The study by Bhandari et al. was very well designed and conducted with high-quality data collection. In terms of the primary outcome—median time to radiographic healing—there was no significant difference between the placebo group (n=100) and 9 romosozumab groups (n=293 total, testing 3 different dose levels and 3 different frequencies). Additionally, analysis revealed no differences between placebo and romosozumab groups in median time to clinical healing or in changes in physical function from baseline. (See related OrthoBuzz post about a recent randomized trial investigating romosozumab for hip fractures.)
Kudos to Amgen for funding the trial and for allowing the 66-center, international academic consortium that conducted it to publish the results, warts and all. Such negative findings appropriately inform decisions about which compounds to investigate and about study designs for retesting the same compounds. For example, Bhandari et al. encourage further study of romosozumab in tibial-fracture patients at high risk of poor fracture healing, such as those with diabetes or patients undergoing treatment with corticosteroids.
We are likely to see many such “failures” in the search for pharmacological adjuncts to improve fracture healing, but it seems our orthopaedic community has laid out a clear roadmap for studying this important question further.
Marc Swiontkowski, MD
JBJS Editor-in-Chief