PRAGUE—Last week, more than 2,200 people from around the globe gathered to share, learn and discuss the latest in autism research at the 25th International Society for Autism Research (INSAR) annual meeting. Over the past quarter century, the gathering has expanded beyond a core group of basic scientists and clinicians studying the condition’s genetics and neurodevelopmental trajectories to also include many autistic researchers and self-advocates exploring adult health concerns and lived experiences. Compared with the inaugural meeting in 2001—which included about 200 abstracts and 500 attendees—this year’s meeting had 2,536 submissions from 67 countries on a broad range of topics.
At last year’s conference, INSAR’s 2025 science chair noted that the meeting was happening during “unprecedented political times”—a sentiment echoed this year. “For many of us, this continues to be a difficult time, between global conflict, political uncertainty and the persistent barriers families face in accessing essential services,” said Christine Wu Nordahl, professor of psychiatry and behavioral sciences at the University of California, Davis MIND Institute, in her opening remarks. Amid unproven and harmful claims about causes of and treatments for autism, “science as a source of truth” needs to be affirmed, she added.
Here are a few themes that emerged during the four-day meeting.
The search for subtypes
Many posters and presentations explored an ongoing and much-discussed goal: the stratification of autism into meaningful subtypes. Researchers shared findings from studies that used a wide range of techniques, such as behavioral phenotyping, genomics, transcriptomics and neuroimaging.
For example, Christine Ecker, professor of child and adolescent psychiatry, psychosomatics and psychotherapy at Goethe University Frankfurt, described using MRI scans combined with transcriptomics data to identify three novel subgroups of autistic people. Other presenters talked about their use of longitudinal data to identify neural and behavioral predictors for support needs and cognitive growth, among other outcomes.
Several speakers acknowledged the challenges of parsing autism’s heterogeneity. “Is there one subtyping scheme that rules them all? It depends on what your goals are,” said Michael Lombardo, senior researcher at the Italian Institute of Technology. For instance, there may be different approaches to stratification that are best suited for investigating biological mechanisms as opposed to predicting treatment outcomes, he explained.
Autism genetics in the spotlight
During her keynote, Angelica Ronald, professor of psychology and genetics at the University of Surrey, described efforts to understand neurodevelopmental conditions by examining the role of genetics in infant and toddler development. She shared findings from a recent meta-analysis of genome-wide association studies that identified genes linked with the age at onset of walking in more than 70,000 infants with European roots. Genes linked with earlier walking were associated with an increased likelihood of attention-deficit/hyperactivity disorder, whereas those tied with later walking were associated with a higher chance of early-diagnosed (but not late-diagnosed) autism.
In a later session, Jakob Grove, professor of biomedicine at Aarhus University, described distinct genetic profiles of people with early and late diagnosis. Other researchers shared findings from studies aimed at understanding the genetic landscape of autism in early life—for example, looking at the role of genes in the delay or loss of social communication skills and the overlap between autism and preterm birth. Genetics also cropped up in presentations about therapeutics and precision medicine.
Autistic self-advocates have raised ethical questions about the goals and values of genetics studies in recent years. Ronald and others spoke of this tension, while also emphasizing the potential benefits of such research, such as understanding why certain traits co-occur, providing opportunities for early diagnosis and intervention, and fighting misinformation.
Sensory over-responsivity in the brain
Multiple researchers shared new findings on the neuroscience of sensory over-responsivity, a heightened response to sensory stimuli that is common among autistic people and linked to anxiety.
Olivia Surgent, a postdoctoral scholar in Nordahl’s lab, described a “sensory superhighway,” consisting of subcortical brain areas involved in sensory relay and integration, that shows unique differences in autistic people with a specific phobia that emerge as early as age 3. Shulamite Green, associate professor of psychiatry and biobehavioral sciences at the University of California, Los Angeles’ Semel Institute, presented work revealing differences in the brain and bodily responses to sensory overwhelm and anxiety in autistic children.
The search for biomarkers
Several investigators from the Autism Biomarkers Consortium for Clinical Trials (ABC-CT) Study presented findings from their decade-long, multicenter effort to track more than 400 autistic and 300 non-autistic children. Of the many eye-tracking and electroencephalography (EEG) measures the group examined, just two—an EEG measure (N170 latency) and oculomotor index of gaze to human faces, or OMI (an eye-tracking marker)—emerged as robust, meaning they could be reliably collected, showed clear differences between autistic and non-autistic people, and remained fairly stable over time. However, because these two biomarkers showed only modest relationships with phenotypes such as differences in the ability to process faces, it’s still unclear whether they will be useful for clinical trials.
Some biomarker presentations highlighted other possible directions, including proteomics, metabolomics and other “omics” techniques. And David Amaral, director of research at the University of California, Davis MIND Institute, described the identification of extra brain fluid (extra-axial fluid) as a potential early biomarker of autism.
Neuroimaging profound autism
Profound autism received much attention in several dedicated sessions. Speakers emphasized that people who fall under the umbrella of profound autism are critically underrepresented in both basic and clinical research. Some described efforts to conduct neuroimaging studies in this population. Nordahl, for example, shared preliminary findings from a neuroimaging study that included autistic children with IQs in the range of profound autism, which revealed that those who engaged in severe self-injurious behavior have smaller left amygdala volumes than those who did not.
Stewart Mostofsky, director of the Center for Neurodevelopmental and Imaging Research at the Kennedy Krieger Institute, presented preliminary findings from studies that used a novel form of functional near-infrared spectroscopy known as high-density diffuse optical tomography to image the brains of minimally speaking autistic people.
Beyond basic science
Many sessions explored clinical and day-to-day challenges that people with autism experience. For example, Damian Milton, senior lecturer in intellectual and developmental disabilities at the University of Kent, gave a keynote speech about the “double empathy problem” that autistic and non-autistic people encounter in conversation.
In another keynote speech, David Mandell, professor of psychiatry at the University of Pennsylvania Perelman School of Medicine, presented findings revealing that interventions that succeed in clinical trials are often much less effective once they’re rolled out to the community.
“What we’re learning in research isn’t translating into practice,” Mandell said.
