Longitudinal Electroencephalogram and Serum Neurotrophic Factor Profiling for Early Prediction of Autism Spectrum Disorder in East Mediterranean Infants

Abstract

Autism Spectrum Disorder (ASD) affects approximately 1.8% of children in the East Mediterranean region, with diagnosis often delayed until after age 3 due to limited screening resources. In this prospective cohort study, 862 infants aged 6 to 24 months were enrolled across four pediatric centers from January 2023 to December 2024. Serial electroencephalogram (EEG) recordings and serum levels of brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF) were measured at baseline (6 months), mid-point (12 months), and follow-up (24 months).

By age 24 months, 72 infants (8.3%) met criteria for ASD based on standardized developmental assessments. Early EEG showed increased theta-band power (4–7 Hz) in 68% of later-diagnosed infants compared to 21% of neurotypical peers (p < 0.001). Median BDNF levels in the ASD group were elevated by 27% at 12 months (median: 24.9 ng/mL; IQR: 21.1–29.3) versus controls (19.6 ng/mL; IQR: 17.2–22.0). NGF levels showed a corresponding increase of 19% (p = 0.003).

A combined predictive model using theta power increase ≥15%, BDNF ≥22 ng/mL, and NGF ≥18 ng/mL yielded an AUC of 0.91 (95% CI: 0.88–0.94), sensitivity 86%, and specificity 89%. Notably, this model identified ASD risk at a median age of 14.8 months—approximately 1.5 years earlier than typical clinical diagnosis (p < 0.001). Early identification enabled targeted behavioral interventions for 53 infants (73% of ASD group), who showed a 32% improvement in adaptive behavior scores (Bayley-III) at 36 months.

This study demonstrates that combining EEG and serum neurotrophin profiling enables early and accurate ASD risk prediction in infants. This cost-effective, easily scalable screening protocol could transform neurodevelopmental care in the East Mediterranean by enabling earlier diagnosis, intervention, and improved long-term outcomes.