According to the CDC, parents report that 17% of children from ages 3-17 years — approximately 1 in 6 — are diagnosed with a developmental disorder, including autism spectrum disorder (ASD). Children with ASD account for 1 in 54 children overall, but the diagnosis is often subjective and difficult. The average age of ASD diagnosis of children in the U.S. is 4 years old. As is the case with many diseases and conditions, diagnosing ASD as early as possible is highly beneficial. The earlier the diagnosis, the sooner a child with ASD can begin therapies to improve behavior flexibility and communications and social skills.

Researchers at UT Southwestern report they have discovered autism blood biomarkers that can assist with ASD diagnosis. The researchers published a report on their discovery in PLOS One this month. The UT Southwestern team collected blood samples from 76 boys with ASD and 78 boys with typical development. The research entailed examining more than 1,100 proteins from the blood serum samples using the SomaLogic protein analysis platform. According to SomaLogic, the platform produces “7,000 reproducible measurements of circulating proteins from a single sample.” The UT team identified nine proteins most likely to predict ASD and validated the data with machine learning.

According to Dwight German, Ph.D., UT Southwestern professor of psychiatry and the senior author of the study, “The more significantly affected the child is, the higher or lower than normal the blood marker is.” Next steps in the UT Southwestern research include further studies to validate the current findings.

Speeding up the age of ASD diagnosis is a wonderful prospect with the potential to significantly improve the lives of children with ASD as well as their siblings and parents. Taking two steps further, we’re encouraged by the further potential of machine learning in detecting additional disorders and conditions based on blood serum samples. We don’t see this coming anytime soon, but the possibility of home blood sample testing kits to discover otherwise undetectable information about individual health is exciting indeed. If a wearable could monitor blood biomarkers continuously the impact on the health system would be industry-changing.