An international team of scientists has found evidence that autism may include at least two distinct biological subtypes, each defined by a different pattern of communication across the brain. One subtype is marked by unusually high levels of connectivity between brain regions, while the other shows reduced connectivity. The discovery could help pave the way for more personalized approaches to autism diagnosis, care, and treatment.
The study was led by researchers at the Istituto Italiano di Tecnologia (IIT-Italian Institute of Technology) in Rovereto, Italy, and the Child Mind Institute in New York, with additional contributions from the University of Trento. Their findings were published in Nature Neuroscience.
Brain Connectivity Reveals Hidden Autism Subtypes
The research was coordinated by Alessandro Gozzi, PhD, director of the Center for Neuroscience and Cognitive Systems (CNCS) at IIT, and Adriana Di Martino, MD, founding director of the Autism Center at the Child Mind Institute.
According to the researchers, this is the first large-scale effort to systematically connect patterns seen in human brain imaging (via fMRI) with their underlying biological causes using mouse models. By linking specific brain connectivity patterns to distinct molecular processes, the work provides a foundation for future precision medicine strategies in autism.
To conduct the study, the team examined functional brain connectivity in 20 different mouse models and analyzed brain scans from 940 children and young adults with autism. These results were compared with scans from more than 1,000 neurotypical individuals.
The analysis revealed two consistent autism subtypes. One showed reduced communication between brain regions, known as hypoconnectivity, and was associated with synaptic pathways. The second showed increased communication between brain regions, known as hyperconnectivity, and was linked to immune-related biological systems. Together, these two groups represented about 25% of the individuals with autism included in the study.
“For decades, we’ve observed tremendous variability in how autism manifests, but we lacked direct evidence that these differences reflected distinct underlying biology,” said Dr. Alessandro Gozzi, at Italian Institute of Technology. “Our approach enabled us to isolate specific genetic and immune factors, then translate those signatures to human brain scans, showing that different connectivity patterns encode different mechanistic pathways underlying autism.”
Mouse Models Provide Biological Clues
The researchers combined brain imaging data with genetic and biochemical analyses in mice. This allowed them to connect specific patterns of brain connectivity with changes occurring at the cellular level.
Their work showed how molecular mechanisms involving synapses and the immune system can produce distinct connectivity patterns that can be detected using fMRI. These findings enabled the team to establish biological reference signatures in mice and then search for matching patterns in human brain scans.
“The mouse models gave us a biological ‘Rosetta Stone,'” said Dr. Adriana Di Martino at the Child Mind Institute. “We could see which biological pathways drive which connectivity signatures, then search for those same patterns in humans.”
Human Brain Imaging Confirms the Findings
The human imaging data came from the Autism Brain Imaging Data Exchange (ABIDE), a large international neuroimaging initiative co-founded by Dr. Di Martino that combines datasets from research centers around the world, as well as from the Child Mind Institute.
When the researchers analyzed the human data, they found the same hyperconnectivity and hypoconnectivity patterns identified in the mouse models.
Additional gene expression analyses strengthened the findings. Brain regions associated with hypoconnectivity showed enrichment of synaptic genes, while hyperconnected regions were enriched for immune-related genes. These results closely matched the biological mechanisms observed in the mouse studies.
Importantly, the same subtypes appeared consistently across multiple independent datasets, demonstrating that the findings were reproducible.
“Finding the same subtypes reproducible across dozens of independent research sites was critical validation,” added Dr. Gozzi.
Toward More Personalized Autism Care
The two subtypes also displayed differences in overall brain organization and showed modest differences on standard autism assessments. Individuals in the hyperconnectivity group tended to score somewhat higher on measures of autism severity.
“Brain-based biological markers reveal distinctions that current behavioral assessments don’t fully capture,” noted Dr. Di Martino.
The researchers caution that these two connectivity patterns likely represent only part of autism’s biological diversity. They believe additional subtypes may emerge as larger datasets become available and analytical methods continue to improve.
The study was supported through an international collaboration coordinated by the Italian Institute of Technology and the Child Mind Institute. Funding was provided by the Simons Foundation Autism Research Initiative, the European Research Council through the #DISCONN and #BRAINAMICS projects, the Brain and Behavior Foundation, Fondazione Telethon, and the US National Institute of Mental Health.