SEARCH

SEARCH BY CITATION

Keywords:

  • breast cancer;
  • carbohydrate;
  • cohort studies;
  • diet;
  • epidemiology;
  • glycemic load

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

High-glycemic load diets have been hypothesized to increase the risk of breast cancer but epidemiologic studies have yielded inconsistent findings. We examined the associations of carbohydrate intake, glycemic index and glycemic load with risk of overall and hormone receptor-defined breast cancer in the Swedish Mammography Cohort, a population-based cohort of 61,433 women who completed a food frequency questionnaire at enrollment in 1987–1990. During a mean follow-up of 17.4 years, we ascertained 2,952 incident cases of invasive breast cancer. Glycemic load but not carbohydrate intake or glycemic index was weakly positively associated with overall breast cancer risk (p for trend = 0.05). In analyses stratified by estrogen receptor (ER) and progesterone receptor (PR) status of the breast tumors, we observed statistically significant positive associations of carbohydrate intake, glycemic index and glycemic load with risk of ER+/PR− breast cancer; the multivariate relative risks comparing extreme quintiles were 1.34 [95% confidence interval (CI) = 0.93–1.94; p for trend = 0.04] for carbohydrate intake, 1.44 (95% CI = 1.06–1.97; p for trend = 0.01) for glycemic index and 1.81 (95% CI = 1.29–2.53; p for trend = 0.0008) for glycemic load. No associations were observed for ER+/PR+ or ER−/PR− breast tumors. These findings suggest that a high carbohydrate intake and diets with high glycemic index and glycemic load may increase the risk of developing ER+/PR− breast cancer. © 2009 UICC

There is a growing recognition that breast cancer may be promoted by insulin resistance and hyperinsulinemia. Overweight and obesity are important determinants of insulin resistance and hyperinsulinemia and are related to an increased risk of postmenopausal breast cancer.1 Furthermore, an increased risk of breast cancer has been observed among women with type 2 diabetes,2 which is characterized by hyperglycemia and insulin resistance. Elevated insulin concentrations may affect breast cancer risk either directly, by stimulating insulin receptors in breast tissue, or indirectly, by increasing the bioactivity of insulin-like growth factor-I (IGF-I), which stimulates cell proliferation and inhibits apopotosis.3 Furthermore, there is evidence that both insulin and IGF-I stimulate the synthesis of sex steroids, particularly androgens, and decrease the concentration of sex-hormone-binding globulin.3 The increase in bioavailable androgens may be a major cause of increased tissue concentrations of estrogens, formed by local conversion of the androgens.3

The amount, type and rate of digestion of dietary carbohydrate influence postprandial glycemia, insulin secretion and average insulin concentrations.4–6 Hence, the quantity and quality of carbohydrate consumed could have a role in breast carcinogenesis. In 1981, Jenkins et al.7 introduced the concept of the glycemic index, which is a means to quantify the glycemic response to carbohydrates in various foods. A related measure, the glycemic load, is the product of the glycemic index of the food and carbohydrate content of the portion ingested.8 The glycemic load thus quantifies the overall glycemic effect and insulin demand induced by a serving of food.

Epidemiologic studies of glycemic index and glycemic load in relation to breast cancer risk have yielded inconsistent results.9–11 Few studies have assessed the associations stratified by hormone receptor status,12–14 although breast cancers classified by estrogen receptor (ER) and progesterone receptor (PR) expression likely have different etiology.15 We examined the associations of carbohydrate intake, glycemic index and glycemic load with the incidence of breast cancer, overall and stratified by ER and PR status of the tumor, in a population-based prospective cohort of Swedish women.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

Study cohort

The Swedish Mammography Cohort was established in 1987–1990 when 66,651 women (74% of the source population) born between 1914 and 1948 and residing in central Sweden completed a questionnaire about diet, weight, height, reproductive factors and other factors. After exclusion of women with an erroneous or missing National Registration Number, those with implausible values for total energy intake (i.e., 3SDs from the loge-transformed mean energy intake) and those with a previous cancer diagnosis (except nonmelanoma skin cancer), 61,433 women remained for analysis. A second questionnaire was sent to all 56,030 women who were still alive and residing in the study area in the late autumn of 1997; 39,227 women (70%) responded to this questionnaire. The study was approved by the ethics committees at the Uppsala University Hospital and the Karolinska Institutet in Sweden.

Dietary assessment

A food frequency questionnaire (FFQ) with 67 and 96 food items was sent to participants at baseline and in 1997, respectively. In these questionnaires, women were asked to report how often, on average, they had consumed each food item during the previous 6 months (baseline FFQ) or the previous year (1997 FFQ). The FFQs had 8 mutually exclusive predefined categories for frequency of consumption, ranging from “never/seldom” to “4 or more times per day” (baseline FFQ) or “3 or times per day” (1997 FFQ). Carbohydrate intake was calculated by multiplying the frequency of consumption of each food item by its carbohydrate content per age-specific serving, using food composition values from the Swedish National Food Administration Database.16 The age-specific serving sizes were based on mean values obtained from 213 randomly selected women from the study area who weighed and recorded their food intake for a mean of 27.8 days. Glycemic index values of foods were obtained from international tables.17 We calculated the average glycemic load for each participant by multiplying the carbohydrate content (grams per serving) of each food by its glycemic index; multiplying that product by the frequency of consumption (servings of that food per day), and summing the values from all foods. Each unit of glycemic load represents the equivalent of 1 g of carbohydrate from white bread. The overall glycemic index for each participant was calculated by dividing the participant's glycemic load by the total grams of carbohydrate consumed.

In a validity study among 129 women randomly chosen from the cohort, the correlation coefficient between the baseline FFQ and the mean intakes assessed by four 1-week weighted diet records was 0.54 for carbohydrate intake (Wolk, unpublished data). The validity of nutrient intake as assessed by the second FFQ has been examined among 248 men in the study area; the correlation coefficient for carbohydrate intake was 0.73 between the FFQ and the average of fourteen 24-hr recall interviews.18

Case ascertainment and follow-up

We ascertained histologically confirmed incident cases of invasive breast cancer by linkage of the study cohort with the national and regional Swedish Cancer registers. The completeness of cancer follow-up was estimated to be almost 100%.19 Information on ER and PR status of breast tumors was obtained by reviewing pathology laboratory work logs stored at Uppsala University Hospital (from 1987 to 1994) and by linkage to the clinical database (the Quality Register) at the Regional Oncology Centre in Uppsala (from January 1992 through December 2007), which was based on the patients' original medical records. ER and PR status was evaluated by using an Abbott immunoassay until 1997 and an immunohistochemical method thereafter. Cases with ≥0.1 fmol/μg cytosol DNA were considered hormone receptor-positive when using the Abbott immunoassay. By the immunohistochemical method, cases were considered as receptor-positive when the percentage of positive cells was ≥10%, and receptor-negative when the percentage of positive cells was <10%. The Department of Pathology and Cytology at Uppsala University Hospital and Västerås Central Hospital were involved in this evaluation. Information on dates of death for deceased participants was obtained from the Swedish Death Registry.

Statistical analysis

Person-time of follow-up was calculated from the date of entry into the cohort to the date of breast cancer diagnosis, death from any cause or December 31, 2007, whichever came first. In analyses of hormone-receptor status, for women in Västmanland County, we calculated entry date from January 1, 1998 because routine evaluation of ER and PR status was implemented in Västmanland County first in 1997. Carbohydrate intake, glycemic index and glycemic load were energy-adjusted using the residual method,20 and categorized into quintiles. We used data from the baseline questionnaire in the main analyses. We also conducted analyses using simple updating of diet. In these analyses, breast cancer incidence from baseline through 1997 was related to dietary intake at baseline and outcomes from 1998 through 2007 were related to dietary intake in 1997.

Cox proportional hazards models21 were used to estimate relative risks (RRs) and 95% confidence intervals (CIs). To control as finely as possible for age and calendar time, and possible two-way interactions between these 2 time scales, we stratified the models by age in months at start of follow-up and year of enrollment. Multivariate models were adjusted for age, education, body mass index (BMI), height, parity, age at first birth, age at menarche, age at menopause, use of oral contraceptives, use of postmenopausal hormones, family history of breast cancer, and intakes of alcohol, dietary fiber and total energy. To test for trend, we assigned the median value to each quintile and modeled this value as a continuous variable. The likelihood ratio test was used to assess statistical interaction. We used the Wald statistic22 to assess whether associations differed among breast cancer subtypes defined by ER and PR status. All statistical analyses were performed with SAS software (version 9.1; SAS Institute Inc., Cary, NC). All p values were two sided; p values <0.05 were considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

Compared with women with a low-glycemic load, women with a higher glycemic load tended to be older and were less likely to have a postsecondary education and use oral contraceptives (Table I). Women with a high-glycemic load consumed less alcohol but more fiber than did women with a low-glycemic load.

Table I. Age-Standardized Baseline Characteristics of 61,433 Women in the Swedish Mammography Cohort for the Whole Study Cohort and by Quintiles of Glycemic Load in 1987–1990
Characteristics1Whole cohortQuintile of glycemic load
<164164−176177−186187−199≥200
  • 1

    All values are means if not otherwise indicated.

  • 2

    Parous women only.

Age at baseline (yr)53.750.752.053.655.157.1
Body mass index (kg/m2)24.724.724.724.824.824.8
Postsecondary education (%)12.614.113.512.611.710.9
Family history of breast cancer (%)7.26.97.46.87.47.1
Parous (%)89.189.190.189.989.087.6
No. children22.42.42.42.42.42.4
Age at first birth (yr)224.124.024.124.223.923.9
Age at menarche (yr)13.213.213.213.213.213.2
Age at menopause (yr)50.750.750.750.750.750.7
Oral contraceptive use (%)54.055.255.153.852.252.7
Postmenopausal hormone use (%)44.345.145.443.644.243.1
Total energy intake (kcal/day)158415471599160516081555
Alcohol intake (g/day)2.53.72.92.42.01.5
Dietary fiber (g/day)24.320.923.224.425.527.2

During a mean follow-up of 17.4 years of 61,433 women, we ascertained 2,952 incident cases of invasive breast cancer. Among cases, 596 were below 55 years of age at diagnosis and 2,356 were 55 years of age or older at diagnosis. Information on ER and PR status was available for 2,062 cases. Of these breast cancers, 1,286 cases (62.4%) were ER+/PR+, 417 (20.2%) were ER+/PR−, 266 (12.9%) were ER−/PR− and 93 (4.5%) were ER−/PR+. About 49% of breast cancer cases were detected at mammography screening.

Glycemic load, but not carbohydrate intake or glycemic index, was significantly positively associated with risk of overall breast cancer (Table II). When we stratified the analysis by ER and PR status, we observed significant dose-response relationship between glycemic index and glycemic load and risk of ER+/PR− breast cancer. Compared with women in the lowest quintile, those in the highest quintile of glycemic index and glycemic load had a 44% and 81% increased relative risk, respectively, of ER+/PR− tumors. Carbohydrate intake was also positively associated with ER+/PR− tumors, but there was no clear dose-response relationship. Further adjustment for intakes of dietary folate, calcium and saturated fat did not change the results. Carbohydrate intake, glycemic index and glycemic load were not associated with risk of ER+/PR+ or ER−/PR− tumors. Tests for heterogeneity between observed risk estimates for ER+/PR+, ER−/PR− and ER−/PR− tumors was statistically significant for the highest category of glycemic index (p = 0.01) and glycemic load (p = 0.003). There were no significant differences in the associations of carbohydrate intake, glycemic index and glycemic load with overall and ER/PR-defined breast cancer in subgroups of age at diagnosis (<55 years or ≥55 years).

Table II. Relative Risks and 95% Confidence Intervals of Breast Cancer by Quintiles of Carbohydrate Intake, Glycemic Index and Glycemic Load Among 61,433 Women in the Swedish Mammography Cohort, 1987−20071
QuintilesAll invasive tumors (n = 2,952)ER+/PR+ tumors (n = 1286)ER+/PR− tumors (n = 417)ER−/PR− tumors (n = 266)
CasesAge-adjustedMultivariate2Multivariate2Multivariate2Multivariate2
  • The values are given as RR (95% CI).

  • 1

    RRs and 95% CIs were estimated by using Cox proportional hazards models.

  • 2

    Adjusted for age, education (primary school, high school, university), body mass index (<18.5, 18.5–24.9, 25–29.9, ≥30 kg/m2), height (in cm; continuous), parity (nulliparous, 1–2, ≥3), age at first birth (nulliparous, <26, 26–30, ≥31 years), age at menarche (≤12, 13, ≥14 years), age at menopause (<51, ≥51 years), use of oral contraceptives (ever, never), use of postmenopausal hormones (ever, never), family history of breast cancer (yes, no), and intakes of alcohol (nondrinkers, <3.4, 3.4–9.9, ≥10.0 g/day), dietary fiber (in quartiles) and total energy (kcal/day; continuous).

  • 3

    The test for trend was calculated by using median intake of carbohydrate intake, glycemic index or glycemic load in each quintile as a continuous variable.

Carbohydrate (g/day)      
 <2115721.00 (referent)1.00 (referent)1.00 (referent)1.00 (referent)1.00 (referent)
 211−2226111.06 (0.94–1.19)1.08 (0.96–1.22)1.16 (0.97–1.39)1.20 (0.88–1.65)1.03 (0.68–1.55)
 223−2335640.98 (0.87–1.10)1.04 (0.92–1.17)1.09 (0.91–1.32)1.41 (1.02–1.94)1.28 (0.85–1.91)
 234−2456181.06 (0.94–1.19)1.14 (0.99–1.29)1.09 (0.90–1.32)1.53 (1.10–2.14)1.14 (0.74–1.74)
 ≥2465870.98 (0.87–1.11)1.09 (0.95–1.25)1.08 (0.88–1.33)1.34 (0.93–1.94)1.14 (0.73–1.79)
  p for trend3 0.810.150.630.040.50
Glycemic index      
 <75.86041.00 (referent)1.00 (referent)1.00 (referent)1.00 (referent)1.00 (referent)
 75.8−78.36141.00 (0.89–1.12)1.02 (0.91–1.14)0.91 (0.77–1.08)0.98 (0.72–1.34)1.54 (1.04–2.26)
 78.4−80.65440.92 (0.82–1.03)0.94 (0.84–1.06)0.81 (0.68–0.97)1.08 (0.79–1.48)1.35 (0.90–2.02)
 80.7−83.35801.00 (0.89–1.12)1.04 (0.92–1.17)0.92 (0.78–1.10)1.21 (0.89–1.65)1.27 (0.84–1.92)
 ≥83.46101.04 (0.92–1.16)1.08 (0.96–1.21)0.89 (0.74–1.06)1.44 (1.06–1.97)1.29 (0.85–1.96)
  p for trend3 0.610.200.320.010.62
Glycemic load      
 <1645871.00 (referent)1.00 (referent)1.00 (referent)1.00 (referent)1.00 (referent)
 164−1755931.00 (0.89–1.12)1.03 (0.92–1.16)0.96 (0.80–1.14)1.12 (0.82–1.55)1.22 (0.83–1.82)
 176−1865730.97 (0.86–1.09)1.03 (0.91–1.16)0.95 (0.79–1.14)1.47 (1.07–2.01)1.20 (0.79–1.81)
 187−1995900.98 (0.87–1.10)1.06 (0.94–1.20)0.99 (0.82–1.19)1.24 (0.88–1.74)1.19 (0.78–1.80)
 ≥2006091.02 (0.91–1.15)1.13 (1.00–1.29)0.94 (0.77–1.13)1.81 (1.29–2.53)1.23 (0.79–1.90)
  p for trend3 0.840.050.590.00080.45

Because adiposity is an important determinant of insulin resistance, the relations between carbohydrate intake, glycemic index, and glycemic index and breast cancer risk may vary by BMI. Glycemic load was positively associated with overall breast cancer risk in women with a BMI <25 kg/m2 (RR for the highest versus lowest quintile = 1.26; 95% CI = 1.06–1.50; p for trend = 0.01) but not in women with a BMI ≥25 kg/m2 (RR = 1.08; 95% CI = 0.88–1.33; p for trend = 0.46). However, a test for interaction was not statistically significant (p for interaction = 0.70). The corresponding RRs for ER+/PR− tumors were 2.03 (95% CI = 1.35–3.06; p for trend = 0.001) in women with a BMI <25 kg/m2 and 1.80 (95% CI = 0.92–3.53; p for trend = 0.12) in women with a BMI ≥25 kg/m2 (p for interaction = 0.61).

To examine whether carbohydrate intake, glycemic index and glycemic load close in time to breast cancer diagnosis is important, we used simple updating of diet. In these analyses, the multivariate RRs of overall breast cancer comparing extreme quintiles were 1.28 (95% CI = 0.90–1.84; p-trend = 0.26) for carbohydrate intake, 1.24 (95% CI = 0.90–1.70; p-trend = 0.24) for glycemic index and 1.49 (95% CI = 1.05–2.12; p-trend = 0.17) for glycemic load. The corresponding RRs for ER+/PR− tumors, were 1.31 (95% CI = 0.93–1.84; p-trend = 0.18) for carbohydrate intake, 1.31 (95% CI = 0.97–1.78; p-trend = 0.08) for glycemic index and 1.55 (95% CI = 1.12–2.15; p-trend = 0.05) for glycemic load.

In a sensitivity analysis, we related dietary intake in 1997 to breast cancer incidence from 1998 through 2007. The multivariate RR of overall breast cancer comparing extreme quintiles of glycemic load was 1.02 (95% CI = 0.82–1.27). When we stratified the analysis by screen-detected and nonscreen-detected cancer, the multivariate RRs for the highest versus lowest quintile of glycemic load were 0.98 (95% CI = 0.71–1.35) for screen-detected and 1.03 (95% CI = 0.76–1.41) for nonscreen-detected breast cancer.

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References

In this cohort study of Swedish women, we observed positive associations of carbohydrate intake, glycemic index and glycemic load with risk of ER+/PR− breast cancer, but not with other breast cancers jointly defined by ER and PR status. We also found a positive association between glycemic load and overall breast cancer risk.

Glycemic index and glycemic load are measures that assess different aspects of dietary carbohydrates. Glycemic index is a measure of the overall quality of carbohydrates in the diet whereas glycemic load incorporates both quality and quantity of carbohydrates consumed. If breast cancer risk is related to the overall insulin demand of the diet, a stronger association would be expected for glycemic load,23 as was observed in the present study.

Major strengths of this study include the prospective and population-based design, a large sample size, detailed information on diet, data on hormone receptor status of the breast tumor and nearly complete follow-up. It is unlikely that our findings were biased by the way of detection of breast cancer, because we observed similar results for screen-detected and nonscreen-detected breast cancer. A limitation of our study is that diet was assessed using a self-administered FFQ, which will inevitably lead to measurement error and, consequently attenuated risk estimates. Thus, as in previous studies, our results may have been affected by misclassification of carbohydrate intake, glycemic index and glycemic load due to errors of recall. In addition, estimates of glycemic index and glycemic load from the FFQ may not accurately reflect the glycemic and insulinemic effects of consumption and metabolism of mixed dishes and prepared foods. Glycemic index values have been determined for a relatively limited number of foods and some values are derived from small studies or based on different methods. Our results may also have been affected by differential systematic error because of underreporting of intake within specific population subgroups, such as overweight women.24 Finally, our findings may be subject to chance because many subgroup analyses was performed.

Previous epidemiologic studies of glycemic index and glycemic load in relation to overall breast cancer risk have yielded inconsistent, but mostly null, results.9–11 Summary results of a recent meta-analysis of 7 prospective studies showed that compared with the lowest category, breast cancer risk was not increased for women in the highest category of glycemic index (RR = 1.06; 95% CI = 0.98–1.15) or glycemic load (RR = 0.99; 95% CI = 0.94–1.06).10

Data on glycemic index and glycemic load in relation to hormone-receptor-defined breast cancer are limited and inconsistent.12–14 Our results are consistent with those of an Australian cohort study, which found a positive, but nonsignificant, relation between glycemic load and risk of ER+/PR− breast cancer (n = 39 cases; RR for an increase of 1SD = 1.32; 95% CI = 0.60–2.90).14 In a French cohort study, glycemic load was positively associated with risk of ER− tumors (n = 279 cases) but not ER+, PR+, or PR− tumors.12 A Danish cohort study found a positive association between glycemic index and ER− tumors (n = 122 cases) but not ER+ tumors; no association was found for glycemic load.13

The biological mechanisms for the observed differential associations with breast cancers jointly defined by ER and PR status are unclear. High-glycemic load diets may increase breast cancer risk via increased concentrations of insulin, IGF-I and sex hormones (androgens and estrogens). Estrogens and insulin/IGF-I have potent positive effects on the proliferation of mammary epithelial cells and estrogen-dependent breast cancer cells.25 Results of an in vitro study showed that ERalpha is a critical requirement for IGF signaling.26 Thus, an association between glycemic load and breast cancer risk may be stronger or limited to ER+ tumors. We have no explanation for the lack of observed association between glycemic load and risk of ER+/PR+ tumors.

We would have expected a stronger association between glycemic load and breast cancer risk among the overweight women with underlying insulin resistance. In the present study, the association between glycemic load and breast cancer risk was slightly stronger in women who were not overweight, but a test for interaction was not significant. Although this finding is likely due to chance it is consistent with the results of a stratified analysis of 298 breast cancer cases in the ORDET Study.27 That study showed a positive association between glycemic load and risk of breast cancer in nonoverweight women (RR for highest versus lowest quintile = 5.79; 95% CI = 2.60–12.9) but not in overweight women (corresponding RR = 1.31; 95% CI = 0.66–2.61; p for interaction = 0.006).27 Likewise, in the Canadian National Breast Screening Study, the observed positive association between glycemic index and breast cancer risk in postmenopausal women was somewhat stronger in nonoverweight women (RR for highest versus lowest quintile = 1.99; 95% CI = 1.06–9.72).28 In contrast, 2 other cohort studies found a statistically significant or a nonsignificant increased risk of breast cancer associated with a high-glycemic index12 or glycemic load29 only in overweight women. Other cohort studies showed no difference in association of glycemic index or glycemic load with breast cancer risk across strata of BMI.30–32

In conclusion, findings from this cohort study suggest that carbohydrate intake, glycemic index and glycemic load are positively associated with risk of ER+/PR− breast cancer. Further research is warranted to assess the differential associations of glycemic index and glycemic load with hormone receptor-defined breast cancer.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. References
  • 1
    Renehan AG,Tyson M,Egger M,Heller RF,Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 2008; 371: 56978.
  • 2
    Larsson SC,Mantzoros CS,Wolk A. Diabetes mellitus and risk of breast cancer: a meta-analysis. Int J Cancer 2007; 121: 85662.
  • 3
    Kaaks R,Lukanova A. Energy balance and cancer: the role of insulin and insulin-like growth factor-I. Proc Nutr Soc 2001; 60: 91106.
  • 4
    Ludwig DS. The glycemic index: physiological mechanisms relating to obesity, diabetes, and cardiovascular disease. JAMA 2002; 287: 241423.
  • 5
    Bjorck I,Granfeldt Y,Liljeberg H,Tovar J,Asp NG. Food properties affecting the digestion and absorption of carbohydrates. Am J Clin Nutr 1994; 59: 699S705S.
  • 6
    Augustin LS,Franceschi S,Jenkins DJ,Kendall CW,La Vecchia C. Glycemic index in chronic disease: a review. Eur J Clin Nutr 2002; 56: 104971.
  • 7
    Jenkins DJ,Wolever TM,Taylor RH,Barker H,Fielden H,Baldwin JM,Bowling AC,Newman HC,Jenkins AL,Goff DV. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981; 34: 3626.
  • 8
    Salmeron J,Manson JE,Stampfer MJ,Colditz GA,Wing AL,Willett WC. Dietary fiber, glycemic load, and risk of non-insulin-dependent diabetes mellitus in women. JAMA 1997; 277: 4727.
  • 9
    Gnagnarella P,Gandini S,La Vecchia C,Maisonneuve P. Glycemic index, glycemic load, and cancer risk: a meta-analysis. Am J Clin Nutr 2008; 87: 1793801.
  • 10
    Barclay AW,Petocz P,McMillan-Price J,Flood VM,Prvan T,Mitchell P,Brand-Miller JC. Glycemic index, glycemic load, and chronic disease risk—a meta-analysis of observational studies. Am J Clin Nutr 2008; 87: 62737.
  • 11
    Mulholland HG,Murray LJ,Cardwell CR,Cantwell MM. Dietary glycaemic index, glycaemic load and breast cancer risk: a systematic review and meta-analysis. Br J Cancer 2008; 99: 11705.
  • 12
    Lajous M,Boutron-Ruault MC,Fabre A,Clavel-Chapelon F,Romieu I. Carbohydrate intake, glycemic index, glycemic load, and risk of postmenopausal breast cancer in a prospective study of French women. Am J Clin Nutr 2008; 87: 138491.
  • 13
    Nielsen TG,Olsen A,Christensen J,Overvad K,Tjonneland A. Dietary carbohydrate intake is not associated with the breast cancer incidence rate ratio in postmenopausal Danish women. J Nutr 2005; 135: 1248.
  • 14
    Giles GG,Simpson JA,English DR,Hodge AM,Gertig DM,Macinnis RJ,Hopper JL. Dietary carbohydrate, fibre, glycaemic index, glycaemic load and the risk of postmenopausal breast cancer. Int J Cancer 2006; 118: 18437.
  • 15
    Althuis MD,Fergenbaum JH,Garcia-Closas M,Brinton LA,Madigan MP,Sherman ME. Etiology of hormone receptor-defined breast cancer: a systematic review of the literature. Cancer Epidemiol Biomarkers Prev 2004; 13: 155868.
  • 16
    Bergström L,Kylberg E,Hagman U,Erikson H,Bruce Å. The food composition database KOST: the National Food Administration's Information System for nutritive values of food. Vår Föda 1991; 43: 43947.
  • 17
    Foster-Powell K,Holt SH,Brand-Miller JC. International table of glycemic index and glycemic load values: 2002. Am J Clin Nutr 2002; 76: 556.
  • 18
    Messerer M,Johansson SE,Wolk A. The validity of questionnaire-based micronutrient intake estimates is increased by including dietary supplement use in Swedish men. J Nutr 2004; 134: 18005.
  • 19
    Mattsson B,Wallgren A. Completeness of the Swedish Cancer Register. Non-notified cancer cases recorded on death certificates in 1978. Acta Radiol Oncol 1984; 23: 30513.
  • 20
    Willett W,Stampfer MJ. Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 1986; 124: 1727.
  • 21
    Cox DR,Oakes D. Analysis of survival data. London: Chapman and Hall, 1984.
  • 22
    Greenland S,Rothman KJ. Introduction to stratified analysis. In: RothmanKJ,GreenlandS, eds. Modern epidemiology, vol. 2. Philadelphia, PA: Lippincott-Raven, 1998. 25379.
  • 23
    Key TJ. Glycemic index, hyperinsulinemia, and breast cancer risk. Ann Oncol 2001; 12: 15079.
  • 24
    Ferrari P,Slimani N,Ciampi A,Trichopoulou A,Naska A,Lauria C,Veglia F,Bueno-de-Mesquita HB,Ocke MC,Brustad M,Braaten T,Jose Tormo M, et al. Evaluation of under- and overreporting of energy intake in the 24-hour diet recalls in the European Prospective Investigation into Cancer and Nutrition (EPIC). Public Health Nutr 2002; 5: 132945.
  • 25
    Lanzino M,Morelli C,Garofalo C,Panno ML,Mauro L,Ando S,Sisci D. Interaction between estrogen receptor alpha and insulin/IGF signaling in breast cancer. Curr Cancer Drug Targets 2008; 8: 597610.
  • 26
    Oesterreich S,Zhang P,Guler RL,Sun X,Curran EM,Welshons WV,Osborne CK,Lee AV. Re-expression of estrogen receptor alpha in estrogen receptor alpha-negative MCF-7 cells restores both estrogen and insulin-like growth factor-mediated signaling and growth. Cancer Res 2001; 61: 57717.
  • 27
    Sieri S,Pala V,Brighenti F,Pellegrini N,Muti P,Micheli A,Evangelista A,Grioni S,Contiero P,Berrino F,Krogh V. Dietary glycemic index, glycemic load, and the risk of breast cancer in an Italian prospective cohort study. Am J Clin Nutr 2007; 86: 11606.
  • 28
    Silvera SA,Jain M,Howe GR,Miller AB,Rohan TE. Dietary carbohydrates and breast cancer risk: a prospective study of the roles of overall glycemic index and glycemic load. Int J Cancer 2005; 114: 6538.
  • 29
    Cho E,Spiegelman D,Hunter DJ,Chen WY,Colditz GA,Willett WC. Premenopausal dietary carbohydrate, glycemic index, glycemic load, and fiber in relation to risk of breast cancer. Cancer Epidemiol Biomarkers Prev 2003; 12: 11538.
  • 30
    Higginbotham S,Zhang ZF,Lee IM,Cook NR,Buring JE,Liu S. Dietary glycemic load and breast cancer risk in the Women's Health Study. Cancer Epidemiol Biomarkers Prev 2004; 13: 6570.
  • 31
    Jonas CR,McCullough ML,Teras LR,Walker-Thurmond KA,Thun MJ,Calle EE. Dietary glycemic index, glycemic load, and risk of incident breast cancer in postmenopausal women. Cancer Epidemiol Biomarkers Prev 2003; 12: 5737.
  • 32
    Holmes MD,Liu S,Hankinson SE,Colditz GA,Hunter DJ,Willett WC. Dietary carbohydrates, fiber, and breast cancer risk. Am J Epidemiol 2004; 159: 7329.