Functional Network Mapping Reveals State-Dependent Response to IGF1 Treatment in Rett Syndrome

Rett Syndrome (RTT) is a neurodevelopmental disorder associated with mutations in the gene MeCP2, which is involved in the development and function of cortical networks. The clinical presentation of RTT is generally severe and includes developmental regression and marked neurologic impairment. Insulin-Like growth factor 1 (IGF1) ameliorates RTT-relevant phenotypes in animal models and improves some clinical manifestations in early human trials. However, it remains unclear whether IGF1 treatment has an impact on cortical electrophysiology in line with MeCP2’s role in network formation, and whether these electrophysiological changes are related to clinical response. We performed clinical assessments and resting-state electroencephalogram (EEG) recordings in eighteen patients with classic RTT, nine of whom were treated with IGF1. Among the treated patients, we distinguished those who showed improvements after treatment (responders) from those who did not show any changes (nonresponders). Clinical assessments were carried out for all individuals with RTT at baseline and 12 months after treatment. Network measures were derived using statistical modelling techniques based on interelectrode coherence measures. We found significant interaction between treatment groups and timepoints, indicating an effect of IGF1 on network measures. We also found a significant effect of responder status and timepoint, indicating that these changes in network measures are associated with clinical response to treatment. Further, we found baseline variability in network characteristics, and a machine learning model using these measures applied to pretreatment data predicted treatment response with 100% accuracy (100% sensitivity and 100% specificity) in this small patient group. These results highlight the importance of network pathology in RTT, as well as providing preliminary evidence for the potential of network measures as tools for the characterisation of disease subtypes and as biomarkers for clinical trials.