1 Servizio di Epidemiologia, Centro di Riferimento Oncologico, Istituto Nazionale Tumori, Aviano, Italy; 2 Department of Nutritional Sciences, Faculty of Medicine, University of Toronto and the Clinical Nutrition and Risk Factor Modification Centre, St Michaels Hospital, Toronto, Ontario, Canada; 3 Istituto di Ricerche Farmacologiche "Mario Negri", Milan; 4 Istituto di Statistica Medica e Biometria, Università degli Studi di Milano, Milan; 5 Servizio Integrato di Epidemiologia e Sistemi Informativi, Rome; 6 Istituto Tumori "Fondazione Pascale", Naples, Italy; 7 International Agency for Research on Cancer, Lyon, France
Received 3 April 2002; Revised 3 July 2002; accepted 18 July 2002
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Dietary carbohydrates vary in their ability to raise blood glucose and insulin levels, which, in turn, influence levels of sex hormones and insulin-like growth factors. We analyzed the effect of type and amount of carbohydrates on ovarian cancer risk, using the glycemic index (GI) and the glycemic load (GL) measurement in a large casecontrol study conducted in Italy.
Materials and methods:
Cases included 1031 women with incident, histologically confirmed epithelial ovarian cancer, from four Italian regions. Controls included 2411 women admitted to the same hospital networks for acute, non-neoplastic conditions. Average daily GI and GL were calculated from a validated food frequency questionnaire. Odds ratios (OR) and the corresponding 95% confidence intervals (CI) were computed using multiple logistic regression.
Results:
Ovarian cancer was directly associated with dietary GI (OR for highest versus lowest quartile = 1.7, 95% CI 1.32.1) and GL (OR = 1.7, 95% CI 1.32.1). The associations were observed in pre- and postmenopausal women, and they remained consistent across strata of major covariates identified.
Conclusions:
This study supports the hypothesis of a direct association between GI and GL and ovarian cancer risk and, consequently, of a possible role of hyperinsulinemia/insulin resistance in ovarian cancer development.
Key words: carbohydrate, casecontrol study, glycemic load, ovarian cancer
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
With reference to specific nutrients, descriptive epidemiology and ecological studies have reported positive relationships between fat, protein and total calory intake, and ovarian cancer risk [7]. Data from analytical, mainly casecontrol studies supported the hypothesis of a possible increased risk in relation to various types of fat [2, 6]. Carbohydrates have also been shown to increase the risk of epithelial ovarian cancer [8].
Different carbohydrates, however, affect blood glucose and insulin levels to varying degrees depending on the nature of the carbohydrate and the type and extent of food processing [9]. On this basis they have been ranked using the glycemic index (GI) and glycemic load (GL). Foods with high GI tend to increase glucose and insulin levels to a greater extent than low GI foods [9]. In turn, it has been proposed that insulin may be directly or indirectly involved in the carcinogenic process by modulating hormonal levels such as sex hormones and insulin-like growth factors (IGF) [10]. High-GL diets have been directly associated with risk of various Western chronic conditions, including diabetes [11], coronary heart disease [12], colorectal [13] and breast [14] cancer, and high insulin levels may be one of the mechanisms of action of risk factors shared by these diseases [15].
We thus evaluated the possible differential effects of carbohydrate-rich foods on epithelial ovarian cancer risk by means of the GI and GL measurements in a large casecontrol study.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Cases included women with incident, histologically confirmed epithelial ovarian cancer diagnosed within 1 year prior to interview and with no previous diagnosis of cancer. Overall, 1031 women aged 1879 years (median age 56 years) were included. Controls included patients with no history of cancer who were admitted to hospitals serving the same areas as those where cases had been identified. Eligible diagnoses were acute, non-neoplastic, non-gynecological conditions, unrelated to hormonal or digestive tract diseases, or associated with long-term modifications of diet. They included 2411 women, aged 1779 years (median age 57 years), belonging to the following diagnostic categories: trauma, mostly fractures and sprains (26%); other orthopedic disorders, such as low back pain and disc disorders (28%); acute surgical conditions (15%); and other illnesses, such as eye, ear, nose, skin and dental conditions (31%). Cases were frequency matched to controls according to quinquennium of age and area of residence. Approximately 4% of cases and controls approached for interview during their hospital stay refused to participate.
The same structured questionnaire and coding manual were used in each center, and all interviewers were centrally trained and routinely supervised. The checking of data for consistency and reliability was also conducted centrally. The questionnaire included information on sociodemographic characteristics, such as education and occupation, lifelong smoking habits, physical activity at various ages, anthropometric measures, a problem-oriented personal medical history, family history of selected cancers, menstrual and reproductive history, history of use of oral contraceptives, hormone replacement treatment, and female hormone-containing drugs for other indications. Dietary habits were investigated through an interviewer-administered food frequency questionnaire (FFQ) that included 78 items. This questionnaire was used to assess the subjects habitual diet during the 2 years prior to cancer diagnosis or hospital admission (for controls), and included questions on the average weekly frequency of consumption of foods or food groups, as well as complex recipes. Satisfactory reproducibility [16] and validity [17] of the FFQ have been reported. Details on methodology used have been described elsewhere [13, 14]. To compute energy and nutrient intake, an Italian food-composition database was used. For each food, we expressed GI as a percentage of the glycemic response elicited using white bread as a standard food. We then calculated daily average GI by summing the products of the carbohydrate content per serving for each food or recipe, multiplied by the average number of servings of that food per week, multiplied by its GI, all divided by the total amount of available carbohydrate weekly intake. This represents the quality of the carbohydrates, namely slow versus fast absorbable carbohydrates. A score for the daily average GL was computed as the GI, but without dividing by the total amount of carbohydrates. For these calculations we used the carbohydrate content of 50 foods or recipes, since 28 foods or recipes, chiefly cheese, meat and fish-based, contained a negligible amount of carbohydrate. With respect to GI values, we chiefly used international tables. In order to take into account Italian cooking habits (e.g. pasta al dente), Italian sources were used for a few local recipes. Food items for which a GI had not been determined were assigned the GI of the nearest comparable food (e.g. tangerines were assigned the same GI as oranges).
Odds ratios (ORs) and the corresponding 95% confidence intervals (CI) for quartiles of GI and GL intake were computed using unconditional multiple logistic regression models [18]. The regression equations included terms for quinquennia of age, study center, years of education, occupational physical activity, history of diabetes, oral contraceptive use, parity, menopausal status, number of daily meals, intakes of fiber, alcohol and total energy intake. Adjustment for energy was made using the residuals method. The modifying effect of various covariates was evaluated comparing the differences between the 2 log likelihood of the model with and without interaction terms, and referring it to the chi square distribution with degrees of freedom equal to the number of interaction terms minus one.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Table 2 shows the ORs of epithelial ovarian cancer according to the quartiles of GI and GL, and total carbohydrate intake by menopausal status. Dietary GI and GL were directly associated with ovarian cancer risk, and the ORs, for the highest versus the lowest quartile, were 1.7 (95% CI 1.32.1) and 1.7 (95% CI 1.32.1), respectively. However, ORs by quartile of GI and GL did not show linear trends, but were already elevated in the second quartile and tended to plateau thereafter. Associations, particularly for GI, were appreciably stronger in postmenopausal compared with premenopausal women, although no significant heterogeneity emerged. Total carbohydrate intake was also associated with ovarian cancer (OR = 1.8, in the highest quartile, 95% CI 1.32.4) in postmenopausal women (Table 2).
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Diets with high GI or GL are associated with a high consumption of refined carbohydrates, which are quickly absorbed and are capable of elevating blood glucose and insulin level to a greater extent than slowly absorbed ones, such as pulses and whole grains, which are low GI foods [20]. The main sources of carbohydrates in the Italian population are bread, a high GI food, and pasta, a medium-low GI food, representing 20.5% and 13.4% of total carbohydrate intake, respectively [13].
High insulin levels have been suggested as a potential unifying mechanism for the risk of several Western chronic diseases related to high intakes of energy, fat, refined carbohydrates, and low physical activity and obesity [15]. Central obesity (i.e. high W/H ratio) was associated with ovarian cancer risk in this study [21]. Diabetes, which is characterized by high insulin levels in its early stages, was considered as a possible correlate of ovarian cancer risk. However, in line with other studies [22, 23] a history of diabetes was not found to consistently affect ovarian cancer risk in the present study. This could, however, have resulted from the small absolute number of diabetic subjects (4%).
Insulin is a growth factor for cancer cells, and it has been shown to act as a cancer promoter in in vitro and in animal studies [15, 24]. Insulin also has affinity for IGF receptors, particularly the IGF-1 receptor, which has strong mitogenic effects on normal and neoplastic cells, including ovarian carcinoma cell lines, where it has been found at higher levels than in non-malignant cells [25]. Epidemiological evidence suggests a promoting effect of hyperinsulinemia [15] and of IGF-1 in carcinogenesis [25].
Insulin and IGF-1 are also powerful negative regulators of sex hormone-binding globulin (SHBG) synthesis in vitro, and they may stimulate ovarian cancer proliferation through a hormonal pathway [26]. An interaction between insulin, IGFs and sex hormones has also been suggested for breast cancer [27].
As in most casecontrol studies, recall and selection biases are possible [18]. However, awareness about any dietary hypotheses, and particularly those related to GI and GL, for ovarian cancer was limited in the Italian population when the study was conducted. While it is conceivable that dietary habits of hospital controls may have differed from those of the general population, great attention was paid in this study to minimize bias by excluding control subjects admitted for conditions that might have been associated with special dietary habits. Of greater concern is the early weight loss often occurring in ovarian cancer patients, which may have led cases to increase their energy and, hence, carbohydrate intake [21]. We had, however, information on weight loss during the year prior to cancer diagnosis or interview. Stratification and adjustment for weight or recent weight loss did not modify the association with GI and GL. Interviewing all subjects in a hospital setting may have allowed greater comparability of dietary history between cases and controls [28], and adjustment for total energy intake should have controlled for potential dietary over- and under-reporting. Furthermore, participation among eligible patients was practically complete and the catchment areas for cases and controls were highly comparable.
GI estimates have some limitations. Some GI estimates have been derived from small samples and their variability is unclear [13]. Statistics on the average dietary GI and GL in the general Italian population are not available, but intakes of bread and pasta in the present study were similar to those reported in the Italian population [29]. In addition, it would be important to confirm the association between GI, GL and ovarian cancer in different populations, since the genotype for insulin resistance may vary between ethnic groups [30].
The major strength of this study is its uniquely large dataset, which allowed reasonably precise risk estimates. Other strengths include consistency of findings, when major categories of controls were used separately, and its reliance on a validated food frequency questionnaire [16, 17].
In conclusion, this study found associations between dietary GI, GL and ovarian cancer risk, thus supporting a possible role of insulin and insulin-related factors in ovarian carcinogenesis. Similar associations were observed for colorectal [13] and breast [14] cancer, indicating the potential role of these factors on several common neoplasms.
![]() |
Acknowledgements |
---|
![]() |
Footnotes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Kushi LH, Mink PJ, Folsom AR et al. Prospective study of diet and ovarian cancer. Am J Epidemiol 1999; 149: 2131.[Abstract]
3. Engle A, Muscat JE, Harris RE. Nutritional risk factors and ovarian cancer. Nutr Cancer 1991; 15: 239247.[ISI][Medline]
4. Shu XO, Gao YT, Yuan JM et al. Dietary factors and epithelial ovarian cancer. Br J Cancer 1989; 59: 9296.[ISI][Medline]
5. Bosetti C, Negri E, Franceschi S et al. Diet and ovarian cancer risk: a casecontrol study in Italy. Int J Cancer 2001; 93: 911915.[CrossRef][ISI][Medline]
6. Cramer DW, Welch WR, Hutchison GB et al. Dietary animal fat in relation to ovarian cancer risk. Obstet Gynecol 1984; 63: 833838.[Abstract]
7. Byers T, Marshall J, Graham S et al. A casecontrol study of dietary and non-dietary factors in ovarian cancer. J Natl Cancer Inst 1993; 71: 681686.
8. Bidoli E, La Vecchia C, Montella M et al. Nutrient intake and ovarian cancer: an Italian casecontrol study. Cancer Causes Control 2002; 13: 255261.[CrossRef][ISI][Medline]
9. Jenkins DJA, Wolever TMS, Taylor RH et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981; 34: 362366.[Abstract]
10. Stoll BA. Western nutrition and the insulin resistance syndrome: a link to breast cancer. Eur J Clin Nutr 1999; 53: 8387.[ISI][Medline]
11. Salmeron J, Manson JE, Stampfer MJ et al. Dietary fibre, glycemic load, and risk of non-insulin-dependent diabetes mellitus in women. JAMA 1997; 277: 472477.[Abstract]
12. Liu S, Willett WC, Stampfer MJ et al. A prospective study of dietary glycemic load, carbohydrate intake, and risk of coronary heart disease in US women. Am J Clin Nutr 2000; 71: 14551461.
13. Franceschi S, Dal Maso L, Augustin L et al. Dietary glycemic load and colorectal cancer risk. Ann Oncol 2001; 12: 173178.[Abstract]
14. Augustin LS, Dal Maso L, Franceschi S et al. Dietary glycemic index and glycemic load and breast cancer risk: a casecontrol study. Ann Oncol 2001; 12: 15331538.[Abstract]
15. Giovannucci E. Insulin and colon cancer. Cancer Causes Control 1995; 6: 164179.[ISI][Medline]
16. Franceschi S, Barbone F, Negri E et al. Reproducibility of an Italian food frequency questionnaire for cancer studies. Results for specific nutrients. Ann Epidemiol 1995; 5: 6975.[CrossRef][Medline]
17. Decarli A, Franceschi S, Ferraroni M et al. Validation of a food-frequency questionnaire to assess dietary intakes in cancer studies in Italy. Results for specific nutrients. Ann Epidemiol 1996; 6: 110118.[CrossRef][ISI][Medline]
18. Breslow NE, Day NE. Statistical methods in cancer research. Vol. I. The analysis of casecontrol studies. IARC Scientific Publications No. 32. Lyon: International Agency for Research on Cancer 1980.
19. Risch HA, Marrett LD, Jain M, Howe GR. Differences in risk factors for epithelial ovarian cancer by histologic type. Results of a casecontrol study. Am J Epidemiol 1996; 144: 363372.[Abstract]
20. Jenkins DJ, Wolever TM, Jenkins AL. Starchy foods and glycemic index. Diabetes Care 1988; 11: 149159.
21. Dal Maso L, Franceschi S, Negri E et al. Body size indices at different ages and epithelial ovarian cancer risk. Eur J Cancer 2002; 38: 17691774.[CrossRef][ISI][Medline]
22. Parazzini F, Moroni S, La Vecchia C et al. Ovarian cancer risk and history of selected medical conditions linked with female hormones. Eur J Cancer 1997; 33: 16341637.[CrossRef][ISI][Medline]
23. Adler AI, Wiss NS, Kamb ML, Lyon JL. Is diabetes mellitus a risk factor for ovarian cancer? A casecontrol study in Utah and Washington (United States). Cancer Causes Control 1996; 7: 475478.[ISI][Medline]
24. Tran TT, Medline A, Bruce WR. Insulin promotion of colon tumors in rats. Cancer Epidemiol Biomarkers Prev 1996; 5: 10131015.[Abstract]
25. Yu H, Rohan T. Role of insulin-like growth factor family in cancer development and progression. J Natl Cancer Inst 2000; 92: 14721489.
26. Risch HA. Hormonal etiology of epithelial ovarian cancer, with a hypothesis concerning the role of androgens and progesterone. J Natl Cancer Inst 1998; 90: 17741786.
27. Yee D, Lee AV. Crosstalk between the insulin-like growth factors and estrogens in breast cancer. J Mammary Gland Biol Neoplasia 2000; 5: 107115.[CrossRef][ISI][Medline]
28. DAvanzo B, La Vecchia C, Katsouyanni K et al. An assessment, and reproducibility of food frequency data provided by hospital controls. Eur J Cancer Prev 1997; 6: 288293.[ISI][Medline]
29. Turrini A, Leclercq C, DAmicis A. Patterns of food and nutrient intakes in Italy and their application to the development of food-based dietary guidelines. Br J Nutr 1999; 81 (Suppl 2): S83S89.[ISI][Medline]
30. Kasim-Karakas SE. Ethnic differences in the insulin resistance syndrome. Am J Clin Nutr 2000; 71: 670671.