MBOAT7 encephalopathy: Characterizing the neurology and epileptology

Abstract

Objective

Biallelic pathogenic MBOAT7 variants are associated with neurodevelopmental disorders, intellectual disability (ID), epilepsy, and neuropsychiatric disorders such as attention‐deficit/hyperactivity disorder and autism spectrum disorders. We aimed to characterize the epilepsy phenotype in a cohort of patients affected by this syndrome.

Methods

We describe epilepsy features, electroencephalography, magnetic resonance imaging (MRI) findings, antiseizure treatment response, and neurodevelopment of 15 patients with biallelic MBOAT7 variants.

Results

All 15 patients had ID or developmental delay (DD). Twelve suffered from epilepsy, with mean age at seizure onset of 36 months (range = 2 months–6.5 years) and 10 of 12 showing signs of DD before seizure onset. Patients with epilepsy presented with focal motor seizures with impaired awareness (n = 3), focal tonic–clonic seizures and epileptic spasms (n = 1), focal to bilateral tonic–clonic seizures (n = 1), unknown onset bilateral tonic–clonic seizures (n = 2), myoclonic seizures (n = 4), myoclonic–atonic seizures (n = 1), atonic seizures (n = 1), tonic seizures (n = 1), and myoclonic absences (n = 2). Seizure freedom was achieved in 66.7% (8/12), with variable antiseizure treatment regimes. We reviewed electroencephalograms of the patients with epilepsy. Background activity was normal in 64%, whereas 36% had either a generalized or a focal slowing. Interictal epileptiform discharges (IEDs) were reported in 83%. Generalized spikes/polyspikes were found in 53%, multifocal IEDs in 23%, and parasagittal focal IEDs in 26%. The most frequent abnormal brain MRI findings, reported in 58% of patients, included high‐intensity signal in T2 and fluid‐attenuated inversion recovery (FLAIR) sequences in dentate nuclei and globus pallidus. Biallelic missense variants seemed to be associated with better cognitive and motor outcomes compared to truncating variants and in‐frame deletions.

Significance

Biallelic MBOAT7 variants are associated with global developmental impairment in all affected patients and epilepsy in the majority. The seizure semiology is heterogenous. One third of our cohort had persistent seizures despite treatment. The most frequent MRI findings were hyperintensities in T2/FLAIR sequences in dentate nuclei and globus pallidus.

Key points.

• MBOAT7 encephalopathy is a rare autosomal recessive cause of neurodevelopmental disorders and epilepsy.

• A gene‐specific electroclinical syndrome remains to be characterized.

• The seizure semiology is heterogeneous, and one third of patients have treatment‐resistant seizures.

1. INTRODUCTION

The mammalian membrane‐bound o‐acyltransferase (MBOAT) protein family comprises several acyltransferases, and each of them has a unique preference toward specific acyl donors and acceptors. MBOAT7 encodes the lysophosphatidylinositol acyltransferase 1, which transfers arachidonic acid from arachidonoyl‐CoA to lysophosphatidylinositol. Biallelic disease‐causing variants in MBOAT7 have been associated with intellectual disability (ID), developmental delay (DD), autism spectrum disorders (ASDs), attention‐deficit/hyperactivity disorder, movement disorders, muscle tone abnormalities, and epilepsy. ¹

Johansen et al. ² initially described 16 patients with MBOAT7 encephalopathy, of whom six patients (37%) had infantile onset epilepsy. As of October 2024, 64 patients with biallelic pathogenic MBOAT7 variants are described, of whom 49 (77%) presented with epileptic seizures. ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ Among these, 15 experienced only febrile seizures. ² , ³ , ⁵ , ⁶ , ⁸ However, for most patients limited data were provided concerning the epileptology. Thus, the epileptology of MBOAT7 encephalopathy, including evolution and treatment response, remains poorly understood.

We, therefore, set out to provide a more in‐depth characterization of the epileptic, neurologic, and developmental phenotypes of 14 novel and one previously reported patient with MBOAT7 encephalopathy.

2. MATERIALS AND METHODS

Patients with biallelic MBOAT7 variants were recruited through GeneMatcher ¹⁴ or an international network of epilepsy and genetics departments or research groups. We also contacted the corresponding authors of previous publications and asked for updates on published patients. We obtained clinical data on the age at seizure onset and offset, seizure semiology, developmental trajectory, medical history, examination, electroencephalography (EEG), and neuroimaging for each patient. Additional information about genetic findings, pregnancy and delivery, neonatal period, feeding, developmental milestones, behavior, and psychiatric comorbidities was also included. Data were collected via standardized pro forma from their clinicians. When possible, EEGs were reviewed by a trained epileptologist (S.O.D.l.R.). Developmental and cognitive levels and the diagnosis of ASD were based on clinical and, whenever available, standardized assessments, and were classified as per the Diagnostic and Statistical Manual of Mental Disorders 5th edition. DD was classified as mild, moderate, severe, or profound in children 5 years or younger. Individuals 6 years or older were classified as having mild, moderate, severe, or profound ID. An antiseizure medication (ASM) was considered effective if the patient achieved a > 50% reduction in seizures for a period of >6 months. The patient was considered to have refractory epilepsy if, at any point throughout the duration of follow‐up, they had failed to become seizure‐free for at least 12 consecutive months despite two adequate trials with ASMs. The seizure and epilepsy types were classified according to the International League Against Epilepsy (ILAE) classification. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ For patients affected by epilepsy, differentiation of developmental encephalopathy with epilepsy versus developmental and epileptic encephalopathy was based on clinical evidence of adverse effects of seizures or epileptiform activity on development. ²⁰ Specific epilepsy syndromes were classified according to the diagnostic criteria of the new ILAE classification. ¹⁷ , ¹⁸ , ¹⁹ If a specific epilepsy syndrome could not be identified, each patient was classified as having focal, generalized, or combined epilepsy.

Descriptive analysis was performed for the demographics, and Fisher exact test was performed to evaluate differences between the proportions of the categories in two‐group variables of the cohort.

MBOAT7 variants were annotated using the transcript NM_024298.5 (GRCh37/hg19) and Human Gene Variation Society nomenclature (https://mutalyzer.nl/). Included in the variant evaluation were searches in the Human Gene Mutation Database (2024.3), ²¹ the Genome Aggregation Database (gnomAD; v.4.1.0), ²² National Center for Biotechnology Information PubMed, and ClinVar (2024), as well as in silico evaluation using SpliceAl, sorting intolerant from intolerant (SIFT), PolyPhen‐2, MutationTaster, combined annotation dependent depletion (CADD), and rare exome variant ensemble learner (REVEL) to predict the functional effect of the variants. Only variants with a population allele frequency < 1% in the heterozygous state and absent in the homozygous state in gnomAD fulfilled our filter criteria. Variant classification was performed according to the 2015 American College of Medical Genetics and Genomics (ACMG) guidelines ²³ and 2024 Association for Clinical Genomic Science (ACGS) best practice guidelines for variant classification in rare diseases. ²⁴

2.1. Standard protocol approvals, registration, and patient consent

The study was conducted in agreement with the Declaration of Helsinki and approved by the local ethics committees. The authors obtained consent from each patient using research protocols approved by their local human research ethics committees. Informed consent was given from parents or legal guardians (when applicable).

3. RESULTS

We identified 15 patients (six females and nine males) with MBOAT7 encephalopathy, of whom 12 were diagnosed with epilepsy. Mean age was 10.1 years (range = 14 months to 24 years). Fourteen patients were unpublished, whereas one had previously been reported. ⁷ We found two pairs of affected siblings, but the remaining patients were unrelated. The first pair of siblings (F6) comprised a 10‐year‐old sister and a 15‐month‐old brother; the sister had moderate DD and ID and developed epilepsy at approximately 7 years of age, whereas the younger brother was reported as having severe DD and had not developed epilepsy at the time of last follow‐up. The second pair of siblings (F7) were twins with similar clinical phenotypes, with the exception that only one of them presented with treatment‐resistant epilepsy. See Table 1 for a general overview of clinical features.

TABLE 1.

Clinical and genetic characteristics of our cohort with MBOAT7 encephalopathy.

Characteristic	Family
	Family 1	Family 2	Family 3	Family 4	Family 5	Family 6	Family 6	Family 7	Family 7	Family 8	Family 9	Family 10	Family 11	Family 12	Family 13
ID number	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Sex	Female	Male	Male	Male	Male	Female	Male	Male	Male	Female	Female	Female	Male	Female	Male
Age	14 months	96 months	36 months	96 months	204 months	132 months	15 months	144 months	144 months	180 months	120 months	108 months	204 months	288 months	35 months
DNA change (NM_024298)	c.400_412del/c.519C>A	c.482G>A/c.77‐2A>C	c.497C>T/c.971_985del	c.761G>A (homozygous)	c.680_690del (homozygous)	c.757_769del/c.472_477dupTGCTAC	c.757_769del/c.472_477dupTGCTAC	c.758_778del (homozygous)	c.758_778del (homozygous)	c.823G>A/c.1031G>A	c.823G>A (homozygous)	Variant 1: c.912C>A (maternal); variant 2: c.596C>G (paternal); variant 3: c.1250T>C (paternal)	c.977G>A (homozygous)	c.1057_1058delGCinsCA (Homozygous)	c.1154_1172del (homozygous)
Protein change (NP_077274)	p.(Phe134Profs*12)/p.(Tyr173*)	p.(Gly161Glu)/p.(?)	p.(Pro166Leu)/p.(Val324_Leu328del)	p.(Cys254Tyr)	p.(Leu227Profs*65)	p.(Glu253Leufs*66)/p.(Cys158_Tyr159dup)	p.(Glu253Leufs*66)/p.(Cys158_Tyr159dup)	p.(Glu253_Ala259del)	p.(Glu253_Ala259del)	p.(Gly275Ser)/p.(Arg344Gln)	p.(Gly275Ser)	Variant 1: p.(Cys304*); variant 2: p.(Pro199Arg); variant 3: p.(Leu417Pro)	p.(Trp326*)	p.(Ala353His)	p.(Gly385Alafs*12)
Type of variants	Frameshift/nonsense	Missense/splice site	Missense/in‐frame deletion	Missense	Frameshift	Frameshift/in‐frame duplication	Frameshift/in‐frame duplication	In‐frame deletion	In‐frame deletion	Missense/missense	Missense	Nonsense/missense/missense	Nonsense	Missense	Frameshift
Classification based on ACMG 2015 and ACGS 2024 guidelines	Pathogenic/pathogenic	Likely pathogenic/pathogenic	VUS/likely pathogenic	VUS	Pathogenic	Pathogenic/likely pathogenic	Pathogenic/likely pathogenic	Pathogenic	Pathogenic	Likely pathogenic/VUS	Likely pathogenic	Pathogenic/likely pathogenic/likely pathogenic	Pathogenic	Likely pathogenic	Pathogenic
(Criteria)	(PVS1, PM2, PM3)/(PVS1, PM2, PM3)	(PM1_SUP, PM2, PM3, PP3, PP4)/(PVS1, PM2, PP3, PP4)	(PM1_SUP, PM2, PP3, PP4)/(PM1_SUP, PM2, PM4, PP4)	(PM1_SUP, PM2, PP3, PP4)	(PVS1, PM2, PM3_SUP, PP4)	(PVS1, PM2, PP4)/(PM1_SUP, PM2, PM3, PM4_SUP, PP4)	(PVS1, PM2, PP4)/(PM1_SUP, PM2, PM3, PM4_SUP, PP4)	(PM1_SUP, PM2, PM4, PP1_STRONG, PP4)	(PM1_SUP, PM2, PM4, PP1_STRONG, PP4)	(PM1_SUP, PM2, PP1_SUP, PP3, PP4)/(PM1_SUP, PM2, PP3, PP4)	(PM1_SUP, PM2, PP1_SUP, PP3, PP4)	(PVS1, PM2, PP4)/(PM1_SUP, PM2, PM3, PP3, PP4)/(PM2, PM3, PP3, PP4)	(PVS1, PM2, PM3_SUP, PP4)	(PS4_SUP, PM1_SUP, PM2, PP3, PP4)	(PVS1, PM2, PM3_SUP, PP4)
gnomAD v.4.1.0. heterozygous state	–/–	–/–	18/1466958/–	2/1588276	3/1591266	–/–	–/–	83/1579556	83/1579556	9/1558070/4/1613256	9/1558070	–/–/2/1599106	–	–	–
ClinVar (2024)	–/–	–/–	–/VUS	VUS	Pathogenic	Pathogenic/–	Pathogenic/–	–	–	VUS/VUS	VUS	–/–/–	–	–	–
HGMD 2024.3	–	–	–	–	DM (Yalnızoǧlu et al., 2019 4 , PMID: 30701556)	–	–	DM (Johansen et al., 2016 2 , PMID: 27616480; Khan et al., 2019 6 , PMID: 31852446; Sun et al., 2020 5 , PMID: 33335874; Nazmina et al., 2024 3 , PMID: 38088234)	DM (Johansen et al., 2016 2 , PMID: 27616480; Khan et al., 2019 6 , PMID: 31852446; Sun et al., 2020 5 , PMID: 33335874; Nazmina et al., 2024 3 , PMID: 38088234)	–	–	–/–/–	–	DM (Farné et al., 2020 7 , PMID: 32744787)	–
SpliceAl	–/–	Loss of splice site	–/–	–	–	–/–	–/–	–	–	–	–	–	–	–	–
	–/–	(.86)	–/–	–	–	–/–	–/–	–	–	–	–	–	–	–	–
SIFT	–/–	Deleterious/–	Deleterious/–	Deleterious	–	–/–	–/–	–	–	Tolerated/deleterious	Tolerated	–/deleterious/deleterious	–	Deleterious	–
(Score)	–/–	(0/)	(.03/–)	(.01)	–	–/–	–/–	–	–	(.08/.01)	(.08)	(–/0/0)	–	(0)	–
PolyPhen‐2	–/–	Probably damaging/–	Probably damaging/–	Possibly damaging	–	–/–	–/–	–	–	Probably damaging/possibly damaging	Probably damaging	–/probably damaging/probably damaging	–	Probably damaging	–
(Score)	–/–	(1/–)	(.989/–)	(.883)	–/–	–/–		–	(1/.898)	(1)	(–/.992/1)	–	(.999)	–
MutationTaster	–/–	Disease‐causing/–	Disease‐causing/–	Disease‐causing	–	–/disease‐causing	–/disease‐causing	–	–	Disease‐causing/disease‐causing	Deleterious	–/Disease‐causing/disease‐causing	–	Disease‐causing	–
(Score)	–/–	(98/–)	(98/–)	194	–	–/–	–/–	–	–	(56/43)	(56)	(–/103/98)	–	(86)	–
CADD score	–/39	31.0/34	28.5/–	35	35	–/–	–/–	23.3	23.2	26.4/32	26.4	40.0/29.3/25.7	50	–	–
REVEL score	–/–	.882/–	.830/–	.496	–	–/–	–/–	–	–	.461/–	.461	–/.741/.888	–	–	–
Consanguinity	N/A	No	No	No	Yes	N/A	N/A	Yes	Yes	N/A	Yes	No	Yes	Yes	No
Ethnic origin	Caucasian	Caucasian	Caucasian	Caucasian	Iranian	Indian	Indian	Afghan	Afghan	Caucasian	Syrian	Caucasian	Pakistani	Caucasian	Moroccan
Developmental delay before epilepsy onset	No	Yes	Yes	No	Yes	Yes	No	Yes	Yes	Yes	Yes	No	Yes	No	Yes
Intellectual disability/developmental delay	Moderate	Severe	Mild	Mild	Moderate	Moderate	Severe	Moderate	Moderate	Severe	Severe	Severe	Severe	Severe	Severe
Age at walking	N/A	60 months	19 months	17 months	36 months	24 months	Nonambulant	42 months	36 months	12 months	N/A	30 months	N/A	16 months	Nonambulant
Current verbal capability	N/A	Nonverbal	Phrases of >4 words	Simple phrases	Nonverbal	N/A	Nonverbal	Babbling	Babbling	Phrases	Words	Simple phrases	Words	Words	Words
Epilepsy diagnosis	No	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes
Age at onset of seizures	N/A	15 months	10 months	9 months	12 months	79 months	N/A	48 months	48 months	48 months	N/A	N/A	17 months	2 months	3 months
Seizure type	N/A	FIAM	Myoclonic/BTC	Myoclonic	FIAM	Myoclonic absences	N/A	Myoclonic/atonic	Nocturnal tonic, myoclonic–atonic/BTC	Myoclonic absences and BTC	Myoclonic	N/A	F‐BTC	FIAM	Focal tonic or clonic/focal spasms
Refractory epilepsy	N/A	No	No	No	Yes	Yes	N/A	No	Yes	Yes	Yes	N/A	No	No	Yes
EEG findings	N/A	Multifocal spikes (frontal, temporoparieto‐occipital)	Normal background, generalized SW	Focal spikes (frontocentrotemporal)	Normal background activity, spike and wave over central region	Generalized 3‐Hz SW	N/A	Slow background, SW left temporal	Slow background, continuous SW	Background activity normal, generalized 3‐Hz SW/PSW	Normal background activity, SW on bilateral frontal areas	N/A	Generalized SW	SW in posterior quadrant	Slow background + multifocal spikes
Brain MRI	White matter hyperintensities	Hyperintensity in globus pallidus and dentate nuclei	Hyperintensity in globus pallidus	Normal	Hyperintensities in dentate nuclei and globus pallidum, mild signs of hyperintensities in mesial temporal region and possible FCD in temporal pole; mild cerebellar atrophy	Hyperintensities in dentate nuclei	N/A	Normal	Normal prior surgery	Normal	Hyperintensities in globus pallidus and dentate nuclei, mild cerebellar atrophy, polymicrogyria	N/A	Hyperintensities in both globus pallidus and dentate nuclei	Cerebellar vermis atrophy	Hyperintensities of dentate nuclei and polymicrogyria

Abbreviations: ACGS, Association for Clinical Genomic Science; ACMG, American College of Medical Genetics; BTC, bilateral tonic–clonic seizure; DM, disease‐causing mutations; EEG, electroencephalogram; F‐BTC, focal to bilateral tonic–clonic seizure; FCD, focal cortical dysplasia; FIAM, focal impaired awareness motor seizure; gnomAD, Genome Aggregation Database; HGMD, Human Gene Mutation Database; MRI, magnetic resonance imaging; N/A, no data available; PMID, puBMed identifier; PSW, polyspike and wave; SW, spike and wave; VUS, variant of unknown significance.

The 15 patients harbored 19 MBOAT7 variants in homozygous or compound heterozygous states that were present in heterozygous state in parents (Table 1). These included missense variants (n = 8), frameshift variants (n = 4), in‐frame duplications or deletions (n = 3), nonsense variants (n = 3), and splice‐site (n = 1) variants. One recurrent variant (c.823G > A; p.Gly275Ser) was observed in heterozygous and homozygous state, respectively, in two different families (F10 and F11). Three of the variants were previously published. ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ According to the ACMG 2015 and ACGS 2024 guidelines, three variants were classified as variant of unknown significance (VUS), whereas the remaining 16 were classified as pathogenic or likely pathogenic. Patient 12 carried three different variants: two missense variants on the paternal allele and a truncating variant on the maternal allele. Both paternal missense variants were ultrarare and were classified as likely pathogenic. In general, all variants identified in our cohort were either absent or extremely rare in the gnomAD database, and none of them occurred in a homozygous state among healthy controls. ²² Results from in silico prediction tools are provided in Table 1, and an overview of the variants is provided in Figure 1.

FIGURE 1.

Schematic representation of the localization of variants within MBOAT7 in both the present cohort and population and clinical databases (Genome Aggregation Database [gnomAD] and ClinVar), along with zygosity for the patients in our cohort and functional consequence.

3.1. Phenotypic analysis

3.1.1. Neurological and behavioral phenotype

All patients had a DD affecting both motor and verbal skills, and the mean age at first developmental concern was 6 months (range = 1–24 months). Age of sitting and walking was reported in 12 of 15. The mean sitting age was 16.7 months (range = 8–48 months), and mean age was 29.2 months (range = 16–60 months) for walking. Two patients were reported to be nonambulant, the oldest of whom was 35 months old. Most patients spoke their first single words around 31 months of life (range = 11 months to 6 years), with subsequent inadequate speech acquisition. Ten patients showed some level of verbal development. None spoke fluently; their communication skills were limited to babbling (n = 2), single words (n = 4), or simple sentences (n = 6). Three patients remained nonverbal at last follow‐up. None of the patients underwent formal neuropsychological testing. The degree of cognitive impairment was based on clinical impression; among the 10 patients who were 6 years or older, ID ranged from mild (n = 1) to moderate (n = 4) to severe (n = 5). All patients displayed variable behavioral issues. Autistic features were present in seven patients, of whom three had received a clinical diagnosis of ASD. Lack of attention span and poor concentration were seen in four. All patients with mild/moderate ID experienced difficulties in understanding social situations and making friendships. Half of the patients had a low anger threshold with tantrums and occasional aggressive outbursts. Stereotypies were reported in three patients. Three patients had spasticity of extremities, and a movement disorder was reported in six patients, including tremor (n = 3), feet dystonia (n = 1), and dyskinesia (n = 2).

3.1.2. Epilepsy

Twelve of 15 patients suffered from epilepsy, with a mean age at seizure onset at 36 months (range = 2 months to 6.5 years). Seizure types at onset included focal motor seizures with impaired awareness (n = 3), unknown onset myoclonic seizures (n = 4), myoclonic absences (n = 2), and nocturnal tonic seizures (n = 1). Five patients subsequently developed different seizure types during the follow‐up period. These included focal (n = 1), unknown onset (n = 2), bilateral tonic–clonic, myoclonic–atonic (n = 1), and atonic (n = 1) seizures. Only one patient was classified as having infantile epileptic spasms syndrome. Of the 12 patients, two (16.6%) also had febrile seizures, and three of 12 (25%) experienced at least one episode of status epilepticus. Based on the seizure types, eight patients were classified as having generalized epilepsy (with myoclonic, absence, and bilateral tonic–clonic seizures being the most prevalent). In comparison, four patients had focal epilepsy (focal motor seizures with impaired awareness or focal tonic–clonic seizures and spasms). For the patients in our cohort, there was no clear clinical evidence indicating that seizures or EEG activity adversely affected developmental progress.

3.1.3. Epilepsy treatment response

Seizure freedom was reported at last follow‐up in 66.7% (8/12) of patients, although 50% (6/12) of the cohort was classified as having refractory epilepsy at some point before achieving seizure freedom. The four patients for whom age at seizure offset was recorded achieved seizure freedom at an average age of 5 years. The ASMs that reached at least 50% seizure reduction included eslicarbazepine, oxcarbazepine, lamotrigine, levetiracetam, valproic acid, ethosuximide, and topiramate. ASMs that seemed to be beneficial in some patients were ineffective or exacerbated seizures in others. Vagal nerve stimulation was implanted in a single patient with significant improvement in myoclonic seizures but almost no effect on tonic–clonic seizures. The ketogenic diet was implemented in three patients; it was later discontinued in two patients due to lack of efficacy, whereas it reduced the third patient's seizure burden by approximately 50%. One patient underwent a corpus callosotomy with a significant improvement in atonic seizures.

3.1.4. EEG features

We were able to obtain 36 EEG reports from the 12 patients affected by epilepsy. EEGs were recorded between the ages of 8 days and 13 years. The number of recordings per patient varied widely (range = 1–5). The background activity was reported in 25 recordings and was normal in 64% (16/25), whereas 36% (9/25) had either a generalized or a focal slowing. Interictal epileptic discharges (IEDs) were reported in 83% (30/36) of recordings. Generalized spikes/polyspikes followed by slow waves with a frequency between 2 and 4 Hz were reported in 53%, with interictal and ictal discharges described as “atypical” in the two patients with myoclonic absences. Multifocal IEDs were found in recordings from 23% of the patients, and parasagittal focal IEDs in 26% (frontal, n = 2; frontocentral, n = 3; and centroparietal, n = 1). An example of a recording presenting with multifocal central and parasagittal IEDs is shown in Figure 2. Two patients had IEDs in the posterior and temporal quadrants, respectively. Due to the heterogeneity of the findings, we did not find any typical EEG features that could be associated with MBOAT7 encephalopathy.

FIGURE 2.

Interictal electroencephalogram from Patient 5 in our cohort, aged 10 years 2 months, showing interictal epileptiform discharges with right central (A), midline (B), and anteriorly predominant diffuse (C, D) distributions.

3.1.5. Magnetic resonance imaging features

Brain magnetic resonance imaging (MRI) was available for 12 patients; these were reported to be normal in four of 12 (33%) patients. The most frequent abnormal finding was a high‐intensity signal in T2 and fluid‐attenuated inversion recovery (FLAIR) sequences in dentate nuclei and/or globus pallidum (Figure 3), which was reported in 58% (7/12). Additional findings included cerebellar atrophy (3/12, 25%), hypomyelination (1/12, 8%), polymicrogyria (2/12, 16%), and abnormal white matter signal (1/12, 8%). Three of the four patients who underwent magnetic resonance with spectroscopy had normal findings, whereas a single scan showed elevated myoinositol values and reduced N‐acetyl aspartate values.

FIGURE 3.

T2‐weighted magnetic resonance imaging from Patient 5 in our cohort, showing the most frequently reported finding of high‐intensity signal at dentate nuclei (A) and globus pallidus (B), highlighted by arrows, as well as a mild hyperintensity and asymmetry of the left mesial temporal lobe (B), possibly indicative of a malformation of cortical development such as a focal cortical dysplasia.

3.1.6. Other clinical features

Seven of 15 patients had feeding difficulties at 1 month of life; no patient required a nasogastric tube. Four had gastroesophageal reflux, and none is currently under treatment. Three had chronic constipation, and all are currently under treatment.

3.1.7. Genotype–phenotype correlations

We compared our dataset with previously published cases and searched for clinical differences of those with biallelic missense variants compared to in‐frame deletions and biallelic truncating and splice‐site variants (as summarized in Table S1). There did not seem to be a significant difference in the prevalence of treatment‐resistant epilepsy between the different categories, whereas biallelic missense variants did seem to be associated with a slightly lower prevalence of severe ID (Fisher exact test p = .043). Notably, severe motor outcomes were not reported in patients with biallelic missense variants; however, these outcomes appeared to be disproportionately more frequent in patients with the biallelic recurrent in‐frame deletion c.758_778del, even when compared to those with biallelic truncating and splice‐site variants (Fisher exact test p = .018, prevalence ratio = 8.12, 95% confidence interval = 1.04–365.96, p = .024). To validate these trends, it is imperative to consider significantly larger cohorts of patients, to enhance statistical power, minimize potential biases inherent in smaller sample sizes, and perhaps evaluate a larger number of clinical variables.

4. DISCUSSION

We have detailed the diverse epileptic phenotypes and neurological characteristics observed in 15 patients with biallelic MBOAT7 variants, classified as pathogenic, likely pathogenic or ultrarare VUS with in silico prediction of pathogenicity. MBOAT7 encephalopathy, a condition that has thus far not been comprehensively characterized in the literature, is a rare and complex disorder resulting in global DD and early onset, potentially treatable focal and generalized onset seizures, in combination with psychiatric and behavioral comorbidities. Although seizures are common and often onset occurs during the first 3 years of life, a minority of patients escape seizures or start seizing later in childhood. We present additional evidence that MBOAT7 encephalopathy is associated not only with focal epilepsy but also with generalized epilepsy. Common seizure types observed include tonic–clonic seizures, occasionally triggered by fever, as well as myoclonic and absence seizures. No patients had combined focal and generalized epilepsy in our cohort. In addition, half of our cohort had treatment‐resistant seizures.

Because most of the patients of this cohort come from epilepsy centers, there is a potential selection bias possibly leading to an overrepresentation of severe epilepsy phenotypes. However, most reported patients in the literature do have epilepsy as a primary diagnosis.

The presence of DD/ID in all published cases (including our cohort), and the lack of definitive evidence of a detrimental impact of seizures or EEG activity on development, suggest that biallelic MBOAT7 variants are associated, in most cases, with developmental encephalopathy with or without epilepsy. ²⁰ In most of the previously published literature, seizures either were reported as "+/−" or were classified as myoclonic or focal, motor‐unaware events. ⁷ It has previously been reported that approximately 77% of patients with MBOAT7 encephalopathy have epilepsy, with time of onset between 2 and 65 months of age. ² , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ So far, 15 patients have also been reported to have fever‐provoked seizures. ² , ³ , ⁵ , ⁶ , ⁸ However, it remains unclear how many of these cases involved exclusively “febrile seizures” or “febrile seizures plus,” and how many had both unprovoked and fever‐provoked seizures. Additionally, it is still unknown whether some patients in the literature may only present fever‐provoked seizures, preventing them from receiving a definitive diagnosis of epilepsy. ⁸ In our cohort, 16% of patients had febrile seizures, and two continued to have febrile‐triggered seizures after the age of 5 years. The description of seizure semiology in the literature is based on different classifications that have changed over time. ⁸ We encourage all authors in different disciplines to use standardized terminology proposed by the ILAE to help the population analysis. Another weakness of the existing literature is that some papers describe multifocal as a type of seizure, ² , ⁶ and it remains unclear whether those reports refer to multifocal epileptic discharges found on the EEG tracing or different seizure types.

Previous publications have also shown variable but overall reasonable seizure control with ASM. The most commonly used ASMs included phenobarbital and valproic acid, both as monotherapy. In our cohort, half of the patients were classified as treatment‐resistant, and more than one third were not seizure‐free. It remains unknown why some patients remain treatment‐resistant, but this may be attributed to their MBOAT7 variants and the residual MBOAT activity, as well as the difference in genetic background and, thereby, variants in other disease‐modifying genes.

We report MBOAT7 encephalopathy‐related EEG findings and seizure classification and evolution in 12 patients affected by epilepsy. Background slowing, as well as generalized spike–wave discharges, are the most frequent features. However, we could not find distinctive EEG traits associated with MBOAT7. Future studies and larger samples on MBOAT7 encephalopathy are needed to establish whether this syndrome has a specific EEG fingerprint. One possible additional limitation of our study aside from the sample size was that for most patients, EEG data were extracted from reports rather than direct evaluation of EEG tracings, although all reports were reviewed by a trained epileptologist (S.O.D.l.R.).

Regarding neuroimaging findings, the most frequent findings were hyperintensities at globus pallidum and dentate nuclei. When present, these were always independently reported. Mild atrophy remains to be quantified with postprocessing tools. Also, this finding requires follow‐up to be considered a suggested finding for this condition.

Combining data from our case series with the available literature, we identified an apparent association between biallelic missense variants and better cognitive and motor outcomes compared to truncating variants and a recurring in‐frame deletion. However, the number of patients published so far, and the lack of standardized assessments and consistent reporting of clinical information, do not enable us to draw definitive conclusions regarding genotype–phenotype correlations.

5. CONCLUSIONS

Biallelic deleterious MBOAT7 variants cause global developmental impairment in all affected patients and epilepsy in most cases. The seizure semiology includes myoclonic seizures, myoclonic absences, focal motor with impaired awareness seizures, and bilateral tonic–clonic seizures. Seizures were occasionally provoked by fever. Most previous cases are reported to be seizure‐controlled, whereas half of our cohort have refractory epilepsy. There are no definitive EEG patterns for MBOAT7 in our cohort. The most frequent MRI findings are hyperintensities in T2/FLAIR sequences in dentate nuclei and globus pallidus. This study provides some insight into the epileptology and natural history of this condition, but more work needs to be done in understanding the pathophysiology and its implications in genotype–phenotype correlation and treatment response.

AUTHOR CONTRIBUTIONS

Allan Bayat: Conceptualization. Valentina Rizzo, Sebastian Ortiz De la Rosa, Robin‐Tobias Jauss, Allan Bayat: Data curation. Valentina Rizzo, Sebastian Ortiz De la Rosa, Robin‐Tobias Jauss, Allan Bayat, Christina D. Fenger: Formal analysis. Tobias Bartolomaeus, Maria Escolar, Geneviève Bernard, Ralitza Gavrilova, Rebecca Ahrens‐Nicklas, Gabrielle Lemire, Kym M. Boycott, Saadet Mercimek‐Andrews, Paolo Prontera, Cinzia Costa, Bojana Rakic, Cornelius F. Boerkoel, Stephanie Huynh, Linda Huh, Elliott Sherr, Emanuela Argilli, Juan Darío Ortigoza‐Escobar, Didac Casas‐Alba, Tania Nunes, David A. Koolen, Konrad Platzer, Marianne S. Khinchi, Elena Gardella: Investigation. Allan Bayat: Methodology. Allan Bayat: Supervision. Robin‐Tobias Jauss: Visualization. Sebastian Ortiz De la Rosa, Valentina Rizzo, Allan Bayat: Writing—original draft. All authors: Writing—review and editing.

CONFLICT OF INTEREST STATEMENT

C.D.F. is employed by the company Amplexa Genetics. The other authors have no conflicts of interest to disclose. 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

TABLE S1.

De la Rosa SO, Rizzo V, Jauss R‐T, Bartolomaeus T, Escolar M, Bernard G, et al. MBOAT7 encephalopathy: Characterizing the neurology and epileptology. Epilepsia. 2025;66:2379–2390. 10.1111/epi.18376

DATA AVAILABILITY STATEMENT

Anonymized data not published in this article will be made available by request from any qualified investigator.