A 23-year-old right-handed man with behavioral problems presented to our hospital. He was born at term by normal spontaneous delivery with a birth weight of 3200 g. He had been healthy and achieved good grades until 19 years of age when he became unable to focus on his classes in college. After taking a leave of absence, his behavior further deteriorated (Fig. 1). He shoplifted several times and stole a purse in a public place. When he was 21 years old, he lost consciousness several times and developed myoclonic jerks during his sleep. At 22 years of age, he had agraphia and apraxia (e.g., unable to cut a piece of paper with scissors). His parents were not in a consanguineous marriage and none of his family had neurological disorders.
On examination, he appeared to be restless, impulsive, and distracted. He continuously had action myoclonus on both of his hands. Otherwise, neurological examinations shown normal findings for the cranial nerves, muscle strength, deep tendon reflexes, sensation, cerebellar, and extrapyramidal signs.
He scored 46 out of 100 on Addenbrooke’s Cognitive Examination version III (ACE-III) with the following subscores: attention and orientation, 10/18; memory, 12/26; fluency, 2/14; language, 17/26; and aspatial skills, 5/16. Electroencephalogram showed generalized intermittent 3–5-Hz spike-and-slow-wave complexes. Brain MRI showed non-significant findings, except for mild diffuse brain atrophy (Fig. 2). N-isopropyl-p-(123I)-iodoamphetamine (IMP) single-photon emission computed tomography (SPECT) revealed widespread hypoperfusion in the cerebral cortices. The whole-exome sequencing and filtering analysis of the patient and his parents identified de novo H338R mutation in the SERPINI1. The H338R mutation, previously reported as a pathogenic mutation causing neuroserpinosis, confirmed the diagnosis of neuroserpinosis3.
As treatment of the patient’s epileptic seizures, he was prescribed with 800-mg valproic acid daily, which suppressed the epileptic seizures. His cognitive impairment gradually progressed (Fig. 1). At the last follow-up (4 years after the first visit), he became incommunicative. Follow-up MRI showed more advanced diffuse brain atrophy (Fig. 2). IMP SPECT showed diffuse hypoperfusion in the cerebral cortices.
Brain MRI analysis
Single-case voxel-based morphometry (VBM) was performed using 3D T1-weighted images to assess the structural brain changes in the gray matter volume. In our single-case VBM analysis, we used the software “voxel-based specific regional analysis system for Alzheimer’s disease advance” (Eisai, Tokyo, Japan) equipped with SPM8 (The Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College of London, UK) and Diffeomorphic Anatomical Registration using Exponentiated Lie Algebra (DARTEL) based on VBM11. To assess the pattern of gray matter volume reduction in the patient, the original control data in the software were replaced with the data of 15 age-matched disease controls (15 men; mean age 23.6 ± 3.0 years; disorder five cases of sleep disorders, four cases of long coronavirus disease, two cases of narcolepsy, two cases of attention-deficit hyperactivity disorder, one case of chronic fatigue syndrome, and one case of anxiety disorder). The segmented gray matter images were compared with the mean and standard deviation of the images of the 15 disease controls using voxel-by-voxel Z-score analysis with voxel normalization to global mean intensities. The Z-score was calculated as follows: Z-score = ([control mean] − [individual value]) / (control standard deviation). A Z-score > 2 was defined as significant. Single-case VBM analysis using MRI at the first visit showed gray matter volume reductions in the ventromedial prefrontal cortices (vmPFC), occipitoparietal cortices on both sides, and the posterior part of the temporal lobe on the right side (Fig. 3). The single-case VBM analysis using MRI scans obtained at 4 years after the initial one demonstrated more severe and extensive gray matter volume reductions in the vmPFC, occipitoparietal cortices, cingulate gyri, and right temporal gyrus (Fig. 3).
To predict the patient’s brain age, the support regression model implemented in the LIBSVM (http://www.csite.ntu.edu.tw/~cjlin/libsvm/) toolbox with a linear kernel and default set of parameters was used (i.e., in the LIBSVM: C = 1, v = 0.5). The details of the analytical method were described in a previous study12. For the regression model, the chronological age was considered the dependent variable, whereas the principal components derived from the gray matter voxel intensities were considered independent variables. Brain-age prediction analysis using the MRI scan obtained at the first visit revealed a predicted brain age and brain-predicted age difference (brain-PAD: predicted brain age–chronological age) of 60.06 and 36.63, respectively. Brain-age prediction analysis using the MRI scan obtained at 4 years after the initial one revealed a predicted brain age and brain-PAD of 80.87 and 53.58, respectively.