Nutrition and Cancer, 61(4), 466–474 Copyright © 2009, Taylor & Francis Group, LLC ISSN: 0163-5581 print / 1532-7914 online DOI: 10.1080/01635580902725995 A Case-Control Study on Fat-to-Muscle Ratio and Risk of Breast Cancer Alvaro L. Ronco Instituto Universitario CLAEH, Maldonado, Uruguay and Centro Hospitalario Pereira Rossell, Montevideo, Uruguay Heiner Boeing German Institute of Human Nutrition, Potsdam-Rehbr¨ ucke, Germany Eduardo De Stefani Universidad de la Rep ´ ublica, Montevideo, Uruguay Mandy Schulz, Matthias Schulze, and Tobias Pischon German Institute of Human Nutrition, Potsdam-Rehbr¨ ucke, Germany Our objective was to analyze detailed anthropometric char- acterization for risk of breast cancer in Uruguayan women. The design was a case-control study. The setting was Pereira Rossell Women’s Hospital, Montevideo, Uruguay. Subjects were 343 in- cident breast cancer cases and 1,042 frequency-matched healthy controls who were interviewed on menstrual and reproductive story; and a series of skin folds, circumferences, and diameters were measured to calculate fat and muscle fractions and the de- rived fat-to-muscle ratio (FMR). Odds ratio (ORs) coefficients were taken as estimates of relative risk derived from unconditional logis- tic regression. Muscle fraction was negatively associated with risk [OR for highest quartile = 0.23, 95% confidence interval (CI) = 0.15–0.34], fat fraction was positively associated (OR = 3.90, 95% CI = 2.62–5.80), and FMR was positively associated (OR = 4.45, 95% CI = 2.99–6.62). Stratified analyses by body mass index levels also showed risk increases for the highest tertiles of FMR, always displaying significant linear trends. Since increases of risk were found in overweight and in normal weight women, results suggest that fractions and amount of muscle and fat components might be risk factors for breast cancer on the basis of currently exist- ing metabolic and immune interrelationships between adipose and muscular tissue given by glutamine, exercise-derived myokines, and other cytokines produced by these tissues. INTRODUCTION Anthropometric measures are considered to be linked with risk of breast cancer (BC). In the last few years, the association Submitted 7 September, 2007; accepted in final form 8 July 2008. Address correspondence to Alvaro L. Ronco, Convenci´ on 1490 dep. 202, 11100 Montevideo, Uruguay. Phone and fax: (+5982) 901-5343. E-mail: alronco@gmail.com between body mass index (BMI) and BC has been systemati- cally examined in expert evaluations (1,2). The available data have shown a contrast: Heavier women have been found to be at increased risk of postmenopausal BC in most studies, whereas BMI is inversely associated with the risk of cancer among pre- menopausal (3–5) women. Recently, high weight and body fat in elder women were reported as an independent risk factor (6). Also, fat-free mass has been found positively associated with the risk of BC in postmenopausal women (7). Absence of association in premenopausal women has been described for certain anthropometric measures (body size, BMI, fat distribution) in some populations such as Chinese (8,9), Japanese (10), or African American women (11), different from what has been consistently described in the Western and Cau- casian societies. Recently, waist-to-hip ratio was also associated with an increase of risk in premenopausal Nigerian (12) and Asian American women (13). Muscularity, or the proportion of adipose tissue-free body mass such as skeletal muscle, showed no significant differences between Asian, Caucasian, and Hispanic elder subjects. In con- trast, African American subjects had a significantly greater mus- cularity than subjects in the other 3 groups (14). Furthermore, some differences of body mass index between hospitalized women and healthy ones enrolled in a screening program have been reported—both being selected as a control group (15). According to studies that have thoroughly addressed method- ological issues regarding the assessment of anthropometric mea- sures (16,17), studies on body composition have mainly been focused on the estimation of body fat. Nevertheless, there is interest in muscle mass due to its relationship with numerous aspects of human health, for example, protein reserve of the 466