Quantitative brain 18F-FDG PET/CT analysis in seronegative autoimmune encephalitis
Samantha Roman1, Mohammad Sadaghiani2, Luisa Diaz-Arias1, Marion Le Marechal1, Arun Venkatesan1, John Probasco1
1The Johns Hopkins Hospital, 2Radiology, The Johns Hopkins Hospital
Objective:
To determine if PET/CT is abnormal in less differentiated forms of autoimmune encephalitis (AE), including seronegative AE and steroid-responsive encephalopathy with associated thyroiditis (SREAT), and whether specific patterns of dysmetabolism exist.
Background:

Brain 18F-FDG PET/CT (PET) has been established as a useful tool in evaluating AE. Specific patterns of dysmetabolism have been identified in patients with anti-N-methyl-D-aspartate (NMDA) and anti-leucine-rich-glioma-inactivated-1 (LGI-1) AE, and preliminary evidence suggests that PET dysmetabolism may correlate with clinical functional status.

Design/Methods:

We conducted a retrospective cross-sectional analysis of PET scans in people with seronegative AE and SREAT. Utilizing NeuroQ™ software, the Z-scores of 47 brain regions were calculated compared to healthy controls, displayed in heat maps, and visually and statistically compared between study groups and to previous data from NMDA and LGI1 cohorts.

Results:

Sixteen seronegative AE and 4 SREAT patients were identified. All PET scans demonstrated abnormal FDG uptake, while MRI, EEG and CSF were abnormal less frequently in both groups. Both seronegative and SREAT patients demonstrated hypometabolism in the medial frontal, inferior frontal and parieto-temporal regions, and hypermetabolism in the medial temporal lobes. Regions of dysmetabolism were significantly different from NMDA and LGI1 patient data, an effect driven by greater severity of dysmetabolism in the NMDA and LGI1 patients. Follow up scans for 3 seronegative AE patients suggest the degree of dysmetabolism on PET correlates with functional status.

Conclusions:

PET is a useful tool in the evaluation of autoimmune encephalitis. SREAT and seronegative AE represent less differentiated forms of AE.  Though frequently abnormal, as anticipated, the pattern of dysmetabolism in SREAT and seronegative AE were less distinct than in NMDA and LGI-1. The pattern of dysmetabolism observed here in SREAT patients was not markedly distinct from seronegative AE patients, potentially supporting consideration of SREAT as a type of seronegative AE rather than a distinct disease entity.

10.1212/WNL.0000000000202401