Comparative proteome analysis of dried blood spots for high-risk group screening in children with autism spectrum disorder.

Scientists studied blood samples from 60 children (30 with autism, 30 without) to find proteins that might help diagnose autism. They found 7 specific proteins that were different between the two groups. These proteins could predict autism with 96.7% accuracy in this small study. The proteins were linked to how the body's immune system, muscles, and cells work. While promising, more research with larger groups is needed before this could be used clinically.

This study examined protein biomarkers in dried blood spots from 30 children with autism spectrum disorder (ASD) and 30 controls using advanced proteomic analysis. Researchers identified 33 candidate proteins from 849 quantified proteins, with 7 core proteins (PSME1, TPM1 isoforms, TPM3 isoforms, S100A6, and TBCA) showing strong discriminatory potential. These proteins were associated with skin, muscle, immune system, and cellular organization pathways. The core proteins correlated negatively with autism severity measures and positively with intellectual quotient.

A predictive model achieved high accuracy (AUC 0.956, sensitivity 96.7%, specificity 86.7%) in distinguishing ASD from control groups, suggesting potential for biomarker-assisted diagnosis.

While promising for future biomarker-assisted ASD diagnosis, these findings require validation in larger, independent cohorts before clinical implementation. The high accuracy suggests potential utility, but dried blood spot collection methods and proteomic analysis would need standardization for clinical use.

Small sample size (60 children total) limits generalizability. Single-site study without independent validation cohort. Cross-sectional design cannot establish causality. Authors acknowledge need for verification in independent cohorts before clinical application.

Reliable biomarkers that assist in the diagnosis of autism spectrum disorder (ASD) are limited. This study aimed to identify proteins that can differentiate children with ASD from controls. A total of 30 children with ASD and 30 control children participated in the study. Psychological tests and questionnaires to assess cognitive function, adaptive function, autism symptoms, and behavioral problems were administered.

Dried blood spots collected from the participants were analyzed using the SWATH LC-MS platform. Core proteins were identified to build a classifying model to predict ASD and control group status. Among the 849 proteins quantified, 33 candidate proteins were identified by combining two different algorithms. Candidate proteins were involved in biological pathways related to the skin, muscle functioning, immune system, and cytoskeleton organization.

Of the candidate proteins, we selected 7 core proteins that overlapped between different algorithms. The core proteins, PSME1, two isoforms of TPM1, two isoforms of TPM3, S100A6, and TBCA, were negatively correlated with the Childhood Autism Rating Scale, Aberrant Behavior Checklist, and Social Responsiveness Scale, and positively correlated with the Full-scale Intellectual Quotient. Furthermore, a logistic regression model with the core proteins predicted the ASD group with an area under the curve (AUC) of 0.956, sensitivity of 0.967, and specificity of 0.867. We performed a proteomic analysis of dried blood spot (DBS) from ASD and control group children to explore candidate biomarkers.

Our data supports the possibility of using proteins as potential biomarkers for ASD, although further verification is warranted in an independent cohort.