CHD2. In 2013, mutations in CHD2 were reported in various publications including two major studies on epileptic encephalopathies, reinforcing the notion that de novo mutations in this gene are a recurrent cause of epileptic encephalopathies. However, large-scale studies often cannot fully appreciate the complete phenotype of the patient behind the gene finding. Therefore, it is difficult to appreciate similarities between patients and assess whether phenotypes constitute a recognizable entity. In a recent publication in Neurology, the phenotype of CHD2 encephalopathy is explored in detail – it represents a distinct, recognizable disease entity.
The discovery of CHD2. I can only present the story of CHD2 from the EuroEPINOMICS perspective, which culminated in our 2013 publication. I always like to tell this story because the CHD2 gene was a difficult gene to work with in the first place. This was the case not because of the gene itself, but because I didn’t know what to think of it in the first place. Against better advice, I believed for almost a year that de novo mutations in this genes were a false positive finding. It was not a gene that we intuitively understood, the mouse phenotype is entirely unrelated to what our patients had, and it was a big gene to sequence. This all happened before the age of RVIS and access to other large-scale datasets and just makes me remember how helpless we felt at times – we had a finding, but simply didn’t know whether it made sense. We knew that mutations in this gene were rare, so there was only limited hope to find additional cases. The almost parallel publication of the two epilepsy manuscripts on CHD2 were then pivotal for our thought process on how to handle and share data – our data sharing policy, which led to our publications on HCN1, DNM1, STX1B and SCN8A. A flurry of gene discoveries fueled by our initial dissatisfaction to face a gene that we did not understand. To make a long story short, I feel attached to CHD2, and I am very happy that the recent publication by Thomas and collaborators attempts to suss out the phenotype behind this gene.
CHD2 myoclonic encephalopathy. The authors compare the phenotypic features of 10 patients with CHD2 encephalopathy including one patient with a 15q26 deletion that encompasses this gene. This patient cohort also includes one patient of the EuroEPINOMICS-RES cohort who was identified after our initial publication. To sum up the publication by Thomas and collaborators: the phenotypes of all patients had some intriguing features that may make it easy for this syndrome to be recognized in clinical practice. After an initial normal development in the first year of life, patients usually present with some delays before their second birthday. Seizure onset is sometimes explosive and occurs at roughly 2 years with frequent and sometimes intractable myoclonic and absence seizures. Very few well-defined epilepsy syndromes follow this pattern, which may help in identifying patients with CHD2 encephalopathy clinically. Virtually all patients have a very strong photosensitivity and photic induced seizures, sometimes with relatively minor environmental triggers. This extreme photosensitivity is a unique feature of CHD2 encephalopathy and even though photosensitivity is also seen in patients with other epileptic encephalopathies such as Dravet Syndrome or Myoclonic Astatic Epilepsy, it is not as prominent as in CHD2 encephalopathy. Later during childhood, patients may develop generalized tonic-clonic seizures and status epilepticus with a mean onset of 9 years. In contrast to our previous report, fever seems to be less of a trigger than initially thought and only very few patients have seizures associated with fever.
EEG, MRI. The EEG features of patients with CHD2 myoclonic encephalopathy are also remarkable, as they progress from features seen in Idiopathic/Genetic Generalized Epilepsies (IGE/GGE) to features seen in Lennox-Gastaut Syndrome. Initially, the patient may have 3-4 Hertz spike-wave discharges and absence, myoclonic or astatic seizures. These EEG features may then progress to continuous 1-2 Hz slow spike-wave, one of the hallmarks of Lennox-Gastaut Syndrome. Tonic seizures may be seen at this later stage, which go along with this EEG pattern. The MRI in patients with CHD2 encephalopathy shows some progressive atrophy and an intriguing pattern with posteriorly dominant atrophy, which is a very uncommon finding and may hint towards the underlying diagnosis. Also, there is some atrophy of the cerebellum, which usually has a broader differential diagnosis. The developmental trajectory of patients with CHD2 encephalopathy may result in moderate to severe intellectual disability with possible regression during periods of high seizure burden.
AMA seizures. Thomas and collaborators also identify a very uncommon seizure type that has only been found in patients with CHD2 encephalopathy, which they refer to as an astatic-myoclonic-absence (AMA) seizure, which usually start at the age of 2 years. This seizure type is characterized by an initial astatic seizure (loss of tone), followed by a myoclonic-absence phase, with “ratched-like” tonic abduction of the arms. Myoclonic absence seizures are already a rare seizure type and the combination with the initial loss of tone during the atonic component is a unique and distinct seizure type, which may further point toward the diagnosis. Thomas and collaborators provide several videos of seizures in patients with CHD2 encephalopathy online in the Supplement of their publication, which is freely available and worth watching (link).
What you need to know. The publication of Thomas and collaborators has turned CHD2 encephalopathy from a broad phenotype identified through large-scale studies into a distinct disease entity, which should be recognizable in clinical practice. To me, this sequence of events represents an ideal interaction between genomic and clinical research, which can only be achieved by the tight connection between geneticists and clinicians that we have in the epilepsy community. It is also a strong reminder that even though we sometimes like to emphasize the broad phenotypic spectrum of newly discovered genes, we might be slightly biased until we have a closer look.