Genotype-Phenotype Correlations and Functional Outcomes in Pediatric Patients with KCNQ2-Related Epilepsy: A Multicenter Observational Study in Korea

Article information

Ann Child Neurol. 2025;33(2):48-55

Publication date (electronic) : 2025 March 21

doi : https://doi.org/10.26815/acn.2024.00787

Eon Ah Kim, MD¹, Mi-Sun Yum, MD^,1, Seungbok Lee, MD^2,3, Jae So Cho, MD⁴, Jeehun Lee, MD⁵, Byung Chan Lim, MD²

¹Department of Pediatrics, Asan Medical Center Children’s Hospital, University of Ulsan College of Medicine, Seoul, Korea

²Department of Pediatrics, Seoul National University Children's Hospital, Seoul, Korea

³Department of Genomic Medicine, Seoul National University Hospital, Seoul, Korea

⁴Department of Pediatrics, Seoul National University Bundang Hospital, Seongnam, Korea

⁵Department of Pediatrics, Samsung Medical Center, Sungkyunkwan University School of Medicine Seoul, Korea

Corresponding author: Mi-Sun Yum, MD Department of Pediatrics, Asan Medical Center Children’s Hospital, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea Tel: +82-2-3010-3386, Fax: +82-2-2045-4054 E-mail: misun.yum@gmail.com

Received 2024 November 27; Revised 2025 January 15; Accepted 2025 January 16.

Abstract

Introduction

Potassium voltage-gated channel subfamily Q member 2 (KCNQ2)-related epilepsy comprises a spectrum of neonatal or infantile onset epileptic phenotypes that vary from self-limited epilepsy to developmental and epileptic encephalopathy (DEE) [1]. These conditions are caused by mutations in the KCNQ2 gene, which encodes a subunit of the voltage-gated potassium channel that regulates neuronal excitability [2]. The estimated incidence of KCNQ2-related epilepsy ranges from 2.93 to 3.59 per 100,000 live births [3], with seizures typically beginning within the first few days to weeks of life [4].

KCNQ2-related epilepsy exhibits a wide spectrum of clinical phenotypes [5]. Beyond the commonly observed self-limited neonatal epilepsy (SeLNE) and KCNQ2-DEE, there are less common manifestations. These include self-limited neonatal-infantile epilepsy, self-limited infantile epilepsy (SeLIE), neonatal encephalopathy with non-epileptic myoclonus, infantile or childhood-onset DEE, isolated intellectual disability without epilepsy, and myokymia [4]. The spectrum of KCNQ2-related neurodevelopmental disorders ranges from normal cognitive function to severe impairments in motor skills, behavior, and social interaction [6]. This variability carries significant long-term consequences. Adults with KCNQ2 encephalopathy often exhibit a high prevalence of intellectual disability and neuropsychiatric disorders. This occurs despite a reduction or cessation in seizure frequency during adulthood, although there remains a risk of seizure recurrence later in life [1,7,8].

Patients with DEE experience medication-resistant seizures and often develop severe intellectual and motor disabilities. Even when these patients achieve seizure freedom through antiseizure medication (ASM), their long-term outcomes remain generally poor. In contrast, patients with self-limited epilepsy usually respond well to ASMs and experience favorable outcomes [9]. Missense variants in KCNQ2 are frequently linked to more severe forms of DEE, whereas truncating mutations are typically associated with milder phenotypes, such as self-limited epilepsy [10]. However, the relationship between genotype and phenotype is not straightforward [11], suggesting that other genetic modifiers or environmental factors might also play a role in determining clinical outcomes [12]. This complexity underscores the importance of further research to elucidate the mechanisms driving these disorders.

This study investigated the phenotype of KCNQ2-related epilepsy among a pediatric cohort in Korea, specifically examining the relationship between genotype and phenotype and the associated comorbidities. By analyzing clinical data, genetic variants, and long-term functional outcomes comprehensively, the study sought to enhance understanding of KCNQ2-related epilepsy and its effects on neurodevelopment, as well as to investigate how comorbidities might influence patient prognosis.

Materials and Methods

This multicenter, observational study was conducted at three tertiary hospitals in Korea, starting in November 2023. It included patients who experienced seizures and had confirmed KCNQ2 mutations identified through genetic testing. All participants had a follow-up period of at least 6 months, which allowed for the assessment of comorbidities and their Functional Status Scale (FSS). The study evaluated the clinical features, genetic findings, and functional status of each patient using established scales. Additionally, data were collected on patient medical histories, types of seizures, electroencephalography (EEG) findings, and comorbid conditions. This study was approved by Institutional Review Board of Asan Medical Center (approval number:2023-0695), and informed consent was waived by the board.

Results

1. Patient demographics

This study collected data from three tertiary hospitals in South Korea, enrolling 20 patients, 45% of whom were male. The average age at first seizure onset was 484 days, ranging from 1 day to 12 years, with a mean follow-up period of 4 years and 1 month (range, 6 months to 12 years). The ages at seizure onset were categorized into three groups: neonates, infants, and children. Of the 20 patients, 13 (65%) were neonates, three (16%) were infants, and four (21%) were children. The study cohort included nine patients with self-limited epilepsy, comprising four without a confirmed family history and five with a confirmed family history. Among those with a confirmed family history, four experienced their first seizure during the neonatal period, and one during infancy. Additionally, 11 patients were diagnosed with DEE; five were classified in the neonatal seizure onset group, and six in the non-neonatal seizure onset group (Table 1).

Table 1.

Classification of KCNQ2-related epilepsy phenotypes of patients in the study cohort

2. Genetic information

All 20 enrolled patients exhibited unique genetic variants, with the exception of one sibling pair (patients 8 and 9) and another unrelated pair (patients 19 and 20), as detailed in Table 2 [10,14-27]. A comparison between the self-limited epilepsy group and the DEE group showed variations in both the location and type of mutations. In the self-limited epilepsy cases, the majority of variants (6/9, 67%) were located in the C-terminal region (Fig. 1). Nonsense mutations were the most prevalent type (4/9, 44%), followed by missense and frameshift mutations. Among the 11 patients with DEE, mutations were also primarily found in the C-terminal region (5/11, 45%). However, missense mutations were notably more common, representing 90% (9/11) of the cases (Fig. 2). Table 2 provides a comparison of the phenotypes associated with KCNQ2 mutations reported in previous studies and those observed in this patient cohort.

Table 2.

Genetic variants and clinical phenotypes in patients with KCNQ2-related epilepsy

Fig. 1.

Distribution of identified mutations within the potassium voltage-gated channel subfamily Q member 2 (KCNQ2) polypeptide chain. The mutation locations for various genetic variants were marked, ^awith the self-limited group indicated in blue, ^bwhile the developmental and epileptic encephalopathy group was highlighted in red.

Fig. 2.

Overview of (A, C) variant locations and (B, D) mutation types identified in patients with potassium voltage-gated channel subfamily Q member 2 (KCNQ2)-related epilepsy: self-limited and developmental and epileptic encephalopathy (DEE) phenotypes. SeLNE, self-limited neonatal epilepsy; SeLFNE, self-limited familial neonatal epilepsy; SeLFNIE, self-limited familial neonatal-infantile epilepsy.

3. Clinical phenotypes

In the self-limited epilepsy group, which included nine patients, the first seizure occurred between days 3 and 7 of life in four patients, and on day 124 in another patient. Additionally, five patients in this group had a family history of seizures, suggesting a possible hereditary component in this subgroup. In contrast, the DEE group, which consisted of 11 patients with more complex presentations, experienced an average seizure onset at 825 days.

EEG was performed for all patients at their initial presentation. In both groups, 80% (16/20) of the initial EEG backgrounds appeared normal. Among the remaining four patients, only one from the self-limited group exhibited a burst suppression pattern. In the DEE group, two patients showed burst suppression patterns, and one displayed slow background activities. However, EDs were observed in 75% (15/20) of the patients. These discharges were classified as focal (localized to one region), multifocal (spanning multiple regions), or generalized. Specifically, focal, multifocal, and generalized discharges were observed in eight, five, and two patients, respectively. The self-limited epilepsy group demonstrated significantly better seizure control than the DEE group, with all patients remaining seizure-free for over 1 year. Overall, 75% of the cohort (15/20) maintained a seizure-free status for more than 1 year (Table 3).

Table 3.

Clinical information and functional outcomes in patients with KCNQ2-related epilepsy

4. Outcomes

In the self-limited epilepsy group, sodium channel blockers were effective in four out of nine cases (44.4%), with phenytoin (DPH) and carbamazepine proving effective in three and one patient, respectively. For those with neonatal-onset epilepsy within this group, phenobarbital was effective in two patients. Additionally, levetiracetam (LEV), clobazam, and valproic acid each demonstrated efficacy in one patient. In the DEE group, sodium channel blockers were effective in four out of 11 cases (36%), with DPH effective in two patients, oxcarbazepine in one patient, and lamotrigine in one patient. Other medications, including vigabatrin, LEV, and perampanel, were effective in two, one, and one patient, respectively. To further evaluate patient outcomes, the FSS was employed to assess and compare functional status across both groups. Among the 20 patients, the average FSS was 7.6 (Table 4). Eleven patients were assigned a score of six, indicating no functional impairment. All nine patients in the self-limited epilepsy group achieved this score, confirming normal functional status across the group. In contrast, the DEE group exhibited an average FSS of 8.82, indicating mild dysfunction, with individual scores ranging from 6 to 15. Mild dysfunction in this group was often associated with impairments in communication and motor functioning, due to intellectual disability or autism spectrum disorder (ASD), while mental status and respiratory functioning remained normal in most cases. In the self-limited epilepsy group, no comorbidities were observed. However, in the DEE group, six out of 11 patients had intellectual disability or developmental disability, two had ASD, and two exhibited both intellectual disability and movement disorders. In one case, the patient is still an infant, and no comorbidities have been identified to date.

Table 4.

Functional Status Score outcomes of pediatric patients with KCNQ2-related epilepsy

Discussion

The findings of this study provide valuable insights into the genotype-phenotype correlation in KCNQ2-related epilepsy, highlighting distinct differences in genetic variants, clinical outcomes, and treatment responses between self-limited epilepsy and DEE. Consistent with prior research, missense mutations were predominantly associated with the DEE group, especially when located in the C-terminal region of the KCNQ2 gene. This region is commonly linked to more severe developmental impairments and treatment-resistant epilepsy, suggesting that missense mutations may significantly alter protein function, thus contributing to the severe neurological and developmental profiles observed in DEE. Conversely, truncating mutations, including nonsense and frameshift mutations, were more common in the self-limited epilepsy group. Interestingly, certain cases in this study highlighted the complexity of these genotype-phenotype correlations. For example, patient 15 exhibited a DEE phenotype, despite carrying a genetic variant previously associated with self-limited epilepsy. This observation illustrates that the genotype-phenotype correlations in KCNQ2-related epilepsy are not entirely predictable. It also suggests that additional genetic modifiers, epigenetic influences, or environmental factors may also influence the clinical variability in this disorder [28]. Regarding clinical outcomes, comorbidities were predominantly observed in the DEE group, whereas the self-limited epilepsy group had notably fewer associated conditions, indicating a more favorable neurodevelopmental outcome. These findings are consistent with results from previous studies. Functional outcomes, as assessed by the FSS, also reflected this distinction: the self-limited epilepsy group consistently demonstrated normal function, while those in the DEE group exhibited mild impairments, particularly in the communication and motor domains, likely influenced by comorbid intellectual disability or ASD. Sodium channel blockers demonstrated relatively higher efficacy in KCNQ2-related epilepsy by reducing hyperexcitability caused by the functional loss of potassium channels. In this study, they were used in a significant proportion of cases, with potentially responsive ASMs observed in 40% (8/20) of the total cohort. Specifically, in the self-limited epilepsy group, sodium channel blockers were effective in 44.4% (4/9) of patients, while in the DEE group, they were effective in 33.3% (4/11) of patients. However, the DEE group showed less effectiveness in seizure control, with seizures persisting in four out of 11 patients. Nevertheless, considering the intractability of DEE, sodium channel blockers can still be considered a first-choice treatment for seizure control in cases of KCNQ2 epilepsy.

This study has several limitations that warrant consideration. First, the small sample size and the focus on a single ethnic cohort limit the generalizability of the findings to other populations. Second, as this was a multicenter retrospective study that relied on data collected from each hospital, it faced challenges in obtaining detailed information on ASM usage and seizure types, highlighting inherent limitations in data collection. Third, although KCNQ2-related epilepsy typically presents within the first few days to weeks after birth, this study included patients with relatively older ages of seizure onset, introducing variability in phenotype analysis. Future studies should aim to address these limitations by including a larger, more ethnically diverse population and collecting comprehensive data on ASM usage, seizure types, and seizure onset patterns. Additionally, exploring potential differences in seizure onset age based on specific genetic variants could further clarify the genotype-phenotype correlation. By overcoming these challenges, future research could enhance our understanding of KCNQ2-related epilepsy and support the development of tailored therapeutic strategies to optimize patient outcomes.

In conclusion, KCNQ2-related epilepsy exhibits complexities that challenge the assumption that mutation type alone determines the clinical course. This underscores the importance of gaining a broader understanding of additional factors that may influence KCNQ2-related phenotypes. Furthermore, research should focus on identifying potential genetic modifiers or environmental factors that could account for the variability in KCNQ2 phenotypes, thus supporting tailored treatment planning. This study highlights the intricate relationship between genotype and phenotype in KCNQ2-related epilepsy, emphasizing the necessity for a personalized approach that adapts therapeutic strategies to individual phenotypes. Such an approach is aimed at optimizing neurodevelopmental outcomes and enhancing the quality of life for patients with KCNQ2-related epilepsy.

Article information Continued