SEARCH

SEARCH BY CITATION

Keywords:

  • breast cancer;
  • biomarkers;
  • fatty acid

Abstract

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The use of the fatty acid composition of adipose tissue, erythrocyte membranes, serum and plasma as biological markers of fatty acid intake was recently introduced in epidemiological studies. The biomarkers of fatty acid intake have the advantage of providing quantitative measurement independent of energy intake and of the subject's memory. We performed a meta-analysis of published results of epidemiological studies of the composition of fatty acids in biological samples and breast cancer risk. The analysis was based on 3 cohort and 7 case-control studies including 2,031 cases and 2,334 controls. The summary statistic used was the average of the relative risk estimated for each level of the fatty acid on study, weighted by the inverse of its variance. Random effect models were assumed when the test for heterogeneity was significant. Overall relative risks were estimated for studies including pre- and post-menopausal breast cancer and separately for post-menopausal women. In cohort studies, a significant protective effect was found for total n-3 polyunsaturated fatty acids, while total monounsaturated fatty acids, oleic acid (C18:1 n-9c) and palmitic acid (C16:0) were significantly associated with an increase of breast cancer risk. Total saturated fatty acids were significantly associated with breast cancer risk in cohort studies only in postmenopausal women. For case-control studies, the only finding was for alpha linolenic acid (C18:3, n-3), which showed an inverse association bordering on statistical significance. The findings of cohort studies fit well with hypotheses derived from experimental animal studies. More epidemiological cohort studies that integrate biological markers of dietary fatty acid intake are needed in order to determine the contribution of different types of fatty acids in the etiology of breast cancer. © 2004 Wiley-Liss, Inc.

Breast cancer is the most widespread cancer for women and account for more than 25% of female cancers. The incidence rate still remains very high in the Western countries and increases quickly in Asian countries classified as countries at very low risk. There is now considerable evidence that the wide geographical heterogeneity in incidence and mortality rate among population can not be explained only by genetic factors and suggest that environmental factors and particularly dietary fat can have an important impact on breast cancer development1 The first evidence on the effect of dietary fatty acids on mammary carcinogenesis comes from studies of migrant populations2 and from animal experiments in which the role of diet in the development of spontaneous and/or chemically induced breast cancer have been investigated. Studies in rodents showed that not only the quantity but also the quality of fat is an important modulator of breast cancer risk. The results suggested that polyunsaturated fatty acids of the linoleic group (n-6 PUFA) stimulate mammary tumor development,3, 4 whereas polyunsaturated fatty acids of the linolenic group (n-3 PUFA) and especially those from marine origin (eicosapentaenoic acid and docosahexaenoic acid) inhibit tumor growth in mice.5 Saturated fatty acids were associated with an increase of breast cancer risk in experimental animal studies,6 whereas monounsaturated fatty acids and especially oleic acid have been shown to protect against breast cancer.7

Although the exact mechanism by which dietary fatty acids can modulate mammary carcinogenesis remain still unknown, metabolites of n-6 PUFA are believed to have an important impact. Indeed, a diet rich in linoleic acid (for example, corn oil) increase the synthesis of the n-6 fatty acids and consequently lead to the formation of a greater quantity of arachidonic acid (C20: 4, n-6), which is the precursor of leucotrienes and prostaglandin. The latter metabolites are very active and are able to act at various levels of the carcinogenesis process.8 The use of Indomethacin, an inhibitor of arachidonic acid metabolism (Cyclo oxygenase inhibitor), in some studies showed an inhibition of the stimulatory effect of n-6 PUFA in chemically induced tumors.9, 10 An increase of the n-3 fatty acids at the expense of n-6 could also counteract the tumor promoting effect of n-6 PUFA by decreasing the production of n-6 metabolites.5

Polyunsaturated fatty acids of n-6 family and particularly linoleic acid can also enhance mammary tumorigenesis by inhibiting the cellular gap junctions. The latter allows cellular communication and homeostasis. Once the communication is interrupted with surrounding cells, the isolated cell becomes anarchistic: it can undergo apoptosis or can become immortal, multiply in a permanent way and be at the origin of a tumoral process. Two studies showed that linoleic acid stimulates the cell proliferation by inhibiting 80% of gap junction 30 min after its addition to cell culture.11, 12 Dietary fat may also influence mammary carcinogenesis by affecting steroid hormone levels.13, 14

The possible role of dietary fat intake in the aetiology of breast cancer in human was widely investigated by epidemiological case control and cohort studies. Total and saturated fats were associated with breast cancer risk as shown in meta-analysis of case control studies and in a recent meta-analysis of cohort studies.15, 16 Other prospective cohort studies, however, have been less supportive for the fat-breast cancer hypothesis.17, 18, 19 These case-control and cohort studies estimated the fatty acids (SFA, MUFA and PUFA) from food frequency questionnaire or diet history. The estimation of fatty acid intake from reported dietary intake is exceptionally complex for many reasons, ranging from the difficulty to convert food items into their fatty acid content to the variation of fatty acid composition of the same food over the year, according to the cooking method and/or industry supply. In contrast, biomarkers of fatty acids offer quantitative measures of bioavailable amounts of these substances irrespective to the source of food, the subject's memory and the capacity to describe food consumed.20 In this regard, the use of biological markers of fat intake would be of major interest.

Dietary fatty acids intake is reflected by the concentration of particular fatty acids in adipose tissue, triglycerides, phospholipids and cholesteryl ester fractions of serum, plasma or erythrocyte membranes as well as free fatty acids.21, 22 As the half-life of fatty acids in adipose tissue is about 2 years,23 the fatty acid composition of this tissue can provide precious information about long-term intake. Subcutaneous fat aspiration, however, requires a local anaesthetic and may be painful and time-consuming. The fatty acid profile of serum or plasma phospholipids reflects medium-term intake (weeks to months) of dietary fat20, 24 and a blood sample can be obtained easily. These blood fractions can therefore be used as a biological marker of habitual dietary intake of fatty acids in large-scale epidemiological studies.25

In this article, we review the epidemiological evidence on the association between fatty acid profile of adipose tissue or blood fraction in relation to breast cancer risk. A meta-analytical approach was applied in order to clarify the possible role of dietary fatty acids in the aetiology of breast cancer.

METHODS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

Identification of studies

Included in this meta-analysis were case-control and cohort studies on biological markers of fatty acid intake and risk of breast cancer published from 1966 to 2002. Studies were identified by a search on the MEDLINE English language literature using the following mesh terms: biomarkers, dietary fat and breast cancer. Only published studies providing relative risk and 95% CI were used in the analysis. The summary statistic used was the average of the relative risk estimated for each level of the fatty acid on study, weighted by the inverse of its variance. Random effect models were assumed because studies results were heterogeneous (p for heterogeneity <0.01). All statistical analyses were performed using SAS statistical software version 8.02 (SAS Institute, Inc., Cary, NC).

Meta-analysis was conducted for case-control and cohort studies separately. Only 1 case-control and 1 cohort studies provided data for premenopausal breast cancer. Overall relative risks were estimated for studies including pre- and post-menopausal breast cancer and separately for postmenopausal breast cancer. In order to estimate overall relative risk for different levels of fatty acids, we pooled the relative risk estimated of the highest category of fatty acid from all studies (all noted as Q4 in Results). For the intermediate categories, we averaged the RR of second categories (noted as Q2), and the penultimate category (noted as Q3).

With the exception of 1 case-control study26 that analyzed the fatty acid composition of serum phospholipids, all other case-control studies involved in this meta-analysis were based on the analysis of fatty acid profile in adipose tissue and expressed the results as percentage of the total area. A sensitivity test by eliminating this study was performed in order to increase the homogeneity among these studies.

RESULTS

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The main characteristics of studies that analyzed the fatty acid composition of adipose tissue or blood lipids are summarized in Table I. We identified a total of 13 articles that contained information on the fatty acid composition of blood or adipose tissue and on estimated relative risk of breast cancer.26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38 Eight studies were carried out in European countries,26, 31, 32, 33, 34, 35, 36, 37 3 in USA,29, 30, 38 one in Moscow28 and another one in Israel.27 Fatty acids were quantified by gas chromatography in all studies but one.28 Controls were hospital-based in 7 case-control studies27, 28, 29, 30, 31, 35, 36 and population-based in the remaining. Fatty acid composition of adipose tissue was used in 8 case-control studies.27, 29, 30, 31, 32, 33, 35, 36 One case-control28 and one cohort study37 analyzed the fatty acid composition of erythrocyte membrane. Serum was used in one case-control26 and 2 cohort studies.34, 38 Among studies including pre- and post-menopausal women, only 228, 38 provided complete data for these 2 groups separately.

Table I. Characteristics of Studies Based on the Analysis of Biomarkers of Fat Intake in Relation to Breast Cancer Risk
Fatty acid: RR (95% CI) Highest vs. lowest category
Author and countrySaturated fatty acidsMonounsaturated fatty acidsTotal PUFAn-6 PUFAn-3 PUFA
Study design
Number of cases/controls
Biomarker
  • *

    Studies included in the meta-analysis. The RR is for the highest vs. lowest level of fatty acid unless otherwise noted.

  • ALA: Alpha linolenic acid; EPA: Eicosapentaenoic acid; DHA: Docosahexaenoic acid; AA: Arachidonic acid, SFA: Saturated fatty acids; MUFA: Monounsaturated fatty acids; PUFA: Polyunsaturated fatty acids.

Case-control studies     
 Eid and Bery27    Nonassociation
 Israël     
 1988     
 Hospital based case-control     
 32/27     
 54 years old in average     
 Subcutaneous adipose tissue     
 Zaridze28PremenopausalPremenopausalPremenopausalPremenopausal 
 MoscowPalmitic acid:Oleic acid:Stearic/Oleic:Linoleic acid: 
 19900.48 (0.14–1.64)0.71 (0.21–2.41)2.60 (0.77–8.78)0.25 (0.07–0.93) 
 Hospital based case-controlStearic acid: PUFA/SFA:AA: 
 46/530.84 (0.25–2.83) 0.74 (0.22–2.52)0.33 (0.08–1.35) 
 Pre and postmenopausalPostmenopausalPostmenopausalPostmenopausalPostmenopausal 
 Erythrocyte membranePalmitic acid:Oleic acid:Stearic/Oleic:Linoleic acid: 
 0.38 (0.12–1.21)0.63 (0.21–1.88)1.35 (0.44–4.13)0.46 (0.14–1.48) 
 RR by 2 categoriesStearic acid: PUFA/SFA:AA: 
 0.51 (0.17–1.57) 1.1 (0.36–3.40)0.23 (0.07–0.78) 
 *London et al.29Palmitic acid:Oleic acid:Total PUFA:Linoleic acid:ALA: 0.9 (0.6–1.5)
 Etats-Unis1.1 (0.7–1.8)1.2 (0.7–1.9)1.0 (0.6–1.6)0.9 (0.6–1.5)EPA: 0.7 (0.4–1.1)
 1993Stearic acid:Total MUFA: AA: 1.0 (0.6–1.6)DHA: 1.1 (0.6–1.7)
 Hospital based case-control1.0 (0.6–1.7)1.2 (0.7–1.9)   
 380/397Total SFA:    
 Postmenopausal1.0 (0.6–1.7)    
 Subcutaneous adipose tissue     
 RR by quantile     
 *Vatten et al.26Palmitic acid:Oleic acid:Total PUFA:Linoleic acid:ALA: 0.6 (0.3–1.4)
 Norway0.6 (0.3–1.2)0.6 (0.3–1.3)0.6 (0.3–1.2)0.4 (0.2–1.0)EPA: 0.9 (0.4–2.0)
 1993Stearic acid:Total MUFA:Total fat:AA: 0.7 (0.3–1.6)DHA: 0.6 (0.3–1.3)
 Population based case-control0.7 (0.3–1.5)0.6 (0.3–1.4)0.5 (0.2–1.2)Total n-6 PUFA:Total n-3 PUFA:
 87/235Total SFA: PUFA/SFA:0.5 (0.2–1.0)0.7 (0.3–1.6)
 Pre and postmenopausal 55 years or younger0.6 (0.3–1.2) 0.6 (0.3–1.3)  
   n-3/n-6:  
 Serum PL  1.0 (0.4–2.1)  
 RR by quartile     
 *Petrek et al.30Total SFA:Oleic acid: Linoleic acid:Total n-3 PUFA:
 Etats-Unis1.19 (0.58–2.37)1.47 (0.71–3.03) 0.58 (0.29–1.16)1.16 (0.59–2.34)
 1994     
 Hospital based case-control     
 154/125     
 Pre and postmenopausal     
 Breast adipose tissue     
 RR by quartile     
 EURAMIC     
 Postmenopausal     
 Subcutaneous     
 Adipose tissue     
 *Kohlmeier et al.31  Total PUFA  
 1997  1.26 (1.04–1.53)  
 Hospital based case-control     
 209/407     
 RR for the difference between 75th and 25th percentiles     
 *Simonsen et al.32, 33 Oleic acid: Total n-6 PUFA:ALA: 0.92 (0.53–1.60)
  0.64 (0.51–0.81)n-3/n-6:0.70 (0.38–1.19)1.42 (0.69–2.94)DHA: 1.09 (0.58–2.07)
  Palmitoleic/Oleic:  Total n-3 PUFA:
 1998 1.08 (0.90–1.29)EPA + DHA/n-6: 1.07 (0.56–2.05)
 Population based case-control  0.65 (0.41–1.03) Total EPA, DHA:
 291/351 Palmitoleic acid:  0.93 (0.50–1.70)
 RR for the difference between the 75th and 25th percentiles in the control population 0.68 (0.49–0.93)   
 *Klein et al.35Palmitic acid:Oleic acid: Linoleic acid:ALA: 0.36 (0.12–1.02)
 France0.90 (0.32–2.50)1.26 (0.48–3.34) 0.70 (0.27–1.82)Total n-3 PUFA:
 2000Stearic acid:  Total n-6 PUFA:2.54 (0.88–7.29)
 Hospital based case-control2.01 (0.69–5.89)  0.57 (0.22–1.49) 
 123/59     
 Pre and postmenopausal     
 Breast adipose tissue     
 RR by quartile     
 *Maillard et al.36Palmitic acid:Oleic acid: Linoleic acid:ALA: 0.39 (0.19–0.78)
 France0.44 (0.22–0.88)0.41 (0.21–0.82) 2.31 (1.15–4.67)DHA: 0.31 (0.13–0.75)
 2002Stearic acid:  AA: 0.98 (0.42–2.29)Total n-3 PUFA:
 Hospital based case-control1.17 (0.57–2.40)  Total n-6 PUFA:0.40 (0.17–0.94)
 241/88   2.29 (0.12–4.69) 
 Pre and postmenopausal     
 Breast adipose tissue     
 RR by tertile     
Cohort studies     
 *Chajes et al.34Palmitic acid:Oleic acid: 1.02 (0.98–2.25) Linoleic acid:ALA: 1.36 (0.63–2.96)
 Sweden2.09 (0.95–4.63)  1.41 (0.67–2.94)EPA: 0.51 (0.25–1.03)
 1999Stearic acid:Palmitoleic acid: AA:DHA: 0.92 (0.42–2.02)
 Prospective0.49 (0.22–1.08)0.73 (0.41–1.32) 0.51 (0.24–1.09)Total n-3 PUFA:
 196/390Total SFA:Total MUFA: Total n-6 PUFA:0.58 (0.27–1.28)
 Postmenopausal1.15 (0.46–2.85)1.78 (0.81–3.92) 0.91 (0.40–2.06) 
 Serum PL Stearic/Oleic:   
 RR by quartile 0.50 (0.23–1.10)   
 *Pala et al.37Palmitic acid:Oleic acid:Total PUFA:Linoleic acid:ALA: 1.38 (0.70–2.70)
 Italy1.49 (0.75–2.96)2.79 (1.24–6.28)0.34 (0.15–0.79)0.44 (0.20–1.00)EPA: 0.76 (0.35–1.62)
 2001Stearic acid:Palmitoleic acid: AA:DHA: 0.48 (0.23–1.00)
 Prospective0.68 (0.32–1.48)2.32 (1.03–5.20) 1.40 (0.64–3.10)Total n-3 PUFA:
 71/141Total SFA:Total MUFA: Total n-6 PUFA:0.53 (0.26–1.08)
 Postmenopausal1.01 (0.45–2.29)5.21 (1.95–13.91) 0.49 (0.22–1.06) 
 Erythrocyte membrane     
 RR by tertile Stearic/Oleic: 0.29 (0.13–0.64)   
 *Saadatian et al.38Pre-menopausalPre-menopausalPre-menopausalPre-menopausalPre-menopausal
 USAPalmitic acid:Oleic acid:Total PUFA:Linoleic acid:ALA: 0.97 (0.41–2.26)
 20011.20 (0.45–3.21)0.96 (0.37–2.45)0.60 (0.24–1.54)1.11 (0.42–2.94)EPA: 0.82 (0.32–2.11)
 ProspectiveStearic acid:Total MUFA:PUFA/SFA:Total n-6 PUFA:DHA: 0.83 (0.27–2.58)
 197/1972.31 (0.69–7.78)1.13 (0.42–3.04)0.45 (0.17–1.23)0.64 (0.22–1.86)Total n-3 PUFA:
 Pre- and postmenopausalTotal SFA:Stearic/Oleic:  0.79 (0.29–2.18)
 Serum PL1.66 (0.56–4.89)1.29 (0.50–3.28)   
 RR by quantilePost-menopausalPost-menopausalPost-menopausalPost-menopausalPost-menopausal
 Palmitic acid:Oleic acid:Total PUFA:Linoleic acid:ALA: 0.64 (0.26–1.57)
 2.57 (0.99–6.61)1.84 (0.72–4.71)0.42 (0.17–1.08)1.04 (0.40–2.67)EPA: 0.91 (0.90–0.99)
 Stearic acid:Total MUFA:PUFA/SFA:Total n-6 PUFA:DHA: 0.67 (0.27–1.70)
 1.01 (0.42–2.47)1.38 (0.55–3.49)1.17 (0.49–2.76)0.69 (0.28–1.73)Total n-3 PUFA:
 Total SFA:Stearic/Oleic:  0.68 (0.26–1.74)
 1.96 (0.73–5.25)0.60 (0.24–1.49)   

We included in the meta-analysis studies that published relative risks estimates and its confidence interval. These studies are marked with an asterix in Table I. One case-control study was excluded because it did not provide odds ratio and its confidence intervals.27 Another case-control study28 could not be included because fatty acids were categorized only into 2 categories due to the small study size (46 cases) and these categories were not comparable to those in other studies. We therefore used for the present article, 3 cohort and 7 case-control studies including 2,031 cases and 2,334 controls. Relative risk estimates and 95% confidence interval by category of fatty acids are shown in Table II.

Table II. Relative Risks Estimated for Case-Control and Cohort Studies
Fatty acid*Case-control (pre- and post-menopausal women)Number of studiesCohort (pre- and post-menopausal women)Cohort post-menopausal
Q2Q3Q4Q2Q3Q4Q4
  • *

    Reference group: Lowest level

  • β

    RR after the exclusion of the study.26 The number of cohort studies is 3.

Total SFA1.10 (0.78–1.57)0.91 (0.48–1.74)0.91 (0.63–1.32)31.07 (0.59–1.92)1.41 (0.58–3.47)1.36 (0.84–2.19)1.26 (1.10–1.45)
 β1.28 (0.87–1.87)β1.16 (0.79–1.71)β1.14 (0.52–2.49)     
Stearic acid0.84 (0.61–1.17)0.79 (0.42–1.49)0.90 (0.62–1.29)40.86 (0.54–1.37)0.69 (0.31–1.57)0.84 (0.47–1.47)0.68 (0.61–0.76)
 β0.91 (0.63–1.30)β1.72 (0.60–4.96)β0.93 (0.69–1.26)     
Palmitic acid0.81 (0.60–1.10)0.69 (0.50–1.00)0.91 (0.66–1.28)41.07 (0.72–1.60)1.24 (0.58–2.64)1.74 (1.15–2.63)1.89 (1.70–2.10)
 β0.81 (0.58–1.14)β0.72 (0.46–1.13)β0.85 (0.61–1.20)     
Total MUFA0.76 (0.35–1.65)0.73 (0.25–2.13)0.91 (0.46–1.76)21.09 (0.47–2.56)0.83 (0.46–1.49)1.93 (1.03–3.61)2.20 (1.93–2.52)
Palmitoleic acid0.89 (0.62–1.28)0.91 (0.60–1.39)1.06 (0.73–1.55)10.88 (0.59–1.29)1.00 (0.63–1.61)1.12 (0.75–1.69)1.12 (1.03–1.23)
Oleic acid0.80 (0.60–1.06)0.82 (0.59–1.13)0.90 (0.67–1.22)61.09 (0.61–1.98)0.98 (0.46–2.07)2.15 (1.68–2.74)1.45 (1.09–1.94)
 0.89 (0.71–1.44)1.01 (0.69–1.46)0.98 (0.70–1.36)     
Total n-6 PUFA0.88 (0.57–1.34)0.76 (0.38–1.49)0.96 (0.57–1.61)41.34 (0.83–2.18)0.88 (0.56–1.38)0.67 (0.44–1.02)0.67 (0.59–0.75)
 1.26 (0.76–2.14)0.76 (0.38–1.491.51 (0.77–2.97)     
Linoleic acid0.89 (0.62–1.27)1.04 (0.74–1.46)0.94 (0.68–1.30)50.91 (0.87–0.96)1.09 (0.67–1.76)0.91 (0.53–1.57)0.88 (0.78–0.98)
 β1.03 (0.69–1.54)β1.24 (0.87–1.77)β1.11 (0.78–1.58)     
Arachidonic acid0.81 (0.51–1.14)0.9 (0.58–1.40)0.93 (0.63–1.35)30.62 (0.38–0.99)0.95 (0.48–1.88)0.95 (0.48–1.88) 
 β0.82 (0.56–1.19)β0.9 (0.60–1.50)β0.99 (0.65–1.52)     
Total n-3 PUFA0.89 (0.61–1.29)0.87 (0.56–1.36)0.90 (0.59–1.36)50.68 (0.46–1.02)0.84 (0.52–1.38)0.61 (0.40–0.93)0.58 (0.52–0.64)
 β1.02 (0.66–1.58)β0.96 (0.63–1.48)β0.97 (0.60–1.58)     
Alpha linolenic acid0.86 (0.60–1.24)0.71 (0.52–0.98)0.64 (0.46–0.89)51.11 (0.75–1.64)1.24 (0.75–2.03)1.10 (0.74–1.63)1.14 (1.03–1.26)
 β0.99 (0.64–1.52)β0.81 (0.57–1.15)β0.65 (0.45–0.93)     
Eicosapentaenoic acid0.8 (0.53–1.20)0.86 (0.63–1.17)0.75 (0.49–1.15)21.19 (0.92–1.55)0.73 (0.45–1.19)0.69 (0.45–1.05)0.91 (0.86–0.95)
Docosahexaenoic acid0.83 (0.61–1.12)1.20 (0.70–1.89)0.81 (0.59–1.13)40.79 (0.42–1.52)0.80 (0.45–1.42)0.68 (0.44–1.04)0.66 (0.59–0.73)
 β0.89 (0.64–1.23)β1.30 (0.8–1.21)β0.88 (0.61–1.27)     

Saturated fatty acids (SFA)

Overall, total saturated fatty acids were not related to the risk of breast cancer in case-control studies. Cohort studies were suggestive of a direct association but the results were not statistically significant. Among the saturated fatty acids reported, cohort studies provided evidence of a statistically significant association between palmitic acid and breast cancer risk (RR=1.74, 95% CI: 1.15–2.63) for the highest versus the lowest category. The individual analysis of each cohort showed the same tendency although none of them reached statistically significant levels. Overall, case-control studies failed to find an association between palmitic acid and breast cancer risk. (RR=0.91, 95% CI: 0.66–1.28).

Monounsaturated fatty acids (MUFA)

As for saturated fatty acids, no relationship was found between total monounsaturated fatty acids and the risk of breast cancer in case-control studies, whereas cohort studies showed a statistically significant increase of almost 2-fold for the highest vs. the lowest level of monounsaturated fatty acids. Among individual monounsaturated fatty acids, oleic acid significantly increased breast cancer risk in cohort studies (RR=2.15, 95% CI: 1.68–2.74) but not in case control studies. No association was evidenced for palmitoleic acid in both case control and cohort studies.

n-6 polyunsaturated fatty acids (n-6 PUFA)

With the exception of EURAMIC study for which n-6 PUFA were associated with risk increase (RR=1.42, 95% CI: 0.69–2.93), all the other case-control and cohort studies consistently highlighted a protective effect. The overall relative risks for case control and cohort studies were not statistically significant but approached the limit of significance in cohort studies (RR=0.67, 95% CI: 0.44–1.02). The essential fatty acid of this group, linoleic acid, was not related to breast cancer risk, neither in case-control studies (RR=0.94, 95% CI: 0.68–1.30) nor in cohort studies (RR=0.91, 95% CI: 0.53–1.57).

-n-3 polyunsaturated fatty acids (n-3 PUFA)

The overall relative risk for cohort studies was statistically significant (RR= 0.61, 95% CI: 0.40–0.93), suggesting a protective effect of total n-3 PUFA against breast cancer risk. The result was, however, not found by case control studies (RR=0.90, 95% CI: 0.59–1.36). The results for alpha linolenic acid, the precursor of n-3 PUFA family, were in the opposite direction, although not statistically significant in cohort studies. The overall RR for cohort studies was higher than 1, while case-control studies provided relative risks estimated lower than 1. Cohort studies were suggestive of a protective, but not statistically significant, effect of eicosapentaenoic acid and docosahexaenoic acid, the very long chain fatty acids of this family (RR= 0.69, 95%CI: 0.45–1.05 and RR=0.68, 95% CI: 0.44–1.04), respectively. Results from case-control studies less strongly support the protective effect of these essential fatty acids.

Post-menopausal breast cancer

The analysis for case-control studies was not possible because only 2 reported the RR for postmenopausal women. Of the 3 cohort studies included in our meta-analysis, 2 were carried out in postmenopausal women34, 36 and the third37 provided data for pre- and post-menopausal women separately. We therefore could calculate the estimated relative risk for postmenopausal women in cohort studies. Our previous finding on the protective effect of n-3 PUFA remained after the exclusion of premenopausal women. The inverse association with the 2 very long chain n-3 PUFA, EPA and DHA became also statistically significant. Besides, we observed a statistically significant direct association between total SFA and total MUFA, oleic acid, palmitoleic acid and breast cancer risk. A significant protective effect was observed for total n-6 PUFA and linoleic acid. Stearic acid showed also a significant inverse association with breast cancer risk in postmenopausal women (RR=0.68, 95%CI: 0.6–0.76). The overall RR for the stearic/oleic acid ratio (called the saturation index; SI) in postmenopausal women was 0.43, 95%CI: 0.30–0.49.

DISCUSSION

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES

The principal objective of this analysis was to determine the extent to which the fatty acid composition of blood or adipose tissue is related to breast cancer risk. We limited our meta-analysis to studies involving biochemical markers of fatty acid intake because they are not subject to recall bias and measurement errors of dietary intake which can be relevant in both case-control and cohort studies and which may lead to a misclassification of dietary exposure.

In cohort studies, a significant protective effect was found for total n-3 PUFA while total MUFA, oleic acid (C18: 1 n-9c) and palmitic acid (C16: 0), the most abundant saturated fatty acid, were significantly associated with an increase of risk. Total SFA were significantly associated with breast cancer risk in cohort studies in post-menopausal women but not in the only cohort study in pre-menopausal breast cancer. For case-control studies, the only significant association was for alpha linolenic acid, which had a protective effect on breast cancer risk.

Except 1 case-control study,26 all the others analyzed the fatty acid composition of adipose tissue. In this case-control study, serum samples were stored at a relatively high temperature (−25°C) for several years before the laboratory analysis. Highly polyunsaturated fatty acids with several doubles bounds are very sensitive to the storage conditions and can be the subject of degradation. A sensitivity test by eliminating this case-control study from the meta-analysis increased the overall RR for SFA, but it was not statistically significant (RR=1.14, 95% CI: 0.52–2.49). Our previous finding on the protective effect of alpha linolenic acid remained after the exclusion of this study (RR=0.65, 95%CI: 0.45–0.93).

There are several possible explanations for the discrepancy between case-control and cohort studies results. These studies used different biomarkers (erythrocytes, serum and adipose tissue) with different time-frame exposures. Indeed, the majority of case-control studies involved in the present meta-analysis investigated fatty acid concentration in adipose tissue which reflect the relative intake over 2 years. The 3 cohort studies investigated fatty acids composition of plasma and erythrocyte membrane that indicate recent and medium intake respectively. In case-control studies the analysis of fatty acid profile is carried out when the cancer is already identified. The presence of the tumor might change the fatty acid composition because of the change in lipid metabolism related to the disease. Furthermore, the study populations have different characteristics (i.e., post and premenopausal together vs. postmenopausal).

Other sources of heterogeneity that can affect the results of case-control and cohort studies are differences in adjustment for confounding factors such as obesity, weight loss, tobacco and alcohol, which are known to be related to both cancer risk and fatty acid concentration. In addition, the results are not always adjusted for major breast cancer risk factors like parity, age at menarche, age at first full term pregnancy and age at menopause. The difference in study design across studies limit also the interpretation of the results. The results of our meta-analysis are also limited by publication bias toward positive finding, the reduced number of studies carried out on biomarkers of fat intake and breast cancer risk, in particular cohort studies, and multiple comparison of different fatty acids.

The findings of this meta-analysis fit well with the results derived from experimental animal studies for total SFA and n-3 PUFA, but not with those for n-6 PUFA. A summary of studies carried out on dietary fat intake and breast cancer risk is given in Table III. Dietary saturated fatty acids have been shown to be associated with breast cancer risk in 2 meta-analysis of case-control studies15, 16 and in a recent meta-analysis of cohort studies while we found an increased risk in cohort studies, statistically significant in postmenopausal women. Our meta-analysis was suggestive of a protective effect of stearic acid and the stearic/oleic acid ratio in breast cancer risk in cohort studies. Previous meta-analyses have not investigated this ratio. Stearic acid has already been found to act directly by inhibiting the proliferation of human tumors cells in vitro.39, 40 Most oleic acid in human body is derived from the saturated stearic acid by desaturation of this latter by the enzyme delta 9 desaturase (Δ9-d). Overexpression of this enzyme and consequently high levels of oleic acid has been shown to be required for tumor development in experimental animal studies.41, 42, 43 Δ9-d activity is modulated by several dietary factors: saturated fatty acids, cholesterol and carbohydrates stimulate this enzyme, whereas dietary polyunsaturated fatty acids and fasting are depressors.37

Table III. A Summary of Reviews and Meta-Analysis of Dietary Fat and Breast Cancer1
Fatty acidExperimental animal studiesEpidemiological studies based on dietary intakeMeta-analysis of studies based on biological markers
Case-controlCohortCase-controlCohortCohort postmenopausal
  • 1

    +no significant increase, ++significant increase, −no significant decrease, −−significant decrease. NA, nonassociation. SFA, saturated fatty acids; MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids; LA: linoleic acid; AA: arachidonic acid; ALA: alpha linolenic acid; EPA: eicosapentaenoic acid; DHA: docosahexaenoic acid, Total PUFA: n-6 and n-3 polyunsaturated fatty acids.

Total fat+53++15++16   
 +3++16NA17   
 ++54++56NA57   
 ++55 NA56   
   NA58   
Total SFA+6++15++16NA+++
 +53++16NA17   
 +59 NA58   
Palmitic acid   NA++++
Stearic acid   NANA−−
Total MUFA6++15NA16NA++++
  ++16NA12   
   NA58   
Palmitoleic acid   NA+++
Oleic acidNo consistency49  NA++++
 No consistency3     
Total PUFA n-6++6  NA−−
 +47     
 ++59     
 +60     
 +4     
 +54     
LA+47NA46NA46NANA−−
 ++55     
 +54     
 +3     
AA   NANA 
Total PUFA n-36     
 −−53  NA−−−−
 49     
 3     
 59     
 60     
ALANo consistency49  NA++
EPA3  −−
DHA3  −−
Total PUFA NA15NA16   
  NA16NA17   
   NA58   

Individual animal3 and epidemiological studies44, 45 have generally shown that olive oil has either no effect or a protective effect on breast cancer risk, which were initially attributed to its high content of the monounsaturated fatty acid, oleic acid. Our results did not show any significant association between oleic acid and breast cancer risk in case-control studies, whereas in cohort studies, we found significantly higher risk for women with high levels of this fatty acid. The conflicting results of the literature on the role of this fatty acid on breast cancer risk could underline the fact that the saturation index (stearic/oleic acid ratio), rather than the oleic acid component alone, may be further related to the risk of breast cancer.

Linoleic acid, the essential fatty acid of n-6 PUFA family, showed a significant inverse association in cohort studies restricted to postmenopausal women but was not associated with breast cancer risk in case-control studies. Epidemiological studies on dietary linoleic acid have reported no association46 while experimental animal studies have generally reported that linoleic acid can promote the growth of mammary tumors in rodents.47 However, one must note that in experimental models, linoleic acid promote mammary tumorigenesis only under certain conditions.48

Our analysis was suggestive of a significant protective effect of total n-3 PUFA, and the 2 highly n-3 PUFA of marine origin, EPA and DHA. A large number of experimental animal data also showed inhibition of mammary tumorigenesis in mice or rats fed high dietary levels of fish oils49 but the data from epidemiological studies on dietary fat intake remain inconclusive. Lipid peroxidation products can act as cytotoxic compounds and therefore inhibit the growth of malignant cells.50 n-3 PUFA and especially the very long chain fatty acids of this family; EPA and DHA have several double bounds and are excellent sources for lipid peroxidation. The ratio between n-3 and n-6 PUFA can also be an important factor in modulating cancer risk. Experimental studies reported that the tumor inhibitory effect of n-3 PUFA is observed only for optimal ratio of these 2 PUFA families.51 Only a few studies published results for the ratio of n-3/n-6 in studies of biological markers of fat intake.

In summary, the results of our meta-analysis of cohort studies, in conjunction with experimental animal studies, suggest that total and very long chain n-3 PUFA play a protective effect, while total SFA, total MUFA, palmitic and oleic acid are associated with an increased risk of breast cancer. Epidemiological case-control and cohort studies in human are relevant and provide useful information on diet-disease relationship. However, in these studies fatty acids are often measured indirectly using food frequency questionnaires that can lead to underestimation of disease-exposure association due to incompleteness of the questionnaires.52 There are also considerable amounts of hidden fat in foods, which are in general not known for the studied subjects. In addition, Food composition tables and nutrient databases are often incomplete for specific fatty acids. In contrast with data provided by dietary questionnaires, fatty acid composition of adipose tissue, erythrocyte membranes, serum and plasma used as biological markers of fatty acid intake give quantitative measurement independent of energy intake and reflect bioavailaible and postabsorptive amounts of fat consumed and therefore eliminate the inadequacies of dietary questionnaires and nutrient databases. The fatty acid profile of adipose tissue or blood would be therefore of major interest in nutritional epidemiological studies on the relationship between dietary fat and breast cancer that can complete estimation of dietary intake. More epidemiological prospective cohort studies on biomarkers of fatty acid intake should be conducted in order to determine the exact role of the dietary fat in the etiology of breast cancer. It is crucial to present results adjusted for the main known risk factors and to provide results separately according to the menopausal status.

REFERENCES

  1. Top of page
  2. Abstract
  3. METHODS
  4. RESULTS
  5. DISCUSSION
  6. REFERENCES
  • 1
    WCRF/AICR. Food, nutrition and the prevention of cancer: a global perspective. Washington, DC: World Cancer Research Fund/ American Institute of Cancer Research, 1997.
  • 2
    Ziegler RG, Hoover RN, Pike MC, Hildesheim A, Nomura AM, West DW, Wu-Williams AH, Kolonel LN, Horn-Ross PL, Rosenthal JF. Migration patterns and breast cancer risk in Asian-American women. J Natl Cancer Inst 1993; 85: 181927.
  • 3
    Welsch CW. Relationship between dietary fat and experimental mammary tumorigenesis: a review and critique. Cancer Res 1992; 52: 2040s8s.
  • 4
    Bartsch H, Nair J, Owen RW. Dietary polyunsaturated fatty acids and cancers of the breast and colorectum: emerging evidence for their role as risk modifiers. Carcinogenesis 1999; 20: 220918.
  • 5
    Connolly JM, Gilhooly EM, Rose DP. Effects of reduced dietary linoleic acid intake, alone or combined with an algal source of docosahexaenoic acid, on MDA-MB-231 breast cancer cell growth and apoptosis in nude mice. Nutr Cancer 1999; 35: 449.
  • 6
    Fay MP, Freedman LS, Clifford CK, Midthune DN. Effect of different types and amounts of fat on the development of mammary tumors in rodents: a review. Cancer Res 1997; 57: 397988.
  • 7
    Newmark HL. Squalene, olive oil, and cancer risk. Review and hypothesis. Ann N Y Acad Sci 1999; 889: 193203.
  • 8
    Lupulescu A. Prostaglandins, their inhibitors and cancer. Prostaglandins Leukot Essent Fatty Acids 1996; 54: 8394.
  • 9
    Morecki S, Yacovlev L, Slavin S. Effect of indomethacin on tumorigenicity and immunity induction in a murine model of mammary carcinoma. Int J Cancer 1998; 75: 8949.
  • 10
    Kundu N, Yang Q, Dorsey R, Fulton AM. Increased cyclooxygenase-2 (cox-2) expression and activity in a murine model of metastatic breast cancer. Int J Cancer 2001; 93: 6816.
  • 11
    de Haan LH, Bosselaers I, Jongen WM, Zwijsen RM, Koeman JH. Effect of lipids and aldehydes on gap-junctional intercellular communication between human smooth muscle cells. Carcinogenesis 1994; 15: 2536.
  • 12
    Hayashi T, Matesic DF, Nomata K, Kang KS, Chang CC, Trosko JE. Stimulation of cell proliferation and inhibition of gap junctional intercellular communication by linoleic acid. Cancer Lett 1997; 112: 10311.
  • 13
    Wu AH, Pike MC, Stram DO. Meta-analysis: dietary fat intake, serum estrogen levels, and the risk of breast cancer. J Natl Cancer Inst 1999; 91: 52934.
  • 14
    Zhu BT, Conney AH. Functional role of estrogen metabolism in target cells: review and perspectives. Carcinogenesis 1998; 19: 127.
  • 15
    Howe GR, Hirohata T, Hislop TG et al. Dietary factors and risk of breast cancer: combined analysis of 12 case-control studies. J Natl Cancer Inst 1990; 82: 5619.
  • 16
    Boyd NF, Stone J, Vogt KN, Connelly BS, Martin LJ, and Minkin S. Dietary fat and breast cancer risk revisited: a meta-analysis of the published literature. Br J Cancer 2003; 89: 167285.
  • 17
    Hunter DJ, Spiegelman D, Adami HO, Beeson L, van den Brandt PA, Folsom AR, Fraser GE, Goldbohm RA, Graham S, Howe GR. Cohort studies of fat intake and the risk of breast cancer: a pooled analysis. N Engl J Med 1996; 334: 35661.
  • 18
    Willett WC. Specific fatty acids and risks of breast and prostate cancer: dietary intake. Am J Clin Nutr 1997; 66: 1557S63S.
  • 19
    Holmes MD, Hunter DJ, Colditz GA, Stampfer MJ, Hankinson SE, Speizer FE, Rosner B, Willett WC. Association of dietary intake of fat and fatty acids with risk of breast cancer. JAMA 1999; 281: 91420.
  • 20
    Riboli E, Ronnholm H, Saracci R. Biological markers of diet. Cancer Surv 1987; 6: 685718.
  • 21
    Melchert HU, Limsathayourat N, Mihajlovic H, Eichberg J, Thefeld W, Rottka H. Fatty acid patterns in triglycerides, diglycerides, free fatty acids, cholesteryl esters and phosphatidylcholine in serum from vegetarians and non-vegetarians. Atherosclerosis 1987; 65: 15966.
  • 22
    Dougherty RM, Galli C, Ferro-Luzzi A, Iacono JM. Lipid and phospholipid fatty acid composition of plasma, red blood cells, and platelets and how they are affected by dietary lipids: a study of normal subjects from Italy, Finland, and the USA. Am J Clin Nutr 1987; 45: 44355.
  • 23
    Dayton S, Hashimoto S, Dixon W, Pearce ML. Composition of lipids in human serum and adipose tissue during prolonged feeding of a diet high in unsaturated fat. J Lipid Res 1966; 7: 10311.
  • 24
    Zock PL, Mensink RP, Harryvan J, de Vries JH, Katan MB. Fatty acids in serum cholesteryl esters as quantitative biomarkers of dietary intake in humans. Am J Epidemiol 1997; 145: 111422.
  • 25
    Zeleniuch-Jacquotte A, Chajès V, Van Kappel AL, Riboli E, Toniolo P. Reliability of fatty acid composition in human serum phospholipids. Eur J Clin Nutr 2000; 54: 36772.
  • 26
    Vatten LJ, Bjerve KS, Andersen A, Jellum E. Polyunsaturated fatty acids in serum phospholipids and risk of breast cancer: a case-control study from the Janus serum bank in Norway. Eur J Cancer 1993; 29A: 5328.
  • 27
    Eid A, Berry EM. The relationship between dietary fat, adipose tissue composition, and neoplasms of the breast. Nutr Cancer 1988; 11: 1737.
  • 28
    Zaridze DG, Chevchenko VE, Levtshuk AA, Lifanova YE, Maximovitch DM. Fatty acid composition of phospholipids in erythrocyte membranes and risk of breast cancer. Int J Cancer 1990; 45: 80710.
  • 29
    London SJ, Sacks FM, Stampfer MJ, Henderson IC, Maclure M, Tomita A, Wood WC, Remine S, Robert NJ, Dmochowski JR. Fatty acid composition of the subcutaneous adipose tissue and risk of proliferative benign breast disease and breast cancer. J Natl Cancer Inst 1993; 85: 78593.
  • 30
    Petrek JA, Hudgins LC, Levine B, Ho M, Hirsch J. Breast cancer risk and fatty acids in the breast and abdominal adipose tissues. J Natl Cancer Inst 1994; 86: 536.
  • 31
    Kohlmeier L, Simonsen N, van't V, Strain JJ, Martin-Moreno JM, Margolin B, Huttunen JK, Fernandez-Crehuet Navajas J, Martin BC, Thamm M, Kardinaal AF, Kok FJ. Adipose tissue trans fatty acids and breast cancer in the European Community Multicenter Study on Antioxidants, Myocardial Infarction, and Breast Cancer. Cancer Epidemiol Biomarkers Prev 1997; 6: 70510.
  • 32
    Simonsen NR, Fernandez-Crehuet NJ, Martin-Moreno JM, Strain JJ, Huttunen JK, Martin BC, Thamm M, Kardinaal AF, van't Veer P, Kok FJ, Kohlmeier L. Tissue stores of individual monounsaturated fatty acids and breast cancer: the EURAMIC study: European Community Multicenter Study on Antioxidants, Myocardial Infarction, and Breast Cancer. Am J Clin Nutr 1998; 68: 13441.
  • 33
    Simonsen N, van't Veer P, Strain JJ, Martin-Moreno JM, Huttunen JK, Navajas JF, Martin BC, Thamm M, Kardinaal AF, Kok FJ, Kohlmeier L. Adipose tissue omega-3 and omega-6 fatty acid content and breast cancer in the EURAMIC study: European Community Multicenter Study on Antioxidants, Myocardial Infarction, and Breast Cancer. Am J Epidemiol 1998; 147: 34252.
  • 34
    Chajes V, Hulten K, Van Kappel AL, Winkvist A, Kaaks R, Hallmans G, Lenner P, Riboli E. Fatty-acid composition in serum phospholipids and risk of breast cancer: an incident case-control study in Sweden. Int J Cancer 1999; 83: 58590.
  • 35
    Klein V, Chajès V, Germain E, Schulgen G, Pinault M, Malvy D, Lefrancq T, Fignon A, Le Floch O, Lhuillery C, Bougnoux P. Low alpha-linolenic acid content of adipose breast tissue is associated with an increased risk of breast cancer. Eur J Cancer 2000; 36: 33540.
  • 36
    Maillard V, Bougnoux P, Ferrari P, Jourdan ML, Pinault M, Lavillonniere F, Body G, Le Floch O, Chajes V. N-3 and N-6 fatty acids in breast adipose tissue and relative risk of breast cancer in a case-control study in Tours, France. Int J Cancer 2002; 98: 7883
  • 37
    Pala V, Krogh V, Muti P, Chajes V, Riboli E, Micheli A, Saadatian M, Sieri S, Berrino F. Erythrocyte membrane fatty acids and subsequent breast cancer: a prospective Italian study. J Natl Cancer Inst 2001; 93: 108895.
  • 38
    Mitra Saadatian-Elahi, Paolo Toniolo, Pietro Ferrari, Joëlle Goudable, Arslan Akhmedkhanov, Anne Zeleniuch-Jacquotte, Elio Riboli. 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 Prevention. 2002; 11: 135360.
  • 39
    Fermor BF, Masters JR, Wood CB, Miller J, Apostolov K, Habib NA. Fatty acid composition of normal and malignant cells and cytotoxicity of stearic, oleic and sterculic acids in vitro. Eur J Cancer 1992; 28A: 11437.
  • 40
    Habib NA, Wood CB, Apostolov K, Barker W, Hershman MJ, Aslam M, Heinemann D, Fermor B, Williamson RC, Jenkins WE. Stearic acid and carcinogenesis. Br J Cancer 1987; 56: 4558.
  • 41
    Lu J, Pei H, Kaeck M, Thompson HJ. Gene expression changes associated with chemically induced rat mammary carcinogenesis. Mol Carcinog 1997; 20: 20415.
  • 42
    de Alaniz MJ, Marra CA. Role of delta 9 desaturase activity in the maintenance of high levels of monoenoic fatty acids in hepatoma cultured cells. Mol Cell Biochem 1994; 137: 8590.
  • 43
    Marzo I, Martinez-Lorenzo MJ, Anel A, Desportes P, Alava MA, Naval J, Pineiro A. Biosynthesis of unsaturated fatty acids in the main cell lineages of human leukemia and lymphoma. Biochim Biophys Acta 1995; 1257: 1408.
  • 44
    La Vecchia C, Negri E, Franceschi S, Decarli A, Giacosa A, Lipworth L. Olive oil, other dietary fats, and the risk of breast cancer (Italy). Cancer Causes Control 1995; 6: 54550.
  • 45
    Wolk A, Bergstrom R, Hunter D, Willett W, Ljung H, Holmberg L, Bergkvist L, Bruce A, Adami HO. A prospective study of association of monounsaturated fat and other types of fat with risk of breast cancer. Arch Intern Med 1998; 158: 415.
  • 46
    Zock PL, Katan MB. Linoleic acid intake and cancer risk: a review and meta-analysis. Am J Clin Nutr 1998; 68: 14253.
  • 47
    Rose DP. Effects of dietary fatty acids on breast and prostate cancers: evidence from in vitro experiments and animal studies. Am J Clin Nutr 1997; 66: 1513S22S.
  • 48
    Ip C. Controversial issues of dietary fat and experimental mammary carcinogenesis. Prev Med 1993; 22: 72837.
  • 49
    Ip C. Review of the effects of trans fatty acids, oleic acid, n-3 polyunsaturated fatty acids, and conjugated linoleic acid on mammary carcinogenesis in animals. Am J Clin Nutr 1997; 66: 1523S9S.
  • 50
    Bougnoux P. n-3 polyunsaturated fatty acids and cancer. Curr Opin Clin Nutr Metab Care 1999; 2: 1216.
  • 51
    Cohen LA, Chen-Backlund JY, Sepkovic DW, Sugie S. Effect of varying proportions of dietary menhaden and corn oil on experimental rat mammary tumor promotion. Lipids 1993; 28: 44956.
  • 52
    Prentice RL. Measurement error and results from analytic epidemiology: dietary fat and breast cancer. J Natl Cancer Inst 1996; 88: 173847.
  • 53
    Freedman LS, Clifford CK. Meta-analysis of animal experiments: elucidating relationships between dietary fat and mammary tumor development in rodents. Adv Exp Med Biol 1994; 364: 93100.
  • 54
    Rogers AE. Diet and breast cancer: studies in laboratory animals. J Nutr 1997; 127: 933S5S.
  • 55
    Wynder EL, Cohen LA, Muscat JE, Winters B, Dwyer JT, Blackburn G. Breast cancer: weighing the evidence for a promoting role of dietary fat. J Natl Cancer Inst 1997; 89: 76675.
  • 56
    Clavel-Chapelon F, Niravong M, Joseph RR. Diet and breast cancer: review of the epidemiologic literature. Cancer Detect Prev 1997; 21: 42640.
  • 57
    Willett WC, Hunter DJ. Prospective studies of diet and breast cancer. Cancer 1994; 74: 10859.
  • 58
    Smith-Warner SA, Spiegelman D, Adami HO, Beeson WL, van den Brandt PA, Folsom AR, Fraser GE, Freudenheim JL, Goldbohm RA, Graham S, Kushi LH, Miller AB, et al. Types of dietary fat and breast cancer: a pooled analysis of cohort studies. Int J Cancer 2001; 92: 76774.
  • 59
    Stoll BA. Breast cancer and the western diet: role of fatty acids and antioxidant vitamins. Eur J Cancer 1998; 34: 18526.
  • 60
    Cave WT, Jr. Dietary omega-3 polyunsaturated fats and breast cancer. Nutrition 1996; 12: S3942.