Effects of romosozumab on bone strength around a pedicle screw as evaluated by biomechanical computed tomography-based virtual stress tests in postmenopausal women

<h2>ABSTRACT</h2><h3>Background Context</h3> Osteoporosis is a significant risk factor for complications after spinal fusion surgery. Romosozumab, a sclerostin inhibitor, is approved for treating osteoporosis in postmenopausal women at high fracture risk. <h3>Purpose</h3> Here we investigated the theoretical effects of romosozumab on changes in vertebral strength around virtually-implanted pedicle screws, using biomechanical computed tomography (BCT). <h3>Study Design/Setting</h3> This retrospective secondary analysis utilized CT scans from two randomized clinical trials of osteoporosis drug treatments in postmenopausal women: a phase 2 trial (12-month romosozumab vs. placebo vs. teriparatide) in women with low bone mineral density (BMD) and a phase 3 trial (12-month romosozumab followed by 12-month alendronate vs. 24-month alendronate alone) in women with osteoporosis. <h3>Patient Sample</h3> Subsets of 82 patients from the phase 2 trial who <h3>received</h3> subcutaneous romosozumab 210 mg/mo, placebo, or subcutaneous teriparatide 20 μg/d and 90 patients from the phase 3 trial were included in respective imaging substudies. Patients from the substudies in the present analysis had to have baseline and ≥ 1 postbaseline BCT measurement at month 12 in the phase 2 trial, and at months 6, 12, and 24 in the phase 3 trial. <h3>Outcome Measures</h3> The primary endpoints were percent changes (follow-up vs. baseline) in shear bone strength and volume of failed tissue compared between treatments; the secondary endpoint was percent change from baseline in periprosthetic BMD. <h3>Methods</h3> For each participant, the baseline CT scan was used to create a finite element model for the L1 vertebra, which was then virtually implanted with pedicle screws and virtually loaded to failure in a "shear pullout" configuration; follow-up CT scans (at 6, 12, or 24 months) were processed in the same way. <h3>Results</h3> For the phase 2 trial (N=79), the mean (95% confidence interval) percent change from baseline to Month 12 in shear bone strength was 24.7% (20.8–28.6%) for romosozumab, which was greater (<i>p</i> < 0.001) than the changes of -2.2% (-5.8–1.5%) for placebo and 14.8% (11.3–18.4%) for teriparatide. For the phase 3 trial (N=79), percent change in shear bone strength was greater (<i>p</i> < 0.001) for romosozumab than alendronate at both Month 6 (21.6% [17.8–25.4%] vs. 6.1% [4.2–8.1%]) and Month 12 (26.3% [22.1–30.6%] vs. 7.3% [5.0–9.6%]), and also at Month 24 after switching at Month 12 from romosozumab to alendronate (25.2% [19.9–30.5%] vs. 5.7% [3.2–8.2%]). Similar trends occurred for volume of failed tissue and periprosthetic BMD. <h3>Conclusions</h3> In BCT-based virtual stress testing of pedicle screws that were virtually implanted in postmenopausal women with low BMD or osteoporosis, shear bone strength, periprosthetic BMD, and amount of failed tissue were significantly improved over time after treatment with romosozumab compared to placebo, teriparatide, and alendronate. Recognizing that these study participants were not candidates for spinal fusion, these findings nevertheless suggest that romosozumab might rapidly and substantially improve the biomechanical integrity of pedicle screw fixation in postmenopausal women with low BMD or osteoporosis.