Association of Fat Body Mass With Vertebral Fractures in Postmenopausal Women With Early Breast Cancer Undergoing Adjuvant Aromatase Inhibitor Therapy

IMPORTANCE Aromatase inhibitors induce a profound depletion in serum estrogen levels. Postmenopausal obese women receiving aromatase inhibitor therapy may be at increased risk of bone fractures owing to the detrimental association of adiposity with bone quality and the loss of the protective effect of estrogens on bone mineral density. OBJECTIVE To determine whether fat body mass (FBM), as measured by dual-energy x-ray absorptiometry, is associated with vertebral fracture prevalence in postmenopausal women undergoing adjuvant aromatase inhibitor therapy for breast cancer. DESIGN, SETTING, AND PARTICIPANTS In this single-center, cross-sectional study, 556 postmenopausal women with early-stage breast cancer were consecutively enrolled from October 15, 2013, to June 30, 2018, and stratified according to whether they were aromatase inhibitor-naive or aromatase inhibitor-treated for at least 2 years. The database was locked on December 31, 2018, and data analysis was completed on February 28, 2019. Eligible patients in both groups had normal renal function, no metabolic diseases, and no previous or current treatment with antiosteoporotic drugs or glucocorticoids. Previous chemotherapy, but not tamoxifen, was permitted. Data were gathered once, at baseline. MAIN OUTCOMES AND MEASURES Vertebral fracture prevalence associated with FBM in aromatase inhibitor-naive and aromatase inhibitor-treated patients. RESULTS Of the 556 women enrolled, the mean age was 63.0 years (95% CI, 62.2-63.8 years). The 195 aromatase inhibitor-treated patients were older than the 361 aromatase inhibitor-naive patients (mean age, 66.1 vs 61.3 years; P<.001), had a higher body mass index (mean, 26.4 vs 25.3; P=.009), were less likely to engage in physical activity (65.3% vs 73.7%; P=.03), and were less likely to consume alcoholic beverages (68.4% vs 80.9%; P=.001). Among the aromatase inhibitor-naive patients, the vertebral fracture prevalence was higher in the subgroup with FBM below the median value than in those with high FBM, but the difference was not statistically significant (19.2% vs 13.3%; P=.13). Conversely, the proportion of vertebral fractures in the aromatase inhibitor-treated group was 20.0% in patients with low FBM vs 33.3% in patients with high FBM (P=.04). An opposite trend in the association of FBM with vertebral fracture prevalence according to aromatase inhibitor group was shown by multivariable analysis in the propensity score-matched sample: odds ratio, 0.38 (95% CI, 0.12-1.19) and 1.94 (95% CI, 0.67-5.64) in the aromatase inhibitor-naive and aromatase inhibitor-treated groups, respectively (odds ratio for the interaction, 5.77 [ 95% CI, 1.08-30.81]; P for interaction term = .03). CONCLUSIONS AND RELEVANCE Fat body mass may be associated with fragility-related fractures in patients with breast cancer who undergo aromatase inhibitor therapy. If these data are confirmed, obesity could be included in the algorithm for assessing fracture risk and selecting patients to receive bone resorption inhibitors.