The Spectrum Brief
Autism research, in plain language
← Back to the feed

Diagnosis & ScreeningResearch

New Biomarker Tests Could Revolutionize Early Autism Diagnosis

Eye-tracking, microbiome analysis, and digital tools show promise for earlier, more accurate autism screening, but challenges remain.

By The Spectrum Brief newsroom · 2 hours agoPeer-reviewed
Share

The Promise of Biomarkers for Early Autism Diagnosis

Early diagnosis of autism spectrum disorder (ASD) can dramatically improve outcomes for children, but current screening methods often rely on subjective observations or lengthy evaluations. Now, a convergence of new biomarker-based approaches—from eye-tracking to stool analysis—is offering hope for faster, more objective screening tools.

Eye-Tracking: Measuring Social Engagement

Eye-tracking technology measures how children visually engage with social stimuli, such as faces or interactive scenes. A recent study in JAMA Network Open found that this method achieved 78% sensitivity and 85% specificity in aiding autism diagnosis within primary care settings. While promising, experts note that false positives and negatives can occur, and the technology requires specialized equipment. Pilot programs, like those highlighted by Contemporary Pediatrics, are testing its feasibility in clinics.

The Microbiome Connection: Exploring Gut-Brain Links

Research into the gut microbiome has identified distinct microbially-derived metabolites—chemical byproducts of gut bacteria—that may signal a specific ASD subtype. A study published in Nature found elevated levels of certain metabolites in children with autism, but these findings are preliminary. Factors like diet and environment can influence microbiome composition, and independent validation is lacking. Projects like the CUHK pilot stool test aim to explore this further, but clinical use is likely years away.

Digital Phenotyping and AI: Scalable Screening Tools

Digital phenotyping uses smartphones or wearables to track behavioral patterns, such as vocalizations or movement. A Nature Medicine study showed that these tools could predict autism risk with 82% accuracy in toddlers. Meanwhile, AI analysis of medical records, as demonstrated in a 2021 Science study, reduced false positives by 30% in screening. However, ethical concerns about data privacy and over-reliance on algorithms remain unresolved.

Challenges and Next Steps

While these advancements are exciting, significant hurdles remain. A Frontiers in Neuroscience review emphasizes the need for larger, more diverse studies to validate biomarkers across populations. Cost and accessibility are also barriers, particularly for low-resource settings. Additionally, early diagnosis carries risks, such as labeling or unnecessary interventions, which must be weighed against the benefits of timely support.

What These Tools Mean for Families

For parents, earlier diagnosis could mean access to interventions like speech or behavioral therapy during critical developmental windows. However, false positives—incorrectly identifying a child as at risk—can lead to unnecessary stress. Resources like Autism Speaks provide guidance for navigating uncertain results. As research progresses, transparency about the limitations and timelines of these tools will be key for families.

#biomarkers#earlydiagnosis#autismresearch#eye-tracking#microbiome#digitalphenotyping
Share

Behind the brief

Adversarial editorial review

Approved70/100 consensus· 2 rounds

Open thread

Discussion

0 comments · The editorial board joins in. Be kind and cite sources where you can.

Loading comments…