Kinetic reconstruction reveals time-dependent effects of romosozumab on bone formation and osteoblast function in vertebral cancellous and cortical bone in cynomolgus monkeys

Romosozumab, a humanized monoclonal sclerostin antibody under development for the treatment of osteoporosis, has a unique mechanism of action on bone increasing bone formation and decreasing bone resorption. The effects on bone formation are transient, eliciting a rapid increase in bone formation that attenuates with continued treatment. Although bone formation attenuates, bone mineral density (BMD) continues to increase. To explore potential tissue-level mechanisms that could contribute to a progressive increase in spine BMD, we used kinetic reconstruction techniques to examine the effects of romosozumab on modeling and remodeling units in vertebral cancellous bone from adult cynomolgus monkeys administered romosozumab for 10 and 28 weeks. The 10-week study duration captured a period of high modeling-based bone formation, and the 28-week study duration followed the self-regulation or attenuation of bone formation in cancellous bone that occurs with longterm treatment. Sequential fluorochrome labels applied for the kinetic reconstruction were also used to evaluate treatment effects on osteoblast function as early as 3 weeks, and on bone formation and bone accrual in the vertebral cortex over 28 weeks. Kinetic reconstruction of remodeling and modeling formation sites in vertebral cancellous bone revealed that romosozumab effected significant transient increases in mineral apposition rate in remodeling sites at week 3 that was not sustained with continued treatment. However, romosozumab treatment caused sustained improvement in fractional labeling of osteoid, an index of osteoblast efficiency, at remodeling formative sites at both weeks 10 and 28 that was the major contributor to significant increases in final wall thickness (W.Th) of remodeling packets. Remodeling W.Th matched the final W.Th of modeling packets at week 10. At both weeks 10 and 28, romosozumab significantly decreased eroded surface (ES/BS). At week 28, romosozumab also significantly reduced resorption period (Rs.P) and final resorption depth (Rs.De). The reduced final Rs.De combined with the increased W.Th resulted in a significant increase in bone balance (BB) at the level of the remodeling unit. Assessment of bone formation on the vertebral periosteal and endocortical surfaces following 28 weeks of treatment revealed that romosozumab significantly increased bone formation on these surfaces, which had attenuated by week 28, resulting in significant increases in new periosteal and endocortical bone by week 28. These data suggest that multiple factors potentially contribute to the increase in spine BMD with romosozumab treatment. In the early period of treatment, increased modeling-based bone formation, increased W.Th at remodeling sites, a decrease in remodeling space secondary to decreased ES/BS in vertebral cancellous bone, and increased periosteal and endocortical bone formation in the vertebral cortex contribute to the early increase in spine BMD. Following the self-regulation of bone formation when modeling-based bone formation has attenuated, a decrease in remodeling space secondary to reduced ES/BS and a positive BB secondary to decreased final Rs.De and increased W.Th contribute to the progressive increase in spine BMD with long-term treatment. (C) 2017 Amgen Inc. Published by Elsevier Inc.