Abstracts

Functional Connectivity as an Emerging Biomarker to Predict Treatment Response for LGS

Abstract number : 1.104
Submission category : 2. Translational Research / 2C. Biomarkers
Year : 2023
Submission ID : 501
Source : www.aesnet.org
Presentation date : 12/2/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Blanca Romero Mila, BS – University of Calidornia Irvine

Virginia Liu, MD – Children's Hospital of Orange County; Natalie Benneian, Undergraduate Student – University of California Irvine; Maija Steenari, MD – Children's Hospital of Orange County; Donald Phillips, MD, MPH – Children's Hospital of Orange County; David Adams, MD – Children's Hospital of Orange County; Clare Skora, MD – Children's Hospital of Orange County; Daniel Shrey, MD – Children's Hospital of Orange County; Beth Lopour, PhD – University of California Irvine

Rationale:
Some 30-60% of patients with Infantile Epileptic Spasms Syndrome (IESS) will eventually develop Lennox-Gastaut Syndrome (LGS), which is characterized by the presence of a triad of encephalopathy, multiple seizure types, and a specific interictal EEG pattern with bursts of slow spike-wave (SSW) complexes and/or generalized paroxysmal activity (GPFA). This progression is gradual and difficult to quantify, with many patients experiencing cognitive impairment, intellectual problems, and psychiatric disorders by the time they receive treatment. In particular, the subjective assessment of the EEG has low interrater reliability and requires a binary decision based on the reader’s individual threshold for diagnosis. A quantitative biomarker could provide an objective, continuous measure, accounting for multiple factors. Based on prior reports of EEG-based functional connectivity as a biomarker for IESS, we hypothesized that this computational metric would also indicate the emergence of LGS and be modulated by treatment response.

Methods:
This prospective study included fifteen children diagnosed with IESS at the Children’s Hospital of Orange County who later progressed to LGS. Each patient had an EEG recording at the time of IESS and LGS diagnoses and a varying number of recordings in between. Functional connectivity networks were calculated using cross-correlation with strict statistical testing, as in prior studies. Each EEG recording was visually assessed for the presence of four standard clinical markers: hypsarrhythmia, spasms, generalized paroxysmal activity, and slow-spike-and-wave complexes. Functional connectivity strength was computed and correlated to disease progression, treatment response, presence of standard clinical markers, age, and amplitude of interictal spikes.

Results:
The connectivity strength was high at the time of IESS and LGS diagnosis. After treatment, the connectivity strength of non-responders was significantly higher than responders (p< 0.001). Subjects with strong pre-treatment functional connectivity networks (defined as having 50 or more connections that exceed threshold) were less likely to respond to treatment. Seven out of eight subjects with strong pre-treatment networks were non-responders, while only one out of five subjects with weak networks was a non-responder. Overall, connectivity strength was not correlated to age, but it was associated with the presence of clinical markers for LGS. However, no single clinical marker could be used as a predictor of connectivity strength.
Translational Research