590 - Altered Visually Evoked Potential in Children with Fragile X Syndrome

Publication Number: 3571.590

Janani C. Elumalai, Boston Children's Hospital, Brookline, MA, United States; Winko An, Boston Children's Hospital, Boston, MA, United States; Carol Wilkinson, Boston Children's Hospital, Brookline, MA, United States

Background: Fragile X Syndrome (FXS) is the most common monogenetic cause of autism and is the most prevalent inherited cause of intellectual disability. Understanding the neurobiology underlying FXS is critical to identifying effective therapies. Electroencephalography (EEG) is a non-invasive neuroimaging method that can measure network-level responses to sensory stimuli. The visually evoked potential (VEP), measured by EEG, evaluates visual sensory processing and has potential for both elucidating underlying neurobiology and serving as a biomarker for clinical outcomes within neurodevelopmental disorders.

Objective: (1) To characterize VEP alteration in FXS compared to age-matched typically developing (TD) children.
(2) To explore the association between VEP measures and the verbal and nonverbal development quotient (VDQ and NVDQ) in FXS.

Design/Methods: EEG data were collected from 2-6.5 years old children with FXS (n=19) and TD children (n=19). A maximum of 200 trials were collected using the checkerboard paradigm reversing every 500ms for approximately 2 minutes. Video coding or eye-tracking data was used to remove trials with noncompliant behaviors. We compared VEP component amplitudes as well as inter-trial phase coherence (ITPC) between groups. The Preschool Language Scales were administered to both cohorts and the Mullen Scales of Early Learning were administered to the FXS cohort (n=19) and the TD cohort (n=17). VDQ and NVDQ were compared with VEP component amplitudes.

Results: Children with FXS exhibited a significantly larger P1-N1 difference (p=0.007, Fig 1B) and a significantly increased P2 component at ~200ms (p=0.013, Fig 1B) compared to TD children. No significant group differences were found for N1, P1, or N2 components. Aligned with P1 and P2 timing, children with FXS exhibited increased ITPC at 10-20Hz 100ms after stimulus presentation and at 5Hz between 100ms and 200ms (Fig 2). Cohort significantly moderated the relationship between the N1 component and NVDQ (p=0.042). In TD children, this relationship had a significant positive association (p=0.041), but was not significant in FXS (Fig 3A). Cohort also significantly moderated the relationship between the P1 component and VDQ (p=0.047), but this relationship was not significant within each cohort (Fig 3B).

Conclusion(s): Children with FXS exhibit an atypical VEP with significantly increased P1-N1 difference and an additional peak at 200ms. However, VEP components were not associated with nonverbal or verbal development. Future analyses will explore associations with other clinical outcome measures to inform VEP's utility as a potential biomarker.

VEP Component Differences
Fig1_VEP Component Differences_PAS Abstract Figures.pdfFigure 1: A) VEP Waveform determined by Spatial Principal Component Analysis for FXS (n=19) and age-matched TD cohorts (n=19). B) VEP Component Amplitudes by Cohort.

ITPC in Children with FXS and TD Children
Fig2_ITPC_PAS Abstract Figures.pdfFigure 2: Inter-Trial Phase Coherence for FXS (n=19) and TD (n=19) cohorts.

Linear Regression Models for VEP Components and NVDQ/VDQ
Fig3_NVDQ_VDQ_PAS Abstract Figures.pdfFigure 3: A) Linear Regression of N1-NVDQ Relationship for FXS (n=19) and TD (n=17) cohorts. B) Linear Regression of P1-VDQ Relationship for FXS (n=19) and TD (n=19).