Intake of the major carotenoids and the risk of epithelial ovarian cancer in a pooled analysis of 10 cohort studies Anita Koushik 1 , David J. Hunter 1–4 , Donna Spiegelman 2,5 , Kristin E. Anderson 6 , Julie E. Buring 2,7 , Jo L. Freudenheim 8 , R. Alexandra Goldbohm 9 , Susan E. Hankinson 2,3 , Susanna C. Larsson 10 , Michael Leitzmann 11 , James R. Marshall 8 , Marjorie L. McCullough 12 , Anthony B. Miller 13 , Carmen Rodriguez 12 , Thomas E. Rohan 14 , Julie A. Ross 6 , Arthur Schatzkin 11 , Leo J. Schouten 15 , Walter C. Willett 1–3 , Alicja Wolk 10 , Shumin M. Zhang 2,7 and Stephanie A. Smith-Warner 1,2 * 1 Department of Nutrition, Harvard School of Public Health, Boston, MA 2 Department of Epidemiology, Harvard School of Public Health, Boston, MA 3 Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 4 Harvard Center for Cancer Prevention, Boston, MA 5 Department of Biostatistics, Harvard School of Public Health, Boston, MA 6 Division of Epidemiology and Community Health, School of Public Health, University of Minnesota, Minneapolis, MN 7 Division of Preventive Medicine, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 8 Department of Social and Preventive Medicine, University at Buffalo, The State University of New York, Buffalo, NY 9 Department of Food and Chemical Risk Analysis, TNO Quality of Life, Zeist, The Netherlands 10 Division of Nutritional Epidemiology, National Institute of Environmental Medicine, Karolinska Institute, Stockholm, Sweden 11 Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD 12 Epidemiology and Surveillance Research, American Cancer Society, Atlanta, GA 13 Department of Public Health Sciences, Faculty of Medicine, University of Toronto, Toronto, ON, Canada 14 Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY 15 Department of Epidemiology, NUTRIM, Maastricht University, Maastricht, The Netherlands Carotenoids, found in fruits and vegetables, have the potential to protect against cancer because of their properties, including their functions as precursors to vitamin A and as antioxidants. We exam- ined the associations between intakes of a-carotene, b-carotene, b-cryptoxanthin, lutein/zeaxanthin and lycopene and the risk of invasive epithelial ovarian cancer. The primary data from 10 pro- spective cohort studies in North America and Europe were ana- lyzed and then pooled. Carotenoid intakes were estimated from a validated food frequency questionnaire administered at baseline in each study. Study-specific relative risks (RR) were estimated using the Cox proportional hazards model and then combined using a random-effects model. Among 521,911 women, 2,012 cases of ovar- ian cancer occurred during a follow-up of 7–22 years across studies. The major carotenoids were not significantly associated with the risk of ovarian cancer. The pooled multivariate RRs (95% confi- dence intervals) were 1.00 (0.95–1.05) for a 600 lg/day increase in a-carotene intake, 0.96 (0.93–1.03) for a 2,500 lg/day increase in b-carotene intake, 0.99 (0.97–1.02) for a 100 lg/day increase in b-cryptoxanthin intake, 0.98 (0.94–1.03) for a 2,500 lg/day increase in lutein/zeaxanthin intake and 1.01 (0.97–1.05) for a 4,000 lg/day increase in lycopene intake. These associations did not appreciably differ by study (p-values, tests for between-studies heterogeneity >0.17). Also, the observed associations did not vary substantially by subgroups of the population or by histological type of ovarian can- cer. These results suggest that consumption of the major carote- noids during adulthood does not play a major role in the incidence of ovarian cancer. ' 2006 Wiley-Liss, Inc. Key words: ovarian cancer; carotenoids; pooled analysis; meta- analysis; cohort studies Because ovarian cancer is usually diagnosed at advanced stages and has poor survival, 1–3 prevention offers an approach to reduc- ing ovarian cancer mortality. Factors that have consistently been associated with a reduced risk of ovarian cancer include increasing parity, oral contraceptive use, lactation and tubal ligation. 4–6 The consumption of foods with cancer preventive nutrients also may play a role in reducing ovarian cancer risk. Carotenoids are found in many fruits and vegetables and contribute to the yellow to red color of these foods. 7,8 The most common dietary carotenoids are a-carotene, b-carotene and b-cryptoxanthin, which have provita- min A activity, as well as lycopene, lutein and zeaxanthin, which do not have provitamin A activity. 9 Vitamin A activity is impor- tant for the normal control of cellular differentiation and prolifera- tion, 10 thus, inadequate levels of vitamin A could enhance carcino- genesis. Carotenoids also have strong antioxidant properties and may therefore play a role in preventing cancer by reducing the potential for damage to cells caused by cellular oxidative reac- tions. 8 Food composition databases for individual carotenoids have not been available until relatively recently. 9,11 Thus, there have been a limited number of studies examining the association between intakes of the major carotenoids and the risk of ovarian cancer, 12–20 and the results for each carotenoid have been inconsistent. Food composition tables that existed before databases for individual caro- tenoids were available used various assumptions to estimate the vitamin A activity of carotenoids in foods, and often expressed this as b-carotene equivalents. 11 Thus, in earlier reports of b-carotene intake and ovarian cancer risk, 21–25 the reported associations re- flected intakes of multiple carotenoids assessed by inconsistent means. In a meta-analysis of these earlier studies that included 1 cohort study and 4 case–control studies, women in the highest ver- sus the lowest category of intake had a statistically significant 16% lower risk of ovarian cancer. 26 Overall, the associations between the major dietary carotenoids and ovarian cancer risk have been minimally examined. 12–20 Only 2 previous studies used a prospective design to analyze the most common dietary carotenoids separately. 17,20 In the present analysis, we examined the association between intakes of a-carotene, b-car- otene, b-cryptoxanthin, lutein and zeaxanthin combined, and lyco- pene and the risk of invasive epithelial ovarian cancer in a pooled analysis of 10 large North American and European prospective cohort studies, 17,20,25,27–33 including 3 prospective studies that have previously published results on these associations. 17,20,25 Using the primary data from each study, we categorized individual carote- noid intakes and potential confounding factors using standardized definitions across studies and conducted multivariate analyses for Grant sponsor: NIH; Grant number: CA55075. *Correspondence to: Department of Nutrition, Harvard School of Pub- lic Health, 665 Huntington Avenue, Boston, MA 02115, USA. E-mail: pooling@hsphsun2.harvard.edu Received 31 January 2006; Accepted 12 April 2006 DOI 10.1002/ijc.22076 Published online 5 July 2006 in Wiley InterScience (www.interscience. wiley.com). Int. J. Cancer: 119, 2148–2154 (2006) ' 2006 Wiley-Liss, Inc. Publication of the International Union Against Cancer