Meat Intake, Heterocyclic Amines, and Risk of Breast ...To study the effects of meat intake,...

Vol. 6, 573-581, August 1997 Cancer Epidemiology, Biomarkers & Prevention 573

Meat Intake, Heterocyclic Amines, and Risk of Breast Cancer:

A Case-Control Study in Uruguay1

Eduardo De Stefani,2 Alvaro Ronco,

Maria Mendilaharsu, Martha Guidobono, andHugo Deneo-Pellegrini

Registro Nacional de Cancer, Avenida Brasil 3080, Department 402 [H. D. S.,

A. R., M. M., M. G., H. D-P.], and Departamento de Anatomia Patologicala,

Hospital de Clinicas [H. D-P.], Montevideo, Uruguay I 1300

Abstract

To study the effects of meat intake, including heterocydic

amine exposure, on the risk of breast cancer, weconducted a hospital-based case-control study involving352 patients with breast cancer and 382 controls. Astrong effect of red meat, total meat, beef, fried meat,and heterocydic amine exposure was found, after

controffing for potential confounders. The odds ratio for

the highest quartile of 2-amino-3-methylimidazo[4,5-fiquinoline exposure was 3.34 (95% confidence

interval 1.85-6.02). According to these results, meatintake and chemicals formed during the cooking process

appear to be strong risk factors in human breast

carcinogenesis.

Introduction

Breast cancer is a major public health problem in Uruguayanwomen, with an age-adjusted incidence rate of9l.0 per 100,000

(1). This very high rate is unusual in developing countries like

Uruguay, and the reasons for it are unknown.The Uruguayan population is exposed to a diet with large

amounts of red meat, mainly beef (2). According the Food andAgriculture Organization, the consumption of red meat in Urn-

guay was 80.1 kg/year in 1977, compared with 54.6 kg/year in

United States in the same year (2). It has been suggested thatmeat intake could be a risk factor for this disease (3). Moreover,

the pyrolisis of amino acids in meat during cooking at hightemperature leads to the formation of heterocyclic aromatic

amines. These amines are highly mutagenic and induce tumors

in several sites, including breast, in different animal species (4).

Therefore, we undertook a case-control study on breast cancerand diet to estimate the risk of this disease associated with meat

intake and with heterocycic aromatic amines resulting from the

process of cooking meat.

Received 1/8/97; revised 4/15/97; accepted 4/21/97.

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.

I This work was supported by a grant from the Comision Uruguaya de Lucha

contra el Cancer.2 To whom requests for reprints should be addressed.

Subjects and Methods

In the time period between May 1994 and November 1996, allnewly diagnosed cases of breast cancer occurring in the sixmajor hospitals of Montevideo, Uruguay, were considered el-igible for this study. A total of 365 cases were identified, 10 ofwhom refused the interview and 3 of whom were too ill to

answer the questionnaire, leading to a total of 352 interviewedcases, with a response rate of 96.4%. In the same time periodand in the same hospitals, 390 patients hospitalized for non-neoplastic diseases were considered eligible for the study. Eight

patients refused to be interviewed, leading to a control series of

382 patients (response rate, 97.9%). The diseases among thecontrol series are shown in Table 1 . Controls were frequencymatched to cases on the following variables: age (5-year inter-vals), sex, and residence (Montevideo or other counties).

Both cases and controls were subjected to a face-to-face

interview by two trained interviewers to complete a question-naire containing the following sections: (a) demographic data;(b) occupational history; (c) smoking history; (d) a history ofalcohol consumption, including beer, wine, and hard liquor; (e)

anthropometric variables (height, weight, and body mass index

calculated as the Quetelet index); and (f) family history ofcancer in first-degree relatives.

Furthermore, the interview included a food frequency

questionnaire covering 64 food items plus vitamin and mineralsupplements and questions about soft drinks, coffee, coffeewith milk, tea, tea with milk, and mate ingestion. Participantswere asked how often, on average, they consumed the amount

of each food over the 2 years prior to the interview or to theonset of symptoms. Because the questionnaire did not includedportion sizes, these were determined by a nutritionist accordingto local practices. These portion sizes were specific for womenin Uruguayan population. The questionnaire allowed the cal-

culation of total energy intake, as well as intake of a number ofmacro- and micronutrients. Among them, protein, carbohy-

drates, total fat, saturated fat, cholesterol, vitamin A, vitamin C,carotenoids, and vitamin E were calculated. Furthermore, the

Table I Distribution of controls by d isease category

ICD-9” Disease No. %

550-553 Abdominal hernia 90 23.6

800-829 Traumatic fractures 70 18.3

360-379 Eye disorders 58 15.2

451-456 Varicose veins 38 9.9

540 Acute appendicitis 32 8.4

122 Hydatid cyst 29 7.6

830-959 Trauma 23 6.0

680-709 Diseases of the skin 20 5.2

710-739 Osteoarticular diseases 16 4.2

280-289 Diseases of the blood 6 1.6

Total 382 100.0

a lCD International C lassification of Diseases.

on March 17, 2020. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from

http://cebp.aacrjournals.org/

574 Meat Intake, Heterocydic Amines, and Breast Cancer

Table 2 Distribution of cases and controls for se lected variables (1)

VariableNo.

Cases ControlsP

(%)No.(%)

Age (years)

20-29 4 (1.1) 9 (2.4)

30-39 33 (9.4) 42 (11.0)

40-49 50 (14.2) 58 (15.2)

50-59 85 (24.1) 86 (22.5)60-69 87 (24.7) 88 (23.0)

70-79 78 (22.2) 83 (21.7)

80-89 15 (4.3) 16 (4.2) 0.85

Residence

Montevideo 183 (52.0) 201 (52.6)

Other counties 169 (48.0) 181 (47.4) 0.92

Urban/rural status

Urban 292 (83.0) 328 (85.9)

Rural 60 (17.0) 54 (14.1) 0.32

Education (years)

0-5 174 (49.4) 199 (52.1)

�6 178 (50.6) 183 (47.9) 0.52

Monthly income (United States dollars)

�147 116 (33.0) 125 (32.7)

�148 117 (33.2) 123 (32.2)

Unknown 1 19 (33.8) 134 (35.1) 0.93

Family history of breast cancer

No 265 (75.3) 342 (89.5)

Yes 87 (24.7) 40 (10.5) <0.001

Body mass index

�18.7 98 (27.8) 101 (26.4)

18.8-21.2 70 (19.9) 98 (25.7)

21.3-24.2 97 (27.6) 86 (22.5)

�24.3 87 (24.7) 97 (25.4) 0.19

Alcohol drinking

Never drinkers 249 (70.7) 307 (80.4)

Ever drinkers 103 (29.3) 75 (19.6) 0.009

Menopausal status

Premenopausal 75 (21 .3) 99 (25.9)

Postmenopausal 277 (78.7) 283 (74.1) 0.17

Age at menarche (years)

�15 32 (9.1) 60 (15.7)

12-14 234 (66.5) 264 (69.1)

�11 86 (24.4) 58 (15.2) 0.006

Parity (no. of live births)

Nuliparous 76 (21.6) 49 (12.8)

1-2 135 (38.4) 130 (34.0)

3-4 84 (23.9) 105 (27.5)�5 57 (16.2) 98 (25.7) 0.0004

Age at first pregnancy (years)

�19 81 (23.0) 1 19 (31.2)

20-24 88 (25.0) 121 (31.7)

�25 108 (30.7) 93 (24.3) 0.01

History of benign breast disease

No 309 (87.8) 354 (92.7)

Yes 43 (12.2) 28 (7.0) 0.03

Total energy intake (kcal)

�l577 74 (21.0) 110 (28.8)

1578-2080 76 (21.6) 108 (28.3)

2081-2592 103 (29.3) 82 (21.5)

�2593 99 (28.1) 82 (21.5) 0.01

Total fat intake (g)

�68.7 57 (16.2) 127 (33.2)

68.8-82.3 87 (24.7) 96 (25.1)

82.4-96.5 89 (25.3) 95 (24.9)

�96.6 1 19 (33.8) 64 (16.8) <0.001

Vegetables (servings/year)

�538 98 (27.8) 87 (22.8)

539-727 88 (25.0) 94 (24.6)

728-948 84 (23.9) 100 (26.2)

�949 82 (23.3) 101 (26.4) 0.05

No. of patients 352 (100) 382 (100)



Cancer Epidemio1oaj�, Biomarkers & Prevention 575

Table 3 Mean concen tration of heterocy cic amines in f oods (in �g/g)a

Food IQ MeIQx PHiP

Fried beef

Broiled beef

Broiled lamb

Broiled chicken

Fried fish

Broiled fish

Fried egg

3.79

0.50

NAb

NA

0.61

1.80

0.10

2.35

2.1 1

1.01

2.33

6.44

2.20

NA

13.19

15.70

4230

38.10

19.60

37.50

NA

a According to values reported in Refs. 9-19.

b NA, not available.

questionnaire included queries concerning the cooking method(frying, broiling, barbecuing, and boiling) of the foods, includ-ing meats and eggs. This questionnaire was previously used in

case-control studies on diet and cancer (3, 5, 6). Nutrient intakewas calculated using specially designed software that uses local

food composition tables (7). All nutrients were energy adjustedby the residual method of Willett and Stampfer (8) and latercategorized into four levels, according to the quartile distribu-tion among the combined sample of cases and controls. Intake

of specific foods was recorded as servings per year.The food composition table for heterocycic amine content

was based on data in the literature (9-19) and total estimated

content of IQ,3 2-a.mino-3,8-dimethylimidazo[4,5-fjquinoxa-line, and PHiP for this data base was calculated as continuousvariables in ng/g. Because the questionnaire assessed the pro-portion ofdifferent meats cooked by frying, broiling, or boiling,

total intake of arilamines was calculated as follows: frequency

of consumption of fried and broiled meats X [(portion size ingrams X arylamine content for each fried and broiled meataccording to literature data)/100 g]. The frequency of consump-tion of each fried and broiled meat was obtained from the food

frequency questionnaire. The three arilamines were categorized

according to the quartile distribution among the combinedsample of cases and controls.

ORs and 95% CIs were calculated by unconditional logis-tic regression (20). ORs were estimated for all women, pre-menopausal women, and postmenopausal women. We classi-

fled a woman as postmenopausal when she reported naturalmenopause or hysterectomy with bilateral oophorectomy. Forwomen who reported hysterectomy without bilateral oophorec-

tomy, menopause was assigned to the median value of age atmenopause for the combined sample of cases and controls.Potential confounders were included in the multivariate models.Decisions on which covariates to include in the final modelswere based on (a) biological plausibility, (b) whether the co-variate entered the model at the 0.10 level of significance, and

(c) whether the covariate acted as a confounder of the primaryassociation of interest (confounding was considered present if

the regression coefficient of the primary independent variablechanged more than 10% after addition of the potentially con-founding variable to the model). These confounders were age(continuous), residence (Montevideo or other counties), familyhistory of breast cancer in a first-degree relative (no or yes), ageat menarche, parity, previous history of benign breast disease

(no or yes), total energy intake (continuous), vegetable intake,

and total fat intake. Age at menarche, parity, vegetable intake,

3 The abbreviations used are: IQ, 2-amino-3-methylimidazo[4,5-flquinoline;

PHiP, 2-amino-1-methyl-6-phenylimidazo[4,5-flpyridine; OR, odds ratio; CI,

confidence interval.

and total fat intake were entered in the models as categoricalvariables (Table 2). All calculations were performed using theEGRET software (21).

Results

Compared to controls, significantly more cases had a family

history of breast cancer in first-degree relatives, early age at

menarche, high parity, late age at first full-term pregnancy, ahistory of benign breast disease, and a history of alcohol con-

sumption (Table 2). Age, residence, urban/rural status, educa-tion, and income were similar among both series of patients.

Concentrations of heterocycic amines in various fooditems are shown in Table 3. Elevated concentrations of PHiPwere reported in fried beef, broiled beef, and broiled lamb.

These food items are frequently consumed by the Uruguayan

population.ORs of breast cancer associated with different meats are

shown in Table 4. Whereas red meat, beef, lamb, and total meatintake were associated with significantly increased risks of

breast cancer, white meat (poultry plus fish), poultry, and fishintake were associated with negative dose-response patterns. Inparticular, white meat displayed an OR of 0.59 (95% CI,0.36-0.97) for the highest quartile of intake. Beef intake was

associated with the highest increase in risk among postmeno-pausal women (OR, 4.75; 95% CI, 2.30-9.79); the dose-re-

sponse pattern was also highly significant. When differentmeats (beef, lamb, poultry, and fish) were analyzed accordingto the cooking method (frying, broiling, or boiling), a signifi-cant association for fried meat was evident, with an OR of 2.71for the highest quartile of intake (Table 5). The effect of broiled

meat was not significant, and boiled meat was associated witha significant reduction in risk for the uppermost quartile of

consumption (OR, 0.42; 95% CI, 0.23-0.78). ORs of breastcancer for heterocycic amine exposure, according to the cook-ing methods and to the values reported in the literature, areshown in Table 6. All heterocycic amines were associated withsignificant increases in risk and IQ displayed a 3-fold increasein risk for all women (pre- and postmenopausal). PHiP was

associated with a positive gradient of risks and a highly signif-icant dose-response. In particular, postmenopausal women

were associated with a 3-fold increase in risk for the highestquartile of estimated consumption of PHiP.

ORs of breast cancer for total fat and vegetable intake,with and without adjustment for meat intake in a multivariate

model, are shown in Table 7. Whereas total fat intake wasassociated with a positive gradient of risks, vegetable intakedisplayed a protective effect. Total fat intake reduced risk by65% after further adjustment for beef intake. On the other hand,

vegetable intake risk estimates remained unchanged after in-

clusion of a term for beef in the model.

Discussion

The present study shows increased risks of breast cancer for

increasing intake of red meat, beef, and fried meat and esti-mated intake of heterocycic amines. These risk estimates werefully controlled for confounders, including total energy and

total fat intake, which were strong risk factors in this data set.A number of previous studies showed an effect of meat con-sumption on the risk of breast cancer (22-31), and this effectwas attributed mostly to the fat content of meat. In our study,

we found increasing risks of breast cancer associated with

increasing meat consumption, controffing for fat intake, sug-

gesting another mechanism for meat in breast carcinogenesis.



Table 4 Relative risks of breast cancer associated with meat variables

All risks are adjusted for age, residence, familyenergy, vegetable intake, and fat intake.

history of breast cancer in a first-degree relative, age at menarche, parity, previous history of benign breast disease, total

.Meat variable

Quartile P for

trendI

�601

II ifi Iv

Total meat’�” s336 337-468 469-600

All subjectsCases 59 79 94 120

Controls 127 108 84 63

OR 1.0 1.29 1.63 2.26

95% CI 0.80-2.09 0.97-2.75 1 .24-4.12 0.006

prelncnopausalc

Cases 11 11 19 34

Controls 29 27 22 21

OR I .0 0.90 1.27 2.42

95% CI 0.28-2.95 0.37-4.32 0.75-7.85 0.11

Postmenopausal�

Cases 48 68 75 86


OR 1.0 1.33 1.58 2.19

95% CI 0.77-2.29 0.87-2.89 1.06-4.51 0.03

Red meaf2e �24l 242-386 387-520 �521

All subjects

Cases 56 76 99 121

Controls 128 107 88 59

OR 1.0 1.25 1.76 2.62

95% CI 0.77-2.05 1.04-2.99 1.41-4.85 0.001

PremenopausalCases 9 10 24 32


OR 1.0 1.41 2.13 3.01

95% CI 0.38-5.29 0.59-7.60 0.77-1 1.7 0.09

Postmenopausal

Cases 47 66 75 89

Controls 101 82 63 37

OR 1.0 1.29 1.57 2.79

95% CI 0.75-2.23 0.86-2.89 1.35-5.75 0.006

Whitemeata.f �36 37-70 71-104 �105

All subjects

Cases 110 78 103 61

Controls 83 90 129 80OR I .0 0.67 0.60 0.59

95% CI 0.42-1.06 0.39-0.93 0.36-0.97 0.02

Premenopausal

Cases 23 19 26 7

Controls 27 30 21 21OR 1.0 0.62 1.36 0.28

95% CI 0.22-1.78 0.50-3.66 0.08-1.03 0.31Postmenopausal

Cases 87 59 77 54

Controls 56 60 108 59OR 1.0 0.60 0.48 0.63

95% CI 0.35-1.04 0.29-0.78 0.36-1.09 0.03

Processed meat” �52 53-104 105-206 �207

All subjects

Cases 92 68 85 107

Controls 114 93 91 84OR 1.0 0.80 0.85 0.88

95% Cl 0.50-1.27 0.53-1.34 0.54-1.43 0.64

Premenopausal

Cases 17 12 13 33


OR 1.0 0.66 0.48 1.3095% CI 0.20-2.16 0.15-1.58 0.43-3.92 0.56

Postmenopausal

Cases 75 56 72 74

Controls 88 71 65 59OR 1.0 0.80 0.91 0.7395% CI 0.47-135 0.54-1.53 0.42-1.30 0.38

Beer �154 155-234 235-364 �365All subjects

Cases 54 85 98 115Controls I 26 136 76 44

OR 1.0 1.23 2.09 3.8495% CI 0.76-1.99 1.23-3.55 2.09-7.05 <0.001

576 Meat Intake, Heterocydlic Amines, and Breast Cancer



Cancer Epidemiology, Bionsarkers & Prevention 577

Table 4 Continued

All risks are adjusted for age, residence, family history of b

energy, vegetable intake, and fat intake.

roast cancer in a first -degree relative, age at menarche, panty, previous history of be nign breast disease, total

Meat variableQuartile P for

trendI II ifi IV

Premenopausal

Cases 7 20 21 27


OR 1.0 1.91 2.41 2.60

95% CI 0.57-6.41 0.69-8.41 0.69-9.82 0.16

Postmenopausal

Cases 47 65 77 88

Controls 99 102 56 26

OR 1.0 1.15 2.02 4.75

95% CI 0.67-1.97 1.10-3.73 2.30-9.79 <0.001

L�bag �l2 13-52 �53

All subjects

Cases 276 24 52

Controls 336 26 20

OR 1.0 1.05 2.38

95% CI 0.56-1.99 1.27-4.47 0.01

Premenopausal

Cases 56 7 12

Controls 86 5 8

OR 1.0 1.32 1.4595% CI 0.32-5.36 0.40-5.28 0.53

Postmenopausal

Cases 220 17 40

Controls 250 21 12

OR 1.0 0.88 2.90

95% CI 0.42-1.84 1.34-6.27 0.02

Poultrya;a �24 25-52 �53

All subjects

Cases 150 150 52

Controls 132 183 67

OR 1 .0 0.75 0.78

95% CI 0.52-1.06 0.48-1.27 0.18

Premenopausal

Cases 35 32 8

Controls 40 42 17OR 1.0 0.62 0.32

95% CI 0.27-1.40 0.10-1.10 0.06

Postmenopausal

Cases 115 118 44

Controls 92 141 50

OR 1.0 0.73 0.85

95% CI 0.49-1.09 0.50-1.47 0.35

� �l2 13-52 �53

All subjects

Cases 148 163 41

Controls 136 193 53OR 1.0 0.76 0.64

95% CI 0.54-1.09 0.38-1.09 0.06Premenopausal

Cases 28 43 4

Controls 44 44 11

OR 1.0 1.97 0.54

95% CI 0.88-4.37 0. 1 1-2.65 0.65

Poatmenopausal

Cases 120 120 37

Controls 92 149 42

OR 1.0 0.62 0.63

95% CI 0.41-0.93 0.35-1.12 0.04

a Intake in servings per year.

b Total meat includes red meat, white meat, and processed meat

C Premenopausal was defined as women who reported regular vaginal bleeding over the course of the last year.

d Postmenopausal was defined as women who reported lack of regular vaginal bleeding for more than I year.

e Red meat includes beef, lamb, and processed meat.

1White meat includes poultry and fish.

S Intake of lamb, poultry, and fish was categorized in tertiles.



All risks are adjusted for age, residence, family history of


breast cancer in a first-degree relative, age at menarche, parity, previous history of benign breast disease, total

Cooking methodQuartile P for

trendI II m iv

�104 105-156 157-208 �209

87

125

1.0

65

107

1.0

81

133

1.16

0.73-1.85

84

85

1.36

0.83-2.22

122

57

2.71

1.61-4.55 <0.001

10

27

1.0

17

34

1.68

033-5.37

17

18

2.92

0.88-9.73

31

20

2.82

0.91-8.77 0.05

55

80

1.0

64

99

1.08

0.64-1.84

67

67

1.10

0.63-1.92

91

37

2.57

1.39-4.74 0.004

105-168 �169

83

116

0.83

0.53-1.30

75

82

0.86

0.53-1.39

107

59

1.57

0.95-2.59 0.10

17

31

0.57

0.19-1.76

19

19

1 .24

0.40-3.89

26

24

0.69

0.21-2.22 0.95

66

850.89

0.54-1.46

56

63

0.73

0.42-1.26

81

35

1.88

1.04-3.39 0.12

13

25

1.0

Fried meat”

All subjects

Cases

Controls

OR

95% CI

Premenopausalc

Cases

Controls

OR

95% CI

Postmenopausal’�

Cases

Controls

OR

95% CI

Broiledmeat” �52 53-104

All subjects

Cases

Controls

OR

95% CI

Premenopausal

Cases

Controls

OR

95% CI

Postmenopausal

Cases

Controls

OR

95% CI

Boiled meat”

All subjects

Cases 77 116 119 40

Controls 78 130 114 60

OR 1.0 0.80 0.81 0.42

95% CI 0.50-1.27 0.51-1.29 0.23-0.78 0.02

Premenopausal

Cases 13 25 24 13


OR 1.0 0.91 0.55 0.29

95% CI 0.26-3.17 0.16-1.86 0.07-1.25 0.06

Postmenopausal

Cases 64 91 95 27

Controls 64 103 78 38

OR 1.0 0.85 0.98 0.44

95% CI 0.51-1.42 0.58-1.68 0.21-0.90 0.16

74

100

I .0

�42 43-78 79-130 � 131

a Meats included were beef, lamb, poultry, fish, and bacon.

b Servings per year.

C Premenopausal was defined as women who reported regular vaginal bleeding over the course of the last year.

d Postmenopausal was defined as women who reported lack of regular vaginal bleeding for more than 1 year.

578 Meat Intake, Heterocydlic Amines, and Breast Cancer

Table 5 Relative risks of breast cancer associated with cooking method’�

Because meat and fat intake are highly correlated, it could bedifficult to evaluate separate effects of meat and fat. In our dataset, the correlation between heterocylic amines and energy-adjusted total fat was only 0.43. Therefore, we included in allmodels a term for energy-adjusted fat intake and a term for totalenergy intake. As a result, the correlation between meat van-ables and fat dropped from 0.43 to 0.39. This allowed us toestimate separate ORs for meat variables (including heterocy-

clic amines) and for total fat intake.

Whereas fried meat was strongly associated with the riskof breast cancer, broiled meat showed only a marginal increase

in risk. Moreover, boiled meat intake was associated with asignificant reduction in risk. The risk estimates for boiled meatcould be due to the presence of cooked vegetables in stew,

which is the richest source of this food item in this population.The findings concerning fried meat are in accordance with

previous studies (32-33), which suggest that this food item isthe richest source of heterocycic amines. These chemicals are



All risks are adjusted for age, residence, family history of


breast cancer in a first-degree relative. age at menarche, parity, previous history of benign breast disease, total

Heterocyclic

amine

Quartile P for

trendI II m N

�0.42 0.43-0.66 0.67-1.01 �l.02

92

81

1.87

1.10-3.15

130

54

3.34

1.85-6.02

18

21

2.70

0.76-933

32

20

2.15

038-8.05

<0.001

0.24

74

60

1.75

0.96-3.19

98

34

3.80

1.90-7.60

3.43-4.52 �433

99

85

1.60

0.99-238

120

62

2.13

I .27-3.55

IQ

All subjects

Cases 53 77

Controls 124 123

OR 1.0 1.22

95% CI 0.75-1.99

Premenopausal’�

Cases 7 18

Controls 27 31

OR 1.0 1.89

95% CI 0.54-6.52

Postmenopausal”

Cases 46 59

Controls 97 92

OR 1.0 1.11

95% CI 0.64-1.92 <0.001

MeIQxc �2.50 231-3.42

All subjects

Cases 63 70

Controls 123 112

OR 1.0 1.12

95% CI 0.70-1.79 0.002

Premenopausal

Cases 11 22

Controls 33 29

OR 1.0 2.03

95% CI 0.69-5.97 0.23

Postmenopausal

Cases 52 48

Controls 90 83

OR 1.0 0.92

95% CI 0.54-1.78 0.006

PHiP �11.07 11.08-15.26

All subjects

Cases 68 68 102 114

Controls 119 112 83 68

OR 1.0 1.06 1.81 2.59

95% CI 0.65-1.73 1.06-3.09 1.42-4.70 <0.001

Premenopausal

Cases 13 15 19 28


OR 1.0 1.19 1.24 1.16

95% CI 0.38-3.71 0.34-4.49 0.34-4.01 0.84

Postmenopausal

Cases 55 53 83 86


OR 1.0 1.02 1.85 3.31

95% CI 0.58-1.79 0.99-3.46 1.60-6.87 <0.001

16

19

233

0.77-8.28

26

18

1.97

0.60-6.42

83

66

1.49

0.86-237

94

44

2.14

1.17-3.89

15.27-20.14 �20.l5

a Premenopausal were defined as women who reported regular vaginal bleeding over the course of the last year.

b Postmenopausal were defmed as women who reported lack of regular vaginal bleeding for more than 1 year.

C MeIQx, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline.

Cancer Epidemiology, Biomarkers & Prevention 579

Table 6 Relative risks of breast cancer associated with heterocyclic amine exposure (ng/g)

strong carcinogens in experimental animals (32-36), leading to studies, heterocycic amine effect was suggested by the cookingthe production of colonic, pulmonary, renal, and mammary method (frying or broiling) or through queries concerning thetumors. It has been suggested (32-33) that heterocyclic amines doneness of meat. The present study combined both the cook-could be important chemicals in human breast carcinogenesis, ing method and the estimated content of heterocyclic amines inpossibly as initiating agents. the diet. Although the validity of our estimation of the content

This study provides some support to the carcinogenic of heterocyclic amines is open to question, it represents, to ourpotential of heterocyclic amine exposure in breast cancer. Pre- knowledge, the first attempt to estimate these substances, and itvious studies in other cancer sites, mainly colon and lung, were could represent an important step to obtain reliable estimates of

partially supportive of such an association (6, 37-40). In those heterocyclic amines in the diet. Our results were mostly similar



580 Meat Intake, Heterocyclic Amines, and Breast Cancer

21. EGRET Reference Manual, Revision 3. Seattle, WA: Statistics and Epide-

miology Research Corp., 1992.

Table 7 Relative risks of breast cancer assoc iated with total fat and vegetable intake, with and w ithout adjustment for m eat intake

Variable Cases/controls OR1” 95% CI 0R2” 95% CI

Total fa(

�68.7

68.8-82.3

82.4-963

�96.6

P for trend

57/127

87/96

89/95

1 19/64

1.0

1.60

1.44

3.16

<0.001

1.01-2.55

0.89-2.33

1.90-5.25

1.0

1.37

1.03

2.08

0.03

0.84-2.21

0.62-1.73

1.21-3.57

Vegetables”

�538

539-727

728-948

�949

P for trend

98/87

88/94

84/100

82/101

1.0

0.67

0.51

0.45

<0.001

0.43-1.05

0.32-0.80

0.28-0.72

1.0

0.66

0.52

0.46

0.001

0.41-1.06

0.32-0.84

0.28-0.75

a Adjusted for age (continuous), residence, family history of breast cancer in a first-degree relative, age at menarche, parity, menopausal status, history of benign breast

disease, and total energy intake (continuous).b Adjusted for the same variables as OR1 plus beef intake.

C Intake in grams per day.

d Intake in servings per year.

for pre- and postmenopausal women, although the small sampleof premenopausal women resulted in wider CIs and in a lowerpower to detect differences among both subsets of women.

Recent studies have emphasized the interplay between

heterocyclic amines from meat and the metabolic phenotype inthe etiology of colon cancer (41-43). In particular, rapid N-acetyltransferase metabolizers are at increased risk of coloncancer; rapid metabolizers who eat well-done meat have a6-fold increased risk in risk of colon cancer (42). It would be of

high importance to establish whether the same interactions withthe metabolic phenotype are present in breast cancer. Because

metabolic phenotype was not measured in our study, we wereunable to estimate this joint effect in our patients.

This study has limitations. First, our estimates of hetero-cyclic amines should be considered a first step in constructinga reliable data base for these important substances. Also, the

effect of the disease in hospitalized patients could modify theusual diet, leading to under- or overestimation of the resultingORs. Nevertheless, our estimates for meat variables are in linewith those observed in previous studies (22-31). On the otherhand, controls were selected from the same hospitals as the

cases, they presented similar residence, and they had similareducational levels. Therefore, they could be considered mem-

bers of the same (secondary) population base (44). In conclu-sion, a high intake of fried meat and a high intake of hetero-

cyclic amines were significantly associated with an increased

risk of breast cancer.

References

1. Vassallo, J. A., Dc Stefani, E., Barrios, E., and Ronco, A. Cancer Incidence in

Uruguay, 1991 (in Spanish). Montevideo, Uruguay: Barreiro & Ramos, 1996.

2. Food Agriculture Organization. Food Balance Sheets. Average 1975-77.

Rome, Italy: FAO, 1980.

3. Ronco, A., Dc Stefani, E., Mendilaharsu, M., and Deneo-Pellegrini, H. Meat,

fat and risk of breast cancer: a case-control study from Uruguay. hit. J. Cancer,

65: 328-331, 1996.

4. Felton, J. S., and Knize, M. G. Occurrence, identification and bacterial mu-

tagenicity of heterocyclic amines in cooked food. Mutat. Res., 259: 205-217,

1991.

5. Deneo-Pellegrini, H., Dc Stefani, E., and Ronco, A. Vegetables, fruits, and risk

of colorectal cancer. Nutr. Cancer, 25: 297-303, 1996.

6. Deneo-Pellegnni, H., Dc Stefani, E., Ronco, A., Mendilaharsu, M., and Car-

zoglio, 3. Meat consumption and risk of lung cancer: a case-control study from

Uruguay. Lung Cancer, 14: 195-205, 1996.

7. Mazzei, M. E., and Puchulu, M. R. Table of Chemical Composition of Foods

(in Spanish). Buenos Aires, Argentina: Cenexa, 1991.

8. Willett, W. C., and Stampfer, M. J. Total energy intake: implications for

epidemiologic analyses. Am. J. Epidemiol., 124: 17-27, 1986.

9. Barnes. W. S., Maher, J. C., and Weisburger, J. H. High pressure liquid

chromatographic method for the analysis of 2-amino-3-methylimidazo

[4,5-f]quinoline, a mutagen formed from the cooking of food. J. Agric. Food

Chem., 31: 883-886, 1983.

10. Felton, J. S., Knize, M. G., Wood, C., Wuebbles, B. J., Healy, S. K.,

Stuermer, D. H., Bjeldanes, L. F., Kimble, B. J., and Hatch, F. T. Isolation and

characterization of new mutagens from fried ground beef. Carcinogenesis

(Land.), 5: 95-102, 1984.

I 1. Turesky, R. J., Bur, H., Huynh-Ba, T., Aeschbacher, H. U., and Milon, H.

Analysis of mutagenic heterocycic amines in cooked beef products by high-

performance liquid chromatography in combination with mass spectrometry.

Food Chem. Toxicol., 26: 501-509.

12. Wakabayashi, K., Nagao, M., Esumi, H., and Sugimura, T. Food-derived

mutagens and carcinogens. Cancer Res., 52(Suppl.): 2092s-2098s, 1992.

13. Yamaizunsi, Z., Kasai, H., Nishimura, S., Edmonds, C. G., and McCloskey,

J. A. Stable isotope dilution quantification of mutagens in cooked foods bycombined liquid chromatography-thermospray mass spectrometry. Mutat. Res.,

173: 1-7, 1986.

14. Thang, X. M., Wakabayashi, K., Liu, Z. C., Sugimura, T., and Nagao, M.Mutagenic and carcinogenic heterocyclic amines in Chinese cooked foods. Mutat.

Res., 201: 181-188, 1988.

15. Grose, K. R., Grant, J. L., Bjeldanes, L. F., Andresen, B. D., Healy, S. K.,Lewis, P. R., Felton, J. S., and Hatch, F. T. Isolation of the carcinogen IQ from

fried egg patties. J. Agric. Food Chem., 34: 201-202, 1986.

16. Hargraves, W. A., and Pariza, M. W. Purification and mass spectral charac-terization of bacterial mutagens from commercial beef extract. Cancer Res., 43:

1467-1472, 1983.

17. Gross, G. A., Philipossian, G., and Aeschbacher, H. U. An efficient andconvenient method for the purification of mutagenic heterocycic amines in

heated meat products. Carcinogenesis (Land.), 10: 1 175-1 182, 1989.

18. Turesky, R. J., Forster, C. M., Aeschbacher, H. U., Wurzner, H. P., Skipper,

P. L., Trudel, L. J., and Tannenbaum, S. R. Purification ofthe food-borne car-cinogens 2-amino-3-methylimidazo[4,5-f]quinoline in heated meat products by

immunoaffirnty chromatography. Carcinogenesis (Lond.), 10: 151-156, 1989.

19. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans.

Some Naturally Occurring Substances: Food Items and Constituents, Heterocy-die Aromatic Amines and Mycotoxins. IARC Scientific Publ., Vol. 56. Lyon,

France: IARC, 1993.

20. Breslow, N. E., and Day, N. E. Statistical Methods in Cancer Research. I. The

Analysis of Case-control Studies. IARC Scientific Publ. No 32. Lyon, France:

IARC, 1980.



Cancer Epidemiology, Biomarkers & Prevention 581

22. Goodman, M. T., Nomura, A. M. Y., Wilkens, L R., and Hankin, J. The

association of diet, obesity, and breast cancer in Hawaii. Cancer Epidemiol.,

Biomarkers & Prey., 1: 269-275, 1992.

23. Knekt, P., Steineck, G., Jarvinen, R., Hakulinen, T., and Aaroma, A. Intake

of fried meat and risk of cancer: a follow-up study in Finland. list. J. Cancer, 59:

756-760, 1994.

24. Richardson, S., Gerber, M., and Cenee, S. The role of fat, animal protein and

some vitamin consumption in breast cancer: a case-control study in southern

France. lot. J. Cancer, 48: 1-9, 1991.

25. Toniolo, P., Riboli, E., Protta, F., Charred, M., and Cappa, A. P. M. Calorie-

providing nutrients and risk of breast cancer. J. Natl. Cancer Inst., 81: 278-286,

1989.

26. Zarisde, D., Lifanova, Y., Maximovitch, D., Day, N. E., and Duffy, S. W.

Diet, alcohol consumption and reproductive factors in a case-control study of

breast cancer in Moscow. mt. i. Cancer, 48: 493-501, 1991.

27. Ewertz, M., and Gill, C. Dietary factors and breast-cancer risk in Denmark.

Int. J. Cancer, 46: 779-784, 1990.

28. Levi, F., La Vecchia, C., Gulie, C., and Negri, E. Dietary factors and breast

cancer risk in Vaud, Switzerland. Nutr. Cancer, 19: 327-335, 1993.

29. Gaard, M., Tretli, S., and Loken, E. B. Dietary fat and the risk of breastcancer a prospective study of 25,892 Norwegian women. hit. J. Cancer, 63:

13-17, 1995.

30. Vatten, L. J., Solvoll, K., and Loken, E. B. Frequency of meat and fish intake

and risk of breast cancer in a prospective study of 14,500 Norwegian women. lot.

J. Cancer, 46: 12-15, 1990.

31. Matos, E. L., Thomas, D. B., Sobel, N., and Vuoto, D. Breast cancer in

Argentina: case-control study with special reference to meat-eating habits. Neo-

plasma (Bratisl), 38: 357-366, 1991.

32. Nagao, M., Ushijima, T., Wakabayashi, K., Ochiai, M., Kushida, H., Sug-

imura, T., Hasegawa, R., Shirai, T., and Ito, N. Dietaiy carcinogens and mammary

carcinogenesis. Induction of rat mammary carcinomas by administration of het-

erocyclic amines in cooked foods. Cancer (Phila.), 74: 1063-1069, 1994.

33. Snyderwine, E. G. Some perspectives on the nutritional aspects of breast

cancer research. Food-derived heterocyclic amines as etiologic agents in human

mammary cancer. Cancer (Phila.), 74: 1070-1077, 1994.

34. Ohgaki, H., Hasegawa, H., Kato, T., Suenaga, M., Ubukata, M., Sato, S.,

Takayama, S., and Sugimura, T. Carcinogenicity in mice and rats of heterocyclic

amines in cooked foods. Environ. Health Perspect., 67: 129-134, 1986.

35. Ochiai, M., Ogawa, K., Wakabayashi, K., Sugimura, T., Nagase, S., Esumi,

H., and Nagao, M. Induction of intestinal adenocarcinomas by 2-amino-l-methyl-

6-phenylimidazo[4,5-b]pyridine in Nagase analbuminemic rats. Jpn. J. Cancer

Res., 82: 363-366, 1991.

36. Takahashi, S., Ogawa, K., Ohshima, H., Esumi, H., Ito, N., and Sugimura, T.

Induction of aberrant crypt foci in the large intestine of F344 rats by oral

administration of 2-amino-l-methyl-6-phenylimidazo[4,5-bjpyridine. Jpn. J.

CancerRes., 82: 135-137, 1991.

37. Lyon, J. L., and Mahoney, A. W. Fried foods and the risk of colon cancer.

Am. J. Epidemiol., 128: 1000-1006, 1988.

38. Gerhardsson de Verdier, M., Hagman, U., Peters, R. K., Steineck, G., and

Overvik, E. Meat, cooking methods and colorectal cancer: a case-referent studyin Stockholm. Int. J. Cancer, 49: 520-525, 1991.

39. Schiffman, M. H., Andrews, A. W., Van Tassell, R. L., Smith, L., Daniel, J.,Robinson, A., Hoover, R. N., Rosenthal, J., Weil, R., Nair, P. P., Schwartz, S.,

Pettigrew, H., Batist, G., Shaw, R., and Wilkins, T. D. Case-control study of

colorectai cancer and fees.! mutagenicity. Cancer Res., 49: 3420-3424, 1989.

40. Schiffman, M. H., and Felton, J. S. Re: “Fried foods and the risk of colon

cancer” (Letter). Am. J. Epidemiol., 131: 376-378, 1990.

41. Wohileb, J. C., Hunter, C. F., Blass, B., Kadlubar, F. F., Chu, D. Z. J., and

Lang, N. P. Aromatic amine acetyltransferase as a marker for colorectal cancer:

environmental and demographic associations. Int. J. Cancer, 46: 22-30, 1990.

42. Lang, N. P., Butler, M. A., Massengill, J., Lawson, M., Craig Stotts, R.,

Hauer-Jensen, M., and Kadlubar, F. F. Rapid metabolic phenotypes for acetyl-

transferase and cytochrome P4501A2 and putative exposure to food-borne het-

erocydic amines increase the risk for colorectal cancer or polyps. Cancer Epi-

demiol., Biomarkers & Prey., 3: 675-682, 1994.

43. Vineis, P., and McMichael, A. Interplay between heterocyclic amines in

cooked meat and metabolic phenotype in the etiology of colon cancer. Cancer

Causes & Control, 7: 479-486, 1996.

44. Wacholder, S., McLaughlin, J. K., Silverman, D. T., and Mandel, J. S.

Selection of controls in case-control studies. I. Principles. Am. J. Epidemiol., 9:

1019-1028, 1992.



1997;6:573-581. Cancer Epidemiol Biomarkers Prev E De Stefani, A Ronco, M Mendilaharsu, et al. case-control study in Uruguay.Meat intake, heterocyclic amines, and risk of breast cancer: a

Updated version

http://cebp.aacrjournals.org/content/6/8/573

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cebp.aacrjournals.org/content/6/8/573To request permission to re-use all or part of this article, use this link



http://cebp.aacrjournals.org/cgi/alerts

mailto:[email protected]



Meat Intake, Heterocyclic Amines, and Risk of Breast ...To study the effects of meat intake,...

Documents

Transcript of Meat Intake, Heterocyclic Amines, and Risk of Breast ...To study the effects of meat intake,...