Efficacy of anti‐seizure medications and alternative therapies (ketogenic diet, CBD, and quinidine) in KCNT1‐related epilepsy: A systematic review

Abstract

Objective

KCNT1‐related epilepsies encompass three main phenotypes: (i) epilepsy of infancy with migrating focal seizures (EIMFS), (ii) autosomal dominant or sporadic sleep‐related hypermotor epilepsy [(AD)SHE], and (iii) different types of developmental and epileptic encephalopathies (DEE). Many patients present with drug‐resistant seizures and global developmental delays. In addition to conventional anti‐seizure medications (ASM), multiple alternative therapies have been tested including the ketogenic diet (KD), cannabidiol (CBD—including Epidyolex © and other CBD derivatives) and quinidine (QUIN). We aimed to clarify the current state of the art concerning the benefits of those therapies administered to the three groups of patients.

Methods

We performed a literature review on PubMed and EMBase with the keyword “KCNT1” and selected articles reporting qualitative and/or quantitative information on responses to these treatments. A treatment was considered beneficial if it improved seizure frequency and/or intensity and/or quality of life. Patients were grouped by phenotype.

Results

A total of 43 studies including 197 patients were reviewed. For EIMFS patients (32 studies, 135 patients), KD resulted in benefit in 62.5% (25/40), all types of CBD resulted in benefit in 50% (6/12), and QUIN resulted in benefit in 44.6% (25/56). For (AD)SHE patients (10 studies, 32 patients), we found only one report of treatment with KD, with no benefit noted. QUIN was trialed in 8 patients with no reported benefit. For DEE patients (10 studies, 30 patients), KD resulted in benefit for 4/7, CBD for 1/2, and QUIN for 6/9. In all groups, conventional ASM are rarely reported as beneficial (in 5%–25% of patients).

Significance

Ketogenic diet, CBD, and QUIN treatments appear to be beneficial in a subset of patient with drug‐resistant epilepsy. The KD and CBD are reasonable to trial in patients with KCNT1‐related epilepsy. Further studies are needed to identify optimal treatment strategies and to establish predictive response factors.

Plain Language Summary

We performed an extensive review of scientific articles providing information about the therapeutic management of epilepsy in patients with epilepsy linked to a mutation in the KCNT1 gene. Conventional anti‐seizure treatments were rarely reported to be beneficial. The ketogenic diet (a medical diet with very high fat, adequate protein and very low carbohydrate intake) and cannabidiol appeared to be useful, but larger studies are needed to reach a conclusion.

Key points.

• We report a literature review on the benefits of conventional anti‐seizure medications (ASM), ketogenic diet (KD), cannabidiol (CBD), and quinidine (QUIN) for patients with KCNT1‐related epilepsy.

• Conventional ASM are rarely reported as beneficial.

• For patients with EIMFS, KD and CBD (in addition to ASM) are beneficial in more than half of cases.

• Quinidine may be helpful, but inconsistent findings make it impossible to predict its efficacy or whether adverse effects may manifest.

1. INTRODUCTION

The KCNT1 gene encodes a sodium‐gated potassium channel—K_Na 1.1, also known as Slack channel, or Slo2.2 channel ¹ which is highly expressed in brain. It plays an important role in neuronal excitability regulation ² and interacts with different proteins involved in neurodevelopment. ³ All known pathogenic variants in KCNT1 are gain of function (GoF) mutations resulting in a wide spectrum of developmental and epileptic disorders. Three main types of epilepsy are associated with KCNT1 mutations ⁴ : (i) epilepsy of infancy with migrating focal seizures (EIMFS) characterized by the onset of seizures before the age of six months, with focal seizures arising in both hemispheres and migrating from one cortical region to another within a seizure, and associated with profound neurodevelopment delay, ⁵ (ii) autosomal dominant or sporadic sleep‐related hypermotor epilepsy (AD)SHE, characterized by an onset in childhood with focal seizures with various motor manifestations during sleep, typically associated with mild to moderate intellectual disability, autistic features, or psychiatric comorbidities, (iii) various developmental and epileptic encephalopathies (DEE) including infantile epileptic spasms syndrome (IESS) and early infantile developmental and epileptic encephalopathy (EIDEE).

Pathogenic variants in KCNT1 are frequently associated with drug‐resistant epilepsies. ⁶ Hence, various alternative therapies have been tried, including the ketogenic diet (KD), cannabidiol (CBD), and the cardiac anti‐arrhythmic drug quinidine (QUIN), which is a pore blocker of the Slack channel. ⁷ Numerous scientific articles, primarily case reports or case series, report the use of conventional anti‐seizure medications (ASM), CBD, KD, or QUIN, but there is no consensus on their benefits. We performed a systematic review on the use of ASM, and the most commonly used alternative therapies (CBD, KD, and QUIN), in KCNT1‐related epilepsy.

2. MATERIALS AND METHODS

This systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta‐analyses (PRISMA) recommendations. ⁸

2.1. Data collection procedure

2.1.1. Data source and search

Electronic databases EMBase and PubMed were queried, using the free text term “KCNT1”. The search was restricted to English‐language publications made before June 1, 2023.

2.1.2. Selection of studies

We screened abstracts of manuscripts describing patient with KCNT1‐related epilepsy, and we read in their entirety those that met the inclusion criteria. When the abstracts did not provide sufficient information, we read the entire manuscript to decide whether or not to include the research in our study. All duplicate articles or patients were removed. We decided to include case reports and case series because of the scarcity of publications on treatments for KCNT1‐related epilepsy (Flow chart, Figure 1).

FIGURE 1.

Flow chart.

Inclusion criteria were:

1. Population: epileptic patients with KCNT1 variant—pathogenic, likely pathogenic variants according to the ACMG classification, or variant of uncertain significance with high pathogenicity score (based on PolyPHEN or SIFT scores) and concordant phenotype.

2. Intervention: Any anti‐seizure treatment including CBD (Epidyolex© or CBD derivatives), QUIN, and dietary therapies.

3. Outcomes: at least one of the following outcomes related to the intervention reported: seizure modification (in intensity and/or frequency), quality of life modification, adverse events, therapies at Last Follow‐Up (LFU).

Exclusion criteria were:

1. Patient without information about the KCNT1 variant, or harboring a benign or likely benign variant according to the ACMG classification.

2. Patients with additional pathogenic mutations in genes other than KCNT1 thought to contribute to the epilepsy phenotype.

2.2. Data extraction and evaluation of evidence

Level of evidence of selected papers was assessed with the Oxford Center for Evidence‐Based‐Medicine 2011 Levels of Evidence (OCEBM) (Table S1).

We extracted from the manuscript: sex, epileptic syndrome, types of seizures, age at onset, KCNT1 variants, age at LFU, medications tested, medications doses, and estimated response. When available, estimated response was expressed differently depending on the article (Table 1). We considered treatments as beneficial if they resulted in a reduction in the frequency, duration, intensity of seizures, and/or if they resulted in an improvement in quality of life and/or if they were reported as beneficial (even without details about the benefit) or temporarily effective by physicians. For the purposes of this analysis, we did not utilize a threshold (e.g., 50% reduction) to determine benefit as this information was not available in most published articles. For analyses of benefits provided by treatments, we removed the articles that did not provide information about benefit (e.g., article with information only about treatments at LFU or tried without information about benefits).

TABLE 1.

Details of selected articles and how the therapies' benefit was reported.

Source	Level of proof a	Design	N° of patients	Patient phenotype	Benefit evaluation
Da Xu et al. 7	1 (QUIN)	Systematic review	80	EIMFS, ADSHE, DEE	Systematic review about QUIN
Hughes et al. 9	1 (QUIN)	n‐of‐1 trial	1	EIMFS	QUIN: % of seizure frequency reduction ASM: “temporary response (…) and early developmental progress appeared relatively preserved” (Clonazepam, VPA) KD, CBD: not described
Mullen et al. 10	2 (QUIN)	Controlled randomized trial	6	(AD)SHE	QUIN: % of seizure frequency reduction ASM: ASM at LFU—benefit not evaluated KD, CBD: not described
Poisson et al. 11	3 (CBD)	Open‐label, prospective, interventional study	3	EIMFS	CBD: % of seizure frequency reduction, Caregiver Global Impression of Change (CGIC) ASM, QUIN, KD: noted if tried—asterisk for whom which denoted benefit, without specifying the nature of this benefit
Numis et al. 12	3 (QUIN)	Interventional prospective study	6	EIMFS	QUIN: video‐EEG and ECG telemetry every 3 months (to evaluate seizure burden) when stable dosing of quinidine in blood (goal serum quinidine level between 2 and 5 mg/L) ASM, KD, CBD: noted if tried—asterisk if it denoted benefit, without specifying the nature of this benefit
James et al. 13	4	Case‐report	1	EIMFS	ASM: list of ASM tried which “did not provide much benefit” CBD, KD, QUIN: not described
Ferretti et al. 14	4	Retrospective observational study	3	EIMFS	QUIN: evaluation of plasma level—% seizure frequency reduction—reduction of the seizure intensity—psychomotor improvement ASM, KD: at LFU / tried without benefit observed CBD: not described
Lu et al. 15	4	Case‐report	2	(AD)SHE	ASM: seizure freedom (Levetiracetam = behavioral disorder so dose decreased and Oxcarbamazepine added ➔ seizure stopped) QUIN, KD, CBD: not described
Cherian et al. 16	4	Case‐series of 9 patients from one family	9	(AD)SHE, DEE, EIMFS	ASM: at LFU + ASM tried QUIN: noted for 1 patient with >50% seizure reduction + Quality of Life improvement KD, CBD: not described
Na Xie et al. 17	4	Case‐reports	3	(AD)SHE	ASM: frequency and intensity of seizures (qualitatively) QUIN, KD, CBD: no described
Bonardi et al. 4	4	Observational retrospective study and literature review	248	EIMFS, (AD)SHE, DEE	ASM, QUIN, KD, CBD: noted if tried—underlined if most often reported to be effective, without specifying the nature of this “effectiveness”
Bushlin et al. 18	4	Case‐report	1	EIMFS	QUIN: seizure frequency reduction ASM: “modest benefit” for some, not effective for many – without specifying the nature of this benefit KD, CBD: not described
Borlot et al. 19	4	Observational retrospective study	27	EIMFS, DEE, (AD)SHE	KD, CBD, QUIN: Estimated response scale: marked improvement, >50% reduction in seizures; some improvement, 25%–50% reduction in seizures; minimal, <25% reduction in seizures; nil, no change in seizure frequency. ASM: at LFU, without information about a benefit
Kravetz et al. 20	4	Case‐report	1	EIMFS	QUIN: evaluation of plasma level—seizure frequency reduction ASM: at initiation of QUIN and at LFU—not details about benefit CBD, KD: not described
Takase et al. 21	4	Case‐report	1	EIMFS	QUIN, ASM: seizure frequency reduction CBD, KD: not described Other (antitussive drugs): seizure frequency reduction
Fitzgerald et al. 22	4	Observational retrospective study	43	EIMFS, DEE	QUIN: % seizure frequency reduction ASM: at LFU, without information about the benefit (patients also described by Bonardi et al.)
Kuchenbuch et al. 6	4	Observational retrospective study	17	EIMFS	QUIN: seizure frequency, neurodevelopment ASM, KD: at LFU, without information about the benefit CBD: not detailed
Passey et al. 23	4	Case report	1	EIMFS	QUIN: seizure frequency and intensity/ seizure freedom KD: seizure frequency and intensity ASM: no benefit identified—no detailed on which criteria
Abdelnour et al. 24	4	Case‐series of 3 patients and review of the literature	3	EIMFS	QUIN: % seizure frequency reduction KD: ineffective (not detailed on which criteria) ASM: not detailed (we known the number tried without benefit observed), CBD: not detailed
Yoshitomi et al. 25	4	Case‐series	4	EIMFS, DEE	QUIN: % seizure frequency reduction ASM, KD: tried without benefit CBD: not detailed
Cataldi et al. 26	4	Case‐report	1	(AD)SHE	ASM: seizure frequency/seizure freedom QUIN, KD, CBD: not detailed
Datta et al. 27	4	Case‐series	2	EIMFS, DEE	ASM, QUIN, CBD: seizure frequency reduction KD: not detailed
Gertler et al. 28	4	Case report	1	DEE	ASM, KD: seizure frequency (qualitative) QUIN, CBD: not detailed
Alsaleem et al. 30	4	Case report	1	EIMFS	ASM, KD, QUIN: seizure frequency and/or severity reduction (qualitative assessment) CBD: seizure frequency and/or severity reduction (stopped because of adverse events)
Barcia et al. 31	4	Observational retrospective study	17	EIMFS	QUIN: seizure frequency ASM: Tried—without information about benefit (just one information: “Combination of STP and CZP relatively improved at age of 3 years”) KD: Tried—without information about benefit CBD: not detailed
Dilena et al. 32	4	Case‐series of two patients and literature review	2	EIMFS	QUIN: seizure frequency CBD, ASM, KD: tried, not beneficial (no precision about what was assessed for the benefit)
Patil et al. 29	4	Case‐series	2	EIMFS	QUIN: % seizure frequency reduction ASM, KD: seizure frequency reduction (qualitative) CBD: not detailed
Madaan et al. 33	4	Case‐report	1	EIMFS	QUIN, KD: seizure frequency ASM: all tried—not beneficial (probably based on seizure frequency) CBD: not described
McTague et al. 34	4	Observational retrospective study	12	EIMFS	ASM, KD: detail of treatments for which “best response to treatment” QUIN, CBD: not detailed
Rubboli et al. 35	4	Case‐series	4	(AD)SHE	ASM: at LFU, benefit not detailed CBD, KD, QUIN: not described
Kawasaki et al. 36	4	Case‐series	3	EIMFS	ASM: “ineffective antiepileptic drugs / Transiently effective antiepileptic drugs” CBD, KD, QUIN: not detailed
Baumer et al. 37	4	Case report	1	EIMFS	QUIN, ASM: tried—benefit not detailed CBD, KD: not described
Fukuoka et al. 38	4	Case‐report	1	EIMFS, DEE	QUIN, ASM, KD: seizure frequency (qualitative) CBD: not described
Mori et al. 39	4	Case report	1	EIMFS	ASM, KD: seizure frequency CBD, QUIN: not described
Rizzo et al. 40	4	Case‐report and functional study	2	EIMFS	ASM: seizure frequency, interaction with the environment KD, CBD, QUIN: not detailed
Selioutski et al. 41	4	Case‐series	2	EIMFS	KD: “ineffective for seizure control” CBD, QUIN, ASM: not described
Ohba et al. 42	4	Observational study	9	EIMFS, DEE	ASM, KD: temporarily effective/ineffective drug (not details about to criteria for being effective) CBD, QUIN: not detailed
Mikati et al. 43	4	Case‐report	2	(AD)SHE	QUIN: % seizure frequency reduction KD: “failed” (without details) CBD, ASM: not described
Vanderver et al. 44	4	Case report	1	DEE	ASM: “seizure control” CBD, QUIN, KD: not detailed
Bearden et al. 45	4	Case‐report	1	EIMFS	ASM: tried “without effect on seizures” KD, QUIN: seizure frequency CBD: not described
Ishii et al. 46	4	Case‐series	2	EIMFS	ASM: seizure frequency (qualitative) KD: “ineffective” (without more details) CBD, QUIN: not described
Barcia et al. 47	4	Observational retrospective study	6	EIMFS	ASM: at LFU and ASM tried, without details about benefits CBD, KD, QUIN: not described
Merdariu et al. 48	4	Case‐report	1	EIMFS	ASM: seizure frequency reduction/seizure freedom KD: “failed” (without more details) CBD, QUIN: not described

Abbreviations: (AD)SHE, autosomal dominant or sporadic sleep‐related hypermotor epilepsy; ASM, anti‐seizure medications; CBD, cannabidiol; DEE, developmental and epileptic encephalopathies; EIMFS, epilepsy of infancy with migrating focal seizures; KD, ketogenic diet; LFU, last follow‐up; QUIN, quinidine.

^a According Oxford Center for Evidence‐Based‐Medicine 2011 Levels of Evidence (OCEBM).

2.3. Data analysis and group characteristics

Patients were categorized into groups based on their phenotype: “EIMFS,” “(AD)SHE,” and “other DEE”. Age at last follow up (LFU) was adjusted to the unit “years”. This information was particularly useful for patients with EIMFS as it gave us an indication of the phase of the disease: sporadic isolated seizures at onset, stormy phase, or consolidation. ⁶ Based on Kuchenbuch et al. data, ⁶ we decided to consider patients younger than 1.3 years old as in the “stormy” phase and patients over 1.3 years old as being in the “consolidation” phase. We did not consider the “sporadic isolated seizure” phase as it is not constantly reported in all patients. Conventional anti‐seizure medications were categorized as “ASM,” with the exception of Epidyolex©, which was grouped with “CBD” data.

2.4. Statistical analysis

As the response to some treatments may vary depending on the genotype and where the variant is located within the protein's many functional domains, ⁷ , ²² , ⁴⁹ we performed Fisher's exact tests to check whether this was the case in our investigation.

3. RESULTS

The search yielded 605 papers of which 43 met the inclusion criteria (Figure 1).

Among the 43 studies, two of them had a level of proof of 1 according to the OCEBM regarding the QUIN, ⁷ , ⁹ one had a level of proof of 2 regarding the QUIN, ¹⁰ one had a level of proof of 3 for CBD, ¹¹ one with a level of proof of 3 regarding the QUIN ¹² and the remaining studies had a level of proof of 4.

From the 43 studies, we extracted the data of 197 patients with KCNT1‐related epilepsy. Totally, 135 of them presented with EIMFS, 32 of them with (AD)SHE, and 30 with DEE (Table S2).

3.1. Results in relation to ASM trials

We identified 190 patients for whom information was accessible for ASM (Table 2).

TABLE 2.

Most common ASM treatment prescribed to patients with KCNT1‐related epilepsy.

	EIMFS (n = 129)			DEE (n = 30)			AD(SHE) (n = 31)
Medication	Prescribed at LFU (/72)	Number of patients who have tried this medication (/129)	Number of patients who have observed benefits a	Prescribed at LFU (/15)	Number of patients who have tried this medication (/30)	Number of patients who have observed benefits a	Prescribed at LFU (/16)	Number of patients who have tried this medication (/31)	Number of patients who have observed benefits a
LEV	31	85	9 (15%)	7	22	1 (5.8%)	2	9	1 (12.5%)
PB	23	87	12 (18.4%)	3	12	2 (18.2%)	2	5	1 (33.3%)
TPM	23	70	5 (10.4%)	3	14	3 (25%)	7	13	1 (16.7%)
VPA	13	57	4 (10.5%)	3	14	0	9	16	4 (36.3%)
CLB	17	55	7 (15.9%)	5	12	2 (22.2%)	2	10	3 (37.5%)
CLZ	15	53	5 (16.7%)	1	8	0	5	6	0
GVG	10	39	4 (25%)	2	10	1 (11.1%)	0	0	0
CBZ	3	34	3 (13%)	1	9	1 (11.1%)	12	19	3 (30%)
KBr	8	20	8 (40%)	0	2	0	0	0	0
LTG	6	23	2 (13.3%)	0	10	1 (10%)	1	6	1 (20%)
PHT	4	47	4 (11.8%)	0	4	0	2	10	1 (12.5%)
OXC	1	22	0	4	5	2 (50%)	2	6	3 (50%)
ZNS	3	19	2 (11.7%)	1	4	1 (25%)	0	1	0
LCS	1	16	1 (6.2%)	1	3	0	2	3	0
CZP	2	14	3 (25%)	0	0	0	0	0	0
STP	3	14	4 (50%)	1	3	1 (33.3%)	0	1	0
NZP	3	7	2 (33.3%)	0	2	0	0	0	0
PER	3	4	1 (33.3%)	0	3	0	0	4	1 (25%)
FBM	0	3	0	0	2	0	0	1	100%
GPT	1	5	0	1	4	0	0	1	0
RUF	1	4	0	0	1	0	0	0	0

Note: Colored box: most prescribed treatments.

Abbreviations: (AD)SHE, autosomal dominant or sporadic sleep‐related hypermotor epilepsy; ASM, anti‐seizure medications; CBZ, carbamazepine; CLB, clobazam; CLZ, clonazepam; CZP, clorazepate; DEE, developmental and epileptic encephalopathies; EIMFS, epilepsy of infancy with migrating focal seizures; FBM, felbamate; GPT, gabapentin; GVG, vigabatrin; KBr, potassium bromide; LCS, lacosamide; LEV, levetiracetam; LFU, last follow‐up; LTG, lamotrigine; NZP, nitrazepam; OXC, oxcarbamazepine; PB, phenobarbital; PER, perampanel; PHT, phenytoin; RUF, rufinamide; STP, stiripentol; TPM, topiramate; VPA, valproic acid; ZNS, zonisamide.

^a NB: The percentages next to the figures in the columns (number of patients who have observed benefits) correspond to the percentages in relation to the number of patients for whom information on the benefit provided by ASM was available.

For patients in the EIMFS group (n = 129):

The most commonly prescribed ASM at LFU, including all stages of the disease, were:

1. Levetiracetam: still prescribed in 31/72 patients (mean age at LFU = 3.6 years) and already tested in 85/129 patients before a mean age of 4.4 years. A benefit was noted for 9 of them (Figure 2).

2. Topiramate: still prescribed in 23/72 patients (mean age at LFU = 4.9 years) and already tested in 70/129 patients before a mean age of 5.5 years. A benefit was noted for 5 of them (Figure 2).

3. Phenobarbital: still prescribed in 23/72 patients (mean age at LFU = 2.7 years) and already tested in 87/129 before a mean age of 4.7 years. A benefit was noted for 12 of them (Figure 2).

FIGURE 2.

Reported benefits for the most prescribed anti‐seizure medications for EIMFS patients (all ages and before 1, 3 years old) and (AD)SHE patients. Details of benefits described for the most “beneficial” anti‐seizure medications for EIMFS patients. (AD)SHE, autosomal dominant sleep hypermotor epilepsy; ASM, anti‐seizure medications; EIMFS, epilepsy of infancy with migration focal seizures.

Despite having a very small number of patients, the drugs with the highest rate of observed benefit were:

1. Potassium bromide: benefit noted in 40% (8/20)—described as beneficial (without details about the benefit) for 5 patients, ⁴ , ¹² as a temporarily effective drug for 2 patients, ³⁶ , ⁴² and associated with a partial and uncertain effect for 1 patient. ⁴⁰

2. Stiripentol: benefit noted in 50%, but always described as beneficial in association: seizure‐free period with a combination of Levetiracetam, Clobazam, and Stiripentol ⁶ , ⁴⁸ ; modest benefit observed when Vigabatrin and Stiripentol added ¹⁸ ; improvement observed with combination of Stiripentol and Clonazepam ³¹ ; best response to treatment with Stiripentol, Levetiracetam, and Clonazepam in combination. ³⁴

3. Nitrazepam: benefit noted in 33.3%, described to be the treatment with the best response for 1, ³⁴ and being part of an association leading to seizure control (Topiramate + Nitrazepam + Levetiracetam) for the other one. ²⁰

Regarding patients under 1.3 years old at LFU (n = 34, ages ranging from 2 days to 15 months, with an average age of 9.7 months), assuming that they were still in the “stormy phase,” the most prescribed drugs were:

1. Phenobarbital: still prescribed in 10/20 and already tested in 27/34 before a median age of 9.6 months. A benefit was noted for 5 of them (Figure 2).

2. Levetiracetam: still prescribed in 8/20 patients and already tested in 20/34 before a median age of 10.8 months, with a benefit noted for 2 of them (Figure 2).

3. Clobazam: still prescribed in 7/20 patients and already tested in 18/34 before a median age of 10.8 months. A benefit was noted for 4 of them (Figure 2).

Only two patients had tested Stiripentol, ¹⁸ , ³¹ and one of them experienced modest benefit with a combination of Stiripentol and Vigabatrin. ¹⁸ One patient had tested Nitrazepam without benefit noted with this treatment. ⁴⁶

Potassium bromide (benefit noted in 2/6—described as beneficial without more details for 1, and as transiently effective for the other) and Clobazam (benefit in 4/18—described as temporarily effective for 3, ³⁶ , ⁴² and as being part of an association associated with seizure frequency reduction ³⁰ ) were frequently reported to be beneficial in patients under 1.3 years of age.

For patients in the AD(SHE) group (n = 31):

The three most prescribed ASM at LFU were:

1. Carbamazepine: still prescribed in 12/16 patients and already tested in 19/31. A benefit was noted for 3 of them (Figure 2).

2. Sodium valproate: still prescribed in 9/16 and already tested in 16/31 with a benefit noted for 4 of them (Figure 2).

3. Topiramate: still prescribed in 7/16 patients and already tested in 13/31 with a benefit noted in 1 (Figure 2).

A benefit was also observed in 37.5% (3/8) of cases with clobazam (“most often reported to be effective” for the 3 patients ⁴ ) and 50% (3/6) with oxcarbamazepine (“most often reported to be effective” for 2 patients, ⁴ and controlled seizures for the last one ¹⁵ ).

For patients in the DEE group (n = 30):

The most prescribed ASM at LFU were:

1. Levetiracetam: still prescribed in 7/15 patients (mean age at LFU = 3.8 years.) and already tested in 22/30 before the mean age of 7 years., with a benefit noted for one, in association with other ASM (seizures controlled on oxcarbazepine, topiramate, levetiracetam, and zonisamide ⁴⁴ ).

2. Clobazam, still prescribed in 5/15 patients (mean age at LFU = 3.8 years) and already tested in 12/30 before the mean age of 8.5 years with a benefit noted for 2 of them (reported to be effective, without more details about the benefit for 2 patients ⁴ ).

Finally, stiripentol, zonizamide, and oxcarbamazepine had the best benefit rate, although the samples were very small (respectively 1/3, 1/4, and 2/5) and with a wide variety of different epileptic syndromes (IESS, EIDEE, and unlabeled DEE).

3.2. Results in relation to KD trials

Totally, 25 articles mentioned the use of a KD, involving 82 patients in all (Table S3):

For patients in the EIMFS group (n = 70):

Benefit of the KD was assessed in 40/70 patients and was observed for 25/40 (62.5%) of patients:

A reduction in seizure frequency was documented by clinicians and caregivers for 5 of them, ¹⁹ with 3 patients showing a reduction of more than 50% of seizures frequency, 1 showing a reduction of 25%–50% and 1 a reduction of less than 25%.

For the other patients, the benefit noted was stated as resulting in a nonquantified decrease in seizure frequency and/or intensity (for 3 patients ²³ , ³⁹ , ⁴⁵ ), or was simply described as “effective”/“beneficial” ⁴ , ¹¹ , ²² , ²³ , ³⁰ , ³⁴ , ³⁹ , ⁴⁵ (for 12 patients), or even temporarily effective for 5 of them. ⁴⁶

The age of diet initiation was reported only for 13 patients, ¹⁹ , ²³ , ³⁰ , ³² , ³³ , ³⁹ with an average age of 12 months (ranging from 6 weeks to 43 months; before 1.3 years for 10). For others, having the age at LFU, we were able to deduce that the KD was introduced/tried before 1.3 years in additional 15 patients. In total, 25/28 patients had started the KD before 1.3 years.

Among patients who began the KD prior to 1.3 years, information about benefit of the KD was available for 16/25, with a benefit for 9 of them.

KD ratio was given for 11 patients: Four had a ratio of 4 g of fat to 1 g of protein, one a ratio of 4.25:1, one a ratio of 3.5:1, one a ratio of 3.25:1, and four a ratio of 3:1.

Duration of the diet was given for 11 patients, varying from 0.5 to 49 months, with a mean duration of 11 month (Cf Table S3).

For patients in the (AD)SHE group (n = 1):

Only 1 patient had tried the KD, and it was described as ineffective. ¹⁹

For patients in the DEE group (n = 11):

The benefit of the KD was assessed in 7 of 11 patients. For 4/7 (57.1%), there was a benefit associated with the diet. Details of the KD were given for 3 patients, with a ratio of 4:1 in all three.

In order to investigate if the genotype influences the response to the diet, we combined the three aforementioned groups. All phenotype combined, we found no differences between the benefit rate observed with variants in functional areas (RCK domains, NAD‐binding domain, S5) versus others: 23/36 in the “functional variant group” versus 6/12 in the “other localization” group, p = 0.5 (Fisher's exact test).

3.3. Results in relation to CBD trials

Eight articles mentioned the use of CBD derivatives, including an open‐label, prospective, interventional study classifying it as level 3 of the OCEBM classification. ¹¹ The other studies were classified as level 4. A total of 25 patients had tested different types of CBD derivatives and CBD oil (Table 3).

TABLE 3.

CBD trials: list of article and patients.

References	Level of proof a	Number of patients	Phenotype (N)	Age at onset	Dose (N)	CBD formulation	Beneficial? (N)	Adverse events? (N)
Poisson et al. 11	3	3	EIMFS (3)	15, 21 and 47 months	25 mg/kg/day (2), 30 mg/kg/day (1)	EPIDYOLEX©, synthetic CBD	Yes (1), No (2)	No (2), increase of seizure frequency (1)
Borlot et al. 19	4	7	EIMFS (7)	NA	10 mg/kg/day (3), 7,5 mg/kg/day (1), 4 mg/kg/day (1), 0,5 mg/kg/day (1), 0,25 mg/kg/day (1)	CBD/THC (variable formula)	Yes: “some”(2), “marked” (3); No (2)	No
Dilena 32 (2)	4	1	EIMFS	Between 4 and 9 months	15 mg/kg/day	Not precised	No	No
Alsaleem et al. 30	4	1	EIMFS	NA	NA	EPIDYOLEX©	No	Diarrhea and vomiting resulting in moderate dehydration (1)
Numis et al. 12	4	2 b	EIMFS	NA	NA	EPIDYOLEX©, synthetic CBD	No
Fitzgerald et al. 22	4	5	EIMFS (5)	NA	NA	EPIDYOLEX©	Yes (1), NA (4)	NA
Bonardi et al. 4	4	7	EIMFS (4), DEE (2), SHE (1)	NA	NA	Not precised	NA	NA
Datta et al. 27	4	1	DEE	Around 3 years old	NA	Not precised	NA	Worsening seizure frequency

Abbreviations: (AD)SHE, autosomal dominant or sporadic sleep‐related hypermotor epilepsy; CBD, cannabidiol; DEE, developmental and epileptic encephalopathies; EIMFS, epilepsy of infancy with migrating focal seizures; N, number of patients; NA, nonavailable; THC, tetrahydrocannabinol.

^a According Oxford Center for Evidence‐Based‐Medicine 2011 Levels of Evidence (OCEBM).

^b Two same patients than Poisson et al. 2020 (patient 2 and 3).

For patients in the EIMFS group (n = 21):

In total, 21 EIMFS patients were reported to have tried different types of CBD derivatives:

1. Epidyolex© (n = 8). ¹¹ , ²² , ³⁰

2. Variable formulations of CBD plus THC (n = 7). ¹⁹

3. Synthetic CBD (n = 1). ¹¹

4. CBD without precise details about the formulation (n = 5). ⁴ , ¹⁵

Among the 8 patients treated with Epidyolex ©, the effect of treatment was reported in only 3, with a benefit in one out of three (dosage of 25 mg/kg/day, initiation at 3.9 years, and follow‐up over several years, with an impression of a decrease in seizure intensity, but no change in frequency). ¹¹ One patient experienced digestive side effects (diarrhea and vomiting) ³⁰ and another observed an increase in seizures frequency. ¹¹ Among the 7 patients taking CBD plus THC formulation, a benefit was observed for five of them (>50% seizure frequency reduction for 3 of them, and 25%–50% seizure frequency reduction for 2 of them), without side effects reported. Of the remaining 5 patients for whom we did not have precise details of the CBD formulation used, the benefit was assessed and estimated to be nil in one case. No benefit was reported for the patient taking synthetic CBD either.

Overall, a benefit was noted with these different forms of CBD in 50% (6/12) of cases.

For patients in the (AD)SHE group (n = 1):

Only one patient had previously received CBD, but we had no information about the dose or benefit of this treatment. ⁴

For patients in the DEE group (n = 3):

Three patients had already tried CBD ⁴ , ²⁷ (no information about the formulation), with benefit noted in one of them, ⁴ without any information on the dose or duration of treatment. Another patient had experienced a worsening of his seizures at the time of trying this treatment. ²⁷

3.4. Result in relation to QUIN trials

We identified 26 articles, representing a total of 75 patients who had tried QUIN (Table S2).

For patient in the EIMFS group (n = 57):

Information on the benefit of QUIN was available for 56 patients.

25 (44.6%) patients experienced a benefit with this treatment:

1. 2/25 had become seizure‐free: one with a QUIN + topiramate + nitrazepam treatment, ²² with a sustained response, and for the second one, seizures were controlled with a QUIN + topiramate treatment. ²⁰

2. 16/25 had experienced >50% reduction in seizure frequency. The long‐lasting response was not available for 10 of them, and 7/9 experienced QT interval prolongation during the trial.

3. 7/25 experienced a benefit with the therapy (reduction in seizures <50%, “minor” relief, etc.) and 2 of them experienced QT interval prolongation. ²¹ , ²²

Among patient who did not observed benefit with QUIN, 7/31 experienced QT prolongation ¹⁸ , ²² , ²⁴ , ²⁵ and one experienced profuse diarrhea and vomiting. ⁹

We then investigated whether the nature of the genetic variant could influence the response to QUIN.

We analyzed the response observed for the most frequent variants and we grouped the responses at the level of the different functional domains of the protein (Table 4). Statistically, we did not observe any difference between the benefit rate observed in patients with variants within the functional domains (RCK1 and RCK2, NAD‐binding domain, S5) versus those without (Fisher's exact test performed, 19/44 vs. 6/12, p = 0.75). Interestingly, we observed a statistical difference when RCK2 + NAD were grouped together versus the rest of the mutations (Fisher's exact test performed, 12/16 vs. 13/40, p = 0.007), as suggested by Fitzgerald et al. ²² Fourteen mutations have been examined in vitro, with in vitro and in vivo discrepancies in the response. ¹⁰ , ³² , ⁴⁰ , ⁴³ , ⁵⁰

TABLE 4.

Summary of the number of EIMFS patients having tested QUIN for whom there is benefit or not.

	Number of patients with benefit observed	Number for whom QUIN was ineffective	Number with adverse events	% beneficial
Variant
p.Gly288Ser	3	4	0	43
p.Arg428Gln	3	6	0	33.3
p.Arg474Lys	0	1	0	0
p.Arg474His	1	7	3	12.5
p.Ala934Thr	2	2	0	50
p.Arg950Gln	3	0	0	100
Domain
S5	0	2	0	0
pore	3	5	0	37.5
NAD binding domain	1	1	0	50
RCK1	7	19	3	27
RCK2	11	3	0	78.6
Other	3	1	0	75

Note: Distribution by most frequent variants (A), and by protein domain (B).

Abbreviations: EIMFS, epilepsy of infancy with migrating focal seizures; QUIN, quinidine.

The dosages experienced by the patients ranged from 20 to 126 mg/kg/day, with a standard dosage of around 40 mg/kg/day. Moving toward very high dosages did not seem to increase efficacy, but led to an increase of adverse events, especially QT prolongation. Blood levels were not available for most of the patient.

For patients in the (AD)SHE group (n = 8):

None of the 8 patients observed any benefit from treatment with QUIN. They had four different mutations: p.Arg398Gln (RCK1), p.Tyr796His, p.Arg928Cys (RCK2), and p.Arg950Gln (RCK2).

For patients in the DEE group (n = 10):

A benefit was assessed in 9 patients and was observed in 66.7% (6/9), at doses ranging from 20 to 80 mg/kg/day.

The number of patients was too small to produce descriptive statistics on this sample. Nine different mutations were represented.

4. DISCUSSION

To our knowledge, this is the first literature review to include both a review of ASM and alternative treatments such as the KD, CBD, and QUIN for patients with KCNT1‐related epilepsy.

The mains findings of our study are about the use of KD for KCNT1‐related epileptic patients: KD is the treatment which showed the best benefit rate in EIMFS patients (62.5%) and DEE patients (57.1%). Our study corroborates the results of Lin et al. ⁴⁹ and encourages the implementation of KD in patients with KCNT1‐related epilepsy. Its feasibility, safety, and efficacy for pharmacoresistant epilepsy in infants under 2 years old, especially for those with genetic etiology, has already been shown, ⁵¹ and this also applies to KCNT1‐related epilepsies.

Cannabidiol also showed a significant benefit rate for EIMFS patients (beneficial for half of the patients). We list a wide range of CBD derivatives that were prescribed, including Epidyolex© which was reported in few cases (seven, with benefit assessed in 3/7). Because of the small number of patients taking CBD and the inconsistent dosages (not always specified), further studies are needed. Another limitation in our review was that CBD was always simultaneously prescribed with KD in EIMFS patients, making it difficult to discern the distinct benefit of either. Despite some limitations, these findings motivate us to start these patients on a KD and introduce CBD rapidly.

Concerning QUIN, we found more optimistic results (benefit in 44.6% EIMFS patients and in 66.7% DEE patients) than those of Da Xu et al. ⁷ and Fitzgerald et al. ²² who used a tighter definition of “beneficial” (e.g., reduction of at least 50% of seizures frequency). Modifications in seizures frequency or in quality of life were assessed (qualitatively or quantitively) by clinicians or caregivers depending on the articles. This can lead to biases, such as the placebo effect which can be significant for parents who have been waiting so long for a treatment with high expectations on the benefit for their child. Furthermore, the persistence of benefit at 3 months was almost never documented, and when it was, the persistence of efficacy at 3 months was very low. ²² On a related note, we carried out our analyses by phenotype, and it should be noted that QUIN was not effective in any of the patients with the SHE phenotype. However, only 8 patients with SHE phenotype had been able to test QUIN, including 4 from the same family. For EIMFS patients, interestingly, our study underlines Fitzgerald's finding that QUIN could be more effective in patients with variants in RCK2 and NAD domains. We need to confirm this result with prospective studies on a larger number of patients. However, it is worth noting that the risk of potentially fatal cardiac arrythmias secondary to QUIN use, as well as the need for intensive monitoring, may limit the utility of this drug in practice.

ASM other than Epidyolex© showed the lower benefit rate: around 10% of EIMFS patients (Figure 2), 4%–21% of DEE patients, and 8%–25% of patients with (AD)SHE experienced benefits with the most frequently prescribed drugs for their phenotype. Certain treatments that are not commonly used in clinical routine have therefore been little reported in the literature. Rufinamide, for example, which has already been described as potentially beneficial in the treatment of seizures in patients with EIMFS, ⁵² had been tested in only 5 patients in our review, with information on benefit available for 3 of them. Moreover, considering the number of drugs used in a relatively short period of time, the subjective character of responses and the fact that they were associated with a range of various polytherapies combinations, it was impossible to determine the intrinsic value of each ASM. It was not possible to study the benefits of combination therapies because they were all different and it was not possible to establish recurrently reported beneficial associations.

One limitation of most reports of treatment response in KCNT1‐related epilepsy is that the natural history of the disease is not normally considered when evaluating the benefit of a particular treatment. There is a tendency for most patients with KCNT1‐related epilepsy to have a reduction in seizures over time, and if a therapy is initiated shortly before this natural decrease, both patients and families may attribute a reduction in seizures to a medication, when the effect is actually due to the normal course of the disease. This bias could potentially be averted in future studies by comparing response to trialed therapies against natural history data.

The inclusion of case series and case reports, along with the impossibility to standardize the evaluation of treatment benefit due to the diversity of the studies, also constitutes a significant bias.

Finally, other promising targeted therapies are not detailed in this article, such as antisense oligonucleotides ⁵³ or KCNT1‐targed small molecule inhibitors ⁵⁴ due to lack of published data. Several of these therapies appear promising in cell models or mouse models, and we anticipate that these therapies may move into clinical trials in the near future.

5. CONCLUSION

Our data show that patients with KCNT1 mutations have often drug‐resistant epilepsy, but may benefit from some therapies such as CBD or the KD. Alternative treatments such as QUIN could also be trialed in a research setting with appropriate monitoring for cardiac side effects, although further studies are necessary to determine whether any potential benefit outweigh the risks. In addition, given the low reported efficacy of most conventional ASM in this population, reduction in the total number of sedating medications as high doses of benzodiazepines should be a priority whenever possible.

Further studies are now needed to identify optimal treatment strategies and to establish predictive response factors. The use of N of 1 methodologies and a better knowledge of natural history will help to better evaluate future therapies.

AUTHOR CONTRIBUTIONS

M.G. collected data and wrote the manuscript. R.N. conceived and designed the study and provided study supervision. J.W. and D.B. reviewed the manuscript. All authors edited the manuscript.

CONFLICT OF INTEREST STATEMENT

The authors report no financial disclosures or competing interests relevant to the study.

ETHICS STATEMENT

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Supporting information

Data S1:

Figure S1:

Gras M, Bearden D, West J, Nabbout R. Efficacy of anti‐seizure medications and alternative therapies (ketogenic diet, CBD, and quinidine) in KCNT1‐related epilepsy: A systematic review. Epilepsia Open. 2024;9:1176–1191. 10.1002/epi4.12975