Purpose: Denosumab (an anti RANKL antibody) is known to be associated with an increased risk for osteonecrosis of the jaw (ONJ). Due to the variety of clinical presentation, many ONJ definitions are used. Evaluation of ONJ's frequency during phase III randomized controlled trials (RCTs) is crucial to assess benefit-risk ratio. We verified that phase III RCTs involving denosumab reported the definition of ONJ used.
Methods: We systematically searched in Central, Medline, Cochrane, and Scopus, until 31 August 2015. We included original phase III RCTs, involving denosumab. Post hoc analysis and trial extension were excluded. Articles that did not mention ONJ in their methods or results were excluded. The primary outcome was the prevalence of a complete definition of ONJ. When no definition was provided, ONJ adjudication process was analyzed.
Results: Of 313 articles found, 13 RCTs were included. A definition of ONJ was detailed in two RCTs (15%). For the remaining 11 RCTs, adjudication process was mentioned for nine. In those processes, "blinded," "expert," and "independent" were the most used words.
Conclusion: Most of the published phase III RCTs involving denosumab did not specify the definition of ONJ used to adjudicate events in the study. Instead of definition, non-scientific and non-reproducible expressions were used. Because the chosen definition could impact the ONJ estimated frequency, it should be mandatory to give the precise definition used in each RCT publication involving denosumab.
Keywords: Denosumab; Osteonecrosis of the jaw; Pharmacovigilance; Safety.