Brain Morphological Abnormalities in Children with Cyclin-Dependent Kinase-Like 5 Deficiency Disorder

Abstract

Background:

To quantitatively evaluate the brain MRI morphological abnormalities in patients with cyclin-dependent kinase-like 5 deficiency disorder (CDD) on a group level and longitudinally.

Methods:

We performed surface-based MRI analysis on high-resolution T1-weighted images on three CDD patients scanned at age of three years, and compared with 12 age- and gender-matched healthy controls. We further examined the longitudinal morphological changes in one patient with a follow-up of 5 years.

Results:

CDD patients presented significant reductions in total intracranial volume, total gray matter (GM) volume and subcortical GM volume compared to controls. For subcortical regions, significant GM volume reductions were seen in the brain stem, bilateral thalamus, bilateral hippocampus, bilateral cerebellum and left amygdala. Although GM volume of cortical mantle did not show statistical differences overall, significant reduction was detected in bilateral parietal, left occipital and right temporal lobes. Cortical thickness exhibited significant decreases in bilateral occipital, parietal and temporal lobes, while surface area did not show any significant differences. Longitudinal follow-up in one patient revealed a monotonic downward trend of relative volume in the majority of brain regions. The relative surface area appeared to gain age-related growth, whereas the relative cortical thickness exhibited a striking progressive decline over time.

Conclusions:

Quantitative morphology analysis in children with CDD showed global volume loss in the cortex and more notably in the subcortical gray matter, with a progressive trend along with the disease course. Cortical thickness is a more sensitive measure to disclose cortical atrophy and disease progression than surface area.

1. Introduction

Cyclin-dependent kinase-like 5 (CDKL5) deficiency disorder (CDD) is a well-known genetic cause of infant-onset epilepsy, with an estimated incidence of approximately one in 40,000 to one in 60,000 live births^{(1, 2)}. The CDKL5 gene has the highest expression in the brain during the early postnatal stage, and is involved in the process of neuron and synapse formation and maturation^{(1, 3)}. Patients with CDD commonly have infantile-onset refractory epilepsy, severe global developmental delay, movement disorders, and cortical visual impairments^{(1, 2, 4–6)}. The lack of treatment response to conventional anti-seizure medications and the pervasive global developmental delay in CDD made it a priority to search for biomarkers that would allow for early identification of the disease, better understanding of the pathophysiology, and protocols for tracking disease progression and treatment response⁽²⁾.

CDD was first reported in 2003⁽⁷⁾. Since that time, accumulating evidence from clinical presentation, electroencephalography (EEG), pathology, and animal models indicated that CDD is a typical developmental encephalopathy with involvement of the whole brain^{(1, 2, 8, 9)}. Brain morphology changes in CDD have not yet been fully studied with magnetic resonance imaging (MRI) analysis. The few previously published cases described MRI findings as mainly normal, with some reported cortical atrophy or white matter (WM) hyperintensities on T2/fluid-attenuated inversion recovery (FLAIR) images^{(1, 4, 10–13)}. The need for anesthesia or sedation in young children and those with cognitive and intellectual disabilities makes it intrinsically difficult to obtain high-quality MRI scans in children with CDD.

The advancement of modern MRI techniques has enabled quantitative morphology analysis to become a powerful tool for providing new insights into the underlying substrate of the pathophysiology of childhood epilepsy syndromes^{(14, 15)}. Surface-based techniques, for example, can generate fine measurements of the cortical mantle by calculating cortical thickness and surface area, which are reflective of microstructure changes and commonly influenced by different genetic factors^{(16, 17)}. In this study, we conducted surface-based morphological analyses in patients with CDD to: (1) evaluate comprehensively the gray matter (GM) volume and white matter (WM) volume changes in both cortical and subcortical regions; (2) identify possible microstructural abnormalities in the cortical mantle by extracting cortical thickness and surface area separately; (3) perform longitudinal observation of these quantitative measurements to detect whether any progressive cerebral morphological alteration occurred.

2. Materials and Methods

2.1. Participants

We included a consecutive series of patients with CDD seen in the Cleveland Clinic CDD Centers of Excellence (COE) from 2012 to 2019. Inclusion criteria: (1) clinical and genetic confirmation of the diagnosis of CDD⁽¹⁾,⁽¹⁸⁾; (2) availability of high-resolution T1-weighted images from clinical MRI for analyses. Clinical and demographic data were collected using standardized history and examination forms. Twelve age- and gender-matched healthy control subjects (HCs) were selected from the Pediatric Imaging, Neurocognition and Genetics (PING) Study⁽¹⁹⁾. Data were available to the public in the National Institute of Mental Health Data Archive (NDA). Controls were selected with following inclusion criteria: (1) no neurological, psychological, or systemic diseases; (2) normal development; (3) no abnormalities on 3.0T MRI.

2.2. Standard Protocol Approvals, Registrations, and Patient Consents

This retrospective study included patients with CDD identified from the Cleveland Clinic CDD COE and the CDKL5 Registry. All parents or legal guardians provided written informed consent prior to study enrollment.

2.3. Data Availability Statement

HCs’ data are publically available on NDA [Collection ID: 2607] (https://nda.nih.gov/edit_collection.html?id=2607) after submission of the appropriate data use certificate approval. Patients’ anonymized data are available upon request to the corresponding author; these data are not publically available as they may contain information that could compromise the privacy of research participants.

2.4. MRI Data Acquisition

Patients’ MRI data were obtained using the Magnetization Prepared Rapid Acquisition with Gradient Echo (MPRAGE) sequence in a 3.0T Siemens Trio MRI scanner with a 32-channel phased array head coil. The scan parameters were: slice thickness=0.94 mm (no gap), repetition time=1,860 ms, echo time=3.4 ms, inversion time=1100 ms, flip angle=10°, matrix size=256×256, voxel size = 0.94×0.94×0.94 mm³. HCs’ MRI data included 6 individuals with MPRAGE sequences on 3.0T Siemens scanners and 6 individuals with Inversion Recovery Spoiled Gradient Recalled (IRSPGR) sequences on 3.0T GE scanners.

2.5. MRI Data Processing

Preprocessing was performed using the FreeSurfer 6.0 software package (http://surfer.nmr.mgh.harvard.edu/) using the previously-described sequence⁽²⁰⁾: (1) motion correction and removal of non-brain tissue using a hybrid watershed/surface deformation procedure; (2) automated Talairach space transformation; (3) segmentation and intensity normalization; (4) tessellation of GM and WM boundaries; (5) automated topology correction; (6) surface deformation to optimally place the GM/WM and GM/cerebrospinal fluid (CSF) borders with the greatest shift in signal intensity. The above procedure generated GM volume and WM volume for both cortical and subcortical structures per hemisphere. Since GM volume in the cortical mantle is a product of cortical thickness and surface area⁽¹⁷⁾, these two measurements were further extracted at subvoxel resolution. Then the average GM volume, cortical thickness, and surface area for the frontal, temporal, parietal, occipital, and insular regions in the cortical mantle were reconstructed from the Desikan-Killiany Atlas⁽²¹⁾. As part of quality control, these automated outputs were carefully inspected for segmentation errors and mislabeling prior to statistical analysis following standardized protocols⁽²²⁾.

2.6. Statistical Analysis

SPSS (version 20.0, http://www.spss.com/) was used for group analysis in all extracted morphological measurements, including GM volume, WM volume, cortical thickness and surface area, as well as the demographic parameters. We performed the Student’s t test for continuous data with normal distribution and the Pearson χ2 test for categorical data to compare the CDD group with the HCs group. The level of statistical significance was set at P-value < 0.05, two tailed. A correction for volume differences due to the growing brain size was made for the longitudinal analysis-i.e., the extracted morphological values were all normalized by the total intracranial volume (TIV) to generate relative volume/thickness/surface area, consistent with the literature^{(23, 24)}.

3. Results

A total of 34 patients with CDD were identified from 2012 to 2019. High-resolution T1-weighted images were available for 4 patients at 7 different time points. One patient’s image data sets (at the ages of 1 month and 3 months, respectively) were excluded due to segmentation failure in Freesurfer. In the end, 3 patients with 5 image data sets were included for final analysis: All 3 patients were scanned at 3 years of age; one patient had 2 additional scans at the ages of 8 months and 6 years, for which longitudinal analysis was performed. The demographic and clinical characteristics at each scan time point are detailed in Table 1. All patients demonstrated polymorphous seizure semiology clustered in multiple phases with hypsarrythmia in EEG, global developmental delay, disturbed sleep, cortical visual impairment, and movement disorder. Stereotypes were recorded in two of them.

Table 1.

Clinical and demographic information of patients with CDKL5 deficiency disorder included in this study

Subject Number / Age at MRI scan (months)	Patient 1/35mo	Patient 2/36mo	Patient 3/8mo	Patient 3/37mo	Patient 3/75mo
Gender	F	F	M
Genetic test	c.1354C>T; p.Gln452†	c.175C>T; p.Arg59X	c.351T>A; p.Tyr117Stop
Gestation period	Full term	Premature	Full term
Family history	-	-	-
Age at seizure onset (months)	1	2	8
Seizure type†	1. Multiple phases in serial (Spasm/Tonic/Clonic); 2. Spasm; 3. Tonic	1. Spasm; 2. HTSS-like	1. Spasm; 2. Multiple phases in serial (Clonic/Hypomotor/Tonic/Spasm); 3. Tonic	1. Spasm; 2. Hypermotor; 3. Hypomotor	1. HTSS-like; 2. Multiple phases in serial (Tonic/Hypomotor); 3. Spasm; 4. Hypomotor; 5. Myoclonic
EEG features†	Asymmetric hypsarrhythmia with EDs maximum in Bi Parietoocciptal region	Generalized hypsarrhythmia with multiregional EDs maximum in Bi Parietal region	Intermittent rhythmic generalized slow waves maximum in Bi temporoparietal region	Generalized hypsarrythmia with polyspikes and CS maximum in Bi parietooccipital region	Generalized hypsarrhythmia with generalized EDs and CS
Global development delay†	+	+	-	+	+
Severe intellectual disability†	+	+	-	+	+
Sleep disturbances†	+	+	-	+	+
Movement disorder†	+	+	-	+	+
Cortical visual impairment†	+	+	-	+	+
Brain MRI	Moderate generalized parenchymal volume loss and nonspecific patchy T2 hyperintensity in periatrial white matter.	Normal	Normal	Nonspecific minimal T2/FLAIR hyperintensity in periatrial white matter.	Stable nonspecific mild T2/FLAIR hyperintensity in periatrial white matter.
ASDs†	CNP, PDN	LEV, VPA, OXC, PHB	LEV, DZP	TPM, LTG	DZP, VPA, LTG

^†All electroclinical data were collected at the follow up visit when the MRI scan was performed.

MRI, magnetic resonance image; mo, months; F, female; M, male; +, feature present; −, feature absent. Flair, fluid attenuated inversion recovery; HTSS, hypermotor-tonic-spasms sequence; EEG, electroencephalography; EDs, epileptic discharges; Bi, bilateral; CS, continuous slow; ASDs: antiseizure drugs. CNP, clonazepam; PDN, prednisone; LEV, levetiracetam; VPA, valproic acid; OXC, oxcarbazepine; PHB, phenobarbital; DZP, diazepam; TPM, topiramate; LTG, lamotrigine.

3.1. Group Morphological Analysis

In relation to the age- and gender-matched HCs (male/female 5/7, average age 3.3±0.3 years, details in Table S1), the 3-year-old CDD group demonstrated significant TIV loss (P=0.024). Both GM and WM showed a trend of volume loss, with a statistically significant difference in total GM (P=0.045) and subcortical GM (P=0.014). More details are included in Figure 1B and Table S2.

Figure 1:

Global brain volume loss in 3-year-old CDD patients compared to age-matched HCs. (A) Visualization of the regions with significant volume loss (P < 0.05, two tailed, details in Table S2 and S3) is displayed on the brain of patient 3. Each significant region was assigned a different color. (B-F) Values of morphological measurements were extracted from the specific brain regions of the 3-year-old CDD patients and HCs. The axis is labeled by the ratio of volume of CDD to HCs, shown in Table S2 and S3. The orange line represents 3-year-old HCs, while the blue line represents 3-year-old CDD patients. CDD= CDKL5 deficiency disorder; HCs= healthy control subjects; Bi= bilateral; L= left; R= right; TIV= total intracranial volume; GMV= gray matter volume; WMV= white matter volume. *P<0.05, two-tailed.

All nuclei except the right pallidum in the subcortical regions presented GM volume decreases, with the brain stem (P=0.048), bilateral thalamus (left: P=0.008; right: P=0.023), bilateral cerebellum (left: P=0.04997; right: P=0.049), and left amygdala (P=0.026) reaching statistical significance. The bilateral hippocampus also showed a significant GM volume decrease compared to the HCs (left: P<0.001; right: P<0.001). More details are included in Figure 1C and Table S2.

In the cortical mantle, GM volume for each hemisphere did not show statistical differences overall. However, within lobar regions, significant reduction of volume was seen in the bilateral parietal GM (left: P=0.041; right: P=0.048), left occipital GM (P=0.036), and right temporal GM (P=0.042). Cortical thickness exhibited a decrease in a more widespread distribution than GM volume, in the bilateral parietal (left: P=0.008; right: P=0.005), bilateral occipital (left: P=0.026; right: P=0.033), and bilateral temporal lobes (left: P=0.019; right: P=0.032). Meanwhile, surface area showed a downward trend with no statistical significance. More details are included in Figures 1D–F and Table S3.

3.2. Longitudinal Observation

We observed three different patterns of relative volume changes in the CDD patient who had longitudinal MRIs acquired over a 5-year time span. First, the majority of the GM, both subcortical and cortical regions, showed monotonic decline of relative volume over time (Figure 2B). Second, there were a few areas, including the left amygdala, right putamen, left cerebellar cortex, bilateral cerebral WM, left occipital cortex, and bilateral temporal cortex, that demonstrated transient volume growth in the 8 months to 3 years period, but subsequent volume loss from 3 to 6 years (Figure 2C). Last, there was slight volume increase over time in a small number of areas, such as the corpus callosum, brain stem, bilateral cerebellar WM, and right insula (Figure 2D). More details can be found in Tables S4 and S5.

Figure 2:

Longitudinal changes of relative volume in patient 3 across different brain regions during 5-year follow-up time. (A) Visualization of the regions with 3 different dynamic patterns of relative volume changes: blue for monotonic decrease; grey for increase followed by decrease; yellow for monotonic increase. (B-D) Relative volume alterations extracted from each brain region over time. Each colored line represents one specific region. The y-axis shows the relative volume value extracted from the region, shown in Tables S4 and S5. The x-axis shows time of follow-up/MRI. *indicates the regions with relative volume values displayed with the secondary y-axis at the right side of the chart (hippocampus, amygdala, thalamus and so on), as their ratios were markedly smaller than those displayed on the primary y-axis on the left. CDD= CDKL5 deficiency disorder; L= left; R= right; WM= white matter; 8-mon-old= 8-month-old; 3-yr-old= 3-year-old; 6-yr-old= 6-year-old.

In terms of relative cortical thickness, this CDD patient exhibited an even more outstanding monotonic downward trend over time, with a uniform pattern in all lobes (Figure 3A). The acceleration of cortical thinning appeared to occur at a relatively early age, from 8 months to 3 years. Meanwhile, the relative surface area seemed to gain age-related expansion in all lobes except for the right parietal lobe (Figure 3B). However, the surface area growth appeared to be insufficient to make up for the cortical thinning, resulting in a GM volume loss over time (Figure 3C). More details can be found in Table S5.

Figure 3:

Longitudinal changes of relative morphological measurements in the cortical mantle of patient 3. Each colored line represents one lobe, with a solid line for the left hemisphere and a dotted line for the right hemisphere. The y-axis shows the relative morphological values (details in Table S5). The x-axis shows time of follow-up/MRI. CDD= CDKL5 deficiency disorder; L= left; R: right; GMV= gray matter volume; 8-mon-old= 8-month-old; 3-yr-old= 3-year-old; 6-yr-old= 6-year-old.

4. Discussion

Our study is the first to describe quantitative MRI morphological abnormalities in young children (3 years of age) with CDD. CDD is a rare disorder that has not been well studied with dedicated MRI protocols and quantitative analyses. Children with CDD typically have cognitive and intellectual disabilities, making high-quality MRI scans difficult to obtain. MRI scans with appropriate sequences for surface-based morphological processing have only recently been consistently acquired for children with CDD. These reasons led to the relatively small number of patients that can be include in our study. Nevertheless, our study includes unique data that enabled group comparisons with controls and longitudinal analysis. It is our hope that findings from the current study will provide a basis for more advanced MRI acquisitions and analyses for patients with CDD.

We detected global morphological deterioration in the brains of children with CDD, which supported the nature of whole-brain involvement in the disease process^{(1, 2, 4–6, 9)}. Due to the paucity of MRI data, prior case reports mainly described normal or unremarkable findings in young children with CDD^{(1, 10, 12, 13)}. A few descriptive studies based on visual inspection showed cerebral atrophy, but in a very wide range of ages (1.8–17 years old)^{(4, 11)}. Recently, it was documented that visually apparent MRI changes were presented in approximately 50% of 6-year-old children with CDD, but rarely seen in earlier ages⁽³⁾. Our quantitative analysis identified significant morphological abnormalities in 3-year-old CDD patients, which suggests that surface-based quantitative measurements might serve as potential sensitive markers for evaluating subtle structural abnormalities in these patients. Similarly, these quantitative MRI techniques have been successfully applied to other epileptic neurodevelopmental encephalopathy to reveal the structural abnormalities that are not easily identified by visual inspection, such as in Rett syndrome, West syndrome, and Rasmussen syndrome^{(14, 25, 26)}.

It is noteworthy that the detected volume loss in CDD patients showed an uneven, even asymmetrical, distribution in cortical and subcortical regions. This is consistent with the abnormal regional volume findings in other pediatric epileptic syndromes. For example, patients with Rett syndrome showed a marked decrease in cerebellar volume⁽¹⁴⁾, and MRI-negative patients with West syndrome of unknown etiology had abnormal GM volume in the left temporal lobe⁽²⁵⁾. Some evidence of seizure semiology as well as EEG changes (both slowing and epileptiform activity) in CDD patients did suggest its lateralized or focal features^{(1, 27)}. Our results, however, cannot lead to conclusive lateralized or focal implications due to the limited sample size.

The most striking volume loss in our study was detected in the subcortical regions, which supplemented the cortical atrophy findings by visual inspection of the MRI in CDD patients as reported in previous studies^{(4, 11)}. This result is consistent with the findings in mouse brains that CDKL5 is highly expressed in both the cortex and the subcortical GM⁽⁹⁾. Such subcortical involvement could perhaps explain some common symptoms of CDD patients and CDKL5-knock-out rodent models, e.g., thalamus atrophy might contribute to movement disorders; hippocampus and amygdala atrophy might contribute to intellectual disabilities; cerebellum atrophy might contribute to motor delays; brain stem atrophy might contribute to arrhythmia and respiratory disturbance^{(1, 9, 28)}. All these symptoms were present in the patients included in this study (see Table 1). Cerebellar gliosis, Purkinje cell loss, and axonal torpedoe were reported in histopathological examination of CDD patients⁽⁸⁾. Reduced spine density and abnormalities in the number of mature spines, dendritic length, and dendritic complexity were observed in cerebellar granule cells and hippocampal CA1 pyramidal neurons^{(9, 29, 30)} in CDKL5 knock-out mouse models. Further studies are still needed to elucidate the neuropathological substrates underlying the subtle brain morphological abnormalities associated with CDD.

We found volume loss and cortical thinning in the occipital, parietal, and temporal lobes. This posterior dominance of cortical structural abnormalities is not surprising, since cortical visual impairment is a very common manifestation in CDD patients^{(1, 31)}, and occurred in all patients we enrolled. It also corresponds well with a post-mortem finding in a 5-year-old CDD patient of symmetrical occipital flattening with gliosis, obtained via microscopic examination⁽⁸⁾. EEG results showed focal slowing in the posterior quadrants as mild abnormalities at the early stage of CDD, preceding more diffuse and severe encephalopathies^{(1, 12)}. As shown in Table 1, all patients in our cohort presented epileptic discharges maximally in the posterior quadrants. Corroborating this finding, a previous quantitative EEG study investigated the role of occipito-temporal networks in the pathogenesis of Rett syndrome, showing discrete coherence performance in MeCP2 vs. CDLK5 genetic variants⁽²⁷⁾. Such posterior predominant of brain structural abnormalities have been reported to be associated with earlier onset of infantile epileptic spasms⁽³²⁾. Because CDD is characterized by infantile-onset refractory epilepsy, and brain maturation proceeds in a posterior-to-anterior pattern at the early stage of life⁽³³⁾, our quantitative morphological findings might suggest that disturbance of brain maturation in CDD children also occurred in a posterior-to-anterior pattern.

Recent studies have shown that cortical thickness might be more sensitive than surface area to detect disease-specific cortical microstructure changes in neurodevelopmental disorders, such as schizophrenia and autism^{(34, 35)}. Our results showed that the most pronounced degeneration of the cortical mantle in CDD did show up in cortical thickness. Cortical thickness and surface area are believed to be influenced by different genetic factors. Firstly, the genetic sources are distinct, in that surface area mainly represents the number of columns within a cortical region, and cortical thickness reflects the size, density, and arrangement of neurons, neuroglia, as well as nerve fibers in the cortical columns^{(17, 35)}. Secondly, the genetic organizations are different, as surface area shows great proximity in clusters from the same lobe, and cortical thickness demonstrates a close relationship among clusters with similar maturational timing⁽³⁶⁾. Therefore, it’s reasonable that cortical thickness is a more sensitive measurement for CDD, especially given CDKL5’s important regulatory roles in cell proliferation, neuronal migration, axonal outgrowth, dendritic morphogenesis, and synapse development⁽⁹⁾.

The main trend in our longitudinal observation of one patient with CDD was a progressive volume loss over time; whereas overall increase of brain volume was expected for healthy children in their early age, with different growth rates across different brain regions⁽³³⁾. The age-related morphological alterations in CDD patients also showed three different patterns varying across different brain regions, which might result from uneven regional disturbances by CDD, or regional heterogeneity of volumetric dynamics along with brain development, or a combination of these two factors. For the measurement of longitudinal brain morphological changes, cortical thickness again served as a more sensitive measure by showing a uniform pattern of ongoing monotonic decline in all lobes. Although surface area seemed to gain transient age-related growth, it could not make up for the effects of cortical thinning, and volume loss still occurred over time. The progressive brain morphological degeneration is consistent with the evolution of epilepsy in CDD^{(1, 4, 10, 12)}, i.e., early onset with mild EEG abnormalities progresses to epileptic encephalopathies with infantile spasms and hypsarrythmia; eventually, refractory multifocal and myoclonic epilepsy with diffuse abnormalities in background rhythms and epileptiform activities develops. In order to determine whether the longitudinal brain morphological change patterns indeed reflect CDD-specific abnormalities, comparisons need to be made with healthy brain developmental trajectories in future studies.

Our study has the same limitations as other prior studies that describe imaging findings retrospectively in CDD. Firstly, it is inevitably limited by a small sample size, which might reduce the power to detect some potential differences. Individual heterogeneity must be taken into account, especially for the longitudinal findings. Secondly, the influence of repeated seizure insults and anti-seizure medications on the brain’s morphological abnormalities should be taken into consideration when linking our findings to CDKL5-related brain impairment. Thirdly, the three time points chosen for our longitudinal observations were based on the availability of MRI results (due to the retrospective nature of the study). Although follow-up time points would typically reflect the natural history of CDD itself, such as developmental milestones, stages of epilepsy, and EEG evolutions^{(1, 4, 10, 12)}. Lastly, we did not have controls for the longitudinal comparison at ages of 8 months and 7 years, which limited our ability to exclude confounders for dynamic change patterns such as normal brain development⁽³³⁾. Further prospective neuroimaging cohort studies are needed to measure the developmental trajectories of brains affected by CDD, and for correlating these trajectories with gender, genotype, phenotype, and severity.

In conclusion, we present the first study to use quantitative MRI morphological measures to study children with CDD on a group level and longitudinally. Global GM volume loss was detected with a main trend of progression over time, corresponding to the electroclinical features and neuropathological findings associated with CDD. Cortical thickness stood out as a sensitive measurement for assessing the alterations in the cortical mantle, which might be attributed to the genetic influence of CDKL5. These findings may contribute to the understanding of the relationship among the brain morphological abnormalities, electroclinical phenotypes, and genetic contributions in the CDD population, and lead to an innovative recognition of the brain microstructural substrates underlying the pathophysiology of CDD.

Highlights.

1. Quantitative MRI morphological abnormalities was first reported in children with CDD.

2. Children with CDD showed global volume loss in cortex and more notably in subcortical GM.

3. Longitudinal follow-up showed monotonic downward trend of relative GM volume along with disease course.

4. Cortical thickness was more sensitive to disclose GM atrophy and disease progression than surface area.

Abbreviations:

CDKL5

cyclin-dependent kinase-like 5

CDD

CDKL5 deficiency disorder

CTh

cortical thickness

EEG

electroencephalography

GM

gray matter

GMV

gray matter volume

HCs

healthy control subjects

MRI

magnetic resonance imaging

SA

surface area

TIV

total intracranial volume

WM

white matter

WMV

white matter volume