Abstract
S. K. Z. Ihnen, S. Alperin, J. K. Capal, A. L. Cohen, J. M. Peters, E. M. Bebin, et al. Epilepsia 2025;66(1): 117-133.DOI: 10.1111/epi.18172; https://pmc.ncbi.nlm.nih.gov/articles/PMC11742629 Objective: Epilepsy and intellectual disability are common in tuberous sclerosis complex (TSC). Although early life seizures and intellectual disability are known to be correlated in TSC, the differential effects of age at seizure onset and accumulated seizure burden on development remain unclear. Methods: Daily seizure diaries, serial neurodevelopmental testing, and brain magnetic resonance imaging were analyzed for 129 patients with TSC followed from 0 to 36 months. We used machine learning to identify subgroups of patients based on neurodevelopmental test scores at 36 months of age and assessed the stability of those subgroups at 12 months. We tested the ability of candidate biomarkers to predict 36-month neurodevelopmental subgroup using univariable and multivariable logistic regression. Candidate biomarkers included age at seizure onset, accumulated seizure burden, tuber volume, sex, and earlier neurodevelopmental test scores. Results: Patients clustered into 2 neurodevelopmental subgroups at 36 months of age, higher and lower scoring. Subgroup was mostly (75%) the same at 12 months. Significant univariable effects on subgroup were seen only for accumulated seizure burden (largest effect), earlier test scores, and tuber volume. Neither age at seizure onset nor sex significantly distinguished 36-month subgroups, although for girls but not boys there was a significant effect of age at seizure onset. In the multivariable model, accumulated seizure burden and earlier test scores together predicted 36-month neurodevelopmental group with 82% accuracy and an area under the curve of .86. Significance: These results untangle the contributions of age at seizure onset and accumulated seizure burden to neurodevelopmental outcomes in young children with TSC. Accumulated seizure burden, rather than the age at seizure onset, most accurately predicts neurodevelopmental outcome at 36 months of age. These results emphasize the need to manage seizures aggressively during the first 3 years of life for patients with TSC, not only to promote seizure control but to optimize cognitive function.
Commentary
Tuberous sclerosis complex (TSC) affects 1 in 6000 to 10,000 individuals. Neurological manifestations including drug-resistant epilepsy (DRE) are a major contributor to developmental outcomes. 1 International prospective registries like the “TuberOus SClerosis registry to increase disease Awareness”’ (TOSCA) was established in 2011 to identify knowledge gaps while furthering research in TSC. 2 Early identification and treatment of infantile spasms (IS) is critical. Initial research was focused on IS as a modifiable risk factor for development at 24 months. Results from the PREVeNT trial seemed counterintuitive to the EPISTOP trial at first glance: early IS treatment with Vigabatrin did not change developmental outcomes at 24 months and did not prevent DRE with focal seizures!3,4 In a previous commentary in this journal, highlighting the differences between PREVeNT and EPISTOP, I had referenced work through the TOSCA registry identifying several factors like genetic diagnosis, age at seizure onset, interictal electroencephalogram (EEG) abnormalities, epileptic spasms, lesion burden on magnetic resonance imaging (MRI) that could affect ultimate development.5,6 The recent publication by Ihnen et al 7 attempts to unravel the complex relationships between these factors and neurodevelopmental outcomes at 36 months.
This work, however, has its origins in a prospective, NIH funded national registry established in the United States in 2012 across several academic sites. The Tuberous Sclerosis Complex Autism Center of Excellence and Research Network (TACERN) was established as a collaborative network focused on identifying early biomarkers for autism spectrum in infants with TSC. 8
It is important to appreciate the stepwise and logical progression of research from this group to adequately understand the findings of the present study. Two TACERN study group clinical trials have served as a study base for several papers. One aimed to identify early biomarkers of autism in infants (NCT 01780441) and enrolled infants at 3 to 12 months of age. Another aimed to identify potential EEG biomarkers and antiepileptogenic strategies for epilepsy (NCT01767779) and enrolled infants at 0 to 6 months of age.
In an initial 2017 publication, Capal et al 9 studied 130 patients with completed developmental assessments at a chronological age of 24 months. All developmental assessments were performed by a licensed speech therapist or psychologist blinded to patient's history. Mullens Scales of Early Learning (MSEL), autism-specific evaluation scores, and Vineland Adaptive Behavior Scales (VABS-2) were assessed. The MSEL consists of 5 domains (gross motor, fine motor, expressive language, receptive language, and visual reception) while the VABS is a caregiver report that assesses social, communication, motor, and daily living skills. This study highlighted significant variability among subjects with respect to seizure frequency. Thus, seizures were categorized into ordinal variables by severity and also averaged in binned frequencies assessed during the 6 months prior to developmental assessment time points. Key results included an earlier seizure onset along with higher seizure frequency by 12 months of age as critical factors affecting immediate and future neurodevelopmental outcomes.
A 2021 analysis of 88 out of 129 patients enrolled in both TACERN studies analyzed association between seizure phenotype and development. 10 Keeping in mind the wide variability in seizures identified in 2017; automated unsupervised machine learning (hierarchical clustering) analysis was performed to group seizure characteristics into favorable and unfavorable groups based on seizure burden, proportion of focal seizures, epileptic spasms or both. This study highlighted the value of further systematic analysis of overall seizure burden and other factors using cluster analysis methodology to identify underlying differences in the data.
In 2025, Ihnen et al, 7 building on previously obtained knowledge,9,10 extended the interim analysis of 2017 using the methodology of cluster analysis used in 2021. They analyzed 129 out of 169 patients with at least 70% of daily seizure diary completion over a minimum span of 365 days and 90% completion on developmental subscales including MSEL and VABS-2 scores. Developmental quotients (DQ) were computed to avoid floor effects. Clustering based on neurodevelopmental assessment led to 2 main groups: generally higher scoring versus generally lower scoring group. Tuber volumes were calculated by systematic segmentation of highest contrast T1-weighted structural MRI performed at age greater than 2 years. Neurological variables that could serve as biomarkers included prevalence of epilepsy, infantile spasms and focal seizures, epilepsy surgery, sex, TSC pathogenic variant, tuber volume, age at seizure onset, accumulated seizure burden, and 12-month DQ. Univariable analysis identified candidate biomarkers that were then included in step-wise multivariable logistic regression. Multivariable regression models were further assessed for accuracy.
Main Take-Home Message
DQ at 12 months is likely to stay stable up to 36 months of age since 74% of patients segregated into the same subgroup at 12 months as at 36 months. On univariable analysis, persistently lower scores on DQ at 36 months are associated with larger tuber volume, earlier age at seizure onset, higher accumulated seizure burden, and lower 12-month DQ. Although the lower scoring subgroup had mean age at seizure onset nearly 2 months earlier than the higher scoring subgroup; this variable did not achieve statistical significance in the multivariable logistic regression model. On multivariable analysis, the accumulated seizure burden with focal seizures is more important than presence or absence of epileptic spasms and earlier age of onset such that the number of seizure (focal) days per week per patient was clearly higher in the lower scoring subgroup compared to the higher scoring subgroup. Ninety percent of those in the favorable group experienced seizure reduction following surgery (90% vs. 69%; p = not significant). As a corollary, patients exposed to higher numbers of antiseizure medications segregated into the lower scoring group. There was a higher proportion of TSC1 compared to TSC2 pathogenic variants in the higher scoring group (p = not significant). The combination of accumulated seizure burden and 12-month DQ was more powerful (82% accuracy) than either variable alone in predicting 36-month DQ.
Authors did not analyze EEG variables and burden of subclinical seizures or interictal abnormalities on ultimate development. A biological effect of sex was found on post hoc analysis such that girls in the lower scoring subgroup are likely to have earlier onset of seizures. This finding needs replication in larger studies.
Conclusion
All in all, the most critical variables identified in the study include focal seizure days and DQ at 12 months as actionable variables for the managing clinician.
The sequential analysis of data from these TOSCA and TACERN registries evoked memories of reading newer editions of a reference book with updated information with each new edition. We eagerly await further analysis of ictal and interictal EEG abnormalities and whether they affect ultimate development.
Footnotes
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.