Abstract
BACKGROUND
Fragility fractures are fractures that result from mechanical forces that would not ordinarily result in fracture.
OBJECTIVES
The objectives were to evaluate the clinical effectiveness, safety and cost-effectiveness of non-bisphosphonates {denosumab [Prolia®; Amgen Inc., Thousand Oaks, CA, USA], raloxifene [Evista®; Daiichi Sankyo Company, Ltd, Tokyo, Japan], romosozumab [Evenity®; Union Chimique Belge (UCB) S.A. (Brussels, Belgium) and Amgen Inc.] and teriparatide [Forsteo®; Eli Lilly and Company, Indianapolis, IN, USA]}, compared with each other, bisphosphonates or no treatment, for the prevention of fragility fracture.
DATA SOURCES
For the clinical effectiveness review, nine electronic databases (including MEDLINE, EMBASE and the World Health Organization International Clinical Trials Registry Platform) were searched up to July 2018.
REVIEW METHODS
A systematic review and network meta-analysis of fracture and femoral neck bone mineral density were conducted. A review of published economic analyses was undertaken and a model previously used to evaluate bisphosphonates was adapted. Discrete event simulation was used to estimate lifetime costs and quality-adjusted life-years for a simulated cohort of patients with heterogeneous characteristics. This was done for each non-bisphosphonate treatment, a strategy of no treatment, and the five bisphosphonate treatments previously evaluated. The model was populated with effectiveness evidence from the systematic review and network meta-analysis. All other parameters were estimated from published sources. An NHS and Personal Social Services perspective was taken, and costs and benefits were discounted at 3.5% per annum. Fracture risk was estimated from patient characteristics using the QFracture® (QFracture-2012 open source revision 38, Clinrisk Ltd, Leeds, UK) and FRAX® (web version 3.9, University of Sheffield, Sheffield, UK) tools. The relationship between fracture risk and incremental net monetary benefit was estimated using non-parametric regression. A probabilistic sensitivity analysis and scenario analyses were used to assess uncertainty.
RESULTS
Fifty-two randomised controlled trials of non-bisphosphonates were included in the clinical effectiveness systematic review and an additional 51 randomised controlled trials of bisphosphonates were included in the network meta-analysis. All treatments had beneficial effects compared with placebo for vertebral, non-vertebral and hip fractures, with hazard ratios varying from 0.23 to 0.94, depending on treatment and fracture type. The effects on vertebral fractures and the percentage change in bone mineral density were statistically significant for all treatments. The rate of serious adverse events varied across trials (0-33%), with most between-group differences not being statistically significant for comparisons with placebo/no active treatment, non-bisphosphonates or bisphosphonates. The incremental cost-effectiveness ratios were > £20,000 per quality-adjusted life-year for all non-bisphosphonate interventions compared with no treatment across the range of QFracture and FRAX scores expected in the population eligible for fracture risk assessment. The incremental cost-effectiveness ratio for denosumab may fall below £30,000 per quality-adjusted life-year at very high levels of risk or for high-risk patients with specific characteristics. Raloxifene was dominated by no treatment (resulted in fewer quality-adjusted life-years) in most risk categories.
LIMITATIONS
The incremental cost-effectiveness ratios are uncertain for very high-risk patients.
CONCLUSIONS
Non-bisphosphonates are effective in preventing fragility fractures, but the incremental cost-effectiveness ratios are generally greater than the commonly applied threshold of £20,000-30,000 per quality-adjusted life-year.
STUDY REGISTRATION
This study is registered as PROSPERO CRD42018107651.
FUNDING
This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 24, No. 29. See the NIHR Journals Library website for further project information.
Collapse