Purpose: To quantify Na18F-PET/CT uptake in relation to clinical and biochemical parameters of fibrous dysplasia (FD) severity and healthy bone (HB) metabolism. Secondary aims: comparing normalization for volume of distribution and determining reproducibility of Na18F-PET/CT uptake parameters in HB and FD. Relating Na18F uptake to skeletal burden score (SBS), bisphosphonate therapy and pain measured by Brief Pain Inventory (BPI).
Methods: In a prospective cohort study (n = 20), Na18F-PET/CT parameters of HB and FD were assessed by two independent readers to determine the cutoff defining increased bone uptake, optimized normalization, and interobserver agreement (ICC) and were related to SBS, serum biomarkers, medication, and clinical parameters.
Results: Physiological bone standardized uptake value (SUV) was best normalized but displayed large interpatient variation (total range 4.1-13.7 g/mL), with very high interobserver agreement (ICC = 0.964). FD burden defined by patient-specific SUV cutoffs reached near-perfect agreement for SUVpeak (ICC = 0.994) and total lesion fluorination (TLF) (ICC = 0.999). TLF correlated weakly with SBS (R2 = 0.384, p = 0.047). TLF correlated positively with serum alkaline phosphatase (R2 = 0.571, p = 0.004) and procollagen type 1 N-terminal propeptide (R2 = 0.621, p = 0.002), SBS did not (p > 0.06). SBS and TLF both correlated with increased fibroblast growth factor-23 (R2 = 0.596, p = 0.007 and R2 = 0.541, p = 0.015, respectively). TLF was higher in use of bisphosphonates (p = 0.023), SBS was not. Average BPI scores correlated to increased FGF-23 (R2 = 0.535, p = 0.045), work-related BPI scores to higher SBS (R2 = 0.518, p = 0.024), higher TLF (R2 = 0.478, p = 0.036), and higher levels of FGF-23 (R2 = 0.567, p = 0.034).
Conclusions: Individualized Na18F-PET/CT SUV cutoffs reproducibly discriminated HB from FD and were well-normalized. The strong relations of bone formation serum markers with Na18F-PET/CT FD burden measurements suggest clinical relevance over SBS as an adjunct instrument in FD patients. The correlation of both imaging modalities with increased work-related BPI scores also indicates clinical applicability. Moreover, SBS is known to remain stationary irrespective of use of medication, whereas TLF on Na18F-PET/CT was higher in baseline patients using bisphosphonates. This makes Na18F-PET/CT a promising tool to quantitatively measure treatment efficacy in FD.
Keywords: Bisphosphonates; Fibrous dysplasia; Reproducibility; Serum biomarkers; Sodium fluoride PET/CT.