Psychotropic medications and proton pump inhibitors and the risk of fractures in the teriparatide versus risedronate VERO clinical trial

Background: VERO is a fracture endpoint study in women with established osteoporosis that showed reduction in the risks of new vertebral fractures (VFx) and clinical fractures in women randomized to teriparatide compared with risedronate. Patients on psychotropic drugs (hypnotics, benzodiazepines and antidepressants [selective serotonin- and norepinephrine-reuptake inhibitors: SSRIs and SNRIs]) and proton pump inhibitors (PPIs) may be at a higher risk of fractures. We studied the association of exposure to these medications with the risk of fractures in the VERO study cohort, including an assessment of their potential interactions with the assigned clinical trial drugs. Methods: A total of 1360 postmenopausal women with at least 2 moderate or 1 severe VFx and bone mineral density T-score ≤-1.50 were randomized to subcutaneous daily teriparatide (20 μg) or oral weekly risedronate (35 mg) in a double-blind, double-dummy, 2-year trial. In thispost-hoc analysis, multivariable log-binomial and Cox proportional hazards regression models were used to estimate adjusted risk ratios (RR) or hazard ratios (HR) for the exposure to these concomitant medications with the occurrence of incident fractures. We also assessed treatment effect modifications on anti-fracture efficacy driven by the use of these medications. Results: There were 406 (29.9 %), 347 (25.5 %) and 176 (12.9 %) subjects taking PPIs, benzodiazepines/hypnotics, and SSRIs/SNRIs during the study, respectively. For all fracture endpoints, the greater risk reduction of teriparatide versus risedronate did not significantly differ within the categories of psychotropic drugs and PPIs. Multivariable analysis showed that the risk of pooled new and worsened VFx was higher in PPI users than in non-PPI users (RR: 1.57; p = 0.032), regardless of the study treatment. Benzodiazepine/hypnotic drug users showed an increased risk of clinical fractures (HR: 1.71; p = 0.026) and non-vertebral fragility fractures (NVFFx, HR: 1.89; p = 0.017), regardless of the study treatment. Increases in the risk of clinical fractures (HR: 1.93; p = 0.018) and NVFFx (HR: 2.16; p = 0.011) were also observed in SSRI/SNRI users, regardless of the study treatment. Conclusion: In postmenopausal women with severe osteoporosis, the superior anti-fracture efficacy of teriparatide compared with risedronate was consistent regardless of psychotropic or PPI drugs use. Patients taking psychotropic drugs and PPIs showed a higher risk for NVFFx and VFx respectively, compared to those not on these medications.