Thesis

167 Quantifying skeletal burden in fibrous dysplasia using Sodium Fluoride PET/CT There was no clinically relevant correlation between Na18F-PET parameters or SBS and any of the subdomains of the SF-36 (n=10). Discussion For this cohort of FD patients, we showed that individualized Na[18F]F-PET/CT SUV cut-offs to discriminate normal bone from FD were highly reproducible among two independent readers and were well-normalized for volume of distribution. The methods from the publications by Rohren, et al. and Lapa, et al. to define pathological bone were not reproducible in our cohort (Osirix, version 9.5) [6, 7]. Moreover, both these studies evaluated bone metastases of prostate and breast cancer. Their results for determining pathological bone differ fundamentally from our study as we could not confirm that a single predefined Na18F uptake level (in this case SUV of 10 g/mL) was able to discriminate pathological fromphysiological uptake in our cohort because of the large range in individual physiological Na18F-uptake [6, 7]. Although the total of factors determining interpatient variation in normal bone uptake parameters on Na[18F]F-PET/CT remains unresolved, we speculate that this may be due to several physiological factors (i.e. basal bone turnover determined by amongst others age, sex and hormonal status) as well as pathological or iatrogenic factors, with a potentially substantial role for increased or decreased bone turnover due to antiresorptive therapy (both bisphosphonates and denosumab). Hydration shortly before scanning and kidney function were probably of little impact as the Na18F-uptake rate would remain constant. Considerable variation in incubation time in our cohort showed to be unrelated to the cut-offs for normal bone. Our results advocate a personalized cut-off for normal bone in Na18F-studies. For future studies and eventual clinical practice, we therefore advise to at least validate whether personalized or a single cut-off will suffice for a cohort or population. In the study of Papadakis, et al., exclusion of non FD-bone pathology was manually performed by the author. They excluded voxels in only 3 patients of their study: sacroiliitis in one patient, one dental infection and growth-plate activity in one child [11]. As 14 out of 15 of their patients were adults, wewould have expected a considerable number of joints affected with osteoarthritis (OA), as we experienced in our study. Uptake in FD on Na[18F]F-PET/CT in our experience largely overlaps with uptake seen in OA due to the benign nature of FD and known sensitivity of Na[18F]F-PET/CT for OA: thresholds only will therefore most likely not be able discriminate OA from FD. Subsequently, in their method using exclusion of everything other than FD, also OA needs to be systematically be excluded from FD measurements, which will be 6

RkJQdWJsaXNoZXIy MjY0ODMw