1408P_PR - A study of body fat composition, derived from DXA-scans, in association with cancer incidence in postmenopausal women.

Date 10 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Aetiology, Epidemiology, Screening and Prevention
Basic Scientific Principles
Presenter Line Staunstrup
Citation Annals of Oncology (2017) 28 (suppl_5): v605-v649. 10.1093/annonc/mdx440
Authors L.M. Staunstrup1, J.F. Christensen2, J. Blair1, H.B. Nielsen1, C. Christiansen3, C.L. Bager1
  • 1Proscion, ProScion, 2730 - Herlev/DK
  • 2Centre For Physical Activity Reasearch, Rigshospitalet, 2100 - København/DK
  • 3Nordic Bioscience, Nordic Bioscience, 2730 - Herlev/DK



Obesity is a known cancer risk factor, but epidemiological studies have mainly evaluated the link between obesity and cancer by non-specific assessment, e.g. weight, body mass index (BMI) and/or waist circumference. In women, it is known that menopause initiates a shift of body fat toward higher level of abdominal adiposity, which may mediate obesity-related cancer risk. Dual-energy X-ray absorptiometry (DXA) scan can determine whole-body and regional composition of lean and fat mass, but few studies have reported DXA scan-based body composition as predictors of cancer incidence. Accordingly, the aim of this study was to explore the association between obesity, evaluated by DXA scans, and cancer incidence in a cohort of postmenopausal women.


From 1999-2001, 5,858 women (age 71±6.5) were included in the Prospective Epidemiologic Risk Factor study which involved collection of medical and demographic background information. DXA-scans were collected at time of enrolment. Cancer diagnoses and information on death causes were obtained from different Danish registries ultimo 2012. DXA-scan values were transformed into z-scores and divided into high/low categories. Cox proportional hazard regression models were used to examine the association between body fat distribution and the risk of cancer incidence, adjusted for standard risk factors including BMI.


There were 801 cancer events. In multivariable analysis, trunk fat, but not BMI nor fat percentage, was an independent predictor of cancer incidence (HR: 1.34; 95%, CI: 1.15 to 1.57; p 


Trunk fat is associated with cancer incidence after adjustment for standard risk factors. These results support previous findings and show that information on fat distribution is important when determining cancer risk, whereas BMI and whole-body fat percentage may not be adequate.

Clinical trial identification

Legal entity responsible for the study



The Danish Research Foundation


L.M. Staunstrup, J. Blair, H.B. Nielsen, C.L. Bager: Employed at ProScion A/S. C. Christiansen: Stock owner of ProScion A/S. All authors have declared no conflicts of interest.