fatty acids Monounsaturated fatty acids Polyunsaturated fatty acids N RR (95% CI) I2 (%) p N RR (95% CI) I2 (%) p N RR (95% CI) I2 (%) p N RR (95% CI) I2 (%) p Adjust for exogenous female hormones Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) Adjust for smoking Yes ( ) ( ) ( ) ( ) 0.00 No ( ) ( ) ( ) ( ) Adjust for alcohol Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) Adjust for energy intake Yes ( ) ( ) ( ) ( ) No ( ) ( ) ( ) ( ) Number of the above covariates adjusted or more ( ) ( ) ( ) ( ) < ( ) ( ) ( ) ( ) Study quality stars ( ) ( ) ( ) ( ) stars ( ) ( ) ( ) ( ) Menopausal status Premenopausal ( ) ( ) ( ) ( ) Postmenopausal ( ) ( ) ( ) ( ) Estrogen receptor (ER), progesterone receptor (PR) status ER1PR ( ) ( ) ( ) ( ) 0.00 ER1PR ( ) ( ) ( ) ( ) 0.00 ER2PR ( ) ( ) ( ) ( ) 0.00 ER2PR ( ) ( ) ( ) ( ) Abbreviations: N: number of studies; p: p values from meta-regression with a permutation of 1000 test. Bold signifies p < 0.05.

6 Cao et al Figure 2. Forest plot for dietary saturated fatty acids intake and risk of breast cancer. The size of gray box is positively proportional to the weight assigned to each study, and horizontal lines represent the 95% confidence intervals. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] ( ), I %, Fig. 2]. No publication bias was detected (p ). Sensitivity analysis showed that no individual study had an excessive influence on the pooled effect (Table 1). In subgroup analysis, a positive association was observed with estrogen and progesterone receptor-positive breast cancer, in studies conducted in Europe, studies with a relatively short follow-up duration (<10 years), studies with subjects of mean age >50 years, studies that did not adjust for family history of breast cancer and studies with study quality of <7 stars. Study quality (p ) might contribute to the between-study heterogeneity. Dietary monounsaturated fat acids and breast cancer. Highest versus lowest levels of dietary total monounsaturated fat acids (MUFA) were not associated with the risk of breast cancer [1.08 ( ), I %, Fig. 3]. No publication bias was detected (p ). Sensitivity analysis showed that no individual study had an excessive influence on the pooled effect (Table 1). In subgroup analysis, a positive association was observed in postmenopausal women, in studies that did not adjust for family history of breast cancer, body mass index, smoking and alcohol, studies adjusting for less number of covariates (<6) and studies with study quality of <7 stars. No association was found between oleic acid [n 5 4, 1.11 ( ), I %] and risk of breast cancer. Dietary polyunsaturated fat acids and breast cancer. Highest versus lowest levels of dietary total polyunsaturated fat acids (PUFA) were not associated with the risk of breast cancer [1.05 ( ), I %, Fig. 4]. No publication bias was detected (p ). Sensitivity analysis showed that no individual study had an excessive influence on the pooled effect (Table 1). In subgroup analysis, a positive association was observed in postmenopausal women and studies with study quality of <7 stars. No association was found between n-3 PUFA [n 5 5, 1.02 ( ), I %], n-6 PUFA [n 5 5, 1.10 ( ), I %], eicosapentaenoic acid [n 5 4, 0.93 ( ), I %], docosahexaenoic acid [n 5 4, 0.94 ( ), I %], alpha-linolenic acid [n 5 5, 0.99 ( ), I %], linoleic acid [n 5 9, 1.01 ( ), I %] and arachidonic acid [n 5 5, 1.00 ( ), I %] and risk of breast cancer. Serum fatty acids and breast cancer. No association was found between serum SFA [n 5 7, 1.00 ( ),

7 1900 Fat, fatty acids and breast cancer Figure 3. Forest plot for dietary monounsaturated fatty acids intake and risk of breast cancer. The size of gray box is positively proportional to the weight assigned to each study, and horizontal lines represent the 95% confidence intervals. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] I %], MUFA [n 5 7, 1.41 ( ), I %], PUFA [n 5 4, 0.59 ( ), I %], n-3 PUFA [n 5 6, 0.81 ( ), I %], n-6 PUFA [n 5 7, 0.84 ( ), I %] and risk of breast cancer (Fig. 5). Subgroup analysis was not conducted because of relatively small number of studies included. Dose response analysis. The risk of breast cancer increased 3% [1.03 ( ), p for nonlinearity ], 1% [1.01 ( ), p for nonlinearity ], 5% [1.05 ( ), p for nonlinearity ], 7% [1.07 ( ), p for nonlinearity ] and 11% [1.11 ( ), p for nonlinearity ] for an increment of 10 g/day 8,11,18,19,21,22,29 and 10% energy of dietary total fat, 9,12,19,21,22,41 10 g/day of dietary SFA, 8,11,22,29 5 g/day of dietary MUFA 8,11,22,29 and 5 g/day of dietary PUFA, 8,11,22 respectively (Supporting Information Fig. 3). Discussion Results from this meta-analysis indicated no association between dietary total fat, SFA, MUFA and PUFA intake and risk of breast cancer, after adjusting for traditional risk factors of breast cancer. No association was found between serum SFA, MUFA and PUFA and risk of breast cancer. Moderate between-study heterogeneity was found. As one of the most controversial hypotheses in nutritional epidemiology, the association of fat intake and breast cancer has been receiving considerable attention for decades. Study design and analysis, population characteristics, differential effects by type of fatty acids, potential measurement error and bias might all contribute to this discrepancy. An earlier pooled analysis of cohort studies found no association between dietary total fat, SFA, MUFA and PUFA with risk of breast cancer, 42 and a randomized controlled trial of postmenopausal women also showed that a low-fat dietary pattern did not result in a statistically significant reduction [0.91 ( )] in breast cancer risk. 43 In this meta-analysis, positive associations were found between dietary total fat and SFA and breast cancer risk in eight studies conducted in Europe. Among the eight studies conducted in Europe, five studies 8,19,22,23,27 adjusted their results for six or more of the selected covariates shown in Table 1, and no association was found between dietary total fat and breast cancer risk in these five studies 8,19,22,23,27 (the risk estimates ranged from 1.00 to 1.51). While an increased risk of breast cancer with SFA was observed in one study [1.14 ( )], 8 no association was found in the other four studies 19,22,23,27 (the risk estimates ranged from 0.95 to 1.40). However, the magnitude of risk estimates of breast cancer ranged from 1.43 to 3.47 for dietary

8 Cao et al Figure 4. Forest plot for dietary polyunsaturated fatty acids intake and risk of breast cancer. The size of gray box is positively proportional to the weight assigned to each study, and horizontal lines represent the 95% confidence intervals. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] total fat and from 1.12 to 1.36 for SFA in the other three studies 24,29,32 adjusting for only one or two of the covariates shown in Table 1. The findings were consistent with those of this meta-analysis in which the magnitude of risk estimates was attenuated after adjusting for the covariates. Therefore, the possibility that the observed association was due to confounding should be considered. In addition, among the seven studies on serum fatty acids and breast cancer, four studies were conducted in Europe, and no association was found between SFA, MUFA, PUFA, n-3 PUFA and n-6 PUFA and risk of breast cancer, except for PUFA [0.31 ( ), ( ) 39 ] and n-6 PUFA [0.35 ( )]. 37 In this meta-analysis, positive associations were found between dietary total fat, MUFA and PUFA intake and risk of breast cancer in postmenopausal women. To explore the possibility of confounding, stratified analysis by number of covariates was also conducted. For studies adjusting for six or more of the above-mentioned covariates, the association was 1.09 ( ) for dietary total fat intake (n 5 7), 1.15 ( ) for MUFA intake (n 5 6) and 1.16 ( ) for PUFA intake (n 5 6). For studies adjusting for <6 of the above-mentioned covariates, the association was 1.68 ( ) for dietary total fat intake (n 5 3), 1.51 ( ) for MUFA intake (n 5 3) and 1.31 ( ) for PUFA intake (n 5 3). These findings indicated that confounding was also a major concern for the association between dietary total fat and fatty acids intake and breast cancer in postmenopausal women. Among the seven studies on serum fatty acids and breast cancer, three studies 35,36,39 reported the associations in postmenopausal women, and no association was found between serum SFA, MUFA, n-3 PUFA and n-6 PUFA and risk of breast cancer, but an inverse association was found with serum PUFA [0.34 ( )]. 39 However, the above-mentioned findings were based on a small number of studies and need to be confirmed. Dietary information with FFQ are subject to measurement error, and underreporting is believed to be the source of measurement error and would tend to obscure diet disease relationships. 44 Correlation between the 7-day diaries and FFQs for dietary total fat was r and for SFA r Therefore, dietary records or 24-h recalls may be preferable to the FFQ. 20,45 Within the EPIC Norfolk study, while positive associations were found between dietary total fat and SFA intake and breast cancer with a 7-day food diary, no association was found with an FFQ. 45 Similar findings were also found in Women s Health Initiative Clinical Trial when assessing the associations of dietary total fat, SFA, MUFA and PUFA intake with a 4-day food record and a FFQ, 20 respectively. In this meta-analysis, the magnitude of

9 1902 Fat, fatty acids and breast cancer Figure 5. Forest plot for serum saturated fatty acids (SFA), monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA) intake and risk of breast cancer. The size of gray box is positively proportional to the weight assigned to each study, and horizontal lines represent the 95% confidence intervals. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] risk estimation between dietary total fat, MUFA and PUFA intake was also stronger in studies with dietary information assessed with food records, 24-hr recall, diet history method or interviewer-administered FFQ than self-administered FFQ. In this respect, analysis with serum fatty acids is expected to offer more objective results. In this meta-analysis, no association between serum SFA and breast cancer risk was found, but higher level of serum MUFA might be associated with an increased risk of breast cancer [1.41 ( )]. However, in contrast to the analysis with dietary information, inverse but not significant associations were found between serum PUFA, n-3 PUFA and n-6 PUFA and risk of breast cancer. In addition, higher intake of marine n-3 PUFA 46 and ratio of n-3/n-6 PUFA 47 were found to be associated with a lower risk of breast cancer. Furthermore, while serum proportion of PUFA is considered as biomarker for the dietary intake of these fatty acids, both SFA and MUFA can be synthesized endogenously and cannot be considered as biomarkers for intake. Therefore, this difference in combination with measurement error with FFQ should all be of concern in assessing the fat breast cancer association. Results from the dose response analysis showed a certain degree of significant positive association between dietary total fat, SFA and MUFA, and breast cancer risk; however, only a relative small number of studies provided the required information for this dose response analysis. In addition, pooled

10 Cao et al results from studies with serum fatty acids also did not support an association between serum SFA and MUFA, and breast cancer risk. However, the dose response analysis showed that the departure from a linear relationship between dietary total fat, SFA and MUFA, and breast cancer risk was not significant, respectively, supporting the validity of the results from the main analysis comparing highest vs. lowest intake of dietary total fat and fatty acids. Other limitations should also be considered. First, several analysis like subgroups by estrogen receptor and progesterone receptor status, and analysis with serum fatty acids were based on a relatively limited number of studies. Second, the effects of other types of fatty acids like long-chain n-3 fatty acids 21 and linolenic acid 23 on risk of breast cancer need to be explored. Third, adjustment of several covariates could influence the fat breast cancer association. Although we extracted the RRs that reflected the greatest degree of control for potential confounders, the extent to which they were adjusted varied among studies, which might complicate interpretation of the findings. In conclusion, this meta-analysis suggested that dietary total fat and fatty acids might not be a major concern for breast cancer. However, considering the measurement error, studies with serum fatty acids are still needed to confirm the findings, especially for studies conducted in Europe and studies with postmenopausal women. References 1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, CA Cancer J Clin 2015;65: Fan L, Strasser-Weippl K, Li JJ, et al. Breast cancer in China. Lancet Oncol 2014;15:e279 e Zeng H, Zheng R, Guo Y, et al. Cancer survival in China, : a population-based study. Int J Cancer 2015;136: Colditz GA, Bohlke K. Priorities for the primary prevention of breast cancer. CA Cancer J Clin 2014;64: Kushi LH, Doyle C, McCullough M, et al. American Cancer Society Guidelines on nutrition and physical activity for cancer prevention: reducing the risk of cancer with healthy food choices and physical activity. CA Cancer J Clin 2012;62: World Cancer Research Fund/American Institute for Cancer Research. Food, nutrition, physical activity and the prevention of cancer: a global perspective. Washington, DC: AICR, Kiyabu GY, Inoue M, Saito E, et al. Fish, n - 3 polyunsaturated fatty acids and n - 6 polyunsaturated fatty acids intake and breast cancer risk: the Japan Public Health Center-based prospective study. Int J Cancer 2015;137: Sieri S, Chiodini P, Agnoli C, et al. Dietary fat intake and development of specific breast cancer subtypes. J Natl Cancer Inst 2014;106:dju068. doi: /jnci/dju Farvid MS, Cho E, Chen WY, et al. Premenopausal dietary fat in relation to pre- and postmenopausal breast cancer. Breast Cancer Res Treat 2014;145: Boeke CE, Eliassen AH, Chen WY, et al. Dietary fat intake in relation to lethal breast cancer in two large prospective cohort studies. Breast Cancer Res Treat 2014;146: Sczaniecka AK, Brasky TM, Lampe JW, et al. Dietary intake of specific fatty acids and breast cancer risk among postmenopausal women in the VITAL cohort. Nutr Cancer 2012;64: Park SY, Kolonel LN, Henderson BE, et al. Dietary fat and breast cancer in postmenopausal women according to ethnicity and hormone receptor status: the Multiethnic Cohort Study. Cancer Prev Res (Phila) 2012;5: Murff HJ, Shu XO, Li H, et al. Dietary polyunsaturated fatty acids and breast cancer risk in Chinese women: a prospective cohort study. Int J Cancer 2011;128: Thiebaut AC, Chajes V, Gerber M, et al. Dietary intakes of omega-6 and omega-3 polyunsaturated fatty acids and the risk of breast cancer. Int J Cancer 2009;124: Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement. PLoS Med 2009; 6:e Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003; 327: Orsini N, Li R, Wolk A, et al. Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software. Am J Epidemiol 2012;175: Thiebaut AC, Kipnis V, Chang SC, et al. Dietary fat and postmenopausal invasive breast cancer in the National Institutes of Health-AARP Diet and Health Study cohort. J Natl Cancer Inst 2007;99: Lof M, Sandin S, Lagiou P, et al. Dietary fat and breast cancer risk in the Swedish women s lifestyle and health cohort. Br J Cancer 2007;97: Freedman LS, Potischman N, Kipnis V, et al. A comparison of two dietary instruments for evaluating the fat-breast cancer relationship. Int J Epidemiol 2006;35: Wakai K, Tamakoshi K, Date C, et al. Dietary intakes of fat and fatty acids and risk of breast cancer: a prospective study in Japan. Cancer Sci 2005;96: Wirfalt E, Mattisson I, Gullberg B, et al. Postmenopausal breast cancer is associated with high intakes of omega6 fatty acids (Sweden). Cancer Causes Control 2002;13: Voorrips LE, Brants HA, Kardinaal AF, et al. Intake of conjugated linoleic acid, fat, and other fatty acids in relation to postmenopausal breast cancer: the Netherlands Cohort Study on Diet and Cancer. Am J Clin Nutr 2002;76: Sieri S, Krogh V, Muti P, et al. Fat and protein intake and subsequent breast cancer risk in postmenopausal women. Nutr Cancer 2002;42: Horn-Ross PL, Hoggatt KJ, West DW, et al. Recent diet and breast cancer risk: the California Teachers Study (USA). Cancer Causes Control 2002;13: Velie E, Kulldorff M, Schairer C, et al. Dietary fat, fat subtypes, and breast cancer in postmenopausal women: a prospective cohort study. J Natl Cancer Inst 2000;92: Wolk A, Bergstrom R, Hunter D, et al. A prospective study of association of monounsaturated fat and other types of fat with risk of breast cancer. Arch Intern Med 1998;158: Kushi LH, Potter JD, Bostick RM, et al. Dietary fat and risk of breast cancer according to hormone receptor status. Cancer Epidemiol Biomarkers Prev 1995;4: Gaard M, Tretli S, Loken EB. Dietary fat and the risk of breast cancer: a prospective study of 25,892 Norwegian women. Int J Cancer 1995;63: Toniolo P, Riboli E, Shore RE, et al. Consumption of meat, animal products, protein, and fat and risk of breast cancer: a prospective cohort study in New York. Epidemiology 1994;5: Howe GR, Friedenreich CM, Jain M, et al. A cohort study of fat intake and risk of breast cancer. J Natl Cancer Inst 1991;83: Knekt P, Albanes D, Seppanen R, et al. Dietary fat and risk of breast cancer. Am J Clin Nutr 1990;52: Jones DY, Schatzkin A, Green SB, et al. Dietary fat and breast cancer in the National Health and Nutrition Examination Survey I Epidemiologic Follow-up Study. J Natl Cancer Inst 1987;79: Pouchieu C, Chajes V, Laporte F, et al. Prospective associations between plasma saturated, monounsaturated and polyunsaturated fatty acids and overall and breast cancer risk - modulation by antioxidants: a nested case-control study. PLoS One 2014;9:e Takata Y, King IB, Neuhouser ML, et al. Association of serum phospholipid fatty acids with breast cancer risk among postmenopausal cigarette smokers. Cancer Causes Control 2009;20: Chajes V, Thiebaut AC, Rotival M, et al. Association between serum trans-monounsaturated fatty acids and breast cancer risk in the E3N-EPIC Study. Am J Epidemiol 2008;167: Rissanen H, Knekt P, Jarvinen R, et al. Serum fatty acids and breast cancer incidence. Nutr Cancer 2003;45: Saadatian-Elahi M, Toniolo P, Ferrari P, et al. Serum fatty acids and risk of breast cancer in a nested case-control study of the New York University Women s Health Study. Cancer Epidemiol Biomarkers Prev 2002;11: Pala V, Krogh V, Muti P, et al. Erythrocyte membrane fatty acids and subsequent breast cancer: a prospective Italian study. J Natl Cancer Inst 2001; 93:

11 1904 Fat, fatty acids and breast cancer 40. Chajes V, Hulten K, Van Kappel AL, et al. Fattyacid composition in serum phospholipids and risk of breast cancer: an incident case-control study in Sweden. Int J Cancer 1999;83: Kim EH, Willett WC, Colditz GA, et al. Dietary fat and risk of postmenopausal breast cancer in a 20- year follow-up. Am J Epidemiol 2006;164: Smith-Warner SA, Spiegelman D, Adami HO, et al. Types of dietary fat and breast cancer: a pooled analysis of cohort studies. Int J Cancer 2001;92: Prentice RL, Thomson CA, Caan B, et al. Low-fat dietary pattern and cancer incidence in the Women s Health Initiative Dietary Modification Randomized Controlled Trial. J Natl Cancer Inst 2007;99: Heitmann BL, Frederiksen P. Imprecise methods may both obscure and aggravate a relation between fat and breast cancer. Eur J Clin Nutr 2007;61: Bingham SA, Luben R, Welch A, et al. Are imprecise methods obscuring a relation between fat and breast cancer? Lancet 2003;362: Zheng JS, Hu XJ, Zhao YM, et al. Intake of fish and marine n-3 polyunsaturated fatty acids and risk of breast cancer: meta-analysis of data from 21 independent prospective cohort studies. BMJ 2013;346:f Yang B, Ren XL, Fu YQ, et al. Ratio of n-3/n-6 PUFAs and risk of breast cancer: a meta-analysis of adult females from 11 independent prospective studies. BMC Cancer 2014;14:105.