Background: Fibrous Dysplasia (FD) is a rare bone disease caused by a somatic mosaic mutation in the GNAS gene, leading to overproduction of FGF23 by metabolically active lesions in bone. This mutation can induce anatomical and/or functional disorders in other organs but their association with plasma FGF23 levels remains unknown. A marker of metabolic activity and response to treatment for FD, measurable in clinical routine, is missing. The objectives of this health-related data reuse analysis were to explore the link between FGF23 levels and FD severity, as well as to determine whether there was an association between FGF23 variation and changes in pain symptomatology, especially in patients under bone-targeted treatment such as biphosphonates.
Methods: 24 patients with any form of FD and at least one FGF23 measure were included. We retrospectively analyzed clinical, radiological and biologic data including plasma FGF23 values, blood and urine biological profiles, pain assessment, bone-targeted therapies, and radiological imaging. The first measured value of FGF23 per patient was named FGF23_t0. We separated patients into two groups: those considered with normal FGF23_t0 (< 50 pg/mL) and those considered with an elevated FGF23_t0 (> 50 pg/mL). We compared the parameters between these two groups with the above-mentioned parameters. Univariable analyses were conducted to examine these associations, by using Stata version 14.2 (StataCorp, College Station, Texas, USA) for Windows.
Results: We found a significant association between the group with high FGF23_t0 (> 50 pg/mL) and the polyostotic form of the disease (p = 0.04). No association was established between the evolution of pain symptomatology under bone-targeted therapy, and changes in plasma FGF23 levels. We found no association between involvement of other organs and FGF23_t0 values greater or less than 50 pg/mL.
Conclusion: We recommend the systematic FGF23 measurement at the beginning of follow- up in FD patients, as our study confirms its use as an indicator of the number of bone lesions. New studies are needed to explore the effectiveness of its measurement in clinical routine as a marker of response to therapy.