Objective To use in vivo multimodal [18F]FDG (fluoro-D-glucose, tracer for inflammation) and [18F]sodium fluoride (bone tracer) positron emission tomography/computed tomography (PET-CT) imaging for the monitoring of systemic inflammatory and bone remodelling processes as well as colocalized bone destructions before and after TNF blockade in human tumour necrosis factor transgenic (hTNFtg) mice, an established mouse model of chronic inflammatory, erosive polyarthritis.
Methods 8 week-old hTNFtg mice were treated with anti-TNF antibodies (Infliximab, i.p., 3x times per week, 10 mg/kg body weight) for 4 weeks. Before and after the treatment period isofluran-anaesthetised mice received [18F]FDG or [18F]sodium fluoride (~25 MBq) static PET scans (45 min post injection) followed by whole-body and high resolution leg CT scans (800kV, 500µA, 800 ms, 360 projections) using an Inveon PET/CT/SPECT multimodality system (Siemens Medical Solutions). PET reconstructions were conducted with OSEM3D/MAP, FBP algorithm. Standard uptake values (SUV) were calculated using PMOD software. Joints were fixed in 7% formaldehyde and stored in 70% ethanol for ex vivo high-resolution µCT imaging (Scanco µCT35). Histopathological features were analysed in H and E, TRAP and TB stained joint sections. Tissue regeneration processes were addressed after short-term TNF blockade (2 weeks) in cryo and metacrylate-embedded sections.
Results Before therapeutic intervention, we observed an increased accumulation of [18F]FDG but no changes of [18F]sodium fluoride uptake in various joints of hTNFtg mice including knees, ankles and shoulders compared to wt littermates indicating ongoing inflammatory processes. After four weeks, placebo-treated hTNFtg animals showed significantly increased [18F]FDG SUVs and severe progressive bone destructions in those joints. In contrast, TNF-blockade led to a significant decrease in [18F]FDG SUVs supposing complete resolution of inflammatory processes in those individuals. Comparison of repeated in vivoCT images demonstrated a significant reduction and reversal of bone destructions indicating bone regeneration upon TNF blockade. However, we found no changes in [18F]Fluoride SUVs. Short-term TNF blockade demonstrated fibrocartilaginous tissue remodelling and endochondral ossification processes at sites of structural joint damage.
Conclusion In vivo small animal multimodal [18F]FDG, but not [18F]sodium fluoride, PET-CT imaging provides an objective, non-invasive imaging tool for the (I) monitoring of reversibility of ongoing inflammatory processes and (II) regeneration of bone damage during therapeutic intervention in TNF-driven experimental arthritis.