Early brain development in infants at high risk for autism spectrum disorder.
Level 3 - non-randomized controlled study
Prospective longitudinal controlled cohort study.
PubMed 28202961 · doi:10.1038/nature21369
What was done
In a prospective longitudinal neuroimaging study, researchers followed 106 infants at high familial risk for autism spectrum disorder (ASD) and 42 low-risk control infants. Brain magnetic resonance imaging (MRI) was acquired across early development (including 6, 12, and 24 months), and clinical ASD evaluations were conducted at 24 months. Researchers measured cortical surface area and overall brain volume trajectories, and evaluated a deep-learning algorithm using 6-to-12-month MRI surface area features to predict ASD diagnosis at 24 months.
What was found
Of the high-risk infants, 15 received an ASD diagnosis at 24 months. Hyperexpansion of cortical surface area between 6 and 12 months preceded brain volume overgrowth observed between 12 and 24 months in these 15 infants. Brain volume overgrowth was linked to the emergence and severity of autistic social deficits. A deep-learning algorithm based primarily on 6-to-12-month surface area MRI data predicted 24-month ASD diagnosis with a positive predictive value of 81% and a sensitivity of 88%.
Why it matters
This study shows that measurable structural brain changes emerge in high-risk infants during the first year of life prior to overt clinical diagnosis. These findings identify potential early neuroimaging biomarkers for presymptomatic ASD risk assessment.
Limits
The diagnosed ASD subgroup was small (n = 15). The cohort was restricted to infants with familial ASD risk, meaning findings may not generalize to sporadic cases in the general population. The abstract does not report specificity or negative predictive value for the predictive algorithm.
Cited by
- context Autistic children have neuroinflammation and on MRI scans their brains are physically enlarged.