Delayed Denosumab Injections and Fracture Risk among Patients with Osteoporosis : A Population-Based Cohort Study.
Houchen Lyu,Kazuki Yoshida,Sizheng S. Zhao,Jie Wei,Chao Zeng,Sara K. Tedeschi,Benjamin Z. Leder,Guanghua Lei,Peifu Tang,Daniel H. Solomon
DOI: https://doi.org/10.1136/annrheumdis-2020-eular.5942
IF: 27.973
2020-01-01
Annals of the Rheumatic Diseases
Abstract:Background:Denosumab is effective for osteoporosis, but discontinuation leads to rapid reversal of its therapeutic effect[1].Objectives:To estimate the risk for fracture among users of denosumab who delayed subsequent dosages compared with users who received dosages on time.Methods:Population-based cohort study. We included patients aged over 45 years who initiated denosumab for osteoporosis from UK THIN database, 2010 to 2019. Observational data were used to “emulate a hypothetical trial”[2, 3] with three dosing intervals: subsequent denosumab injection 24-28 weeks after prior dose (“on time”), delay by 4-16 weeks (“short delay”), and delay by over 16 weeks (“long delay”). The primary outcome was a composite of all fracture types. Secondary outcomes included major osteoporotic fracture, vertebral fracture, and hip fracture.Results:The rate of composite fracture per 1000 person-years was 58.9 for on-time, 61.7 for short delay, and 85.4 for long delay of subsequent denosumab injections. Compared to on-time injections, short delay had a hazard ratio (HR) for composite fracture 1.03 (95% CI 0.63-1.69) and long delay HR 1.44 (95% CI 0.96-2.17; p for trend 0.093). For major osteoporotic fractures, short delay had an HR 0.94 (95% CI 0.57-1.55) and long delay an HR of 1.69 (95% CI 1.01-2.83; p for trend 0.056). For vertebral fractures, short delay had an HR 1.48 (95% CI 0.58-3.79) and long delay 3.91 (95% CI 1.62-9.45; p for trend 0.005).Conclusion:While delayed subsequent denosumab dosages over 16 weeks was associated with an increased risk of vertebral and major osteoporotic fracture compared to no delay, composite fracture risk was not increased with longer delays.References:[1]Cummings SR, Ferrari S, Eastell R, et al. Vertebral Fractures After Discontinuation of Denosumab: A Post Hoc Analysis of the Randomized Placebo-Controlled FREEDOM Trial and Its Extension. J Bone Miner Res, 2017.[2]Hernán MA. How to estimate the effect of treatment duration on survival outcomes using observational data. BMJ 2018.[3]Hernán MA, Robins JM. Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available. Am J Epidemiol 2016.Table.Rates and Adjusted Hazard Ratios of FractureOn-timeShort delayLong delayP for linear trendComposite FractureRate (per 1000 person-years)5961.785.4-Unadjusted HR (95 %)Ref1.05 (0.62, 1.76)1.45 (0.95, 2.21)0.097Adjusted HR (95% CI)†Ref1.03 (0.63, 1.69)1.44 (0.96, 2.17)0.093Major Osteoporotic FractureRate (per 1000 person-years)34.831.958-Unadjusted HR (95 %)Ref0.92 (0.55, 1.53)1.67 (0.98, 2.84)0.074Adjusted HR (95% CI)†Ref0.94 (0.57, 1.55)1.69 (1.01, 2.83)0.056Vertebral FractureRate (per 1000 person-years)4.97.319.4-Unadjusted HR (95 %)Ref1.47 (0.58, 3.71)3.93 (1.59, 9.72)0.006Adjusted HR (95% CI)†Ref1.48 (0.58, 3.79)3.91 (1.62, 9.45)0.005Hip FractureRate (per 1000 person-years)10.29.618.3-Unadjusted HR (95 %)Ref0.94 (0.43, 2.04)1.78 (0.80, 3.97)0.18Adjusted HR (95% CI)†Ref0.97 (0.44, 2.12)1.75 (0.81, 3.79)0.173†Adjusted model: adjusted by age, sex, baseline CCI index, major osteoporotic fracture, oral BP duration (years), 10-year risk of major osteoporotic fracture, prior denosumab doses.Acknowledgments:We acknowledge Dr. Dani Prieto-Alhambra for kindly providing Read codes.Disclosure of Interests:Houchen Lyu: None declared, Kazuki Yoshida: None declared, Sizheng Steven Zhao: None declared, Xabier García-Albéniz: None declared, Jie Wei: None declared, Chao Zeng: None declared, Sara Tedeschi: None declared, Benjamin Leder Grant/research support from: Research funding from Amgen, Guanghua Lei: None declared, Peifu Tang: None declared, Daniel Solomon Grant/research support from: Funding from Abbvie and Amgen unrelated to this work