Background/Purpose:   A common complication in rheumatoid arthritis (RA) is osteoporosis (OP) with increased incidence of fragility fractures. RA is the only disease specifically included in the World Health Organization fracture prediction algorithm, FRAX^®. The effect of teriparatide (TPTD) on the incidence of clinical vertebral (VERT) and nonvertebral (NV) fragility fractures, bone mineral density (BMD), and safety was examined in the subgroup of OP patients with RA enrolled in the DANCE trial.

Methods: This was a post hoc analysis of patients with RA enrolled in the DANCE open-label, prospective, observational study, treated with TPTD 20 mg/day for ≤2 years (treatment phase), and followed for up to 2 more years (cessation phase). Mixed model repeated measures analysis was used to evaluate BMD changes over 18 months (mo). Incident rates of new clinical VERT or NV fragility fractures (traumatic fractures excluded) after 6-24 mo vs 0-6 mo of TPTD therapy were compared using a Poisson regression model. Time to new NV fragility fracture was compared using Kaplan-Meier analysis. Rates of incident NV fractures were assessed at intervals over treatment and cessation phases vs the first 6-mo treatment (reference) period.

Results: Of 4085 patients who received ≥1 dose of TPTD, 544 had documented RA. Patients with vs without RA at baseline had similar age (mean [SD] age, 68.6 [11.5] vs 67.8 [11.9] years), but significantly more history of prior NV fractures (60.8% vs 55.4%, p=0.017), higher L1-L4 T-scores (-2.21 [1.44] vs -2.51 [1.36], p<0.001), more baseline clinical conditions (3.1 [1.4] vs 1.6 [1.3], p<0.001), and more glucocorticoid use (32.5% vs 7.1%, p<0.001). Mean TPTD exposure was similar (523.0 [303.9] vs 545.4 [288.4] days). Over 18 mo, bone density in the spine, femoral neck, and total hip increased similarly in patients with vs without RA (all p-values>0.1). Incident rates of new VERT with or without back pain or NV fragility fractures decreased in mo 6-24 vs the first 6 mo. There was no significant difference in the decrease in fracture incidence in mo 6-24 vs the first 6 mo in the patients with RA compared with those without, regardless of the type of fracture (all interaction p>0.15). There was no difference in the time to a new NV fragility fracture (log-rank p>0.2) by RA status. NV fracture incidence rates/100 patient years decreased vs reference baseline during treatment and stayed down during cessation phase vs reference baseline (Figure 1). In the overall DANCE study, TPTD was well tolerated and no new significant safety findings were observed.

Conclusion: Patients with rheumatoid arthritis receiving teriparatide showed similar increases in BMD at the spine, femoral neck, and total hip and similar reduction over time in the incidence of NV fractures compared with osteoporosis patients without RA. Compared to baseline, the incidence of NV fractures remained down after drug cessation.