Purpose
Previous studies have established associations between body mass index and breast cancer, but fat mass is a more direct measure of the amount of fat tissue in the body than body mass index. This study examined the association between body fat mass, fat-free mass, and other anthropometric measures and breast cancer in postmenopausal women according to use of hormone replacement therapy (HRT).

Methods
From the Danish Diet, Cancer and Health cohort established during 1993–1997, 24,219 postmenopausal women were included who had anthropometric and bioimpedance measurements performed by a laboratory technician at baseline. Information on breast cancer incidence (outcome), other cancer diagnoses, and vital status (censoring variables) through 2016 was obtained from nationwide registers. Cox regression models were used to estimate hazard ratios (HR) and 95% confidence intervals (CIs) while adjusting for known breast cancer risk factors and stratifying by HRT use and running age.

Results
During a total of 431,104 person-years, 1919 women developed breast cancer. Among never-users of HRT, the HR for breast cancer at or after age 65 years was 1.06 (95% CI, 1.03–1.08) per 1 kg/m2 higher body fat mass index (BFMI), and 1.30 (95% CI, 1.14–1.47) per 10% higher body fat percentage. The corresponding HRs for breast cancer before age 65 years were close to unity. The HRs were 1.11 (95% CI, 1.02–1.21) and 1.17 (95% CI, 1.10–1.23) for each 1 kg/m2 increase in fat-free mass index, respectively, for breast cancer below and above age 65 years. Mutual adjustment attenuated the HRs for BFMI and body fat percentage, whereas the HRs for fat-free mass index were largely unaffected. Among ever-users of HRT, there was no statistical significant association between any of the body composition measures and breast cancer incidence in the two age groups.

Conclusions
Among postmenopausal women who never used HRT, BFMI was associated with breast cancer in women aged 65 years or older. Fat-free mass index was found to be more strongly associated with postmenopausal breast cancer incidence than BFMI independently of age in never-users of HRT.