ECTS2013 Oral Communications Treatment of osteoporosis (6 abstracts)
Fracture risk factors during treatment with denosumab
Steven Cummings 1 , Amy Feng 2 , Dennis Black 3 , Rachel Wagman 2 , Matt Austin 2 , Andrea Wang 2 , Mona Walimbe 3 , Lucy Wu 3 , Lily Lui 1 & Eric Vittinghoff 3
1CPMC Reseach Institute, San Francisco Coordinating Center, San Francisco, California, USA; 2Amgen, Inc., Thousand Oaks, California, USA; 3University of California, San Francisco, California, USA.
Background: There are no models for estimating risk of fracture in patients taking treatments for osteoporosis. Knowing a patient’s risk of fracture during treatment may help make future treatment decisions; therefore, the development of on-treatment fracture risk models is needed.
Methods: To assess on-treatment fracture risk, the analysis included subjects who received denosumab (DMAb) 60 mg Q6 every 6 months for at least 1 year in either FREEDOM or its study extension through 6 years, missed no more than one dose, and had complete data on clinical risk factors. Baseline risk factors examined included BMI, BMD T-scores, parental hip fracture status, and sCTX. To test the value of assessments during follow-up, we analyzed time-dependent risk factors updated at each year included age, incident vertebral fracture (VFX) and nonvertebral fracture (NVFX), BMD changes and years of exposure to DMAb. A continuation ratio model was used for new or worsening VFX and a Cox regression model was used for NVFX.
Results: The baseline model for VFX included BMI and baseline spine and total hip (TH) T-scores; the time-dependent model added percent change in TH BMD, history of VFX as well as NVFX, and exposure to DMAb during follow-up. The baseline model for NVFX included BMI, baseline TH T-score, parental hip fracture status, and baseline sCTX; the time-dependent model added history of VFX and NVFX and exposure to DMAb during follow-up. Areas under receiver-operating curves indicated better predictive value for models including on-treatment risk factors compared to just the baseline models (0.66 vs. 0.61 for VFX and 0.62 vs. 0.58 for NVFX).
Conclusions: Risk factors at the start of treatment including BMI and BMD predicted fracture risk on treatment. Accounting for incident fractures and changes in TH BMD during treatment may also improve the fracture risk prediction while on treatment with denosumab.
Article tools
My recent searches
My recently viewed abstracts
Authors
Bone Abstracts
ISSN 2052-1219 (online)
© Bioscientifica 2024 |
Privacy policy |
Cookie settings
BiosciAbstracts
Bioscientifica Abstracts is the gateway to a series of products that provide a permanent, citable record of abstracts for biomedical and life science conferences.
Find out more