fluorodeoxy- glucose positron emission tomography pattern and prognostic predictors in patients with anti- GABAB receptor encephalitis

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. CNS Neuroscience & Therapeutics published by John Wiley & Sons Ltd. 1Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, Beijing, China 2China National Clinical Research Center for Neurological Diseases, Beijing, China 3Department of Nuclear Medicine, Beijing Tiantan Hospital, Capital Medical University, Beijing, China 4Collaborative Innovation Center for Brain Disorders, Beijing Institute of Brain Disorders, Capital Medical University, Beijing, China


| INTRODUC TI ON
Part of a group of severe but treatable neurological diseases, antigamma-aminobutyric-acid B (GABAB) receptor encephalitis is a novel form of autoimmune encephalitis (AE) associated with antibody to GABAB receptor of cell surface. 1,2 The majority of patients with anti-GABAB receptor encephalitis present with new-onset seizures, cognitive deficits, and mental and behavioral disorders, all with or without the presence of underlying small cell lung cancer (SCLC). 3 Early recognition of anti-GABAB encephalitis is of vital importance because most patients respond well to timely treatment. [4][5][6] Current diagnostic criteria depend highly on positive GABAB antibody in serum or cerebrospinal fluid (CSF), which can lead to false negatives or unavailable results, causing diagnostic difficulties and treatment delay. 7 Thus, it is necessary to consider an additional novel biomarker in early diagnosis and prognostic evaluation for anti-GABAB receptor in encephalitis patients. Although neuroimaging plays a key role in the routine evaluation of neurological diseases, 8 it has received little attention for anti-GABAB receptor encephalitis. Some prior studies have shown that magnetic resonance imaging (MRI) in these patients have mainly presented with hyperintensity signals in medial temporal lobe (MTL), but only 18%-50% patients have abnormal and specific results. 3,9,1018 F-fluoro-2-deoxy-d-glucose positron emission tomography ( 18 F-FDG-PET) is frequently used for whole-body tumor screening, but recently, it has been reported to demonstrate MTL hypermetabolism in patients with anti-GABAB receptor encephalitis, especially when MRI was negative. 11 Nevertheless, only a limited number of isolated cases have suggested that 18 F-FDG-PET might be useful for early diagnosis in subjects with anti-GABAB receptor encephalitis, and there are no current data available to determine the sensitivity of brain 18 F-FDG-PET in encephalitis associated with GABAB receptor antibody. [11][12][13][14] Moreover, the majority of previous studies of 18 F-FDG-PET in anti-GABAB receptor encephalitis have been restricted to qualitative characterization of FDG-PET findings. 15 In addition, to our knowledge, there is no systematically relevant study to evaluate the correlation between PET index and clinical prognosis in patients with anti-GABAB receptor encephalitis.
Overall, the metabolic pattern and prognostic role of 18 F-FDG-PET in patients with anti-GABAB receptor encephalitis are still not well described.
To address these unclear questions, we conducted a semiquantitative study reviewing the 18 F-FDG-PET data of 21 patients with a definite diagnosis of anti-GABAB receptor encephalitis.
One aim of this study was to identify the 18 F-FDG-PET pattern of anti-GABAB receptor encephalitis, especially in those with unremarkable MRI changes. Further, this study also sought to find an imaging prognostic predictor by evaluating the association between metabolic signature and prognosis in anti-GABAB receptor encephalitis.

| Study participants
The study was approved by the ethics committee of the Beijing Tiantan hospital that was affiliated to the Capital Medical University of the People's Republic of China. The study was conducted in accordance with the Declaration of Helsinki, and all patients and controls provided informed consent for the use of their medical records.

| MRI
All MRI examinations were acquired using a 3.0 Tesla Siemens Trio  18 F-FDG-PET was performed according to previously published methods. 16 PET images were acquired using a PET/CT scanner (Elite Discovery, GE HealthCare). All patients fasted for at least 6 h, and fasting blood glucose levels could not exceed 8 mmol/L.

| 18 F-FDG-PET acquisition and statistical parametric mapping (SPM) analysis
No patients received neuroleptic drugs before undergoing FDG-PET. 18 F-FDG was intravenously injected at a dose of 3.7-5.0 MBq/kg within 1 min, and subsequent uptake required that patients be in a quiet resting status for 1 h prior to scanning in a dedicated room after 18 F-FDG injection. First, a low-dose CT scan (120 kV, 60-180 mA/s, slice thickness 3.75 mm) was performed.
The PET scan was subsequently performed, with a whole-body (including brain region) FDG-PET scanning acquired for approximately 30-35 min. The brain imaging data were reconstructed into trans-axial slices with a matrix size of 128 × 128 and a slice thickness of 3.3 mm.
After acquisition of PET images, all data were preprocessed by SPM12 software implanted in a MATLAB 2018a environment (MathWorks Inc.). The pre-processing steps were as follows: first, the PET images were segmented and spatially normalized into a common Montreal Neurological Institute (MNI) atlas anatomical space following a 12-parameter affine transformation and non-linear transformations, yielding images composed of 2 mm × 2 mm × 2 mm voxels. Then, default SPM smoothing was applied using 12-mm Gaussian kernel to increase the signal-tonoise ratio. For single subject analysis, the statistical basic models were performed between individual patient and controls using two-sample t test model with age and gender as the nuisance variables. For groupwise analysis, FDG uptake was compared voxelby-voxel between patients and controls group using a two-sample t-test of SPM. Significant results were viewed at the height threshold (p < 0.001) and corrected for multiple comparisons (familywise error [FWE] corrected or false discovery rate [FDR] corrected, p < 0.05). If significant clusters were not found, the more liberal threshold was considered (p < 0.001, uncorrected; extent threshold, k = 300). 17

| Follow-up and prognosis analysis based on SPM
The follow-up and clinical outcome information was obtained from outpatient visits and telephone interviews with patients or relatives. The modified Rankin Scale (mRS) was used to assess neurological disability at the last follow-up in patients with anti-GABAB receptor encephalitis 3,18,19 ; patients were considered to have a good outcome if mRS score was ≤2 and poor outcome was defined as mRS score >2. In order to assess the potential imaging predictors on 18 F-FDG-PET that might influence the long-term outcome in subjects with anti-GABAB receptor encephalitis, we performed group comparisons of 18 F-FDG-PET data between patients with poor outcomes (n = 8) and good outcomes (n = 13) by two-sample t-test model of SPM12. The statistically significant and corrected method of multiple comparisons were same as aforementioned standards.

| Comparisons of MRI and 18 F-FDG-PET findings
The MRI and 18 F-FDG-PET findings in subjects with anti-GABAB receptor encephalitis are summarized in Table 2 Of those 9 patients, 7 (77.8%) with mixed metabolism mainly presented with a relatively common metabolic pattern consisting of MTL hypermetabolism and relative cortex hypometabolism (frontal and parietal lobe) (FWE corrected, p < 0.05; Figure 2A). In addition, groupwise analysis also confirmed MTL hypermetabolism in association with relative hypometabolism in frontal or parietal lobe, extending to cingulate gyrus, was a general metabolic pattern in subjects with anti-GABAB receptor encephalitis (FDR corrected, p < 0.05; Figure 2B).

| Correlation analysis between long-term outcome and 18 F-FDG-PET based on SPM
All patients received first-line immunotherapy, including intravenous immunoglobulin and steroid pulse therapy, and 7 patients with tumors were additionally treated with chemotherapy or tumor removal (

| DISCUSS ION
Our study has three major highlights and clinical implications. First, we revealed highly pronounced MTL hypermetabolism in association with relative frontal and parietal hypometabolism detected by semi-quantitative brain 18  Future prospective studies are required to verify these findings and clarify potential pathogenic mechanisms.
There are several limitations in this study. First, a potential selection bias exists based on the retrospective nature of this study.
Second, the sample size may be relatively small in this study due to low prevalence of anti-GABAB receptor encephalitis and high eco-

| CON CLUS IONS
In summary, this study demonstrates a previously unidentified role of 18 F-FDG-PET imaging in the diagnosis and outcome of anti-GABAB TA B L E 3 Baseline comparison between patients with good outcomes and poor outcomes

ACK N OWLED G EM ENTS
The authors thank the following colleagues who contributed to this study providing FDG-PET data: Chengxu Jiang, Wei Zhang. The authors thank AiMi Academic Services (www.aimie ditor.com) for the English language editing and review services.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data are not publicly available due to privacy or ethical restrictions. The data that support the findings of this study are available on request from the corresponding author.