Strides in STXBP1: What Happened Last Month in STXBP1 Research?

So, what was new in October?

October was a banner month for STXBP1 publications with some exceptional articles coming out.

The month started out with a case report study from a group in Morocco describing an STXBP1 patient with a novel splice-site pathogenic variant. The child has neurodevelopmental delay, cognitive impairment, history of epileptic spasms, motor impairments, slight hearing loss, and autism. Genomic testing revealed a novel variant, c.38-2A>G, that had not previously been reported in any genetic database. Interestingly, later in the month researchers from China published a case report describing an STXBP1 patient with a splice-site variant in the same region, c.37+2dupT. This patient demonstrated somewhat atypical phenotypes characterized by motor and cognitive impairments, late onset epilepsy, and leukoaraiosis. Leukoaraiosis is an abnormal change in the appearance of white matter in a particular area of the brain observed in an MRI scan. It is a sign of hypomyelinating leukodystrophy, which is a condition where the nervous system has a reduced ability to form myelin, a fatty substance that covers nerve fibers and acts like electrical insulation. This is the first reported instance of leukoaraiosis associated with a variant in STXBP1.

A study in Turkey evaluated the utility of genetic testing in the diagnosis of early-onset developmental epileptic encephalopathies (EO-DEEs). In cohort of 1450 patients, genetic testing revealed a gene-related cause in 701 individuals (48.3%). Of this group, 21 individuals were identified with a STXBP1 variant (2.9%). The authors concluded that genetic testing was a cost-effective analysis by providing useful causative data in helping to determine therapeutic management plans for patients with EO-DEEs.

Lennox-Gastaut syndrome (LGS) is a severe form of epilepsy that usually begins in early childhood. It involves tonic seizures plus at least one other type of seizure, cognitive and behavioral impairments, and a specific type of EEG signature. There is no specific underlying cause, rather it can be caused by a variety of conditions, including the presence of monogenic variants. A group from China performed a retrospective study of LGS 46 patients to determine if seizure types had any correlation with eventual prognosis, included in the study were STXBP1 patients as well as patients with 4 other gene known epileptic gene variants. The group found that patients that experienced spasm-tonic seizures exhibited poorer treatment responses and neurodevelopment than patients who experienced tonic or myoclonic-tonic seizures. The group did not report that the type of gene variant had any impact on treatment response or prognosis.

Juliet Knowles, who leads our Stanford STARR site and Zach Grinspan, who leads our Weill-Cornell site, were the lead and senior authors on a paper that reviewed the challenges developing efficacious therapies for individuals with LGS. Since variants in STXBP1 are associated with LGS, these challenges affect a significant portion of our population. Among the challenges discussed were a poor understanding of the natural history of LGS, lack of standardizing LGS treatment based on expert consensus, and the narrow focus on seizures involving “drops” in clinical trials as opposed to patient-centered outcomes.

Peter Galer and the CHOP group released their findings on EEG analysis of genetic epilepsies including STXBP1, SCN1A, and SYNGAP1. This retrospective study used spectral analysis, a standard mathematical method used to extract information from clinical EEGs, and machine learning to identify features that differ between the three genetic epilepsies and to determine if EEGs correlate with a functional motor assessment in individuals. First, they extracted from each EEG the ‘relative power (RP)’ of four separate frequencies, delta, theta, alpha, and beta and then examined the relative ratios of these frequencies. The alpha-delta ratio of an EEG is a common quantitative EEG feature that roughly estimates the faster (alpha) to slower (delta) activity in the brain and changes in this ratio have been associated with neurological disorders. The team found that this ratio was significantly lower in the STXBP1 cohort compared to EEGs from controls (patients without epilepsy or other brain disease) across all age ranges and lower compared to either the SCN1A or SYNGAP cohorts, thus this measure could potentially serve as a specific biomarker for STXBP1-RD. Further, the group found that within the STXBP1 cohort, this ratio was lower in those individuals with a missense variant compared to individuals with a protein truncating variant. Next, they took the alpha/delta, beta/delta, alpha/theta, and beta/theta ratio data from five different areas of the brain and developed a machine learning model that could predict the type of DEE. They found that that the model could accurately predict STXBP1 and that this prediction relied primarily on the EEG data obtained from the front of the brain. Finally, the team used another machine learning model to demonstrate that EEG spectral analysis data correlated with motor abilities as separately assessed using GMFM scores, thus showing the potential ability of EEG to predict functional outcomes in patients.

Elena Gardella and colleagues from the Danish Epilepsy Centre as well as their collaborators from AOUI Verona Italy also published their study on EEG analysis of individuals with STXBP1-RD and two other developmental epileptic encephalopathies (DEEs), Dravet and Angelman syndrome. Like the CHOP study, they quantified the RP of the four EEG frequencies, delta, theta, alpha, and beta in both the front and back of the brain. They found a significant increase in the delta RP in all three DEEs compared to age and sex matched neurotypical controls. Further they found that the delta RP was significantly higher in the STXBP1 cohort compared to the other two DEEs. Interestingly, the data also appears to show a substantial decrease in alpha RP in the STXBP1 cohort compared to the other DEEs and controls. A low alpha RP and high delta RP would lead to a decreased alpha/delta ratio as observed in the CHOP study. When comparing the delta RP from the front of the brain (anterior) to the back of the brain (posterior) the team found that the anterior delta RP was significantly greater than the posterior delta RP in the STXBP1 cohort, but not in the other two DEEs, suggesting that this specific slowing in the anterior part of the brain may be specific to STXBP1 and could be a potential biomarker.

Elena Gardella and Francesca Furia from the Danish Epilepsy Centre also published their findings on early mortality in STXBP1-related disorder; this study was funded by the Foundation. The study analyzed the data on 40 individuals, from around the world, with pathogenic STXBP1 variants who died between the ages of 11 months and 46 years. The most common cause of death was SUDEP (36%), followed by pulmonary infections and respiratory complications (33%). The incidence of SUDEP peaked in mid-childhood (6-12 years) while non-SUDEP causes peaked in early childhood and adulthood. The overall mortality rate for STXBP1-RD was calculated to be 3.2% or 3 deaths/1000 person-years, which is higher than the general epilepsy population (1.2%-1.3%) and similar to many other developmental epileptic encephalopathies, though less than that observed for Dravet syndrome (8.6%). Though genetic data wasn’t available for every case examined, for the data that was available there appeared to be no effect of variant position on cause of death but a higher percentage of SUDEP was associated with missense variants.