This article is more than 5 years old. Information may no longer be current.

'Real breakthrough' discovered in early diagnosis of autism

Hyperexpansion of cortical surface areas from ages 6 to 12 months may play a critical role in the development of autism, according to a research letter that recently appeared in Nature.

The researchers’ findings could also help develop new autism treatments, as well as help parents with one child who has autism determine the risk for future children acquiring the condition, according to a press release.

"The results of this study are a real breakthrough for early diagnosis of autism," Robert T. Schultz, PhD, Center for Autism Research, Children’s Hospital of Philadelphia, said in the release. "While we have known for some time that autism emerges in subtle, gradual ways over the first few years of life, this study offers the first firm evidence before a child's first birthday predicting whether certain high-risk children are likely to be diagnosed with autism.”

Researchers prospectively studied MRIs from 106 infants at high-risk for autism spectrum disorder (ASD) and 42 low-risk infants who had MRI scans obtained at ages 6, 12 and 24 months. They reported that hyperexpansion of the cortical surface area between ages 6 to 12 months preceded the brain volume overgrowth that was observed between 12 and 24 months in 15 high-risk infants who were diagnosed with autism at 24 months, and that this overgrowth was connected to the emergence and severity of autistic social deficits. They also stated that a deep-learning algorithm that primarily uses surface area information from MRIs of the 6- to 12-month-old infants’ brains predicted the diagnosis of autism in individual high-risk children at 24 months (with a sensitivity of 88% and a positive predictive value of 81%). According to the researchers, these findings show that early brain changes occur during the period in which autistic behaviors are first emerging.

In addition, researchers stated the clinical best estimate for ASD group showed a significantly increased surface area growth rate from 6 to 12 months of age compared to both low familial risk and those that did not meet the ASD criteria groups, with the most robust increases observed in the right lingual gyrus, right cuneus and left/right middle occipital gyrus areas. Researchers also stated that they saw a significant correlation between surface area growth rate of 6 to 12 months and enlargement in total brain volume at 24 months of age in all subjects (r₁₉₂ =  0.59, P <  0.001), as well as in the combined high familial risk subgroup (r₁₃₉ =  0.63, P <  0.001), and a significant correlation between the 24-month autism diagnostic observation schedule severity score and the 12 to 24 month total brain volume change rate (r₁₉₃ =  0.16, P =  0.03). Analyses intended to look into the components of overall autism severity autism diagnostic observation schedule severity score during the 12 to 24 month total brain volume change rate revealed a significant correlation between this group and the 24-month autism diagnostic observation schedule social affect score (r₁₉₄ =  0.17, P =  0.01), but not the autism diagnostic observation schedule restricted/repetitive behavior score (r₁₉₄ =  0.07, P =  0.31). According to researchers, the Communication and Symbolic Behavior Scale social composite score was significantly correlated with a more rapid total brain volume change rate at 12 to 24 months (r₁₅₈ =  0.18, P =  0.02) in high familial risk subjects. The researchers also reported a significant group (best estimate criteria for ASD vs. those who did not meet the criteria for ASD) × time (12-24 months) interaction for the Communication and Symbolic Behavior Scale social composite score (F _2,130 =  10.0, P <  0.0001). This finding was further supported by the observation that the Communication and Symbolic Behavior Scale effect size almost tripled from 12 (d =  0.39) to 24 (d =  1.22) months. Researchers also stated they used a 10-fold cross-validation to determine classification performance, whereby the whole classification procedure, including network training was performed separately in each fold. This classification scheme set the best-estimate for ASD group apart from the group that did not meet the ASD criteria in the cross-validation with 94% accuracy (n =  168 out of 179), 88% sensitivity (n =  30 out of 34), 95% specificity (n =  138 out of 145), 81% positive predictive value (n =  30 out of 37) and 97% negative predictive value (n =  138 out of 142).

Schultz noted that the findings need to be replicated in future studies before they can be put into clinical practice, but if they are validated, could impact how autism screenings are conducted.

“If we are able to replicate these results in further studies, these findings promise to change how we approach infant and toddler screening for autism, making it possible to identify infants who will later develop autism before the behavioral symptoms of autism become apparent," he said in the release. – by Janel Miller

Disclosure: The researchers report no relevant financial disclosures.