Adults with osteogenesis imperfecta treated with teriparatide followed by zoledronic acid did not have a statistically significant reduction in imaging-confirmed fractures compared with those who received standard care, despite greater gains in bone mineral density, according to findings from the randomized TOPaZ trial published in JAMA.
The open-label, parallel-group trial enrolled 350 adult patients with a clinical diagnosis of osteogenesis imperfecta at 27 referral centers for bone disease in the United Kingdom and Europe. Patients were randomly assigned to receive 20 micrograms of teriparatide daily by subcutaneous injection for 24 months followed by a 5-mg infusion of zoledronic acid or to receive standard care. Bone anabolic agents were prohibited in the standard-care group, but other licensed osteoporosis therapies could be used at investigator discretion.
The comparison was clinically notable because nearly all patients assigned to teriparatide followed by zoledronic acid adhered to treatment, whereas fewer than half of those assigned to standard care received any bone-targeted therapy. In the intervention group, 98.3% of patients adhered to assigned treatment. In the standard-care group, 44.5% received 1 or more bone-targeted medications, most commonly vitamin D supplements and bisphosphonates.
The mean patient age was 43.7 years, and 53.9% of patients were female. Most patients had type I osteogenesis imperfecta, the mildest clinical subtype. Pathogenic variants in COL1A1 or COL1A2 were identified in 86.2% of patients, and quantitative variants were more common than qualitative variants. Mean baseline T scores were in the osteopenic range in both groups, with low bone mass most pronounced at the lumbar spine: -2.18 in the teriparatide plus zoledronic acid group and -2.11 in the standard-care group.
The primary endpoint was the number of patients with imaging-confirmed incident fractures adjudicated by reviewers who were blinded to treatment assignment. The endpoint included new fractures and new or worsening vertebral fractures identified on end-of-study spine radiographs. Secondary endpoints included total fractures, patient-reported fractures not confirmed by imaging, bone mineral density, biochemical markers of bone turnover, health-related quality-of-life measures, and adverse events.
Incident fractures occurred in 65 of 176 patients, or 36.9%, assigned to teriparatide followed by zoledronic acid and 63 of 173 patients, or 36.4%, assigned to standard care. The absolute risk reduction was negative 1.57%, reflecting a small numerical excess of patients with fractures in the intervention group rather than a numerical reduction. The between-group difference was not statistically significant.
Median time to first fracture was 27.1 months in the teriparatide plus zoledronic acid group and 23.9 months in the standard-care group. The difference was not statistically significant.
Secondary fracture analyses also did not show a statistically significant benefit with teriparatide followed by zoledronic acid. Total confirmed fractures numbered 191 in the intervention group and 146 in the standard-care group. This total included incident fractures during follow-up as well as new or worsening vertebral fractures identified at the end of the study. Confirmed vertebral fractures accounted for much of the numerical imbalance, with 125 vertebral fractures in the intervention group and 85 in the standard-care group; this difference was not statistically significant.
Patient-reported fractures that were not confirmed on imaging were far more common than imaging-confirmed fractures and did not differ meaningfully between groups. These were reported by 131 of 176 patients, or 74.4%, in the intervention group and 131 of 173 patients, or 75.5%, in the standard-care group. Patients reported 860 fractures in the teriparatide plus zoledronic acid group and 932 fractures in the standard-care group.
Bone mineral density improved despite the absence of fracture reduction. Lumbar spine bone mineral density was statistically significantly higher in the teriparatide plus zoledronic acid group at 24 months and at the end of the study compared with standard care. Total hip bone mineral density was statistically significantly higher at the end of the study and overall, although not at 24 months. Femoral neck bone mineral density did not differ statistically significantly between groups.
Markers of bone turnover rose during teriparatide treatment. Serum procollagen type 1 N-terminal propeptide and C-terminal telopeptide of type 1 collagen were higher at 12 and 24 months in the teriparatide plus zoledronic acid group, but levels were similar between groups by the end of the study, consistent with the antiresorptive effect of zoledronic acid following teriparatide.
Several patient-reported outcomes favored teriparatide followed by zoledronic acid, including Brief Pain Inventory interference scores and EQ-5D-3L scores, with differences that exceeded thresholds considered clinically important. Pittsburgh Sleep Quality Index scores also favored the intervention, and Health Assessment Questionnaire scores differed at 1 year but not at other time points. SF-36 physical and mental summary scores did not differ between groups. The researchers cautioned that these findings are difficult to interpret because the trial was open label and perceived benefit from injectable treatment may have influenced patient-reported outcomes.
Prespecified analyses did not show evidence that clinical subtype, baseline bone mineral density, sex, molecular diagnosis, recent fracture history, recent bisphosphonate therapy, or age modified the treatment effect on the primary outcome. However, the study population was predominantly composed of patients with type I osteogenesis imperfecta and COL1A1 or COL1A2 variants, limiting generalizability to patients with more severe subtypes or rarer genetic causes.
The researchers noted several limitations, including fewer primary outcome fractures than expected. The trial was designed with 88% power to detect a 25% reduction in the number of patients experiencing a new incident fracture, assuming an annual first-fracture rate of 16%. The observed number of primary outcome fractures was lower than expected, which may have limited the ability to detect a treatment effect, although the researchers noted that there was no trend toward fracture prevention with teriparatide followed by zoledronic acid. Other limitations included no data collection on falls, the possibility that smaller gains in total hip bone mineral density were insufficient to reduce fractures at cortical bone sites, and limited generalizability to treatment-naive patients because about 70% of participants had previously received bisphosphonates at some point in life.
The researchers said the findings may shift attention away from bone density alone as a therapeutic target in osteogenesis imperfecta. They noted that a prior baseline analysis of TOPaZ found similar fracture rates in the 2 years before enrollment among patients with normal bone mineral density, osteopenia, and osteoporosis, supporting the view that bone density does not correlate well with fracture risk in this disease.
“These results signal the need for a paradigm shift in the medical management of osteogenesis imperfecta to move away from strategies that increase bone density to those that improve bone quality,” wrote lead study author Jannie Dahl Hald, PhD, of Aarhus University Hospital, and colleagues.
Disclosures can be found in the published investigation.
Source: JAMA
