Abstract 650: Association between obesity, body composition and risk of female thyroid cancer

Abstract

Abstract Obesity increases the risk of many cancers, including endometrial, colorectal, kidney, and postmenopausal breast cancer. Some reported a positive association between obesity measured in body mass index (BMI, kg/m2) and risk of thyroid cancer, but others observed no association. We aimed to examine the association between various obesity and body composition measures and the risk of female thyroid cancer in Korea. A total of 158 incident papillary thyroid cancer and 144 non-cancer controls from one university hospital in Seoul were interviewed on their life-style factors between April, 2006 and December, 2007. Controls included patients with incidental benign thyroid nodules, goiter, and other minor surgical conditions. Weight, height, and BMI were measured. Among the total participants, 93 patients and 102 controls agreed to be measured for their body composition using In-Body 3.0 (BioSpace). Diagnostic procedural thyroid stimulating hormones (TSH) levels were abstracted from the clinical records of 155 cases and 138 controls. Odds ratios (ORs) and 95% confidence intervals (CIs) of thyroid cancer were calculated using logistic regression adjusted for age, physical exercise (total MET value), smoking, alcohol intake amount, menopausal status, and serum TSH levels. The level of obesity was not high in this population. The range of BMI was 16.6 - 36.8 kg/m2, with the mean of 23.4 ± 3.0 kg/m2. There were no differences in means of BMI, weight, height, and WC between cases and controls. The proportion of participants with BMI >= 27.5 kg/m2 was 9.5 % in cases and 6.9% in controls. The OR of thyroid cancer for BMI>=27.5 kg/m2 compared to 20<=BMI<25 was 2.67 (95%CI= 0.89- 6.83, ptrend = 0.01). When we separated body composition into lean mass and fat mass, lean mass increased (q5 vs. q1, OR = 4.05, 95%CI= 1.48 - 11.10, ptrend <0.001) while fat mass decreased (q5 vs. q1, OR = 0.32, 95%CI= 0.11 - 0.98, ptrend = 0.01) the risk of thyroid cancer. In correlation analysis, BMI was more strongly correlated with fat mass (r=0.80) than with lean mass (r=0.46). The association between BMI and risk of thyroid cancer seemed slightly stronger among premenopausal than postmenopausal women and non-alcohol drinkers than alcohol drinkers. However, the increased risk by lean mass and decreased risk by fat mass were stronger among postmenopausal than premenopausal women and alcohol drinkers than non-drinkers. In conclusion, lean mass may increase the risk of thyroid cancer. A larger, confirmatory study is warranted. Independent roles of lean and fat body mass should be considered when evaluating the association between obesity and risk of thyroid cancer, at least among a population with low grade of obesity. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 650. doi:1538-7445.AM2012-650