The Autism Innovation Coalition is a collaborative effort of leading scientists, clinicians, and policy advocates working to redefine autism understanding and care through systems biology, precision medicine, and innovation. The coalition brings together expertise across immunology, neurology, metabolism, genetics, psychiatry, and artificial intelligence to uncover the biological mechanisms that drive autism and its co-occurring conditions.
Our mission is to accelerate biological research to improve clinical practice and advance real-world solutions that improve health outcomes for individuals with autism. By integrating cutting-edge science, data, and technology, we can support existing behavioral frameworks using a more comprehensive, biologically grounded model of autism, one that recognizes its diversity, medical complexity, and potential for treatment. We aim to develop policy recommendations that can translate into actionable steps for adopting a medical based standard of care that utilizes a review of systems for identifying co-occurring and underlying conditions in autism.
Founding coalition partners include Dr. Robert Naviaux, Dr. Judy Van de Water, Dr. Sylvia Fogel, Dr. John Gaitanis, and Laura Cellini, representing a shared commitment to advancing scientific discovery, medical innovation, and compassionate care.
How to Read This Framework
The Autism Innovation Coalition approaches autism as a biologically heterogeneous neurodevelopmental condition, shaped by dynamic interactions across immune, metabolic, neurological, and developmental systems. Although autism is diagnosed behaviorally, individuals differ substantially in underlying biology, medical complexity, and response to intervention. We have adopted the terminology “Autism with Co-Occurring Medical Conditions” (ACMC) to describe individuals whose autism is accompanied by underlying medical, neurological, immunological, or metabolic conditions that affect functioning, health, and overall quality of life. ACMC distinguishes medically complex presentations from uses of the term “autism” that refer solely to neurodivergent identity without associated medical or biological conditions.
Throughout this website, the terms autism and ACMC may be used interchangeably, however, the focus of our work is on individuals whose symptoms reflect underlying medical complexity and therefore warrant biological investigation, clinical evaluation, and targeted treatment. We seek to address biologically mediated symptoms that may benefit from medical intervention, while recognizing that autism is described and understood in different ways across clinical, research, and community contexts.
In most cases, autism does not arise from a single gene, exposure, or event acting in isolation. Risk reflects the cumulative influence of multiple biological contributors, each adding incrementally to overall vulnerability. These contributors vary in both type and magnitude; ranging from immune dysregulation, mitochondrial or metabolic constraints, prenatal and placental factors, maternal autoantibodies, and genetic variants that shape resilience rather than determine outcome.
The content across this site is intentionally integrated, just like our biological systems. Each section, from genetics to maternal immune activation, from microglial dysfunction to mitochondrial metabolism, connects to the others because biological systems operate as an interconnected network rather than independently. Genetic variants influence metabolic capacity; metabolic stress alters immune signaling; immune activation shapes microglial behavior; microglial activation affects synaptic development and gut–brain signaling. Understanding any one domain requires understanding how it interacts with the rest. We advise taking in the entirety of the content in order to grasp a more complete understanding.
This framework is inherently individualized. What can make treating autism especially challenging is that no two individuals with autism share an identical biological profile. One child may have maternal autoantibody exposure combined with redox or NRF2 pathway vulnerability and subsequent microglial priming; another may have primary mitochondrial dysfunction amplified by recurrent infections without autoantibody involvement. Even shared clinical features, such as regression, fever-associated behavioral change, seizures, or gastrointestinal symptoms, can arise through different underlying mechanisms. Each individual’s primers, triggers, and amplifiers combine uniquely, which is why precision approaches require mapping biological profiles rather than applying one-size-fits-all explanations or interventions.
Our goal is to assist in translating biological findings into clinical insight. Rather than simply cataloging research findings, this site contextualizes what specific biological patterns may mean for real children and families. Why might one child regress after illness and temporarily improve during fever, while another shows opposite patterns? What do positive folate receptor autoantibodies suggest about CNS folate transport and potential responsiveness to leucovorin? Why do some children with identical genetic variants develop autism while their siblings do not? How do certain genetic variants increase vulnerability to specific immune or metabolic stressors? We connect molecular mechanisms to observable clinical phenomena, regression patterns, co-occurring medical conditions, treatment responses, and developmental symptoms, to support more informed assessment and medical decision-making.
Why This Framework?
Most pediatric training still presents autism primarily through behavioral, developmental, or broad genetic lenses. As a result, many clinicians receive limited exposure to the immune, metabolic, and neurobiological mechanisms increasingly documented in the research literature. Pediatricians routinely refer for therapies to treat the behavioral or communication aspects, but fail to recognize the underlying mechanisms that may be contributing to these symptoms. This gap is not a failure of individual clinicians, but a reflection of how medical education, guidelines, and specialty silos have historically framed autism. The consequence is that underlying biological contributors, such as immune dysregulation, mitochondrial stress, or neuroinflammatory processes, are often under-recognized in routine care, even when children present with clear medical comorbidities.
A Path Forward
Recognizing autism as a multi-system neurobiological condition opens new opportunities for clinical insight and care. When immune, metabolic, gastrointestinal, or neurological contributors are identified, clinicians may better understand why a child presents the way they do and why responses to interventions vary so widely. In some cases, addressing co-occurring medical conditions can reduce symptom burden, improve functional capacity, or enhance responsiveness to developmental and educational supports. Even modest improvements in sleep, gastrointestinal function, immune stability, or energy metabolism can meaningfully improve a child’s health, learning, behavior, and overall quality of life; and, in turn, reduce strain on families and caregivers. This approach does not replace behavioral or developmental care; it complements it by broadening the clinical lens and expanding the range of meaningful, individualized interventions.
Beyond immediate clinical application, this framework has implications for research priorities and health policy. When biological heterogeneity is recognized, research can move beyond searching for single causes toward mapping vulnerability patterns, and policy can support medical evaluation as standard practice rather than exception.
More than Two Decades of Evidence. No Medical Standard of Care.
For more than twenty years, research has documented that autism is not a brain-isolated condition. Studies published in leading medical journals have identified immune activation and neuroinflammation, gastrointestinal pathology, mitochondrial dysfunction, maternal autoantibodies affecting fetal brain development, cerebellar abnormalities, neuronal signaling imbalances, abnormal cytokine profiles, folate blocking autoantibodies, and fever-responsive symptom changes in biologically defined subsets of individuals with autism. These findings include systematic reviews, meta-analyses, and expert consensus statements, and have been replicated across countries and disciplines.
Yet clinical care has not kept pace with the biological science. Children who regress after infection often receive no immune evaluation. Chronic gastrointestinal symptoms often go untreated. Metabolic or mitochondrial dysfunction is rarely considered. Temporary improvement during fever is dismissed rather than investigated. Autism is still treated as if it exists apart from the biological underpinnings that may be driving the presentation.
The Autism Innovation Coalition exists to help close this gap. We bring together scientists, clinicians, and policy experts to translate established biological evidence into a modern, systems-based medical framework for autism. Our goal is to accelerate the adoption of a standard of care that reflects what the science already shows: autism is biologically heterogeneous, medically complex, and responsive to informed investigation and treatment.
The Framework
Autism cannot be explained by a single gene, exposure, or event. The complexity of the condition demands a framework that accounts for primers, triggers, and amplifiers and recognition that the trajectory is unique for each individual. Not every path to development of autism is exactly the same, but a framework helps us understand how various impacts can shape it. This framework does not blame one factor as a singular cause. Rather, it reveals autism as a complex process shaped by interactions across multiple biological systems, with outcomes that depend on timing, context, and contributors that confer exceptional vulnerability.
We understand that for many (especially the regressive subtype in a vulnerable child) autism is the result of a cumulative cascade: genetic predisposition + early immune/metabolic stressors + failure of normal resolution. It explains why no two children with autism are exactly alike; different combinations of genes, different triggers, timing, and various amplifying factors converge on common biological pathways and derail normal development. It is the cumulative and interactive effect of multiple risk factors that may be most consequential for tipping a child into the autism spectrum. Future research should be prioritized to examine the synergistic effects of multiple exposures and timing for exceptionally vulnerable populations.
Primers
Every child is born with a biological threshold: the extent of the body’s ability to achieve homeostasis. Some infants have greater resilience, and some have increased vulnerability. Primers are the various factors that can lower this threshold. Genes, prenatal immune activation, maternal nutrition, metabolic health, and environmental exposures are examples of primers that determine a child’s biological threshold. Primers do not necessarily cause autism by themselves. They lower the system’s defenses and influence how energy, oxidative balance, and immune regulation is managed.
Triggers
Triggers are acute biological stressors that push a vulnerable system past its threshold. When a child with underlying primers encounters additional strain, the system may shift towards dysregulation. Stressors include infections, environmental toxicants, anesthesia, or tightly timed immune-stimulus events that activate latent vulnerabilities. Triggers occurring during prenatal development, infancy, or early toddlerhood have greater effects because microglia, mitochondria, and synaptic networks are undergoing rapid maturation. In these early windows, the same stressor that would be inconsequential later in life can redirect developmental trajectories.
Amplifiers
Amplifiers sustain cascades and promote looped patterns after a trigger has occurred. They are not typically initiating events; they are feedback cycles that prevent the system from returning to balance. Once a biological threshold is crossed, cycles such as chronic inflammation, oxidative stress, mitochondrial energy deficits, impaired synaptic pruning, and altered gut-immune signaling can maintain the system in a defensive state. These amplifiers create a pattern where each subsequent stressor evokes a larger, more prolonged reaction. In this stage, even mild immune stimuli or metabolic challenges can feel destabilizing because the underlying biology has shifted from normal growth and repair.
Inside the Biology
Cell Danger Response (CDR)
When cells sense a threat, they shift from building to defense. If the signal never turns off, development stalls, but it can be reset.
Maternal Immune Activation
Immune signals in pregnancy can program in the fetal brain. NIH data show ~12-17% of Autism cases trace to this pathway, and that it's preventable.
Autoantibodies
Sometimes the immune system mistakes itself for the enemy. Targets such as FRAA and CRMP1/2 link directly to speech and motor.
Cytokines and Microglia
The brain's immune cells guide pruning and learning. When over-activated, they cause a regression in learned skills, but a fever can temporarily restore those skills.
Gut-Brain Axis
The gut talks to the brain every day. Inflammation, dysbiosis, and barrier leaks send stress signals that affect behavior.
Genetics in Autism
How Variants Influence Metabolism, Immunity, Redox Balance, Neurodevelopment, and Clinical Vulnerability
Mitochondria & Redox Metabolism
Mitochondria generate more than energy, they orchestrate redox balance, metabolic flexibility, and cellular responses to stress.
Neuroimmune and Autoimmune Disorders
Abrupt changes in behavior or cognition, movement abnormalities, catatonia-like shutdown, or seizure activity following infections, immune activation, metabolic stress, or environmental exposures.
Seizure Disorders & Electrical Instability
Seizure disorders affect a substantial portion of individuals with autism. When longitudinal data are considered, up to 70% demonstrate overt seizures, subclinical epileptiform activity, sleep-dependent electrical disturbances, or electrical instability.
References
Leisman, G., & Melillo, R. (2025). Autism Spectrum Disorder: What Do We Know and Where Do We Go?. *Brain sciences*, 15(9), 1010. https://doi.org/10.3390/brainsci15091010 PMID: 41008370
Frasch, M. G., Yoon, B. J., Helbing, D. L., Snir, G., Antonelli, M. C., & Bauer, R. (2023). Autism Spectrum Disorder: A Neuro-Immunometabolic Hypothesis of the Developmental Origins. *Biology*, 12(7), 914. https://doi.org/10.3390/biology12070914 PMID: 37508346
Naviaux R. K. (2025). A 3-hit metabolic signaling model for the core symptoms of autism spectrum disorder. *Mitochondrion*, 87, 102096. https://doi.org/10.1016/j.mito.2025.102096 PMID: 41242673
Vargas, D. L., Nascimbene, C., Krishnan, C., Zimmerman, A. W., & Pardo, C. A. (2005). Neuroglial activation and neuroinflammation in the brain of patients with autism. *Annals of Neurology*, 57(1), 67-81. https://doi.org/10.1002/ana.20315 PMID: 15546155
Fatemi, S. H., Aldinger, K. A., Ashwood, P., Bauman, M. L., Blaha, C. D., Blatt, G. J., Chauhan, A., Chauhan, V., Dager, S. R., Dickson, P. E., Estes, A. M., Goldowitz, D., Heck, D. H., Kemper, T. L., King, B. H., Martin, L. A., Millen, K. J., Mittleman, G., Mosconi, M. W., Persico, A. M., Sweeney, J. A., Webb, S. J., & Welsh, J. P. (2012). Consensus paper: pathological role of the cerebellum in autism. *Cerebellum*, 11(3), 777-807. https://doi.org/10.1007/s12311-012-0355-9 PMID: 22370873