1708PAnalysis of bone events in patients with aromatase inhibitors (AI)

Background: Nowadays breast cancer (BC) may have high cure rate, so is important how to prevent or resolve adjuvant treatments adverse effects. AI are an essential treatment in patients with early diagnosed estrogen-dependent BC. Our aim is to analyse bone mass loss in BC, focussing on fractures incidence and their risk as well as preventive methods. The latest consensus establishes as intervention criteria a BMD value> -2T Score and 2 or more risk factors. Methods: 114 BC patients and 272 without BC sent to our centre in 2010 to perform densitometry (DXA) were included in our review. Beyond DXA and anthropometry we carried out an extensive clinical questionnaire on co-morbidity, risk factors for osteoporosis, medication, personal and relatives history of fractures. Hospital and primary care records were reviewed in BC patients until May 2017 to determine fractures incidence. Results: There were significant differences in age (BC 59 ± 11 vs 62 ± 11 years, p = 0.01). No differences in weight, height, menarche/menopause or toxic habits. 25% had personal history of fracture (31% control p = 0.05) and 17% relatives history (27% control p = 0.03). 4.3% of corticosteroid intake (16% control p = 0.01). There were no differences in T Lumbar Score (0.97 ± 1.3 vs -0.94 ± -1.2) or in femur neck (-.0.93 ± 1.2 vs-1.1 ± 1) but yes in the fracture risk assessment tool (FRAX) for major fractures with BMDM (BC 7 ± 5% vs 11 ± 6% p0.02) and hip fractures (1.2 ± 2 vs 2.4 ± 4 p 0.02). Patients with AI lost bone mass at two years (BMD 0.96 ± 1.87 to 0.921 ± 0.18 g / cm2 p0.03), without changes in control group. At 7 years follow-up, 8 fractures appeared in patients with AI (3 Colles, 3 vertebral and 2 humerus), 4 patients with Exemestane (E), 2 with Letrozole (L) and 2 with Anastrozole (A) (40% took E, 26% A and 20% L). Only 3 of the 8 had baseline T Lumbar Score