Background: General obesity, as measured using body mass index (BMI), has been linked to the risk of multiple cancers, but limited studies have been conducted to evaluate associations of visceral obesity, fat mass and body fat percentage with cancer risk. The aim of this study was to analyze different measurements and aspect of obesity as potential risk factors for common cancers.

Methods: The UK Biobank cohort was designed to recruit middle and old aged participants in UK between 2006 and 2010, and follow them up via linkage to health records in National Health Service. Cox proportional hazard models were fitted to prospectively investigate associations between common site-specific cancers with BMI, waist-to-hip ratio (WHR), body fat percentage (BF%), whole body fat mass (WBFM), and trunk fat percentage (TF%), adjusting for potential confounders. Detailed measurements were obtained at baseline survey conducted in UK Biobank cohort, among approximately 492,000 participants. BF%, WBFM and TF% were ascertained using bio-impedance measures.

Results: A total of 20,628 incident cancers (except for non-melanoma skin cancer) occurred during 5.6 years of follow-up. Each SD increase in BMI was linearly associated with increased cancer risks of all-cancers combined (HR = 1.05, 95%CI = 1.04-1.07, p < 0.001), esophageal adenocarcinoma (1.43, 1.25-1.63, p < 0.001), colon (1.10, 1.04-1.16, p < 0.001), liver (1.22, 1.06-1.41, p = 0.007), gallbladder (1.36, 1.14-1.64, p = 0.001), pancreas (1.18, 1.07-1.31, p = 0.001), kidney (1.38, 1.27-1.50, p < 0.001), postmenopausal breast (1.10, 1.06-1.14, p < 0.001), and endometrium (1.65, 1.56-1.76, p < 0.001), and with decreased risk of prostate cancer (0.90, 0.87-0.94, p < 0.001). Positive associations for these cancers were also observed with WHR, BF%, WBFM and TF%. The association with WHR (per SD increment) remained statistically significant after further adjustment of BMI using residual method for risks of all cancers combined, esophageal adenocarcinoma, colon, liver, bladder, and kidney. For postmenopausal breast cancer, the positive association remained statistically significant for BF%, WBFM, TF%, but not for WHR after adjusting for BMI; for epithelial ovarian cancer, the positive associations retained for WBFM and TFM, but not for WHR and BF% after adjusting for BMI. For endometrial cancer, prostate cancer, non-Hodgkin lymphoma, multiple myeloma, and leukemia, the negative association remained statistically significant for BF% and TF%, even after adjusting for BMI.

Conclusions: Visceral obesity and specific fat mass measurements, including WHR, BF%, WBFM, and TF% confer additional risk to multiple cancers beyond the level contributed by general obesity measured using BMI. Cancer prevention programs should be designed not only to reduce body weight but also visceral fat and body fat mass.

Citation Format: Ying Liu, Wei Zheng. Prospective investigation of general and visceral obesity, body fat distribution and risk of common cancers using data from the UK Biobank cohort study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5251.