Clinical Outcomes of Denosumab in Osteoporosis

Clinical outcomes of denosumab in postmenopausal osteoporosis (vertebral, nonvertebral, and hip fractures)

Date: Jun 4, 4:44 PM

Reviewed denosumab fracture outcome evidence and refined the search plan.

The pivotal evidence comes from the FREEDOM trial and its 7-year open-label extension, which together follow the same cohort out to 10 years of denosumab exposure.

The 3-year placebo-controlled phase (FREEDOM, n=7,808)

Postmenopausal women aged 60–90 with a lumbar spine or total hip T-score between −2.5 and −4.0 were randomized to denosumab 60 mg SC every 6 months or placebo for 36 months. Relative to placebo, denosumab cut new radiographic vertebral fractures from 7.2% to 2.3% (a 68% relative reduction), hip fractures from 1.2% to 0.7% (40% relative reduction, HR 0.60, 95% CI 0.37–0.97), and nonvertebral fractures from 8.0% to 6.5% (20% relative reduction, HR 0.80, 95% CI 0.67–0.95)Cummings et al. 2009. The vertebral and hip effects are the strongest signals; the nonvertebral effect, while statistically significant, is more modest. Denosumab given subcutaneously twice yearly for 36 months was associated with reductions in vertebral, nonvertebral, and hip fractures in women with osteoporosis.Cummings et al. 2009

Higher-risk subgroups

Post-hoc FREEDOM analyses confirmed the effect held in women at highest baseline risk: vertebral fracture incidence in those with multiple/severe prevalent vertebral fractures fell from 16.6% to 7.5%, and hip fracture incidence in women ≥75 fell from 2.3% to 0.9%Boonen et al. 2011. McCloskey et al.'s FRAX-stratified analysis showed denosumab's benefit was largest in women with moderate-to-high baseline FRAX scoresMcCloskey et al. 2012.

Long-term extension (years 4–10)

Of 5,928 FREEDOM completers, 4,550 enrolled in the open-label extension. Yearly vertebral fracture incidence stayed between 0.90% and 1.86% and nonvertebral between 0.84% and 2.55% — similar to the active-treatment rates in the first 3 years and well below modeled virtual-placebo ratesBone et al. 2017. Crucially, nonvertebral fracture rates kept falling with continued treatment: 2.15/100 subject-years in years 1–3 versus 1.53 in years 4–7 (rate ratio 0.72, p<0.001), and 1.44 in years 4–10 among long-term recipients (rate ratio 0.74, p=0.008)Ferrari et al. 2019. This continued reduction beyond year 3 is unusual for antiresorptives and tracks the non-plateauing BMD gains (lumbar spine +21.7%, total hip +9.2% from baseline at 10 yearsBone et al. 2017).

Safety over 10 years

Adverse-event rates were stable or declined. Across the entire extension, only one atypical femoral fracture occurred in each of the long-term and crossover groups, and 13 total cases of osteonecrosis of the jaw were reported (7 long-term, 6 crossover)Bone et al. 2017. These are rare but real and accumulate with exposure.

Caveats worth flagging

• The FREEDOM nonvertebral effect (HR 0.80) is modest by comparison to vertebral, and the hip effect, while clinically important, rests on a small absolute number of events (CI extends close to 1.0)Cummings et al. 2009.
• Discontinuation rebound — rapid loss of BMD and a documented spike in (especially multiple) vertebral fractures after stopping denosumab — is the major clinical management issue and shapes current guidance to transition patients to a bisphosphonate rather than simply stop. The FREEDOM dataset itself flagged the off-treatment fracture signal earlyRizzo & Tørring 2012, but the abstracts I pulled don't quantify it in detail; that's a worthwhile follow-up if you want the specifics.