Niu et al. 2021

Quick Summary

This 2021 paper described six children with IRF2BPL gene variants who were seen for epilepsy at Peking University First Hospital between May 2017 and September 2020. All six children had seizures beginning in infancy, with seizure onset between 3.5 and 7 months of age. All six were diagnosed with infantile spasms, a seizure syndrome that can appear early in life and is often associated with developmental concerns.

The study is useful because it focuses on a very specific early presentation of IRF2BPL-related disorder: babies whose first major sign was epilepsy, especially spasms. The group included five girls and one boy. Every child had developmental delay, and five of the six had low muscle tone. Two had swallowing difficulty. One child had a broader neurological picture that included microcephaly, nystagmus, chorea, and athetosis.

The paper also showed that the same early epilepsy pattern can be seen with different kinds of IRF2BPL variants. The six children had three frameshift variants, two missense variants, and one in-frame deletion. All six variants were reported as de novo, meaning they were new in the child and were not inherited from either parent.

Why This Paper Matters

Earlier IRF2BPL papers helped establish that this gene can be linked with developmental delay, seizures, regression, movement problems, speech loss, and brain changes. Niu et al. added a focused view of the infantile-spasm end of the condition. Instead of describing a broad age range or a mixed neurological group, this paper centered on six children whose epilepsy began very early in life.

For families, that matters because infantile spasms can be frightening and confusing. A baby may show brief stiffening, bending, or clusters of events that are easy to miss at first. The diagnosis often depends on EEG testing and clinical experience. This paper helps place infantile spasms within the IRF2BPL story by showing that, in this cohort, every child with an IRF2BPL variant had spasms and all had developmental delay.

The paper also matters because it included patient-level treatment and follow-up information. It did not prove which treatment is best, but it did report that four of the six children became seizure-free after treatment combinations that included adrenocorticotropic hormone, topiramate, and vigabatrin. Two children continued to have frequent seizures despite trying multiple anti-seizure medicines. This kind of detail is helpful because it shows both possibility and uncertainty: seizure improvement can happen, but responses were not the same for every child.

Another important contribution was the genetic range. Earlier reports often emphasized truncating variants. This cohort included frameshift variants, missense variants, and an in-frame deletion, all associated with infantile spasms. That supports the idea that IRF2BPL-related epilepsy is not limited to one single variant type.

What The Researchers Studied

This was a primary cohort paper. The researchers retrospectively reviewed six children with epilepsy and IRF2BPL variants who were treated in the Department of Pediatrics at Peking University First Hospital. "Retrospective" means the team looked back at existing clinical records rather than enrolling people in a new treatment trial.

The researchers collected clinical histories, seizure information, developmental details, EEG results, brain MRI findings, genetic testing results, treatment information, and follow-up outcomes. The cohort was small, but the data were detailed enough to show patterns across the six children.

The table in the paper summarized each child by case number. It included sex, variant, seizure onset age, seizure type, other clinical features, MRI result, and treatments. The article text added more detail about EEG monitoring, follow-up age, treatment response, and how the children's development changed over time.

The children were followed to ages ranging from 1 year to 3 years 10 months. That means the paper mostly describes the early childhood course. It does not tell us what happened later in childhood, adolescence, or adulthood for these children.

What Was Learned About Symptoms

The central clinical finding was that all six children had epilepsy beginning in infancy. Seizure onset ranged from 3.5 months to 7 months. All six had epileptic spasms and were diagnosed with infantile spasms. One child also had tonic seizures and tonic-spasms, and one child had focal seizures. EEG monitoring captured spasms in all six children.

Developmental delay was also present in all six. Some children had developmental regression after seizure onset, meaning they lost or slowed in skills after the seizures began. The details varied. One child was later able to sit independently and stand with support, and could say simple meaningful words. Another child could stand and walk independently after seizures came under better control. Others remained much more delayed, with head control only, no independent sitting, and no language development by follow-up.

Low muscle tone was common. Five of the six children had hypotonia. Two had swallowing difficulty. One child had a more complex neurological presentation that included microcephaly, nystagmus, chorea, and athetosis. In plain language, this means that one child had a smaller head size, abnormal eye movements, and involuntary movements.

The EEG findings were striking. All six had hypsarrhythmia, an EEG pattern often associated with infantile spasms. Two children had slow background activity, and two had focal epileptic discharges. The paper also described EEG-captured seizure types, including spasms in all children, a focal seizure in one child, and tonic-spasm in one child.

Brain MRI results varied. One child had cerebral atrophy and corpus callosum dysplasia. Two had delayed myelination. Three had normal brain MRI results. This is important because a normal MRI did not rule out IRF2BPL-related epilepsy in this cohort. The diagnosis depended on the combination of clinical features, EEG, and genetic testing.

The treatment outcomes were mixed. Four children became seizure-free by the last follow-up. Two continued to have frequent seizures. The paper reported use of medications and treatments including valproate, topiramate, lamotrigine, levetiracetam, vigabatrin, and adrenocorticotropic hormone. Because this was not a controlled treatment study, these outcomes should be read as clinical observations, not as proof that one treatment is best for IRF2BPL-related epilepsy.

What Was Learned About Genetics

All six children had variants in IRF2BPL. The gene was listed using transcript `NM_024496.4`, and all variants were in exon 1, which is expected because IRF2BPL is an intronless gene. All six variants were reported as de novo.

Three children had frameshift variants: `c.232delG / p.V78Sfs*73`, `c.244del / p.A82Pfs*70`, and `c.283-308del / p.Ala95Thrfs*29`. A frameshift changes the way the genetic instruction is read from that point onward. This often creates an early stop signal and a shortened protein. In family terms, the instruction becomes disrupted very early in the gene.

Two children had missense variants: `c.1280C>T / p.L474F` and `c.1420C>T / p.S427L`. A missense variant changes one amino acid in the protein. It does not automatically mean the protein is shortened, but it can still affect how the protein works.

One child had an in-frame deletion: `c.1453-c.1455delTTC / p.F485del`. This means one amino acid, phenylalanine at position 485, was deleted while the rest of the reading frame stayed intact.

The article classified some variants as pathogenic and others as likely pathogenic using ACMG-style interpretation. The frameshift variants at `p.V78Sfs*73`, `p.A82Pfs*70`, and `p.Ala95Thrfs*29` were treated as pathogenic in the article text. The missense variants and the in-frame deletion were treated as likely pathogenic. The paper's overall message was that different variant types in IRF2BPL can be associated with infantile-onset epilepsy and developmental delay.

One source detail needed care during extraction. The Chinese abstract has one place where the first frameshift is written as `p.V78Sfs*7`, while the English abstract and Table 1 report `p.V78Sfs*73`. The verified cohort uses `p.V78Sfs*73`, because that is the patient-table value and the English abstract value.

Patient And Cohort Details

Case 1 was a girl with seizure onset at 3.5 months and the variant `c.232delG / p.V78Sfs*73`. The table listed epileptic spasms and focal seizures. She had no language development, microcephaly, nystagmus, hypotonia, and dysphagia. MRI showed cerebral atrophy, and the article text also reported corpus callosum dysplasia. At 3 years 10 months, spasms were controlled, but focal seizures continued about once per month. Development remained severely delayed.

Case 2 was a girl with seizure onset at 7 months and the variant `c.244del / p.A82Pfs*70`. Her seizure types included spasms, tonic seizures, tonic-spasms, and focal seizures. She had no language development, hypotonia, and dysphagia. MRI was normal. Vigabatrin reduced spasms, but focal seizures continued daily at follow-up, and several other medicines had been ineffective.

Case 3 was the one boy in the cohort. He had seizure onset at 5 months and the missense variant `c.1280C>T / p.L474F`. He had epileptic spasms, hypotonia, and delayed development, but his MRI was normal. Vigabatrin controlled seizures, and he was seizure-free for 1 year 4 months by the last follow-up. Development had improved compared with the more severe cases: he could sit independently, stand with support, and intentionally say simple words.

Case 4 was a girl with seizure onset at 3.5 months and the missense variant `c.1420C>T / p.S427L`. She had epileptic spasms and developmental regression after seizure onset. MRI showed delayed myelination. The EEG section reported that a focal seizure was captured, even though Table 1 listed spasms as the seizure type. ACTH was reported as ineffective, but oral topiramate controlled seizures. At follow-up, she had been seizure-free for 10 months and could stand, walk independently, and say simple repeated words.

Case 5 was a girl with seizure onset at 4 months and the in-frame deletion `c.1453-c.1455delTTC / p.F485del`. She had epileptic spasms, mild developmental delay, and hypotonia. MRI was normal. The table listed topiramate and vigabatrin, and the follow-up text reported that ACTH controlled seizures by day 7. She was seizure-free for more than 7 months at last follow-up.

Case 6 was a girl with seizure onset at 7 months and the frameshift variant `c.283-308del / p.Ala95Thrfs*29`. She had epileptic spasms, no language development, inability to sit independently, hypotonia, and delayed myelination on MRI. Topiramate and vigabatrin reduced seizures, and ACTH later controlled them. She was seizure-free for more than 1 year by follow-up, but development remained severely delayed.

Across the cohort, the consistent pattern was early-onset spasms plus developmental delay. The differences were also important. Some children had persistent seizures, while others became seizure-free. Some had normal MRI results, while others had delayed myelination or atrophy. Some made developmental gains after seizure control, while others remained profoundly delayed.

What Families Can Take Away

This paper supports that IRF2BPL-related disorder can present very early in life as infantile spasms with developmental delay. For families, that means an IRF2BPL diagnosis can help connect seizure history, EEG findings, developmental changes, muscle tone, swallowing concerns, and MRI findings into one broader explanation.

The paper also shows why EEG and genetic testing can both matter. In this cohort, all six children had hypsarrhythmia on EEG and all had de novo IRF2BPL variants. MRI was helpful for some children, but it was normal in three of the six, so MRI alone was not enough to exclude the diagnosis.

Treatment response was variable. Some children became seizure-free after treatment, while others continued to have frequent seizures. This does not mean the same treatment will work the same way for every child. It does mean that seizure control is possible in some children with IRF2BPL-related infantile spasms, and it supports careful follow-up with the child's neurology team.

Development also varied. A few children made visible gains after better seizure control, such as standing, walking, or using simple words. Others remained severely delayed even when seizures improved. This is an important reminder that seizure control and developmental outcome are related but not identical.

Families should not treat this paper as a care plan. It is a clinical report from six children, not a guideline. Decisions about medicines, EEG monitoring, swallowing evaluation, therapy, mobility support, and developmental services should be made with the care team that knows the child.

Limits Of The Paper

The main limit is the small cohort size. Six children can show a pattern, but they cannot define the full range of IRF2BPL-related epilepsy. Later reports and larger datasets are needed to understand how often each feature occurs and how outcomes change over time.

The paper focused on children with epilepsy who were seen at one hospital. That means the cohort may be enriched for children with more obvious seizure presentations. It may not represent people with milder IRF2BPL symptoms, later-onset symptoms, movement-dominant presentations, or people without infantile spasms.

Follow-up was short. The last follow-up ages ranged from 1 year to 3 years 10 months. The paper therefore cannot tell us what happened later in childhood, whether seizure control lasted long term, or how movement, speech, swallowing, learning, and daily function changed with age.

The treatment information is useful but not definitive. The children received different combinations of medicines and hormonal therapy, and this was not a randomized or controlled trial. The paper can report what happened in these six children, but it cannot prove which treatment should be preferred.

Some source details required careful extraction. The official Yiigle and PubMed pages expose the abstract and metadata, while the patient-level table was transcribed from a public full-text repost image. The cohort table was cross-checked against the official abstract, article text, and variant list, but the table itself is not available as structured text from the official page.

Source Notes

• Official article metadata and abstract: Yiigle cmaid 1324226, "Clinical features of epilepsy in children with IRF2BPL gene variation."
• PubMed abstract: PMID 34102826, English and Chinese abstracts.
• Public full-text repost: article text and Table 1 image for the six patient rows.
• Table 1: sex, variant, seizure onset age, seizure type, other clinical manifestations, brain MRI, and treatment.
• Main text: cohort identification from Peking University First Hospital, May 2017 to September 2020.
• Main text: EEG findings, MRI summary, ACMG interpretation, treatment response, and follow-up ages.
• Verified extraction files: `extracted-source-table.csv` and `patients.csv` in this article folder.