Recent studies suggest that elevated body iron levels may contribute to breast carcinogenesis; however, epidemiologic evidence is lacking. We used data from a large cohort study of Canadian women to assess breast cancer in association with total iron and heme iron intake. Among 49,654 women ages 40 to 59 followed for an average of 16.4 years, we identified 2,545 incident breast cancer cases. Data from a food frequency questionnaire administered at baseline were used to calculate total dietary iron and heme iron intake. Using Cox proportional hazards models, we found no association of iron or heme iron intake with risk of breast cancer overall, in women consuming 30+ g of alcohol per day, or in women who had ever used hormone replacement therapy. The present study offers no support for an association of iron or heme iron intake with breast cancer risk or for a modification by iron of the effect of alcohol or estrogen. However, further cohort studies with repeated measurement of iron intake are warranted. (Cancer Epidemiol Biomarkers Prev 2007;16(6):1306–8)

Free iron is associated with oxidative DNA damage and lipid peroxidation (1-4), both of which are thought to contribute to breast carcinogenesis (5-11). Iron has been shown to enhance chemically initiated mammary carcinogenesis (12, 13) and may potentiate the effects of two established risk factors for breast cancer (i.e., alcohol and exogenous estrogens). Both alcohol and estrogen are capable of inducing oxidative DNA damage, and both may disturb iron homeostasis by releasing iron from its bound form, further contributing to the production of reactive oxygen species (2, 14). Intake of red meat, an important source of the more bioavailable heme iron, has shown an inconsistent association with breast cancer (15). However, to date, only one report (16) has assessed intake of heme iron in relation to breast cancer risk, reporting a positive association of iron and heme iron intake with postmenopausal breast cancer in women consuming 20+ g of alcohol per day but no overall association. Given the limited epidemiologic evidence currently available, we used data from a large cohort study of Canadian women to assess breast cancer risk in association with total iron intake and heme iron intake.

Our study was conducted in the Canadian National Breast Screening Study, a randomized controlled trial of screening for breast cancer involving 89,835 women ages 40 to 59 at baseline enrolled between 1980 and 1985 (17, 18). Starting in 1982, dietary information was obtained by means of a food frequency questionnaire completed by women attending participating screening centers (19). The food frequency questionnaire elicited information on usual portion size and consumption of 86 food items and included photographs of portion sizes to assist respondents in quantifying intake. A total of 49,654 dietary questionnaires was returned. Data from the food frequency questionnaire were used to calculate total dietary iron intake using a database described elsewhere (19). The values for iron intake presented here are for dietary sources alone because data on iron supplements were not collected. Total intake of meat iron was calculated from the reported intake of 22 meat items and 2 mixed dishes containing meat. Heme iron intake was computed by two different methods using different proportions for heme iron from different types of meat: 69% for beef; 39% for pork, ham, bacon, pork-based luncheon meats, and veal; 26% for chicken and fish; and 21% for liver (20); and, alternatively, using 40% as the average proportion of heme iron in all meats (21). Results were similar for both methods, and we present data using the first approach. Total iron and heme iron intake were calorie adjusted using the residuals method (22). In addition, we assessed meat iron intake and red meat intake.

During an average follow-up of 16.4 years through the end of December 2000, 2,545 breast cancer cases were identified by means of record linkage to the Canadian Cancer Registry. We excluded 947 women with extreme energy intake values (<730 or >6,485 kcal/d) and women whose body mass index was <15 kg/m2 or >50 kg/m2 or who were missing information on body mass index. In addition, 4 women with prevalent breast cancer were excluded, leaving 48,662 women, including 2,491 cases, available for analysis.

Cox proportional hazards models (using age as the time scale) were used to estimate hazard ratios (HR) and 95% confidence intervals (95% CI) for the association between iron intake and breast cancer risk. Quintiles and deciles of iron-related variables were derived based on their distribution in the total population. All multivariate models included covariates shown in the footnotes to Tables 1 and 2. We also examined the associations within strata of alcohol consumption (nondrinker versus drinkers of 30+ g/d) and hormone replacement therapy (HRT) use (ever versus never), and we tested for interactions using likelihood ratio tests. To test for trends in risk with increasing levels of exposure, we assigned the median value for each quintile (or decile) and then fitted the medians as a continuous variable in the risk models. We then evaluated the statistical significance of the corresponding coefficient using the Wald test (23). All analyses were done using Statistical Analysis System version 9 (SAS Institute). Use of the lifetest procedure in Statistical Analysis System showed that the proportional hazards assumption was met in this data set. All tests of statistical significance were two sided.

Table 1.

Multivariate-adjusted HRs and 95% CIs for the association of four measures of dietary iron or meat intake and incident breast cancer

FactorMultivariate HR* (95% CI)
Premenopausal (n cases = 1,171)Postmenopausal (n cases = 993)All women (n cases = 2,491)
Total iron (mg/d)    
    <11.90 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    11.90 to <12.90 1.06 (0.88-1.27) 0.90 (0.73-1.10) 0.98 (0.86-1.11) 
    12.90 to <13.82 1.04 (0.86-1.25) 0.88 (0.72-1.07) 0.96 (0.84-1.09) 
    13.82 to <14.99 0.97 (0.80-1.17) 0.93 (0.76-1.14) 0.97 (0.85-1.10) 
    ≥14.99 1.07 (0.89-1.30) 0.87 (0.71-1.06) 0.97 (0.85-1.10) 
    P for trend 0.82 0.28 0.63 
Heme iron (mg/d)    
    <1.58 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    1.58 to <1.99 1.06 (0.88-1.27) 1.00 (0.82-1.23) 1.03 (0.91-1.18) 
    1.99 to <2.40 1.08 (0.89-1.30) 1.07 (0.87-1.30) 1.10 (0.97-1.25) 
    2.40 to <2.95 1.14 (0.94-1.37) 1.15 (0.94-1.41) 1.15 (1.01-1.31) 
    >2.95 1.03 (0.84-1.25) 0.97 (0.78-1.20) 1.03 (0.90-1.18) 
    P for trend 0.56 0.71 0.25 
Meat iron (mg/d)    
    <3.40 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    3.40 to <4.23 1.17 (0.97-1.40) 1.05 (0.86-1.29) 1.09 (0.96-1.23) 
    4.23 to <5.02 0.91 (0.74-1.58) 1.15 (0.94-1.41) 1.02 (0.90-1.17) 
    5.02 to <6.11 1.13 (0.94-1.37) 1.08 (0.88-1.32) 1.09 (0.96-1.24) 
    ≥6.11 1.13 (0.93-1.37) 1.03 (0.83-1.27) 1.09 (0.96-1.24) 
    P for trend 0.35 0.73 0.26 
Red meat (g/d)    
    <14.25 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    14.25 to <21.02 1.10 (0.91-1.34) 0.99 (0.81-1.19) 0.98 (0.87-1.11) 
    21.02 to <28.74 1.29 (1.07-1.58) 0.87 (0.71-1.06) 1.04 (0.92-1.16) 
    28.74 to <40.30 1.18 (0.97-1.43) 0.89 (0.72-1.09) 1.03 (0.90-1.18) 
    40.30+ 1.14 (0.94-1.39) 0.89 (0.72-1.09) 0.98 (0.86-1.12) 
    P for trend 0.16 0.13 0.91 
FactorMultivariate HR* (95% CI)
Premenopausal (n cases = 1,171)Postmenopausal (n cases = 993)All women (n cases = 2,491)
Total iron (mg/d)    
    <11.90 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    11.90 to <12.90 1.06 (0.88-1.27) 0.90 (0.73-1.10) 0.98 (0.86-1.11) 
    12.90 to <13.82 1.04 (0.86-1.25) 0.88 (0.72-1.07) 0.96 (0.84-1.09) 
    13.82 to <14.99 0.97 (0.80-1.17) 0.93 (0.76-1.14) 0.97 (0.85-1.10) 
    ≥14.99 1.07 (0.89-1.30) 0.87 (0.71-1.06) 0.97 (0.85-1.10) 
    P for trend 0.82 0.28 0.63 
Heme iron (mg/d)    
    <1.58 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    1.58 to <1.99 1.06 (0.88-1.27) 1.00 (0.82-1.23) 1.03 (0.91-1.18) 
    1.99 to <2.40 1.08 (0.89-1.30) 1.07 (0.87-1.30) 1.10 (0.97-1.25) 
    2.40 to <2.95 1.14 (0.94-1.37) 1.15 (0.94-1.41) 1.15 (1.01-1.31) 
    >2.95 1.03 (0.84-1.25) 0.97 (0.78-1.20) 1.03 (0.90-1.18) 
    P for trend 0.56 0.71 0.25 
Meat iron (mg/d)    
    <3.40 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    3.40 to <4.23 1.17 (0.97-1.40) 1.05 (0.86-1.29) 1.09 (0.96-1.23) 
    4.23 to <5.02 0.91 (0.74-1.58) 1.15 (0.94-1.41) 1.02 (0.90-1.17) 
    5.02 to <6.11 1.13 (0.94-1.37) 1.08 (0.88-1.32) 1.09 (0.96-1.24) 
    ≥6.11 1.13 (0.93-1.37) 1.03 (0.83-1.27) 1.09 (0.96-1.24) 
    P for trend 0.35 0.73 0.26 
Red meat (g/d)    
    <14.25 1.00 (reference) 1.00 (reference) 1.00 (reference) 
    14.25 to <21.02 1.10 (0.91-1.34) 0.99 (0.81-1.19) 0.98 (0.87-1.11) 
    21.02 to <28.74 1.29 (1.07-1.58) 0.87 (0.71-1.06) 1.04 (0.92-1.16) 
    28.74 to <40.30 1.18 (0.97-1.43) 0.89 (0.72-1.09) 1.03 (0.90-1.18) 
    40.30+ 1.14 (0.94-1.39) 0.89 (0.72-1.09) 0.98 (0.86-1.12) 
    P for trend 0.16 0.13 0.91 
*

Adjusted for age (time to event variable), body mass index (kg/m2; quintiles), menopausal status (premenopausal, perimenopausal, postmenopausal), parity (parous/nulliparous), age at menarche (continuous), family history of breast cancer in a first-degree relative (yes/no), history of benign breast disease (yes, no), oral contraceptive use (ever/never), HRT (ever/never), total calorie intake (continuous), alcohol intake (never drinker, >0 to <5 g/d, 5 to <10 g/d, 10 to <20 g/d, 20 to <30 g/d, 30 to <40 g/d, 40+ g/d), education (three levels), study center, and randomization group (intervention or control in the original clinical trial).

Three hundred and twenty-seven cases and 6,365 noncases were classified as perimenopausal and therefore excluded.

Table 2.

Multivariate-adjusted HRs and 95% CIs for the association between heme iron intake and breast cancer, stratified by alcohol consumption and use of HRT

Stratification variables
HR* (95% CI)
HR* (95% CI)
Alcohol intakeNondrinker (n cases = 524)>30 g/d (n cases = 193)
Heme iron intake   
    <1.58 1.00 (reference) 1.00 (reference) 
    1.58 to <1.99 0.94 (0.71-1.26) 1.13 (0.76-1.69) 
    1.99 to <2.40 1.07 (0.81-1.42) 1.38 (0.92-2.08) 
    2.40 to <2.95 1.06 (0.81-1.40) 1.32 (0.84-2.06) 
    >2.95 0.99 (0.76-1.30) 0.80 (0.45-1.43) 
    P for trend 0.82 0.78 
   
HRT Never (n cases = 1,831) Ever (n cases = 660) 
Heme iron intake   
    <1.58 1.00 (reference) 1.00 (reference) 
    1.58 to <1.99 1.02 (0.87-1.18) 1.10 (0.86-1.43) 
    1.99 to <2.40 1.05 (0.90-1.22) 1.25 (0.98-1.61) 
    2.40 to <2.95 1.13 (0.98-1.32) 1.23 (0.95-1.58) 
    >2.95 1.06 (0.91-1.24) 0.96 (0.73-1.25) 
    P for trend 0.18 0.92 
Stratification variables
HR* (95% CI)
HR* (95% CI)
Alcohol intakeNondrinker (n cases = 524)>30 g/d (n cases = 193)
Heme iron intake   
    <1.58 1.00 (reference) 1.00 (reference) 
    1.58 to <1.99 0.94 (0.71-1.26) 1.13 (0.76-1.69) 
    1.99 to <2.40 1.07 (0.81-1.42) 1.38 (0.92-2.08) 
    2.40 to <2.95 1.06 (0.81-1.40) 1.32 (0.84-2.06) 
    >2.95 0.99 (0.76-1.30) 0.80 (0.45-1.43) 
    P for trend 0.82 0.78 
   
HRT Never (n cases = 1,831) Ever (n cases = 660) 
Heme iron intake   
    <1.58 1.00 (reference) 1.00 (reference) 
    1.58 to <1.99 1.02 (0.87-1.18) 1.10 (0.86-1.43) 
    1.99 to <2.40 1.05 (0.90-1.22) 1.25 (0.98-1.61) 
    2.40 to <2.95 1.13 (0.98-1.32) 1.23 (0.95-1.58) 
    >2.95 1.06 (0.91-1.24) 0.96 (0.73-1.25) 
    P for trend 0.18 0.92 
*

Adjusted for the following variables (except the stratification variable): age (time to event variable), body mass index (kg/m2; quintiles), menopausal status (premenopausal, perimenopausal, postmenopausal), parity (parous/nulliparous), age at menarche (continuous), family history of breast cancer in a first-degree relative (yes/no), history of benign breast disease (yes/no), oral contraceptive use (ever/never), HRT (ever/never), total calorie intake (continuous), alcohol intake (continuous), education (three levels), study center, and randomization group (intervention or control in the original clinical trial).

In multivariate models, quintiles of total iron intake, heme iron intake, iron intake from meat, and red meat intake showed no association with breast cancer risk in premenopausal women, postmenopausal women, or all women combined (Table 1). Non–heme iron intake also was not associated with breast cancer risk (data not shown). Furthermore, no associations or trends were seen when iron-related variables were categorized by deciles or treated as continuous variables (data not shown).

In this study population, alcohol was associated with increased risk of breast cancer (HR for an intake of ≥40 g/d, 1.20; 95% CI, 0.96-1.51; P for trend = 0.04), whereas HRT use was not (HR, 0.98; 95% CI, 0.89-1.09). There was no association between iron intake or heme iron intake and breast cancer risk within strata defined by alcohol consumption or HRT use (Table 2). Tests for heterogeneity in the association between subgroups were not statistically significant.

Consistent with the cohort data of Lee et al. (16), we found no overall association between dietary iron or heme iron intake and breast cancer risk. However, the prior report did find a significant association of both iron intake and heme iron intake in women consuming 20+ g of alcohol per day. With ∼2,500 incident breast cancer cases ascertained over 16 years of follow-up, our study had excellent statistical power to detect a main effect of iron and heme iron on breast cancer. Specifically, we had 87% power to detect a HR of 1.2 for extreme quintiles with a two-sided α-level of 5%. In the subgroup with the smallest number of breast cancer cases (women who reported drinking 30+ g/d of alcohol at baseline), we had 86% power to detect a HR of 1.8 for extreme quintiles. Two limitations of this study should be mentioned. First, information on use of iron supplements was not available. However, in their studies on heme iron, Lee et al. al. (21, 24) seem to have based their estimates on dietary data only. Second, our dietary data were limited to intake reported at baseline, and changes in dietary intake over the long follow-up period could have led to misclassification of exposure, thereby reducing our ability to detect an association.

Considering the biological plausibility of an association between iron and breast cancer risk, further investigation is warranted given the limited data currently available. Of particular importance are cohort studies with repeated measurement of iron intake. If elevated iron intake is found to contribute to breast carcinogenesis, preventive strategies might include dietary modification, avoidance of iron supplements, and chelation (25).

Grant support: National Cancer Institute of Canada.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Statistics Canada, the provincial and territorial Registrars of Vital Statistics, and the Cancer Registry directors for their assistance in making the cancer incidence and mortality data available.

1
Huang X. Iron overload and its association with cancer risk in humans: evidence for iron as a carcinogenic metal.
Mutat Res
2003
;
533
:
153
–71.
2
Wyllie S, Liehr JG. Release of iron from ferritin storage by redox cycling of stilbene and steroid estrogen metabolites: a mechanism of induction of free radical damage.
Arch Biochem Biophys
1997
;
346
:
180
–6.
3
Reizenstein P. Iron, free radicals and cancer: a review.
Med Oncol Tumor Pharmacother
1991
;
8
:
229
–33.
4
Toyokuni S. Iron-induced carcinogenesis: the role of redox regulation.
Free Radic Biol Med
1996
;
20
:
553
–66.
5
Kang D-H. Oxidative stress, DNA damage, and breast cancer.
AACN Clin Issues
2002
;
13
:
540
–9.
6
Wang M, Dhingra K, Hittelman WN, Liehr JG, de Andrade M, Li D. Lipid peroxidation-induced putative malondialdehyde-DNA adducts in human breast tissues.
Cancer Epidemiol Biomarkers Prev
1996
;
5
:
705
–10.
7
Boyd NF, McGuire V. Evidence of lipid peroxidation in premenopausal women with mammographic dysplasia.
Cancer Lett
1990
;
50
:
31
–7.
8
Boyd NF, McGuire V. The possible role of lipid peroxidation in breast cancer risk.
Free Radic Biol Med
1991
;
10
:
185
–90.
9
Rossner P, Gammon MD, Terry MB, et al. Relationship between urinary 15-F2t-isoprotane and 8-oxodeoxyguanosine levels and breast cancer risk.
Cancer Epidemiol Biomarkers Prev
2006
;
15
:
639
–44.
10
Kumar K, Thangaraju M, Sachdanandam P. Changes observed in antioxidant system in the blood of postmenopausal women with breast cancer.
Biochem Int
1991
;
25
:
371
–9.
11
Huang Y-L, Sheu J-Y, Lin T-H. Association between oxidative stress and changes of trace elements in patients with breast cancer.
Clin Biochem
1999
;
32
:
131
–6.
12
Diwan BA, Kasprzak KS, Anderson LM. Promotion of dimethylbenz[a]anthracene-initiated mammary carcinogenesis by iron in female Sprague-Dawley rats.
Carcinogenesis
1997
;
18
:
1757
–62.
13
Thompson HJ, Kennedy K, Witt M, Juzefyk J. Effect of dietary iron deficiency or excess on the induction of mammary carcinogenesis by 1-methyl-1-nitrosourea.
Carcinogenesis
1991
;
12
:
111
–4.
14
Wright RM, McManaman JL, Repine JE. Alcohol-induced breast cancer: a proposed mechanism.
Free Radic Biol Med
1999
;
26
:
348
–54.
15
Cho E, Chen WY, Hunter DJ, et al. Red meat intake and risk of breast cancer among premenopausal women.
Arch Intern Med
2006
;
166
:
2253
–9.
17
Terry PD, Jain M, Miller AB, Howe GR, Rohan TE. Dietary carotenoid intake and colorectal cancer risk.
Nutr Cancer
2002
;
42
:
167
–72.
18
Miller AB, Baines CJ, To T, Wall C. Canadian National Breast Screening Study. I. Breast cancer detection and death rates among women aged 40-49 years. II. Breast cancer detection and death rates among women aged 50-59 years.
Can Med Assoc J
1992
;
147
:
1459
–88.
19
Jain M, Harrison L, Howe GR, Miller AB. Evaluation of a self-administered dietary questionnaire for use in a cohort study.
Am J Clin Nutr
1982
;
36
:
931
–5.
20
Balder HF, de Vogel J, Jansen MCJF, et al. Heme and chlorophyll intake and risk of colorectal cancer in the Netherlands Cohort Study.
Cancer Epidemiol Biomarkers Prev
2006
;
15
:
717
–25.
21
Lee D-H, Jacobs DR, Jr. Interaction among heme iron, zinc, and supplemental vitamin C intake and risk of lung cancer: Iowa Women's Health Study.
Nutr Cancer
2005
;
52
:
130
–7.
22
Willett WC, Stampfer MJ. Total energy intake: implications for epidemiologic analyses.
Am J Epidemiol
1986
;
124
:
17
–27.
23
Rothman KJ, Greenland S. Modern epidemiology. Philadelphia (PA): Lippincott-Raven; 1998. p. 319.
24
Lee DH, Anderson KE, Harnack LJ, Folsom AR, Jacobs DR, Jr. Heme iron, zinc, alcohol consumption, and colon cancer: Iowa Women's Health Study.
J Natl Cancer Inst
2004
;
96
:
403
–7.
25
Wang F, Elliott RL, Head JF. Inhibitory effect of deferoxamine mesylate and low iron diet on the 13762NF rat mammary adenocarcinoma.
Anticancer Res
1999
;
19
:
445
–50.