Insights into cognitive and behavioral comorbidities of SLC6A1-related epilepsy: five new cases and literature review

Introduction SLC6A1 pathogenic variants have been associated with epilepsy and neurodevelopmental disorders. The clinical phenotype includes different seizure types, intellectual disability, and psychiatric symptoms affecting mood and behavior. Few data regarding neuropsychological features have been described, and details on cognitive profiles are often missing due to the lack of standardized tests. Methods We retrospectively reviewed the neuropsychological assessments of five subjects carrying heterozygous missense genetic variants in SLC6A1. We also collected data on epileptic features, EEGs, and brain MRIs. Additionally, we reviewed neuropsychological data from 204 previously reported patients with SLC6A1 pathogenic variants. Results In our series, at the last evaluation (median 12.6 years), three patients had borderline intellectual functioning, one patient had mild cognitive impairment, and one patient presented with a moderate cognitive disability. Three out of five patients underwent at least two neuropsychological evaluations, which revealed a worsening of cognitive functions over time. We detected attention deficits in all patients. In addition, we observed anxiety, disruptive behavior disorder, emotional instability, and hetero aggressiveness. We also performed a literature review that highlighted that most of the patients with SLC6A1 pathogenic variants have mild-to-moderate intellectual disability and that one-third of cases have autistic traits. Discussion Based on the literature review and the detailed description of our cases, we conclude that patients with SLC6A1-related epilepsy mostly present with mild-to-moderate intellectual disability, often associated with attention disorders. Such symptoms may worsen over time. Periodic standardized neuropsychological tests may be useful tools to follow development over time, and patient-specific rehabilitation programs could be tailored consistently.

Pathogenic variants of SLC6A1 destabilize the global protein conformation, causing a loss of GAT-1 function that leads to alterations in GABA levels in the synaptic cleft (Kahen et al., 2022).This imbalance may independently affect many pathways involved in both neurological and psychiatric diseases, leading to a wide spectrum of clinical phenotypes, including epilepsy, intellectual disability (ID), and behavioral disorders (Bröer and Gether, 2012;Carvill et al., 2015;Poliquin et al., 2021).
In this study, we describe the neuropsychological profile of five new patients carrying SLC6A1 pathogenic variants, detailing the cognitive and behavioral features.The clinical presentation of all subjects was consistent with EMAtS.Moreover, we performed a literature review, focusing on the neuropsychological profiles (cognitive and behavioral details), which includes 204 previously published patients with SLC6A1 pathogenic variants.

Methods
We retrospectively reviewed the medical charts of five patients with SLC6A1 pathogenic variants who were followed at Bambino Gesù Children's Hospital in Rome, Italy.SLC6A1 pathogenic variants were identified in all patients through a trio-based New Generation Sequencing (NGS) epilepsy gene panel.We collected clinical data, including sex, family history, age at seizure onset, seizure semiology and frequency, intellectual disability, behavioral and psychiatric disorders, EEG and brain MRI.
Seizure types and epilepsy syndromes were classified according to the Commission on Classification and Terminology of the International League Against Epilepsy (Fisher et al., 2017;Scheffer et al., 2017;Specchio et al., 2022).
We retrospectively reviewed all cognitive and behavioral evaluations for all patients performed with the following scales: 1) Developmental and cognitive assessment: -Griffiths Mental Development Scales (GMDS).
- This study was performed according to ethical principles for medical research involving human subjects stated in the Declaration of Helsinki and was approved by the local ethics committee.
We also reviewed the literature from January 2012 to February 2023 related to all published patients with SLC6A1 pathogenic variants.We identified a total of 204 patients carrying SLC6A1 genetic variants.The search strategy included the terms "SLC6A1" and/or "Epilepsy" on PubMed.References were also identified manually from relevant articles and by searching through the authors' files.We limited the review to 50 English-language articles.We excluded patients with microdeletions encompassing SLC6A1 and extended the search to other genes that could be causative of misleading data.
For all reported patients, we collected the genetic variants (including encoded protein, type of mutation and inheritance) and clinical data (including gender, age at the study, seizure types and syndromic classification, cognitive and behavioral features).Supplementary Table S1 shows a summary of the findings.We cannot exclude the presence of overlapping patients within this dataset due to incomplete data.This is a 25-year-old male with normal psychomotor development before seizure onset.Epilepsy started at the age of 2 years and 3 months with atonic seizures, followed by absences with eyelid myoclonia, myoclonic-atonic, and myoclonic seizures.He was treated with valproate (VPA), ethosuximide (ETS), and clonazepam (CNZ).The last seizure was at the age of 3 years and 7 months; therefore, in adolescence, anti-seizure medications (ASMs) were discontinued without seizure relapse.

Clinical cases
At the age of 7 years, the Leiter-R scale, exploring the nonverbal cognitive profile, revealed a borderline cognitive level (IQ = 80).At the age of 12 years, we performed a second cognitive evaluation, which detected mild cognitive impairment (IQ = 67).He attended school with a support teacher.He also had a behavioral disorder, including attention deficit and stereotyped behavior.At the age of 25 years, at the last follow-up (FU), he had mild ID requiring support in personal autonomy and self-care.

Case 2
This is an 18-year-old male.Epilepsy started at the age of 6 years with episodes of absences with and without eyelid myoclonia.At the age of 10 years, he presented with tonic seizures, and by the age of 12, he started to present also with atonic seizures.He was treated with several ASMs, such as VPA, ETS, carbamazepine (CBZ), and clobazam (CLB), with poor efficacy.Tonic and atonic seizures have been controlled with ASMs since the age of 15 years, while daily absences persist at the last FU.
Independent gait was reached at the age of 18 months.He had speech delay: first words at 24 months; therefore, speech therapy was started.
The first neuropsychological evaluation was performed at the age of 13 years using the nonverbal cognitive test PM 38, which detected borderline intellectual functioning in terms of fluid reasoning (IQ = 74).In addition, the CBCL showed several difficulties in attention, anxiety, and social domains.He also had difficulties in managing his emotional states, which often led to behavioral dysregulation.A rehabilitation program was indicated.
We re-evaluated his cognitive performance at the age of 16 years with the PM 38.We found cognitive regression (IQ = 62), indicating mild cognitive delay.He needed school support and a differentiated school program.
Finally, we evaluated the patient at the age of 18 years with the WAIS-IV, which showed a moderate cognitive delay (IQ = 44), with greater impairment in the working memory index.He also presented with anxiety, relational disorders, and depressive features, which may have affected cognitive assessments.

Case 3
This is a 6-year-old girl born at 35 weeks of gestation from a twin cesarean section.Independent walking was reached at the age of 16 months.Parents reported speech difficulties since the second year of life.Epilepsy onset was at the age of 3 years with atonic seizures and episodes characterized by psychomotor arrest and oral automatisms.
Seizures were treated with CBZ, which worsened atonic seizures; after the replacement of CBZ with VPA and CLB, she reached complete remission of seizures.
The first neuropsychological evaluation was performed at the age of 3 years and 2 months with GMDS, which showed a developmental age of 32.5 months versus a chronological age of 37 months (DQ = 88): lower scores were seen in language and oculomotor coordination.
A second evaluation was performed at the age of 4 years and 4 months.The PM47 nonverbal cognitive scale showed an IQ of 90, with weakness in the areas of attention skills; this was not confirmed by the CPRS-R:L.
Finally, we evaluated the patient at the age of 5 years and 9 months with the WPPSI III scale, which showed borderline intellectual functioning (IQ = 77), with greater impairment in the processing speed index.We observed a deficit in attentional skills, with worsening of both selective and sustained attention.
She had been treated with psychomotor and speech therapies since the age of 3 years.Currently, the child attends primary school with a support teacher.

Case 4
This is a 6 years and 7 months old girl born from cesarean section after a normal pregnancy.At the age of 6 years and 5 months, she presented with epileptic seizures: TCS and absences with eyelid myoclonia were reported.She was treated with levetiracetam (LEV), lamotrigine (LTG) and CNZ with poor efficacy, while VPA led to significant seizure reduction.
Motor development was normal in the first few years of life, except for delayed speech (first words at the age of 3 years) and phonoarticulatory difficulties.At the age of 6 years and 5 months, she was assessed with the PM47 nonverbal cognitive scale (language barrier was clear), and she had an IQ of 70.At the same age, she also presented with behavioral disturbances characterized by oppositionality, emotional instability, heteroaggressiveness, attention deficit, and impulsivity.Poor frustration tolerance and fatigability were observed during the evaluation.Moreover, the use of CNZ had a negative impact on attention and behavior, inducing irritability, which was restored with drug withdrawal.
The child is not currently undergoing rehabilitation treatment due to poor familial psychosocial conditions.

Case 5
This is a 6-year-and 6-month-old boy born after a normal pregnancy.His mother suffered from generalized epilepsy with absences since the age of 6 years.Neuropsychological development was normal during infancy.Epilepsy started at the age of 3 years with atonic seizures and absences with eyelid myoclonia successfully treated with VPA.
The WISC IV at the age of 6 years and 6 months revealed a borderline cognitive level (IQ = 80) with strengths in processing speed and visual spatial index and weaknesses in working memory index.In addition, we detected a mild language delay.Over time, he also developed behavioral disturbances such as attention and executive function deficits, mild anxiety, and depression.
Neuropsychological assessments were sparsely reported in the literature, although data were often neither detailed nor standardized.
Regarding psychiatric features, ADHD was reported in 20 patients, ASD or autistic features in 64 patients, and behavioral disorders in 41 patients.

Discussion
The clinical phenotype associated with SLC6A1 pathogenic variants includes epilepsy with various seizure types (absences, myoclonic, atonic, and tonic-clonic seizures), neurodevelopmental disorders (i.e., ASD and ADHD), and psychiatric features affecting mood and behavior (i.e., aggressivity and anxiety).To date, few data on neuropsychological features have been reported; however, data on cognition and behavioral disturbances are often missing because of a lack of standardized testing.
Developmental delay has been frequently reported by parents even before epilepsy onset, as in three out of five of our patients (Cai et al., 2019;Goodspeed et al., 2020).Once epileptic seizures appear, those patients are referred to pediatric neurologists, and soon after cognitive evaluations, a wide spectrum of cognitive and behavioral disturbances become evident (Johannesen et al., 2018;Goodspeed et al., 2020).
Data from the literature review report developmental regression in approximately 13% of patients around the age of 3 years  10.3389/fnins.2023.1215684Frontiers in Neuroscience 06 frontiersin.org(Kalvakuntla et al., 2023).The disability involved several domains, including language, motor, social and adaptive skills (Kalvakuntla et al., 2023).Developmental regression may be underestimated as most of the evaluations were done through unstructured interviews with parents, and standardized tests were rarely reported.Additionally, the severity of ID was also seldom detailed in the literature, and when reported, it ranged from mild to moderate in most cases.
In general, no significant differences, in terms of cognitive impairment, were reported as related to seizure control (Johannesen et al., 2018).Accordingly, one patient in our series achieved seizure control around the age of 4 years; however, cognitive disability associated with poor autonomy in activities of daily living persisted into adulthood.
Eleven patients with normal IQ were reported in the literature, even if learning disorders were observed in five of them.Six out of 11 patients with normal IQ had epilepsy and interictal epileptiform discharges at EEGs: these findings support that etiology, rather than epilepsy, has a major role in this condition (Johannesen et al., 2018;Kim et al., 2019;Posar and Visconti, 2019;Goodspeed et al., 2020).Most of our patients had speech disorders before seizure onset, making this condition truly developmental and epileptic encephalopathy (DEE), in which epilepsy may worsen cognitive functions, which are already damaged by the genetic variant itself.The spectrum of SLC6A1 genetic variants encompasses wide phenotypic variability, from epilepsy with or without ID to the DEE phenotype.
Our patients received several cognitive and behavioral tests.At the last neuropsychological assessment, three patients had borderline intellectual functioning, one had mild cognitive impairment, and the last patient had moderate cognitive disability.Three out of five had a worsening of cognitive functions over time.Patient 1 moved from borderline functioning to mild cognitive impairment, patient 2 from borderline to moderate cognitive delay, and patient 3 from normal to borderline functioning over 2 years without developing a clear-cut cognitive disability.What was seen in most patients was a delay in acquiring new skills rather than a clear regression.Adult patients achieved personal independence in daily activities, even if they live with their parents.They do not have jobs, and one of them still attends school with a support teacher and an individualized learning plan.They do not practice any rehabilitative or psychotherapeutic therapy.They achieved personal independence in daily activities with minimal supervision.
These findings underline the importance of longitudinal assessments in patients with epilepsy over time, especially in children of developing ages.Accordingly, periodic cognitive evaluations with standardized testing, jointly with structured behavioral assessments, might be useful to adapt over time for a patient-specific rehabilitation program to identify the best cognitive functioning and reach good daily life autonomies during adulthood.
Regarding behavioral disorders, data from the literature may be misleading.Few authors have properly analyzed these features, reported as aggressive or oppositional behaviors with psychiatric traits (Rauch et al., 2012;Lucariello et al., 2016;Snoeijen-Schouwenaars et al., 2018;Kim et al., 2019).
Consistent with literature data (Poliquin et al., 2021), we found an attention deficit disorder in all analyzed patients together with high levels of anxiety, social and disruptive behavior disorder with a pattern of oppositionality, emotional instability, and heteroaggressiveness.No pharmacological therapy was prescribed for such symptoms.Worthy of mention is the possible role of some ASMs in worsening cognitive abilities and behavioral disturbances (Strzelczyk and Schubert-Bast, 2022).Indeed, among them, drugs such as benzodiazepines may have a negative impact on attention and behavior, provoking irritability, as occurred in one of our patients.
Several studies have reported that SLC6A1 patients with autism, from traits to ASD, are frequently related to developmental regression (Kalvakuntla et al., 2023).In particular, autism was reported in approximately one-third of patients (31.4%); in our series, one patient presented with autistic traits.We did not test patients with ASD-specific scales; therefore, the prevalence of ASD in these patients may be underestimated.
The main limitation is the small size of the sample and the lack of longitudinal data.Regarding the literature review, the unavailability of a single common database for all published cases and the lack of many details may have introduced some bias in the evaluation.In addition, considering the retrospective nature of this study, our data are heterogeneous in terms of age, timing/frequency and type of testing.We used different tests depending upon age and language barriers.It should be considered that RPM and Leiter-R, being two nonverbal cognitive tests, may overestimate patients' IQ, unlike the Wechsler scales, which provide full IQ.
Despite the limited number of patients and the few neuropsychological data reported in the literature, the overall data suggest the need for a specific assessment for ADHD, ASD and cognitive development in patients with epilepsy related to SLC6A1 pathogenic variants.
The high presence of behavioral disturbances and reduced cognitive performance underline the need for a targeted rehabilitation program, including strengthening attention skills.
Finally, early integrated intervention, including speech, psychological and psychomotor therapies during the early years of life, may be crucial to improve cognitive outcomes and support patients' daily life autonomy in adulthood.
The deep phenotyping of SLC6A1-related disorders, including cognitive and behavioral comorbidities, is also important considering the promising pharmacological perspectives as hypothesized in vivo and in vitro studies, using chemical and pharmacological chaperones as possible treatments in deficiency of GAT-1 expression and function (Kasture et al., 2023).
Table 1 reports the demographic, genetic, clinical, EEG, and neuroradiological features in our series of five patients.

TABLE 1
Clinical, genetic and neuropsychological data of our series.