18F-FDG PET detects inflammatory infiltrates in spinal cord experimental autoimmune encephalomyelitis lesions.

UNLABELLED: Multiple sclerosis (MS) is a heterogeneous disease with respect to lesion pathology, course of disease, and treatment response. Imaging modalities are needed that allow better definition of MS lesions in vivo. The aim of this study was to establish an MRI- and PET/CT-based imaging modality and to evaluate approved and promising PET tracers in experimental autoimmune encephalomyelitis (EAE), the animal model of MS.

METHODS: MRI and PET/CT scans were obtained in Dark agouti rats with EAE and healthy control rats. The PET tracers 2-(18)F-fluoro-2-deoxy-d-glucose ((18)F-FDG), 3'-deoxy-3'-(18)F-fluorothymidine ((18)F-FLT), and O-(2-(18)F-fluoro-ethyl)-l-tyrosine ((18)F-FET) were used as surrogate markers of glucose utilization, proliferative activity, and amino acid transport and protein biosynthesis. Immediately after the PET/CT scan, animals were sacrificed for autoradiography, histologic work-up, or RNA expression analysis.

RESULTS: EAE lesions were predominantly located in the spinal cord. With MRI, we were able to detect inflammatory lesions in diseased rats, which correlated well with inflammatory infiltrates as determined by histology. Increased (18)F-FDG uptake was observed in spinal cord lesions in all diseased rats. Further investigation by volume-of-interest analysis demonstrated a correlation between the density of histologically proven cellular infiltrates and the (18)F-FDG signal intensity in PET (F(DF=3) = 5.9, P = 0.001) and autoradiography (F(DF=3) = 4.2, P = 0.008). With (18)F-FET and (18)F-FLT, no definite uptake could be observed on PET scans, whereas autoradiography showed slight radiotracer accumulation in some lesions.

CONCLUSION: Spinal cord inflammatory lesions in the EAE model can be noninvasively visualized in vivo using MRI and (18)F-FDG PET/CT. Localized (18)F-FDG uptake correlates better with a histologically proven abundance of inflammatory cells as a critical marker of disease activity than MRI. Neither (18)F-FET nor (18)F-FLT seems to be a suitable marker for the in vivo detection of inflammatory lesions.