Autism Spectrum Disorder

🏷

Classifications

Harrison's Part

📘

Definitions

1

Contemporary clinical ASD case definition

A clinical diagnosis requires persistent social-communication and social-interaction differences or impairments plus restricted or repetitive behavior, interests, or activities. Features begin in the early developmental period, may become fully apparent only when demands exceed capacities, cause clinically meaningful functional impact, and are not better explained by intellectual disability or global developmental delay alone. Sensory hyperreactivity, hyporeactivity, or unusual sensory interests are included within the restricted/repetitive domain. Support needs and language and intellectual-function specifiers do not define etiologic subtypes.

CASE_DEFINITION

Show evidence (2 references)

PMID:31949163 SUPPORT Other

"Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life."

The primer states the defining domains and early developmental onset.

PMID:31843864 SUPPORT Other

"Core deficits are identified in 2 domains: social communication/interaction and restrictive, repetitive patterns of behavior."

The AAP clinical report independently confirms the two-domain definition.

◈

Mechanistic Hypotheses

2

Subgroup-specific excitation-inhibition and synaptic-signaling model

subgroup_ei_synaptic_model EMERGING

Evidence balance 1 support

Some genetic and cellular ASD subgroups may reach similar circuit-level phenotypes through altered excitatory or inhibitory synaptic signaling. Direction and developmental timing may differ, and the model is not a universal ASD mechanism or a basis for general treatment selection.

Pathograph links

Heterogeneous neurodevelopmental molecular effects -> Excitation-inhibition and synaptic-signaling differences in studied subgroups INDIRECT UNKNOWN INTERMEDIATES Within the subgroup-specific E/I hypothesis, synaptic and transcriptional perturbations may alter excitation or inhibition through genotype-, region-, cell-type-, and developmental-stage-specific routes.

Show evidence (1 reference)

PMID:33076974 SUPPORT In Vitro

"We find that ASD patient-derived neurons with a functional loss of TSC2, in addition to possessing neuronal hyperactivity, develop a dysfunctional neuronal network with reduced synchronisation of neuronal bursting and lower spatial connectivity."

A genotype-specific cellular model motivates, but cannot universalize, the hypothesis.

Immune and microbiome modifiers in selected ASD subgroups

immune_microbiome_modifier_model EMERGING

Evidence balance 1 support

Immune state, gastrointestinal physiology, diet, and microbiome composition may modify symptoms or co-occurring conditions in selected people. Current evidence does not establish that immune activation, blood-brain-barrier disruption, enteric nervous-system dysfunction, or dysbiosis is a primary cause of the broad ASD diagnosis.

Pathograph links

Immune and neuroinflammatory associations in studied groups -> Distributed developmental circuit differences INDIRECT UNKNOWN INTERMEDIATES Within the emerging modifier hypothesis, immune and metabolic differences are associated with synaptic and network measures; directionality and mediation of clinical traits remain unestablished.

Gut microbiome associations in studied groups -> Immune and neuroinflammatory associations in studied groups INDIRECT UNKNOWN INTERMEDIATES Within the emerging modifier hypothesis, dysbiosis may affect immune and neuroinflammatory signaling. Confounding and reverse causation remain major alternatives.

Show evidence (1 reference)

PMID:41550027 SUPPORT Other

"In autism and ADHD, altered tight junction protein profiles imply more subtle or context-dependent barrier dysfunction."

The source itself frames barrier findings as context dependent.

?

Discussions and Knowledge Gaps

6

Which reproducible combinations of genotype, development, physiology, and phenotype define treatment-relevant ASD subgroups without collapsing heterogeneous people into a universal biomarker signature?

KNOWLEDGE GAP OPEN gap_asd_validated_biological_stratification

Attached to

pathophysiology#Polygenic neurodevelopmental liability pathophysiology#Rare coding and copy-number variant liability pathophysiology#Distributed developmental circuit differences

Imaging, EEG, immune, metabolomic, microbiome, and molecular differences are generally group associations with limited replication or specificity. No validated biological test diagnoses broad ASD or selects a general core-symptom treatment. Large, diverse, longitudinal, externally replicated studies with prespecified individual-level performance are needed.

Show evidence (1 reference)

PMID:40274203 SUPPORT Human Clinical

"Overall, certainty of the evidence was low and very low except for ADOS-2 (moderate)."

Even established behavioral tools have important evidence limitations.

Which supports improve self-defined quality of life, health, communication, autonomy, safety, participation, housing, and employment across diverse autistic adults and people with high communication or support needs?

KNOWLEDGE GAP OPEN gap_asd_lifespan_natural_history_and_supports

Attached to

progression#Adulthood and later life

Adult and later-life cohorts, people with intellectual disability or limited speech, and people outside high-income settings remain underrepresented. Outcomes should extend beyond symptom scores and be selected with autistic people and families.

Show evidence (1 reference)

PMID:41926193 SUPPORT Human Clinical

"Thirty-five studies were included (N = 1,631 total participants)."

The modest adult evidence base motivates broader, more representative lifespan research.

Do immune or microbiome changes causally modify defined symptoms or co-occurring conditions in reproducible ASD subgroups, or are they consequences of diet, medication, gastrointestinal disease, stress, sampling, and other confounding?

CONTROVERSY OPEN gap_asd_causal_status_of_immune_and_microbiome_associations

Attached to

pathophysiology#Immune and neuroinflammatory associations in studied groups pathophysiology#Gut microbiome associations in studied groups

Current reviews supply biological plausibility and group associations but not a general causal ASD pathway. Mechanistic trials require prespecified subgroups, appropriate non-autistic and gastrointestinal controls, longitudinal sampling, and clinically meaningful outcomes.

Show evidence (1 reference)

PMID:41550027 SUPPORT Other

"In autism and ADHD, altered tight junction protein profiles imply more subtle or context-dependent barrier dysfunction."

The context-dependent wording illustrates the unresolved causal status.

Does hypothalamic-pituitary-adrenal (HPA) axis dysregulation with impaired glucocorticoid-responsive immune (GRI) signaling in peripheral immune cells (monocytes, natural killer cells, B cells; reported core genes LRFN1, NUAK2, TMEM154, GAPT) causally couple the immune and microbiome associations reported in autism spectrum disorder to those reported in the frequently co-occurring irritable bowel syndrome, or is the shared GRI signature a common downstream stress response in both conditions?

KNOWLEDGE GAP OPEN gap_asd_ibs_hpa_gri_brain_gut_coupling

Attached to

pathophysiology#Immune and neuroinflammatory associations in studied groups pathophysiology#Gut microbiome associations in studied groups Irritable_Bowel_Syndrome:pathophysiology#Gut-Brain Axis Dysfunction Irritable_Bowel_Syndrome:pathophysiology#Immune Activation and Mast Cell Degranulation

Functional gastrointestinal symptoms co-occur with ASD, and a shared glucocorticoid-responsive immune transcriptional signature in peripheral blood mononuclear cells has been proposed as a molecular link between ASD and IBS. The available evidence, however, is an integrative computational reanalysis of existing transcriptomes (ssGSEA, WGCNA, machine-learning feature selection, deconvolution, Connectivity Map inference). It identifies candidate regulators and a correlated signature but establishes neither causal ordering nor a mechanism coupling peripheral GRI dysregulation to the group-level immune and microbiome associations recorded in this entry. This gap is narrower than gap_asd_causal_status_of_immune_and_microbiome_associations, which asks whether those associations are causal at all: it asks specifically whether HPA/GRI signaling is the substrate of the ASD-IBS co-occurrence rather than a bystander stress response shared by both. It deliberately makes no claim about an enteric-nervous-system causal chain, which the 2026-08-05 publication-readiness review removed from this entry as unsupported.

Proposed experiments

Stress-challenge PBMC GRI multi-omics cohort in ASD with and without IBS

longitudinal stress-challenge multi-omics cohort

exp_asd_ibs_gri_stress_challenge_multiomics

Enroll ASD participants stratified by presence or absence of IBS-type gastrointestinal symptoms plus matched non-autistic and IBS-only controls; apply a standardized HPA-axis challenge (for example dexamethasone suppression or an acute psychosocial stressor with serial cortisol); and measure, before and after challenge, single-cell PBMC transcriptomes (monocytes, NK cells, B cells) for the reported GRI core genes LRFN1, NUAK2, TMEM154, and GAPT, circulating cytokines, intestinal-permeability biomarkers, and stool microbiome composition, relating each to prespecified social-communication and gastrointestinal symptom measures.

Model systems

Human ASD/IBS stress-challenge cohort

Patient cohort designed to temporally order HPA activation, peripheral GRI gene dysregulation, systemic immune signaling, gut-barrier and microbiome change, and behavioral readouts across the ASD-IBS co-occurrence.

OTHER

human NCBITaxon:9606

Decision criterion

HPA/GRI signaling is supported as an upstream driver of the ASD-IBS co-occurrence if post-challenge GRI dysregulation in monocytes, NK cells, and B cells precedes and quantitatively mediates gut-barrier, microbiome, and immune change and tracks with gastrointestinal symptom severity; a shared-bystander interpretation is favored if GRI changes neither precede nor mediate those organ-level readouts.

Would support

pathophysiology#Immune and neuroinflammatory associations in studied groups pathophysiology#Gut microbiome associations in studied groups

Show evidence (2 references)

PMID:42424309 SUPPORT Computational

"While dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis and impaired glucocorticoid-responsive immune (GRI) signaling are proposed links between these disorders, the precise molecular mechanisms remain poorly understood."

The source study explicitly frames the HPA/GRI brain-gut link between ASD and IBS as proposed but mechanistically unresolved, defining the gap.

PMID:42424309 SUPPORT Computational

"These findings define a shared GRI-associated molecular signature linking systemic stress adaptation to immune dysregulation along the brain-gut axis."

A computational reanalysis of ASD and IBS PBMC transcriptomes reports a shared GRI signature along the brain-gut axis, motivating the causal question posed by this gap.

Are the reported associations between prenatal or early-childhood lead and mercury exposure and ASD causal, or do they reflect residual confounding, reverse causation, and heterogeneous exposure-assessment methods that also explain why arsenic and cadmium findings are inconsistent?

CONTROVERSY OPEN gap_asd_heavy_metal_exposure_causal_status

Attached to

environmental#Prenatal and early-childhood heavy metal exposure

Existing studies differ in biomarker matrix (blood, hair, urine), exposure window, timing relative to diagnosis, ascertainment, and confounder adjustment, and most measure a single metal rather than the correlated mixtures children are actually exposed to. Lead and mercury show the most consistent signal and arsenic and cadmium the least, but the source review declines to conclude that heavy metals cause ASD and calls for large prospective cohorts. This item concerns whether the human association is real and causal at all; the separate question of whether animal-model mechanisms apply to human neurodevelopment is recorded in mismatch_asd_heavy_metal_model_mechanism.

Proposed experiments

Prospective mother-child cohort with repeated metal biomarkers and mediation analysis

prospective pregnancy and birth cohort with mixture and mediation analysis

exp_asd_prospective_metal_mixture_birth_cohort

Follow a large, diverse pregnancy cohort with repeated maternal and child metal biomarkers across defined prenatal and early postnatal windows, prespecified mixture models that separate correlated metals from each other and from co-exposures, blinded standardized ASD ascertainment, and genetic and socioeconomic confounder control. Sibling or negative-control-exposure comparisons and biomarker path analysis through oxidative-stress and inflammatory intermediates test whether any exposure-to-outcome association survives and is mediated.

Model systems

Diverse prospective pregnancy and birth cohort

Human mother-child cohort with repeated exposure biomarkers, blinded diagnostic ascertainment, and sibling comparisons for unmeasured familial confounding.

OTHER

human NCBITaxon:9606

Perturbations

Measured prenatal and early-childhood metal exposure

environmental#Prenatal and early-childhood heavy metal exposure

exposure to heavy metal ECTO:9002163

Effect: Observed rather than assigned exposure, measured repeatedly by window and modeled as a correlated mixture rather than one metal at a time.

Readouts

Blinded ASD ascertainment and exposure-response gradient

environmental#Prenatal and early-childhood heavy metal exposure

Interpretation: A window-specific exposure-response gradient that survives mixture, sibling, and negative-control analyses is the minimum needed to move this exposure beyond an observational association.

Decision criterion

A causal contribution is supported if window-specific lead or mercury biomarkers predict ASD outcomes with exposure-response gradients that persist in mixture models, sibling comparisons, and negative-control analyses; a confounding interpretation is favored if associations attenuate to the null once familial and socioeconomic factors and correlated co-exposures are accounted for.

Would support

environmental#Prenatal and early-childhood heavy metal exposure

Show evidence (1 reference)

PMID:42540380 SUPPORT Other

"The findings are inconsistent across metals and studies, and should be interpreted with caution due to potential residual confounding and heterogeneity in exposure assessment methods. Large prospective cohort studies are needed to clarify causal relationships."

The source review states the unresolved causal status and the study design needed to resolve it, defining this controversy.

Do the oxidative-stress, synaptic, and glial-activation mechanisms characterized for lead and mercury in rodent, zebrafish, and cell models operate at human-relevant developmental exposure levels, and do they mediate autistic traits in people, or are they model-system toxicology that does not transfer to human ASD?

HUMAN MODEL MISMATCH OPEN mismatch_asd_heavy_metal_model_mechanism

Attached to

environmental#Prenatal and early-childhood heavy metal exposure pathophysiology#Immune and neuroinflammatory associations in studied groups

Metal neurotoxicity mechanisms are described in detail in experimental animals, including reactive oxygen species generation, impaired synaptic plasticity, microglial and astrocytic activation, and lipid and steroid metabolic disruption, and a zebrafish arm of one birth-cohort study nominates a PPAR-linked pathway. None of this has been demonstrated in human neurodevelopment, at the cord-blood concentrations measured in cohorts, or as a mediator of autistic traits, and reviews of developmental mercury toxicity state that no single process explains the observed effects. That is why this entry records the exposure as an environmental association and adds no heavy-metal pathophysiology node: promoting these model findings to a human ASD mechanism node would repeat the animal-only maternal-immune-activation claim that the 2026-08-05 publication-readiness review removed. The prior question of whether the human epidemiological association is causal at all is recorded in gap_asd_heavy_metal_exposure_causal_status.

Proposed experiments

Human cortical organoid and microglia-containing assembloid exposure at cohort-relevant metal concentrations

dose-anchored human organoid exposure experiment

exp_asd_human_cortical_organoid_metal_exposure

Expose human iPSC-derived cortical organoids and microglia-containing assembloids, including lines carrying defined ASD-associated variants and isogenic controls, to lead and methylmercury across a dose range anchored to measured human cord-blood concentrations. Read out oxidative stress, microglial activation state, synaptic density and network activity, and transcriptomic and lipidomic changes including the PPAR pathway nominated by the zebrafish work, and test whether effects appear at human-relevant rather than only supraphysiological doses.

Model systems

Human iPSC-derived cortical organoid and microglia-containing assembloid panel

Human cortical organoids with and without integrated microglia, spanning ASD-associated genotypes and isogenic controls, used to test whether animal-model metal neurotoxicity mechanisms are reproduced in human neural tissue at cohort-relevant exposure levels.

ORGANOID

human NCBITaxon:9606

neural progenitor cell CL:0011020 microglial cell CL:0000129

Perturbations

Cohort-anchored lead exposure

environmental#Prenatal and early-childhood heavy metal exposure

exposure to lead ECTO:9000945

Effect: Applies lead across a dose range spanning measured human cord-blood concentrations, testing human-relevant rather than only high-dose effects.

Cohort-anchored mercury exposure

environmental#Prenatal and early-childhood heavy metal exposure

exposure to mercury ECTO:0001571

Effect: Applies methylmercury across a comparably anchored dose range.

Readouts

Oxidative stress and glial activation

pathophysiology#Immune and neuroinflammatory associations in studied groups

response to oxidative stress GO:0006979 microglial cell activation GO:0001774

Interpretation: Detects whether the oxidative and glial responses reported in animal models occur in human neural tissue at cohort-relevant doses.

Synaptic and network development

pathophysiology#Excitation-inhibition and synaptic-signaling differences in studied subgroups

regulation of synaptic plasticity GO:0048167

Interpretation: Tests whether metal exposure reproduces the synaptic and network measures already curated for genotype-defined subgroups.

Decision criterion

Translational relevance is supported if oxidative, glial, and synaptic effects appear at concentrations within the measured human cord-blood range and scale with dose; the model-system interpretation stands if effects require concentrations well above human exposure or fail to reproduce in human neural tissue.

Would support

pathophysiology#Immune and neuroinflammatory associations in studied groups

Show evidence (4 references)

PMID:42541535 SUPPORT Model Organism

"Once in the brain, lead induces oxidative stress through excessive reactive oxygen species generation, mitochondrial dysfunction, lipid peroxidation, DNA damage, and depletion of antioxidant defenses."

Establishes that a detailed mechanism exists in animal models, which is the model side of the mismatch. The review covers rodent and zebrafish data and makes no human ASD claim.

PMID:42541535 SUPPORT Model Organism

"In parallel, it activates both intrinsic and extrinsic apoptotic pathways and enhances neuroinflammatory signaling through microglial and astrocytic activation, further contributing to neuronal injury."

Names the glial-activation route that would have to be demonstrated in human tissue before a heavy-metal neuroinflammation node could be curated.

PMID:41110787 SUPPORT Model Organism

"Integrated omics analyses identified substantial disruption of lipid and steroid metabolic pathways, particularly involving the peroxisome proliferator-activated receptor (PPAR) signaling cascade."

The candidate PPAR mechanism comes from the zebrafish arm of the study, not from the human cohort arm, so it is a nominated hypothesis awaiting human confirmation.

+ 1 more reference

⚙

Pathophysiology

8

Polygenic neurodevelopmental liability

Common inherited variants of individually small effect contribute substantially to ASD susceptibility. Their aggregate influence is heterogeneous across people and clinical presentations and is neither necessary nor sufficient for an individual diagnosis.

nervous system development GO:0007399

Downstream

Heterogeneous neurodevelopmental molecular effects Polygenic liability influences neurodevelopment through many unresolved regulatory and cellular routes rather than one canonical cascade.

Show evidence (2 references)

PMID:30804558 SUPPORT Human Clinical

"Common genetic variants contribute substantially to ASD susceptibility, but to date no individual variants have been robustly associated with ASD."

The GWAS establishes distributed common-variant susceptibility and argues against deterministic interpretation of any single common variant.

PMID:26709141 SUPPORT Human Clinical

"The meta-analytic heritability estimates were substantial: 64-91%."

Twin-study meta-analysis quantifies the strong genetic contribution while its range prevents presentation as one fixed population parameter.

Rare coding and copy-number variant liability

A minority of autistic people carry rare protein-truncating, damaging missense, or copy-number variants with larger effects. These variants are enriched in clinically ascertained ASD but show variable penetrance and pleiotropy with developmental delay and other neurodevelopmental outcomes; they do not make ASD a collection of simple Mendelian subtypes.

Downstream

Heterogeneous neurodevelopmental molecular effects Different rare variants perturb distinct developmental pathways that may partly converge, but pleiotropy and mutation-specific effects remain.

Convergent transcriptional effects in human stem-cell models Selected rare ASD-associated mutations produce mutation-specific and partly convergent transcriptional effects in human stem-cell-derived neural models; this is a model-system route, not a universal human-brain mechanism.

Show evidence (2 references)

PMID:35982160 SUPPORT Human Clinical

"Some individuals with autism spectrum disorder (ASD) carry functional mutations rarely observed in the general population."

The large sequencing study supports a rare-variant subgroup.

PMID:35982160 SUPPORT Human Clinical

"We discovered 72 genes associated with ASD at false discovery rate (FDR) ≤ 0.001 (185 at FDR ≤ 0.05)."

Quantifies association while avoiding a claim that every listed gene is sufficient to cause the broad ASD phenotype.

Heterogeneous neurodevelopmental molecular effects

Human genetic studies recurrently implicate synaptic formation, transcriptional regulation, chromatin remodeling, neuronal maturation, and corticogenesis. These are pathway-level convergences across subsets, not a molecular lesion demonstrated in every autistic person.

Subgroup-specific excitation-inhibition and synaptic-signaling model

synapse organization GO:0050808 chromatin remodeling GO:0006338

Downstream

Distributed developmental circuit differences Molecular perturbations can alter neuronal development and circuit function through mutation-, cell-type-, and developmental-stage-specific routes.

Excitation-inhibition and synaptic-signaling differences in studied subgroups Within the subgroup-specific E/I hypothesis, synaptic and transcriptional perturbations may alter excitation or inhibition through genotype-, region-, cell-type-, and developmental-stage-specific routes.

Show evidence (1 reference)

PMID:25363760 SUPPORT Human Clinical

"Many of the genes implicated encode proteins for synaptic formation, transcriptional regulation and chromatin-remodelling pathways."

Large-scale exome analysis supports pathway convergence but not universal disruption of each pathway in every individual.

Distributed developmental circuit differences

Post-mortem, imaging, and electrophysiological studies identify subtle, distributed group-level differences rather than a characteristic gross lesion. Direction and location vary by cohort, age, task, method, and subgroup. These observations are not individual diagnostic biomarkers.

brain UBERON:0000955

Downstream

Abnormal Nonverbal Communicative Behavior Developmental circuit differences may contribute to social-communication phenotypes through multiple routes; no single circuit signature is necessary or sufficient.

Restricted and Repetitive Behavior Distributed developmental differences may contribute to repetitive and inflexible behavior, without implying one universal cognitive or circuit cause.

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Although gross brain pathology is not characteristic of autism, subtle anatomical and functional differences have been observed in post-mortem, neuroimaging and electrophysiological studies."

This directly supports a cautious distributed, group-level circuit node and rejects a characteristic structural lesion.

Excitation-inhibition and synaptic-signaling differences in studied subgroups

E/I imbalance is a useful research framework supported by particular genetic syndromes, cellular models, and group-level studies, but direction, cell type, region, and developmental timing differ. mGluR5, GABAergic, and glutamatergic findings must not be generalized from biologically defined subsets to all ASD.

chemical synaptic transmission GO:0007268 ⚠ ABNORMAL

Show evidence (2 references)

PMID:41653294 SUPPORT Other

"Alterations in synaptic scaffolding complexes (SHANK3-Homer-mGluR5 interactions), receptor trafficking, and activity-dependent protein synthesis may contribute to excitatory/inhibitory imbalance and circuit dysfunction."

The review uses conditional language and concerns biologically defined pathways, so it supports only a provisional subgroup mechanism.

PMID:33076974 SUPPORT In Vitro

"We find that ASD patient-derived neurons with a functional loss of TSC2, in addition to possessing neuronal hyperactivity, develop a dysfunctional neuronal network with reduced synchronisation of neuronal bursting and lower spatial connectivity."

The finding is limited to patient-derived neurons with TSC2 loss and is not evidence of universal neuronal hyperactivity in ASD.

Immune and neuroinflammatory associations in studied groups

Studies report group-level cytokine, oxidative-stress, glial, and connectivity associations. Sampling, medication, co-occurring illness, tissue, age, and subgroup differences complicate interpretation. These associations do not establish that neuroinflammation causes core autism in all people, and blood-brain-barrier findings are subtle or context dependent.

Immune and microbiome modifiers in selected ASD subgroups

neuroinflammatory response GO:0150076

Downstream

Distributed developmental circuit differences Within the emerging modifier hypothesis, immune and metabolic differences are associated with synaptic and network measures; directionality and mediation of clinical traits remain unestablished.

Show evidence (2 references)

PMID:41947852 SUPPORT Other

"individuals with ASD show consistent evidence of microglial and astroglial activation, altered cytokine profiles (including IL-1β, IL-6, and TNF-α), and markers of oxidative stress such as glutathione imbalance and lipid peroxidation."

This is review-level group evidence and does not establish individual universality, directionality, or mediation of diagnostic traits.

PMID:41550027 SUPPORT Other

"In autism and ADHD, altered tight junction protein profiles imply more subtle or context-dependent barrier dysfunction."

The review explicitly requires a subtle, context-dependent interpretation.

Gut microbiome associations in studied groups

Microbiome composition and metabolite differences have been reported in some ASD cohorts, especially those selected for gastrointestinal symptoms. Diet, medication, geography, age, constipation, sequencing methods, and reverse causation are major confounders. A causal microbiome-to-core-autism pathway and a generalizable microbial biomarker remain unproven.

Immune and microbiome modifiers in selected ASD subgroups

Downstream

Immune and neuroinflammatory associations in studied groups Within the emerging modifier hypothesis, dysbiosis may affect immune and neuroinflammatory signaling. Confounding and reverse causation remain major alternatives.

Show evidence (1 reference)

PMID:39733842 SUPPORT Other

"Gut microbiota significantly influences behavior and neurodevelopment by regulating the gut-brain axis."

This narrative review supports biological plausibility, not an ASD-specific causal pathway or a finding present in every autistic person.

Convergent transcriptional effects in human stem-cell models

In iPSC-derived neural progenitor and cortical organoid models carrying selected ASD-associated mutations, mutation-specific early changes partly converge later on shared transcriptional networks. The result is a model system observation across selected genotypes, not proof of a universal cortical program in autistic brains.

neural progenitor cell CL:0011020

regulation of DNA-templated transcription GO:0006355 ⚠ ABNORMAL

Show evidence (1 reference)

PMID:41611887 SUPPORT In Vitro

"Early time points harboured the largest mutation-specific changes, but different mutations converged on shared transcriptional changes as development progressed."

Supports developmental convergence within the studied stem-cell models while retaining mutation-specific early effects.

⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.

●

Phenotypes

12

Nervous System 5

Delayed Speech and Language Development HP:0000750

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Psychosocial interventions in children can improve specific behaviours, such as joint attention, language and social engagement, that may affect further development and could reduce symptom severity."

The primer identifies language as a variable developmental domain; it does not make language delay a required feature.

Intellectual Disability FREQUENT HP:0001249

Show evidence (1 reference)

PMID:40232988 SUPPORT Human Clinical

"Among 5,292 (61.4% of 8,613) children aged 8 years with ASD with information on cognitive ability, 39.6% were classified as having an intellectual disability."

Site-based surveillance supports a frequent co-occurrence while the missing cognitive data and ascertainment context limit generalization.

Sleep Disturbance HP:0002360

Show evidence (1 reference)

PMID:37913872 SUPPORT Human Clinical

"Point prevalence of developmental coordination disorder, sleep-wake problem, gastrointestinal problem, ADHD, anxiety disorder, overweight/obesity, feeding and eating disorder, elimination disorder, disruptive behavior, and somatic symptoms and related disorder were the most frequent CCs."

The large meta-analysis supports common co-occurrence without imposing an unsupported universal percentage.

Seizure OCCASIONAL HP:0001250

Show evidence (1 reference)

PMID:34510916 SUPPORT Human Clinical

"About 1/10 autistic individuals also had epilepsy, which was common in the clinical setting, adolescents, adults, females, or patients with intellectual disability and less common in the country with high human development index."

The meta-analysis supports an occasional overall frequency and identifies important modifiers rather than a uniform risk.

Attention Deficit Hyperactivity Disorder HP:0007018

Show evidence (1 reference)

PMID:37913872 SUPPORT Human Clinical

"Point prevalence of developmental coordination disorder, sleep-wake problem, gastrointestinal problem, ADHD, anxiety disorder, overweight/obesity, feeding and eating disorder, elimination disorder, disruptive behavior, and somatic symptoms and related disorder were the most frequent CCs."

The systematic review supports ADHD as a frequent co-occurring condition.

Other 7

Abnormal Nonverbal Communicative Behavior HP:0000758

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life."

The primer directly supports the social-communication domain.

Restricted and Repetitive Behavior Restrictive behavior HP:0000723

Show evidence (1 reference)

PMID:31843864 SUPPORT Other

"Core deficits are identified in 2 domains: social communication/interaction and restrictive, repetitive patterns of behavior."

The AAP report directly supports this diagnostic domain.

Reduced Ability to Form Peer Relationships HP:0000728

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Developing, maintaining and understanding relationships"

The reproduced DSM-5 criteria identify the relationship subdomain; the phenotype remains context-sensitive rather than a uniform social outcome.

Motor Stereotypy HP:0000733

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life."

Supports repetitive behavior generally; motor stereotypy is one non-obligate manifestation.

Inflexible Adherence to Routines HP:0000732

Show evidence (1 reference)

PMID:31843864 SUPPORT Other

"Core deficits are identified in 2 domains: social communication/interaction and restrictive, repetitive patterns of behavior."

Supports the parent domain; this is one variable manifestation.

Sensory Behavioral Abnormality HP:5200046

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life."

Directly supports sensory behavior as part of the spectrum definition.

Gastrointestinal Symptoms FREQUENT Functional abnormality of the gastrointestinal tract HP:0012719

Show evidence (1 reference)

PMID:37474417 SUPPORT Human Clinical

"The prevalence of GI symptoms ranged between 0% and 69%, with an estimated general prevalence of 33% (95% CI, 13%-57%), higher than that reported by a previous meta-analysis for the general paediatric population."

The pooled estimate supports FREQUENT while the range and interval document substantial heterogeneity.

🧬

Genetic Associations

4

CHD8 (Risk Factor)

Gene: CHD8 hgnc:20153

Show evidence (1 reference)

PMID:24387789 SUPPORT Other

"Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and..."

The review identifies CHD8 among rare-variant loci supported by targeted large-scale resequencing; the record is therefore typed as a risk factor, not a sufficient cause of broad ASD.

SCN2A (Risk Factor)

Gene: SCN2A hgnc:10588

Show evidence (1 reference)

PMID:24387789 SUPPORT Other

"Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and..."

The review identifies SCN2A among rare-variant loci supported by targeted large-scale resequencing; the record does not generalize across variant mechanisms or treat the gene as a sufficient cause.

DYRK1A (Risk Factor)

Gene: DYRK1A hgnc:3091

Show evidence (1 reference)

PMID:24387789 SUPPORT Other

"Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and..."

The review identifies DYRK1A among rare-variant loci supported by targeted large-scale resequencing and supports risk-factor, rather than universal causal, typing.

CTNNB1 (Risk Factor)

Gene: CTNNB1 hgnc:2514

Show evidence (1 reference)

PMID:24387789 SUPPORT Other

"Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and..."

The review identifies CTNNB1 among rare-variant loci supported by targeted large-scale resequencing; it does not make CTNNB1 a sufficient or spectrum-wide cause.

💊

Medical Actions

4

Individualized developmental and behavioral intervention

Category: Therapeutic Action: behavioral intervention Ontology label: Behavioral Intervention NCIT:C15184

Goal-directed developmental and behavioral programs can improve selected communication, adaptive, cognitive, play, or social-engagement outcomes in some young children. Program model, intensity, acceptability, access, outcomes, and response vary. Evidence does not support a promise of normalization or one mandatory program for every child; goals should be selected collaboratively and monitored for benefit and burden.

Target Phenotypes: Abnormal nonverbal communicative behavior HP:0000758 Delayed speech and language development HP:0000750

Show evidence (2 references)

PMID:41502379 SUPPORT Human Clinical

"Across the 17 identified studies, we obtained participant data from 15 studies: 341 children received EIBI and 280 were in comparison-groups. All studies had a serious risk of bias due to the lack of random assignment."

The individual-participant meta-analysis supports possible benefit while requiring explicit acknowledgment of serious bias.

PMID:31949163 SUPPORT Other

"Psychosocial interventions in children can improve specific behaviours, such as joint attention, language and social engagement, that may affect further development and could reduce symptom severity."

Supports domain-specific, non-universal benefit.

Communication, educational, environmental, and family supports

Category: Therapeutic Action: supportive care Ontology label: Supportive Care NCIT:C15747

Speech-language therapy, augmentative and alternative communication, occupational and educational supports, sensory or environmental accommodations, caregiver education, transition planning, and assistance with community participation, housing, transport, employment, and benefits should be individualized to communication, function, preferences, and support needs across the lifespan.

Show evidence (2 references)

PMID:31843864 SUPPORT Other

"Children and youth with ASD have service needs in behavioral, educational, health, leisure, family support, and other areas."

The AAP report supports broad, coordinated pediatric service needs.

PMID:31949163 SUPPORT Other

"Families are often the major source of support for people with autism throughout much of life and need to be considered, along with the perspectives of autistic individuals, in both research and practice."

Supports lifespan family support and autistic-person participation.

Pharmacotherapy for severe irritability and co-occurring conditions

Category: Therapeutic Action: pharmacotherapy Ontology label: Pharmacotherapy NCIT:C15986

Agent: risperidone CHEBI:8871 aripiprazole CHEBI:31236

Medication does not treat autism as a whole. Risperidone or aripiprazole can reduce severe irritability, aggression, self-injury, or tantrums in selected children and adolescents, with sedation, weight/metabolic effects, and movement-disorder risk requiring monitoring. ADHD, anxiety, depression, epilepsy, sleep, and gastrointestinal disorders should be assessed and treated on their own evidence base with autism-relevant tolerability and communication considerations.

Show evidence (2 references)

PMID:31949163 SUPPORT Other

"All medications that have evidence of benefit for ASD treat the associated symptoms"

Directly constrains medication claims to associated symptoms and diagnoses.

PMID:31949163 SUPPORT Other

"Risperidone and aripiprazole (both of which are atypical antipsychotics) are approved in the USA to treat"

The cached line fragment identifies the two agents and their US approval context; the source context and bounded treatment description supply the associated-symptom target without claiming core-ASD treatment.

Melatonin for co-occurring sleep disturbance

Category: Therapeutic Action: pharmacotherapy Ontology label: Pharmacotherapy NCIT:C15986

Agent: melatonin CHEBI:16796

Melatonin can improve total sleep time, sleep latency, or sleep efficiency for some autistic children with sleep disturbance after behavioral and environmental contributors are assessed. Trial heterogeneity is substantial; benefit is symptom-targeted rather than treatment of autism as a whole, and dose, formulation, interactions, and adverse effects require clinical review.

Target Phenotypes: Sleep disturbance HP:0002360

Show evidence (1 reference)

PMID:36584862 SUPPORT Human Clinical

"Melatonin has possible efficacy over total time, latency, and efficiency sleep parameters."

The meta-analysis supports possible sleep-domain benefit, while the treatment description records its substantial between-study heterogeneity and symptom-specific scope.

🌍

Environmental Factors

3

Prenatal and perinatal epidemiologic associations

Advanced parental age, maternal metabolic conditions, short interpregnancy interval, and some proxies of perinatal hypoxia have been associated with ASD in observational studies. Residual confounding, genetics, and correlated exposures limit causal inference; these are population risk associations, not explanations for an individual diagnosis. Vaccination is not associated with ASD.

Show evidence (2 references)

PMID:31949163 SUPPORT Other

"considered causal, but could be reactive, independent or contributory"

The primer explicitly cautions against promoting observational risk associations to established causes.

PMID:31949163 SUPPORT Other

"including clear evidence that ASD is not associated with vaccination"

Directly records the well-established negative association.

Prenatal valproate exposure

In a Danish population cohort, maternal valproate use during pregnancy was associated with increased offspring ASD risk after adjustment for maternal epilepsy. This is a population-level medication-exposure association, not a deterministic explanation for an individual diagnosis; pregnancy treatment decisions require individualized assessment of maternal seizure and fetal risks rather than abrupt medication changes.

Show evidence (1 reference)

PMID:23613074 SUPPORT Human Clinical

"Maternal use of valproate during pregnancy was associated with a significantly increased risk of autism spectrum disorder and childhood autism in the offspring, even after adjusting for maternal epilepsy."

The population cohort supports a specific adjusted association while the wording avoids treating exposure as sufficient or inevitable causation.

Mechanism Target:

PREDISPOSES Heterogeneous neurodevelopmental molecular effects — The best-evidenced environmental association in this entry: a population cohort reporting increased risk after adjustment for maternal epilepsy, which is the confounder that would otherwise explain it. Recorded as predisposing and never as triggering, because this entry is careful throughout that these are population-level risk associations and not explanations for an individual diagnosis, and because the exposure's own description warns against abrupt medication changes in pregnancy. No cited sentence follows valproate to any molecular step, so the intermediates are unknown despite the node being molecular.

Show evidence (1 reference)

PMID:23613074 SUPPORT Human Clinical

"Maternal use of valproate during pregnancy was associated with a significantly increased risk of autism spectrum disorder and childhood autism in the offspring, even after adjusting for maternal epilepsy."

Danish population cohort finding maternal valproate use in pregnancy associated with significantly increased offspring risk even after adjusting for maternal epilepsy. An adjusted population association.

Prenatal and early-childhood heavy metal exposure

Candidate mechanisms, recorded as candidates only. Three routes are repeatedly proposed for lead and mercury developmental neurotoxicity: oxidative stress with mitochondrial dysfunction and lipid peroxidation; impaired synaptic plasticity, including altered NMDA receptor subunit composition and reduced synaptic protein expression; and microglial and astrocytic activation with neuroinflammatory signaling. A zebrafish arm of one birth-cohort study additionally nominates PPAR-linked lipid and steroid metabolic disruption. Every one of these is characterized in rodent, zebrafish, or cell models; none has been demonstrated in human neurodevelopment, at the cord-blood concentrations measured in human cohorts, or as a mediator of autistic traits. They are therefore documented here rather than curated as pathophysiology nodes, and the open translational question is tracked in the discussion mismatch_asd_heavy_metal_model_mechanism.

Prenatal and early-childhood exposure to lead, mercury, cadmium, and arsenic has been examined as a possible contributor to ASD risk. The reported associations are metal specific and inconsistent: reviews of human epidemiological studies report a more consistent association for lead and mercury than for arsenic and cadmium, while acknowledging residual confounding and heterogeneous exposure-assessment methods. This is a contested population-level exposure association, not an established cause of an individual diagnosis, and the reviewed evidence does not support a blanket claim that heavy metals cause ASD.

Show evidence (7 references)

PMID:42540380 SUPPORT Other

"In contrast, elevated prenatal and early childhood Pb and Hg concentrations in blood and hair samples showed a significant consistent association with both increased ASD risk and symptom severity, even after adjustment for key demographic and environmental confounders."

A critical review of 36 human epidemiological studies reports a comparatively consistent adjusted association for lead and mercury. It is a narrative synthesis rather than a pooled estimate, so it supports a population-level association only.

PMID:42540380 SUPPORT Other

"We revealed inconsistent associations between prenatal and childhood urinary, blood, and hair, As and Cd exposure, and ASD outcomes."

The same review records that arsenic and cadmium associations are inconsistent, so this exposure class must not be curated as a uniform risk factor across metals.

PMID:42540380 SUPPORT Other

"Rather, lead and mercury may contribute to ASD risk in certain populations or under specific exposure conditions, but these findings require cautious interpretation due to methodological differences and potential biases."

The authors' own summary bounds the claim to specific populations and exposure conditions and requires cautious interpretation.

+ 4 more references

Mechanism Target:

PREDISPOSES Heterogeneous neurodevelopmental molecular effects — Graded below the valproate link and carrying its own contradiction on purpose. The same review that reports a comparatively consistent adjusted association for lead and mercury reports inconsistent associations for arsenic and cadmium, and bounds its own conclusion to certain populations and exposure conditions. All three sentences are carried here, so the reader of the pathograph meets the metal-by-metal disagreement rather than a uniform risk factor. The candidate mechanisms named in this exposure's notes are recorded there as candidates and are not asserted by any link.

Show evidence (3 references)

PMID:42540380 SUPPORT Other

"In contrast, elevated prenatal and early childhood Pb and Hg concentrations in blood and hair samples showed a significant consistent association with both increased ASD risk and symptom severity, even after adjustment for key demographic and environmental confounders."

Reports a significant consistent adjusted association for lead and mercury with risk and symptom severity. A narrative synthesis rather than a pooled estimate.

PMID:42540380 SUPPORT Other

"We revealed inconsistent associations between prenatal and childhood urinary, blood, and hair, As and Cd exposure, and ASD outcomes."

The same review recording inconsistent associations for arsenic and cadmium. Carried so this exposure class is not read as uniform across metals.

PMID:42540380 SUPPORT Other

"Rather, lead and mercury may contribute to ASD risk in certain populations or under specific exposure conditions, but these findings require cautious interpretation due to methodological differences and potential biases."

The authors' own bound: lead and mercury may contribute in certain populations or under specific exposure conditions, with cautious interpretation required.

🔬

Diagnosis

3

Multidisciplinary clinical developmental assessment (Diagnosis is clinical and integrates developmental history, direct observation, current social communication and restricted/repetitive and sensory behavior, language, cognition, adaptive function, medical and psychiatric assessment, and information across settings. Hearing, vision, language, motor, learning, trauma, and co-occurring conditions are evaluated as appropriate. No laboratory or imaging test establishes broad ASD.)

clinical assessment NCIT:C124351

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Diagnosis of ASD is made on the basis of"

The cached PDF line fragment identifies that diagnosis has an assessment basis but is truncated by line-number extraction; the complete case definition and adjunct-tool evidence above supply the substantive clinical diagnostic claim.

Standardized diagnostic instruments as adjuncts (Instruments such as ADOS-2, ADI-R, CARS, and structured rating scales can organize observation and history, but thresholds do not replace expert clinical synthesis. Accuracy varies by tool, age, comparator, setting, and study quality; cultural, sex/gender, language, intellectual, and masking effects require consideration.)

clinical assessment NCIT:C124351

Show evidence (1 reference)

PMID:40274203 SUPPORT Human Clinical

"Diagnostic tools should be regarded as adjunctive aids rather than comprehensive substitutes for diagnosis."

Directly states the intended use and limitation of standardized tools.

Etiologic and co-occurring-condition evaluation (Genetic testing and targeted metabolic, neurologic, sleep, gastrointestinal, hearing, vision, or other evaluation can identify an etiology or separate condition in selected people. A pathogenic finding may change counseling or management but does not serve as a general ASD diagnostic biomarker, and a negative genetic test does not exclude ASD.)

clinical assessment NCIT:C124351

Show evidence (1 reference)

PMID:31843864 SUPPORT Other

"Primary care providers should be familiar with the diagnostic criteria for ASD, appropriate etiologic evaluation, and co-occurring medical and behavioral conditions"

Supports etiologic and co-occurring-condition evaluation as distinct from the clinical diagnosis.

📈

Progression

3

Early developmental emergence

Age: Infancy through early childhood

Features begin in early development, commonly become apparent during the second or third year, and may accumulate gradually. Some children show frank loss of previously acquired skills, whereas others plateau or show gradually widening social-communication and language differences. Neither a neonatal movement pattern nor an imaging, gaze, EEG, molecular, or blood measure is a validated presymptomatic diagnostic biomarker.

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life."

Supports early developmental onset without fixing one trajectory.

Childhood and adolescence

Age: Childhood through adolescence

Preschool diagnoses are often stable, but symptom severity, adaptive function, language, learning, and co-occurring sleep, attention, anxiety, gastrointestinal, or seizure problems can change. Diagnosis may occur only in later childhood or adolescence when social demands exceed capacity; late recognition does not imply late biological onset.

Show evidence (1 reference)

PMID:31843864 SUPPORT Other

"Primary care providers should be familiar with the diagnostic criteria for ASD, appropriate etiologic evaluation, and co-occurring medical and behavioral conditions (such as disorders of sleep and feeding, gastrointestinal tract symptoms, obesity, seizures, attention-deficit/hyperactivity..."

The clinical report supports age-relevant monitoring of changing co-occurring conditions and functional impact.

Adulthood and later life

Age: Adulthood

Autism is generally lifelong, but adult independence, health, employment, relationships, and quality of life are highly heterogeneous and depend on communication, intellectual and adaptive function, co-occurring conditions, opportunity, accommodations, and support. Adult intervention evidence is much thinner than pediatric evidence. Population studies show excess mortality, but risks are not uniform and should not be interpreted as an individual prognosis.

Show evidence (2 references)

PMID:41926193 SUPPORT Human Clinical

"Meta-analyses of behavioral, cognitive, adaptive, and employment outcomes were nonsignificant."

The current adult meta-analysis documents important areas without demonstrated pooled benefit and prevents extrapolation from child studies.

PMID:37042154 SUPPORT Human Clinical

"The results showed that all-cause mortality was higher for the autistic patients (RR=2.32, 95%CI: 1.98-2.72, I2=87.1%, P < 0.001)."

The systematic review supports excess group-level mortality while its high heterogeneity argues against an individual deterministic prognosis.

📊

Prevalence

3

Worldwide

Point Prevalence 800.0 per 100,000 >1 in 1,000

The 2026 frequentist pooled estimate was 0.8%, but between-study heterogeneity was extreme and the prediction interval was very wide. Diagnostic framework, setting, region, ascertainment, and access to evaluation materially affect estimates, so this is not a universal fixed rate.

Show evidence (1 reference)

PMID:42078229 SUPPORT Human Clinical

"The frequentist pooled prevalence of ASD was 0.8% (95% CI: 0.4%-1.7%), with substantial heterogeneity (I² ≈ 100%) and a wide prediction interval (0.03%-17.3%)."

This supplies the normalized global estimate and, critically, its large uncertainty and heterogeneity.

United States, children aged 8 years, 16 ADDM sites, 2022

Point Prevalence 3220.0 per 100,000 >1 in 1,000

Active surveillance estimate across 16 sites; it is not a nationally representative prevalence survey. Site rates ranged from 9.7 to 53.1 per 1,000. Recorded prevalence was 3.4 times as high among boys as girls, which can reflect both liability and ascertainment.

Show evidence (1 reference)

PMID:40232988 SUPPORT Human Clinical

"Among children aged 8 years in 2022, ASD prevalence was 32.2 per 1,000 children (one in 31) across the 16 sites, ranging from 9.7 in Texas (Laredo) to 53.1 in California."

Current ADDM surveillance supplies the site-based 2022 estimate.

Children in mainland China, studies published since 2017

Point Prevalence 700.0 per 100,000 >1 in 1,000

Meta-analysis estimate; boys and girls had different recorded rates, and ascertainment and diagnostic access should be considered when interpreting the sex difference.

Show evidence (1 reference)

PMID:38811881 SUPPORT Human Clinical

"The ASD prevalence among children in mainland China has been 0.7% (95% confidence interval(CI): 0.006-0.008) since 2017."

The meta-analysis directly supports the normalized estimate.

🔀

Differential Diagnoses

4

Conditions with similar clinical presentations that must be differentiated from Autism Spectrum Disorder:

Intellectual disability or global developmental delay without ASD

Overlapping Features Developmental level can explain delayed communication and social behavior; ASD requires social-communication differences beyond those expected for developmental level plus the restricted/repetitive domain. Intellectual disability and ASD can also co-occur.

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"Not better explained by intellectual disability or global developmental delay"

The reproduced diagnostic criteria state this exclusion boundary.

Attention Deficit-Hyperactivity Disorder MONDO:0007743

Overlapping Features Inattention, impulsivity, social difficulty, or dysregulation can resemble parts of ASD, but ADHD does not require the two ASD diagnostic domains. The diagnoses frequently co-occur and one should not be used to erase the other.

Show evidence (1 reference)

PMID:31843864 SUPPORT Other

"Primary care providers should be familiar with the diagnostic criteria for ASD, appropriate etiologic evaluation, and co-occurring medical and behavioral conditions (such as disorders of sleep and feeding, gastrointestinal tract symptoms, obesity, seizures, attention-deficit/hyperactivity..."

Supports ADHD as a distinct, potentially co-occurring condition.

Genetic and neurologic syndromes with autistic features

Overlapping Features Rett syndrome, fragile X syndrome, tuberous sclerosis complex, and other genetic or neurologic conditions can include autistic features or a co-occurring ASD diagnosis. Regression pattern, examination, seizures, dysmorphology, family history, and targeted genetic evaluation can identify an additional etiology; the syndromic diagnosis does not automatically establish or exclude ASD.

Show evidence (1 reference)

PMID:31949163 SUPPORT Other

"other disorders including ADHD, intellectual disability, language delay and genetic syndromes."

Supports dual diagnosis and cautions against treating genetic syndromes as interchangeable with broad ASD.

📊

Related Datasets

8

Disruption of Autism Spectrum Disorder-Susceptibility Genes Predominantly Reduces Functional Connectivity of Isogenic Human Neurons geo:GSE107878

Autism Spectrum Disorder (ASD) is phenotypically and genetically heterogeneous, but genomic analyses have identified candidate susceptibility genes. We present a CRISPR gene editing strategy to insert a protein tag and premature termination sites creating an induced pluripotent stem cell (iPSC) knockout resource for functional studies of 10 ASD-relevant genes (AFF2/FMR2, ANOS1, ASTN2, ATRX, CACNA1C, CHD8, DLGAP2, KCNQ2, SCN2A, TENM1). Neurogenin 2 (NEUROG2)-directed differentiation of iPSCs allowed production of cortical excitatory neurons, and mutant proteins were not detectable.

human BULK RNA SEQ n=86

PMID:30392976

Identified by GEO DataSets index search for Autism Spectrum Disorder (scripts/discover_datasets.py); accession and metadata verified against NCBI E-utilities on 2026-07-31. Title, sample count, and organism are GEO's own values.

Developmental dynamics of the cortical cellular and molecular landscapes in autism spectrum disorder models geo:GSE328363

Recent research has identified over 100 causal genes in autism spectrum disorder (ASD), raising the question of how mutations in genes with diverse functions result in similar clinical presentations. Here, we profiled 251 samples from eleven monogenic ASD mouse models using single-nucleus multi-omic sequencing across three developmental stages, both sexes, and two brain regions. We discovered that, despite wide genetic heterogeneity, ASD-linked mutations converged on perturbations of the radial glial cell lineage. This converging alteration primarily reflects a transient developmental delay rather than a lasting lineage misspecification and resolves by postnatal stages.

mouse SINGLE CELL RNA SEQ n=135

PMID:42310454

Identified by GEO DataSets index search for Autism Spectrum Disorder (scripts/discover_datasets.py); accession and metadata verified against NCBI E-utilities on 2026-07-31. Title, sample count, and organism are GEO's own values.

Autism spectrum disorder associated chromatin modifiers converge on transcription with sex-specific regulatory signatures geo:GSE286067

We sought to understand the transcriptional disruptions and functional impacts of the loss of 9 autism spectrum disorder (ASD) risk genes that encode transcriptional regulators in neurons. In addition to understanding how these signature converge or diverge, we aimed to study how sex could modulate these consequences.

mouse BULK RNA SEQ n=66

PMID:40196547

Identified by GEO DataSets index search for Autism Spectrum Disorder (scripts/discover_datasets.py); accession and metadata verified against NCBI E-utilities on 2026-07-31. Title, sample count, and organism are GEO's own values.

Whole genome sequencing of autism spectrum disorder ega:EGAS00001000556

Autism Spectrum Disorder (ASD) demonstrates high heritability and familial clustering, yet the genetic causes remain only partially understood as a result of extensive clinical and genomic heterogeneity. Whole-genome sequencing (WGS) shows promise as a tool for identifying ASD risk genes as well as unreported mutations in known loci, but an assessment of its full utility in an ASD group has not been performed. We used WGS to examine 32 families with ASD to detect de novo or rare inherited genetic variants predicted to be deleterious (loss-of-function and damaging missense mutations).

human

PMID:23849776

European Genome-phenome Archive study, matched because the disease is named in the study's own title ("Autism Spectrum Disorder"); description-level mentions were not accepted. EGA study_type: Other. Controlled access -- data require a Data Access Agreement. EGA metadata retrieved 2026-08-01.

Detection of Clinically Relevant Genetic Variants in Autism spectrum Disorder by Whole-Genome Sequencing ega:EGAS00001000829

Autism Spectrum Disorder (ASD) demonstrates high heritability and familial clustering, yet the genetic causes remain only partially understood as a result of extensive clinical and genomic heterogeneity. Whole-genome sequencing (WGS) shows promise as a tool for identifying ASD risk genes as well as unreported mutations in known loci, but an assessment of its full utility in an ASD group has not been performed. We used WGS to examine 32 families with ASD to detect de novo or rare inherited genetic variants predicted to be deleterious (loss-of-function and damaging missense mutations).

human

European Genome-phenome Archive study, matched because the disease is named in the study's own title ("Autism Spectrum Disorder"); description-level mentions were not accepted. EGA study_type: Other. Controlled access -- data require a Data Access Agreement. EGA metadata retrieved 2026-08-01.

Histone Acetylome-wide Association Study of Autism Spectrum Disorder ega:EGAS00001001943

H3K27ac ChIP-seq were performed on postmortem samples from autism spectrum disorder and matched control brains. Tissues were chosen from three brain regions: prefrontal cortex, temporal cortex and cerebellum.

human

European Genome-phenome Archive study, matched because the disease is named in the study's own title ("Autism Spectrum Disorder"); description-level mentions were not accepted. EGA study_type: Other. Controlled access -- data require a Data Access Agreement. EGA metadata retrieved 2026-08-01.

Plasma metabolomic profiling of individuals with autism spectrum disorder and their family members. metabolomics_workbench:ST002332

Located via OmicsDI, which aggregates across omics repositories; this record comes from metabolomics_workbench. Only repositories with no other discovery route in this project and with a working accession resolver are curated from OmicsDI -- GEO, ArrayExpress, PRIDE, MetaboLights and EGA hits are excluded as duplicates of dedicated passes. Matched because the disease is named in the dataset's own title ("Autism Spectrum Disorder"). Retrieved 2026-08-02.

Metaproteomics investigation on the gut microbiota of children affected by Autism Spectrum Disorder massive:MSV000085232

Metaproteomics investigation of the Gut Microbiota in young subjects with Autism Spectrum Disorders and their relatives

Located via OmicsDI, which aggregates across omics repositories; this record comes from massive. Only repositories with no other discovery route in this project and with a working accession resolver are curated from OmicsDI -- GEO, ArrayExpress, PRIDE, MetaboLights and EGA hits are excluded as duplicates of dedicated passes. Matched because the disease is named in the dataset's own title ("Autism Spectrum Disorder"). Retrieved 2026-08-02.

🔬

Clinical Trials

1

NCT03504917 PHASE_III TERMINATED

Industry-sponsored randomized double-blind placebo-controlled Phase III trial of oral balovaptan in 322 autistic adults. The registry reports that it was terminated after a futility analysis found the predefined primary objective highly unlikely to be met; no new safety concern was identified. This negative program illustrates why social-neuropeptide hypotheses cannot be promoted to established treatment mechanisms.

Target Phenotypes: Abnormal nonverbal communicative behavior HP:0000758

Show evidence (2 references)

clinicaltrials:NCT03504917 SUPPORT Human Clinical

"This study will evaluate the efficacy, safety, and pharmacokinetics of 10 mg of oral administration balovaptan once a day (QD) compared with matching placebo in adults (18 years and older) with autism spectrum disorder (ASD)."

The registry supports the intervention, comparator, population, and purpose; phase, enrollment, termination, and futility reason were rechecked in the live registry on 2026-08-05.

PMID:31949163 SUPPORT Other

"balovaptan, a vasopressin AVPR1A antagonist in adults with ASD showed negative results on its primary"

The disease primer independently reports the negative primary outcome; the live registry supplies the later termination and futility status.

🧫

Experimental Models

2

Genotype-defined patient iPSC-derived neuronal networks IPSC_DERIVED_MODEL

Neuronal networks derived from autistic participants with defined variants can test cellular excitability, synchrony, connectivity, and rescue. They retain donor/genotype relevance but lack whole-brain development, behavior, immune and environmental context, and population representativeness. The TSC2 result is a genotype-specific mechanistic model, not a model of all ASD.

Cell source

Patient-derived induced pluripotent stem cells

Show evidence (1 reference)

PMID:33076974 SUPPORT In Vitro

"We find that ASD patient-derived neurons with a functional loss of TSC2, in addition to possessing neuronal hyperactivity, develop a dysfunctional neuronal network with reduced synchronisation of neuronal bursting and lower spatial connectivity."

Defines the model finding and its genotype-limited scope.

Multi-genotype human cortical organoid and neural-progenitor panel ORGANOID

A panel spanning selected ASD-associated mutations and idiopathic donor lines can compare mutation-specific and convergent transcriptional effects across neural progenitors and cortical organoids. Immaturity, variable cell composition, absent long-range inputs and vasculature, and selected genotypes limit inference to the human brain or the entire spectrum.

Cell source

Induced pluripotent stem cell-derived, including isogenic perturbations

Show evidence (1 reference)

PMID:41611887 SUPPORT In Vitro

"Early time points harboured the largest mutation-specific changes, but different mutations converged on shared transcriptional changes as development progressed."

Supports the panel's central mutation-specific and convergent readout.

🐁

Animal Models

2

Cntnap2 knockout mouse with maternal immune activation Gene-environment interaction model

Combines a defined genetic perturbation, prenatal immune challenge, and sex to test interaction effects on selected mouse behaviors. "Autistic-like" behavior is not equivalent to the human diagnosis, and results cannot establish maternal immune activation as a human ASD cause.

Species

Mouse

Genotype

Cntnap2 knockout with maternal immune activation exposure

Show evidence (1 reference)

PMID:41898431 SUPPORT Model Organism

"The three-hit theory states that the vulnerability of an individual to develop ASD is modulated by the interplay between genetic predisposition, sex, and environmental insults."

Supports the interaction model while not validating it as a universal human pathway.

Prenatal interferon-alpha exposed rat Prenatal exposure model

Tests how prenatal cytokine exposure affects neurotransmitter measures, histology, and selected offspring behaviors. It is an exposure/toxicity model and does not reproduce ASD's clinical definition or heterogeneous human genetic architecture.

Species

Rat

Genotype

Prenatal interferon-alpha exposure

Show evidence (1 reference)

PMID:41484215 SUPPORT Model Organism

"IFN-α exposure resulted in significant reductions in GABA, 5-HIAA, and GAD-67 levels, particularly in male offspring, indicating neurotransmitter dysregulation."

Documents the model readout without equating it to human ASD.

{ }

Source YAML

click to show

name: Autism Spectrum Disorder
creation_date: "2026-04-10T00:00:00Z"
description: >-
  Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental
  condition defined clinically by persistent differences or impairments in
  social communication and interaction together with restricted or repetitive
  behavior, interests, or activities, including sensory features, beginning in
  the early developmental period. Language, intellectual ability, adaptive
  function, support needs, co-occurring conditions, and developmental course
  vary substantially. Population liability includes many common variants of
  small effect and a minority of rare coding or copy-number variants of larger
  effect. No single molecular, cellular, imaging, electrophysiological, immune,
  or microbiome mechanism is present in or diagnostic of all autistic people.
category: Complex
disease_term:
  preferred_term: autism spectrum disorder
  term:
    id: MONDO:0005258
    label: autism spectrum disorder
parents:
- Neurodevelopmental Disorder
synonyms:
- ASD
- autism spectrum condition
references:
- reference: PMID:31949163
  title: Autism spectrum disorder.
- reference: PMID:31843864
  title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
- reference: PMID:40274203
  title: "Autism diagnosis in children and adolescents: A systematic review and meta-analysis of test accuracy."
notes: >-
  Scope is the broad clinical spectrum MONDO:0005258, not the older/broader
  MONDO:0005260 concept used by the OpenScientist deep-research artifact and
  not any single gene-defined syndrome. DSM-5/ICD-11 support-level and
  language/intellectual-function specifiers describe current clinical needs;
  they are not assumed to be stable biological subtypes. Historical Asperger
  disorder and pervasive-developmental-disorder categories were subsumed into
  the unitary ASD diagnosis because their boundaries were not reliably applied.
  Identity-first and person-first language preferences vary; this entry uses
  both respectfully and does not frame autistic traits themselves as treatment
  targets independent of the person's goals, function, safety, or quality of
  life. The bounded top-level genetic list is illustrative rather than
  exhaustive: it records four directly evidenced pleiotropic risk loci as risk
  factors, never as sufficient Mendelian causes of broad ASD. Common and rare
  risk architecture and pathway convergence are curated as mechanism nodes;
  individual gene-defined syndromes belong in their own disease entries.
classifications:
  harrisons_chapter:
  - classification_value: NEUROLOGIC
    evidence:
    - reference: PMID:31949163
      reference_title: Autism spectrum disorder.
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        Autism spectrum disorder (ASD) is a common, highly heritable and heterogeneous neurodevelopmental
      explanation: >-
        The disease primer explicitly classifies ASD as a neurodevelopmental
        disorder, supporting placement in the neurologic part without asserting
        a single neurological lesion.
definitions:
- name: Contemporary clinical ASD case definition
  definition_type: CASE_DEFINITION
  derivation_basis: ESTABLISHED_CRITERIA
  description: >-
    A clinical diagnosis requires persistent social-communication and
    social-interaction differences or impairments plus restricted or repetitive
    behavior, interests, or activities. Features begin in the early
    developmental period, may become fully apparent only when demands exceed
    capacities, cause clinically meaningful functional impact, and are not
    better explained by intellectual disability or global developmental delay
    alone. Sensory hyperreactivity, hyporeactivity, or unusual sensory interests
    are included within the restricted/repetitive domain. Support needs and
    language and intellectual-function specifiers do not define etiologic
    subtypes.
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
    explanation: >-
      The primer states the defining domains and early developmental onset.
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Core deficits are identified in 2 domains: social communication/interaction and restrictive, repetitive patterns of behavior.
    explanation: The AAP clinical report independently confirms the two-domain definition.
prevalence:
- population: Worldwide
  measure_type: POINT_PREVALENCE
  prevalence_class: ABOVE_1_IN_1000
  rate_per_100000: 800.0
  notes: >-
    The 2026 frequentist pooled estimate was 0.8%, but between-study
    heterogeneity was extreme and the prediction interval was very wide.
    Diagnostic framework, setting, region, ascertainment, and access to
    evaluation materially affect estimates, so this is not a universal fixed
    rate.
  evidence:
  - reference: PMID:42078229
    reference_title: "Global Autism Spectrum Disorder Prevalence Estimates and Associated Covariates: A Systematic Review and Meta-Regression Analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The frequentist pooled prevalence of ASD was 0.8% (95% CI: 0.4%-1.7%), with substantial heterogeneity (I² ≈ 100%) and a wide prediction interval (0.03%-17.3%).
    explanation: >-
      This supplies the normalized global estimate and, critically, its large
      uncertainty and heterogeneity.
- population: United States, children aged 8 years, 16 ADDM sites, 2022
  measure_type: POINT_PREVALENCE
  prevalence_class: ABOVE_1_IN_1000
  rate_per_100000: 3220.0
  notes: >-
    Active surveillance estimate across 16 sites; it is not a nationally
    representative prevalence survey. Site rates ranged from 9.7 to 53.1 per
    1,000. Recorded prevalence was 3.4 times as high among boys as girls, which
    can reflect both liability and ascertainment.
  evidence:
  - reference: PMID:40232988
    reference_title: Prevalence and Early Identification of Autism Spectrum Disorder Among Children Aged 4 and 8 Years - Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Among children aged 8 years in 2022, ASD prevalence was 32.2 per 1,000 children (one in 31) across the 16 sites, ranging from 9.7 in Texas (Laredo) to 53.1 in California.
    explanation: Current ADDM surveillance supplies the site-based 2022 estimate.
- population: Children in mainland China, studies published since 2017
  measure_type: POINT_PREVALENCE
  prevalence_class: ABOVE_1_IN_1000
  rate_per_100000: 700.0
  notes: >-
    Meta-analysis estimate; boys and girls had different recorded rates, and
    ascertainment and diagnostic access should be considered when interpreting
    the sex difference.
  evidence:
  - reference: PMID:38811881
    reference_title: "Prevalence of autism spectrum disorder in mainland china over the past 6 years: a systematic review and meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The ASD prevalence among children in mainland China has been 0.7% (95% confidence interval(CI): 0.006-0.008) since 2017.
    explanation: The meta-analysis directly supports the normalized estimate.
progression:
- phase: Early developmental emergence
  age_range: Infancy through early childhood
  notes: >-
    Features begin in early development, commonly become apparent during the
    second or third year, and may accumulate gradually. Some children show
    frank loss of previously acquired skills, whereas others plateau or show
    gradually widening social-communication and language differences. Neither
    a neonatal movement pattern nor an imaging, gaze, EEG, molecular, or blood
    measure is a validated presymptomatic diagnostic biomarker.
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
    explanation: Supports early developmental onset without fixing one trajectory.
- phase: Childhood and adolescence
  age_range: Childhood through adolescence
  notes: >-
    Preschool diagnoses are often stable, but symptom severity, adaptive
    function, language, learning, and co-occurring sleep, attention, anxiety,
    gastrointestinal, or seizure problems can change. Diagnosis may occur only
    in later childhood or adolescence when social demands exceed capacity;
    late recognition does not imply late biological onset.
  evidence:
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Primary care providers should be familiar with the diagnostic criteria for ASD, appropriate etiologic evaluation, and co-occurring medical and behavioral conditions (such as disorders of sleep and feeding, gastrointestinal tract symptoms, obesity, seizures, attention-deficit/hyperactivity disorder, anxiety, and wandering) that affect the child's function and quality of life.
    explanation: >-
      The clinical report supports age-relevant monitoring of changing
      co-occurring conditions and functional impact.
- phase: Adulthood and later life
  age_range: Adulthood
  notes: >-
    Autism is generally lifelong, but adult independence, health, employment,
    relationships, and quality of life are highly heterogeneous and depend on
    communication, intellectual and adaptive function, co-occurring conditions,
    opportunity, accommodations, and support. Adult intervention evidence is
    much thinner than pediatric evidence. Population studies show excess
    mortality, but risks are not uniform and should not be interpreted as an
    individual prognosis.
  evidence:
  - reference: PMID:41926193
    reference_title: "Interventions for autistic adults: A meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Meta-analyses of behavioral, cognitive, adaptive, and employment outcomes were nonsignificant.
    explanation: >-
      The current adult meta-analysis documents important areas without
      demonstrated pooled benefit and prevents extrapolation from child studies.
  - reference: PMID:37042154
    reference_title: "Early death and causes of death of patients with autism spectrum disorders: A systematic review."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The results showed that all-cause mortality was higher for the autistic patients (RR=2.32, 95%CI: 1.98-2.72, I2=87.1%, P < 0.001).
    explanation: >-
      The systematic review supports excess group-level mortality while its
      high heterogeneity argues against an individual deterministic prognosis.
pathophysiology:
- name: Polygenic neurodevelopmental liability
  description: >-
    Common inherited variants of individually small effect contribute
    substantially to ASD susceptibility. Their aggregate influence is
    heterogeneous across people and clinical presentations and is neither
    necessary nor sufficient for an individual diagnosis.
  mechanism_confidence: ESTABLISHED
  biological_processes:
  - preferred_term: nervous system development
    term:
      id: GO:0007399
      label: nervous system development
  evidence:
  - reference: PMID:30804558
    reference_title: Identification of common genetic risk variants for autism spectrum disorder.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Common genetic variants contribute substantially to ASD susceptibility, but to date no individual variants have been robustly associated with ASD.
    explanation: >-
      The GWAS establishes distributed common-variant susceptibility and argues
      against deterministic interpretation of any single common variant.
  - reference: PMID:26709141
    reference_title: "Heritability of autism spectrum disorders: a meta-analysis of twin studies."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The meta-analytic heritability estimates were substantial: 64-91%.
    explanation: >-
      Twin-study meta-analysis quantifies the strong genetic contribution while
      its range prevents presentation as one fixed population parameter.
  downstream:
  - target: Heterogeneous neurodevelopmental molecular effects
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Polygenic liability influences neurodevelopment through many unresolved
      regulatory and cellular routes rather than one canonical cascade.
    evidence:
    - reference: PMID:30804558
      reference_title: Identification of common genetic risk variants for autism spectrum disorder.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: >-
        Dissecting the polygenic architecture, we found both quantitative and qualitative polygenic heterogeneity across ASD subtypes.
      explanation: >-
        The study supports heterogeneous polygenic architecture but does not
        identify one intervening molecular chain.
- name: Rare coding and copy-number variant liability
  description: >-
    A minority of autistic people carry rare protein-truncating, damaging
    missense, or copy-number variants with larger effects. These variants are
    enriched in clinically ascertained ASD but show variable penetrance and
    pleiotropy with developmental delay and other neurodevelopmental outcomes;
    they do not make ASD a collection of simple Mendelian subtypes.
  mechanism_confidence: ESTABLISHED
  evidence:
  - reference: PMID:35982160
    reference_title: Rare coding variation provides insight into the genetic architecture and phenotypic context of autism.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Some individuals with autism spectrum disorder (ASD) carry functional mutations rarely observed in the general population.
    explanation: The large sequencing study supports a rare-variant subgroup.
  - reference: PMID:35982160
    reference_title: Rare coding variation provides insight into the genetic architecture and phenotypic context of autism.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      We discovered 72 genes associated with ASD at false discovery rate (FDR) ≤ 0.001 (185 at FDR ≤ 0.05).
    explanation: >-
      Quantifies association while avoiding a claim that every listed gene is
      sufficient to cause the broad ASD phenotype.
  downstream:
  - target: Heterogeneous neurodevelopmental molecular effects
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Different rare variants perturb distinct developmental pathways that may
      partly converge, but pleiotropy and mutation-specific effects remain.
    evidence:
    - reference: PMID:35982160
      reference_title: Rare coding variation provides insight into the genetic architecture and phenotypic context of autism.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: >-
        Meta-analysis with cohorts ascertained for developmental delay (DD) (n = 91,605) yielded 373 genes associated with ASD/DD at FDR ≤ 0.001 (664 at FDR ≤ 0.05), some of which differed in relative frequency of mutation between ASD and DD cohorts.
      explanation: >-
        Shared and differing associations support convergence plus pleiotropy,
        not an ASD-specific linear pathway.
  - target: Convergent transcriptional effects in human stem-cell models
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Selected rare ASD-associated mutations produce mutation-specific and
      partly convergent transcriptional effects in human stem-cell-derived
      neural models; this is a model-system route, not a universal human-brain
      mechanism.
    evidence:
    - reference: PMID:41611887
      reference_title: Developmental convergence and divergence in human stem cell models of autism.
      supports: SUPPORT
      evidence_source: IN_VITRO
      snippet: >-
        These data illustrate how risk associated with genetically defined forms of ASD can propagate by means of transcriptional regulation to affect convergently dysregulated pathways, providing new insight into the convergent impact of ASD genetic risk on human neurodevelopment.
      explanation: >-
        The experiment directly links selected genetically defined ASD forms to
        convergent model-system transcriptional effects; PARTIAL preserves the
        limit to eight mutations and in-vitro development.
- name: Heterogeneous neurodevelopmental molecular effects
  description: >-
    Human genetic studies recurrently implicate synaptic formation,
    transcriptional regulation, chromatin remodeling, neuronal maturation, and
    corticogenesis. These are pathway-level convergences across subsets, not a
    molecular lesion demonstrated in every autistic person.
  mechanism_confidence: PROVISIONAL
  biological_processes:
  - preferred_term: synapse organization
    term:
      id: GO:0050808
      label: synapse organization
  - preferred_term: chromatin remodeling
    term:
      id: GO:0006338
      label: chromatin remodeling
  evidence:
  - reference: PMID:25363760
    reference_title: Synaptic, transcriptional and chromatin genes disrupted in autism.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Many of the genes implicated encode proteins for synaptic formation, transcriptional regulation and chromatin-remodelling pathways.
    explanation: >-
      Large-scale exome analysis supports pathway convergence but not universal
      disruption of each pathway in every individual.
  downstream:
  - target: Distributed developmental circuit differences
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Molecular perturbations can alter neuronal development and circuit
      function through mutation-, cell-type-, and developmental-stage-specific
      routes.
    evidence:
    - reference: PMID:31949163
      reference_title: Autism spectrum disorder.
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        subtle anatomical and functional differences have been observed in post-mortem, neuroimaging and electrophysiological studies.
      explanation: >-
        The primer supports group-level neural differences but does not prove
        that the molecular pathways mediate them in all ASD.
  - target: Excitation-inhibition and synaptic-signaling differences in studied subgroups
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - subgroup_ei_synaptic_model
    description: >-
      Within the subgroup-specific E/I hypothesis, synaptic and transcriptional
      perturbations may alter excitation or inhibition through genotype-,
      region-, cell-type-, and developmental-stage-specific routes.
    evidence:
    - reference: PMID:41653294
      reference_title: "Unraveling mGluR5 dysfunction in autism spectrum disorder: a multi-level analysis of genetic, molecular, and neurobiological mechanisms."
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        Alterations in synaptic scaffolding complexes (SHANK3-Homer-mGluR5 interactions), receptor trafficking, and activity-dependent protein synthesis may contribute to excitatory/inhibitory imbalance and circuit dysfunction.
      explanation: >-
        The review supports a conditional route in biologically defined
        pathways; the hypothesis tag prevents universalization.
- name: Distributed developmental circuit differences
  description: >-
    Post-mortem, imaging, and electrophysiological studies identify subtle,
    distributed group-level differences rather than a characteristic gross
    lesion. Direction and location vary by cohort, age, task, method, and
    subgroup. These observations are not individual diagnostic biomarkers.
  mechanism_confidence: PROVISIONAL
  locations:
  - preferred_term: brain
    term:
      id: UBERON:0000955
      label: brain
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Although gross brain pathology is not characteristic of autism, subtle anatomical and functional differences have been observed in post-mortem, neuroimaging and electrophysiological studies.
    explanation: >-
      This directly supports a cautious distributed, group-level circuit node
      and rejects a characteristic structural lesion.
  downstream:
  - target: Abnormal Nonverbal Communicative Behavior
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Developmental circuit differences may contribute to social-communication
      phenotypes through multiple routes; no single circuit signature is
      necessary or sufficient.
    evidence:
    - reference: PMID:31949163
      reference_title: Autism spectrum disorder.
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
      explanation: >-
        The phenotype is definitional, but the proposed mediation by distributed
        circuit differences remains indirect.
  - target: Restricted and Repetitive Behavior
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Distributed developmental differences may contribute to repetitive and
      inflexible behavior, without implying one universal cognitive or circuit
      cause.
    evidence:
    - reference: PMID:31949163
      reference_title: Autism spectrum disorder.
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
      explanation: >-
        The source supports the outcome domain but not a single circuit-level
        causal route.
- name: Excitation-inhibition and synaptic-signaling differences in studied subgroups
  description: >-
    E/I imbalance is a useful research framework supported by particular
    genetic syndromes, cellular models, and group-level studies, but direction,
    cell type, region, and developmental timing differ. mGluR5, GABAergic, and
    glutamatergic findings must not be generalized from biologically defined
    subsets to all ASD.
  mechanism_confidence: PROVISIONAL
  biological_processes:
  - preferred_term: chemical synaptic transmission
    term:
      id: GO:0007268
      label: chemical synaptic transmission
    modifier: ABNORMAL
  evidence:
  - reference: PMID:41653294
    reference_title: "Unraveling mGluR5 dysfunction in autism spectrum disorder: a multi-level analysis of genetic, molecular, and neurobiological mechanisms."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Alterations in synaptic scaffolding complexes (SHANK3-Homer-mGluR5 interactions), receptor trafficking, and activity-dependent protein synthesis may contribute to excitatory/inhibitory imbalance and circuit dysfunction.
    explanation: >-
      The review uses conditional language and concerns biologically defined
      pathways, so it supports only a provisional subgroup mechanism.
  - reference: PMID:33076974
    reference_title: Pharmacological intervention to restore connectivity deficits of neuronal networks derived from ASD patient iPSC with a TSC2 mutation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      We find that ASD patient-derived neurons with a functional loss of TSC2, in addition to possessing neuronal hyperactivity, develop a dysfunctional neuronal network with reduced synchronisation of neuronal bursting and lower spatial connectivity.
    explanation: >-
      The finding is limited to patient-derived neurons with TSC2 loss and is
      not evidence of universal neuronal hyperactivity in ASD.
- name: Immune and neuroinflammatory associations in studied groups
  description: >-
    Studies report group-level cytokine, oxidative-stress, glial, and
    connectivity associations. Sampling, medication, co-occurring illness,
    tissue, age, and subgroup differences complicate interpretation. These
    associations do not establish that neuroinflammation causes core autism in
    all people, and blood-brain-barrier findings are subtle or context dependent.
  mechanism_confidence: HYPOTHETICAL
  biological_processes:
  - preferred_term: neuroinflammatory response
    term:
      id: GO:0150076
      label: neuroinflammatory response
  evidence:
  - reference: PMID:41947852
    reference_title: The relationship between functional brain connectivity and neuroinflammatory processes-new insights into the pathomechanisms of ASD.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      individuals with ASD show consistent evidence of microglial and astroglial activation, altered cytokine profiles (including IL-1β, IL-6, and TNF-α), and markers of oxidative stress such as glutathione imbalance and lipid peroxidation.
    explanation: >-
      This is review-level group evidence and does not establish individual
      universality, directionality, or mediation of diagnostic traits.
  - reference: PMID:41550027
    reference_title: "The gatekeepers breached: claudin dysregulation in psychiatric disorders."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      In autism and ADHD, altered tight junction protein profiles imply more subtle or context-dependent barrier dysfunction.
    explanation: >-
      The review explicitly requires a subtle, context-dependent interpretation.
  downstream:
  - target: Distributed developmental circuit differences
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - immune_microbiome_modifier_model
    description: >-
      Within the emerging modifier hypothesis, immune and metabolic differences
      are associated with synaptic and network measures; directionality and
      mediation of clinical traits remain unestablished.
    evidence:
    - reference: PMID:41947852
      reference_title: The relationship between functional brain connectivity and neuroinflammatory processes-new insights into the pathomechanisms of ASD.
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        These immune and metabolic alterations are associated with changes in synaptic plasticity, neurotransmission, and large-scale neuronal network organization, including altered functional connectivity within the default mode, salience, and executive control networks.
      explanation: >-
        The review reports association rather than causal mediation, hence the
        PARTIAL support and hypothesis-only edge.
- name: Gut microbiome associations in studied groups
  description: >-
    Microbiome composition and metabolite differences have been reported in
    some ASD cohorts, especially those selected for gastrointestinal symptoms.
    Diet, medication, geography, age, constipation, sequencing methods, and
    reverse causation are major confounders. A causal microbiome-to-core-autism
    pathway and a generalizable microbial biomarker remain unproven.
  mechanism_confidence: HYPOTHETICAL
  evidence:
  - reference: PMID:39733842
    reference_title: "IUPHAR review: Targeted therapies of signaling pathways based on the gut microbiome in autism spectrum disorders: Mechanistic and therapeutic applications."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Gut microbiota significantly influences behavior and neurodevelopment by regulating the gut-brain axis.
    explanation: >-
      This narrative review supports biological plausibility, not an ASD-specific
      causal pathway or a finding present in every autistic person.
  downstream:
  - target: Immune and neuroinflammatory associations in studied groups
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - immune_microbiome_modifier_model
    description: >-
      Within the emerging modifier hypothesis, dysbiosis may affect immune and
      neuroinflammatory signaling. Confounding and reverse causation remain
      major alternatives.
    evidence:
    - reference: PMID:39733842
      reference_title: "IUPHAR review: Targeted therapies of signaling pathways based on the gut microbiome in autism spectrum disorders: Mechanistic and therapeutic applications."
      supports: SUPPORT
      evidence_source: OTHER
      snippet: >-
        Dysbiosis in the gut microbiome may impact neuroinflammatory processes linked to autism, negatively affecting immune signaling pathways.
      explanation: >-
        The source itself uses conditional language; this supports only the
        tagged hypothesis edge.
- name: Convergent transcriptional effects in human stem-cell models
  description: >-
    In iPSC-derived neural progenitor and cortical organoid models carrying
    selected ASD-associated mutations, mutation-specific early changes partly
    converge later on shared transcriptional networks. The result is a model
    system observation across selected genotypes, not proof of a universal
    cortical program in autistic brains.
  mechanism_confidence: PROVISIONAL
  cell_types:
  - preferred_term: neural progenitor cell
    term:
      id: CL:0011020
      label: neural progenitor cell
  biological_processes:
  - preferred_term: regulation of DNA-templated transcription
    term:
      id: GO:0006355
      label: regulation of DNA-templated transcription
    modifier: ABNORMAL
  evidence:
  - reference: PMID:41611887
    reference_title: Developmental convergence and divergence in human stem cell models of autism.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      Early time points harboured the largest mutation-specific changes, but different mutations converged on shared transcriptional changes as development progressed.
    explanation: >-
      Supports developmental convergence within the studied stem-cell models
      while retaining mutation-specific early effects.
mechanistic_hypotheses:
- hypothesis_group_id: subgroup_ei_synaptic_model
  hypothesis_label: Subgroup-specific excitation-inhibition and synaptic-signaling model
  status: EMERGING
  description: >-
    Some genetic and cellular ASD subgroups may reach similar circuit-level
    phenotypes through altered excitatory or inhibitory synaptic signaling.
    Direction and developmental timing may differ, and the model is not a
    universal ASD mechanism or a basis for general treatment selection.
  evidence:
  - reference: PMID:33076974
    reference_title: Pharmacological intervention to restore connectivity deficits of neuronal networks derived from ASD patient iPSC with a TSC2 mutation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      We find that ASD patient-derived neurons with a functional loss of TSC2, in addition to possessing neuronal hyperactivity, develop a dysfunctional neuronal network with reduced synchronisation of neuronal bursting and lower spatial connectivity.
    explanation: A genotype-specific cellular model motivates, but cannot universalize, the hypothesis.
- hypothesis_group_id: immune_microbiome_modifier_model
  hypothesis_label: Immune and microbiome modifiers in selected ASD subgroups
  status: EMERGING
  description: >-
    Immune state, gastrointestinal physiology, diet, and microbiome composition
    may modify symptoms or co-occurring conditions in selected people. Current
    evidence does not establish that immune activation, blood-brain-barrier
    disruption, enteric nervous-system dysfunction, or dysbiosis is a primary
    cause of the broad ASD diagnosis.
  evidence:
  - reference: PMID:41550027
    reference_title: "The gatekeepers breached: claudin dysregulation in psychiatric disorders."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      In autism and ADHD, altered tight junction protein profiles imply more subtle or context-dependent barrier dysfunction.
    explanation: The source itself frames barrier findings as context dependent.
phenotypes:
- category: Core diagnostic
  name: Abnormal Nonverbal Communicative Behavior
  description: >-
    Persistent difficulty integrating or using eye gaze, facial expression,
    gesture, body language, or other nonverbal communication in social
    interaction. Manifestation and severity vary with age, language, culture,
    context, and compensatory strategies.
  diagnostic: true
  severity: Variable
  phenotype_term:
    preferred_term: Abnormal nonverbal communicative behavior
    term:
      id: HP:0000758
      label: Abnormal nonverbal communicative behavior
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
    explanation: The primer directly supports the social-communication domain.
- category: Core diagnostic
  name: Restricted and Repetitive Behavior
  description: >-
    Restricted or repetitive motor behavior, speech, object use, interests, or
    insistence on sameness forms one required diagnostic domain, but no single
    narrow behavior is present in every person.
  diagnostic: true
  severity: Variable
  phenotype_term:
    preferred_term: Restrictive behavior
    term:
      id: HP:0000723
      label: Restrictive behavior
  evidence:
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Core deficits are identified in 2 domains: social communication/interaction and restrictive, repetitive patterns of behavior.
    explanation: The AAP report directly supports this diagnostic domain.
- category: Core diagnostic
  name: Reduced Ability to Form Peer Relationships
  description: >-
    Difficulty developing, maintaining, or understanding peer relationships is
    one manifestation of the required social-interaction domain. It varies with
    age, opportunity, culture, communication, masking, and the accessibility of
    the social environment; it is not itself obligatory in one fixed form.
  diagnostic: true
  severity: Variable
  phenotype_term:
    preferred_term: Reduced ability to form peer relationships
    term:
      id: HP:0000728
      label: Reduced ability to form peer relationships
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Developing, maintaining and understanding relationships
    explanation: >-
      The reproduced DSM-5 criteria identify the relationship subdomain; the
      phenotype remains context-sensitive rather than a uniform social outcome.
- category: Core diagnostic
  name: Motor Stereotypy
  description: >-
    Repetitive motor movements may occur within the restricted/repetitive
    domain, but motor stereotypy itself is not required for diagnosis.
  severity: Variable
  phenotype_term:
    preferred_term: Motor stereotypy
    term:
      id: HP:0000733
      label: Motor stereotypy
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
    explanation: Supports repetitive behavior generally; motor stereotypy is one non-obligate manifestation.
- category: Core diagnostic
  name: Inflexible Adherence to Routines
  description: >-
    Insistence on sameness, ritualized patterns, or marked difficulty with
    transitions can satisfy part of the restricted/repetitive domain but varies
    widely and is not individually obligatory.
  severity: Variable
  phenotype_term:
    preferred_term: Inflexible adherence to routines
    term:
      id: HP:0000732
      label: Inflexible adherence to routines
  evidence:
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Core deficits are identified in 2 domains: social communication/interaction and restrictive, repetitive patterns of behavior.
    explanation: Supports the parent domain; this is one variable manifestation.
- category: Associated developmental feature
  name: Delayed Speech and Language Development
  description: >-
    Spoken-language delay occurs in a clinically important subset and can drive
    early referral, but it is not required for ASD and fluent or precocious
    language does not exclude the diagnosis.
  severity: Variable
  phenotype_term:
    preferred_term: Delayed speech and language development
    term:
      id: HP:0000750
      label: Delayed speech and language development
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Psychosocial interventions in children can improve specific behaviours, such as joint attention, language and social engagement, that may affect further development and could reduce symptom severity.
    explanation: >-
      The primer identifies language as a variable developmental domain; it does
      not make language delay a required feature.
- category: Core diagnostic
  name: Sensory Behavioral Abnormality
  description: >-
    Hyperreactivity, hyporeactivity, or unusual interest in sensory input can
    occur within the restricted/repetitive diagnostic domain. Modality,
    direction, severity, and functional impact vary; no universal percentage is
    asserted because estimates depend strongly on instruments and samples.
  severity: Variable
  phenotype_term:
    preferred_term: Sensory behavioral abnormality
    term:
      id: HP:5200046
      label: Sensory behavioral abnormality
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Autism spectrum disorder is a construct used to describe individuals with a specific combination of impairments in social communication and repetitive behaviours, highly restricted interests and/or sensory behaviours beginning early in life.
    explanation: Directly supports sensory behavior as part of the spectrum definition.
- category: Co-occurring
  name: Intellectual Disability
  description: >-
    Intellectual disability co-occurs in a substantial subset but is neither
    necessary nor sufficient for ASD. Cognitive profiles are heterogeneous and
    estimates vary with cohort and test completion.
  severity: Variable
  frequency: FREQUENT
  phenotype_term:
    preferred_term: Intellectual disability
    term:
      id: HP:0001249
      label: Intellectual disability
  evidence:
  - reference: PMID:40232988
    reference_title: Prevalence and Early Identification of Autism Spectrum Disorder Among Children Aged 4 and 8 Years - Autism and Developmental Disabilities Monitoring Network, 16 Sites, United States, 2022.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Among 5,292 (61.4% of 8,613) children aged 8 years with ASD with information on cognitive ability, 39.6% were classified as having an intellectual disability.
    explanation: >-
      Site-based surveillance supports a frequent co-occurrence while the
      missing cognitive data and ascertainment context limit generalization.
- category: Co-occurring
  name: Sleep Disturbance
  description: >-
    Sleep-wake problems are among the most frequent co-occurring conditions,
    with heterogeneous insomnia, circadian, breathing, and movement phenotypes.
    Sleep problems warrant their own clinical evaluation rather than attribution
    to a universal immune-pineal mechanism.
  phenotype_term:
    preferred_term: Sleep disturbance
    term:
      id: HP:0002360
      label: Sleep disturbance
  evidence:
  - reference: PMID:37913872
    reference_title: "Prevalence of co-occurring conditions in children and adults with autism spectrum disorder: A systematic review and meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Point prevalence of developmental coordination disorder, sleep-wake problem, gastrointestinal problem, ADHD, anxiety disorder, overweight/obesity, feeding and eating disorder, elimination disorder, disruptive behavior, and somatic symptoms and related disorder were the most frequent CCs.
    explanation: >-
      The large meta-analysis supports common co-occurrence without imposing an
      unsupported universal percentage.
- category: Co-occurring
  name: Gastrointestinal Symptoms
  description: >-
    Constipation, abdominal pain, diarrhea, and other functional gastrointestinal
    symptoms occur more often than in neurotypical pediatric comparators, but
    prevalence is heterogeneous and their presence does not establish a causal
    gut-to-autism mechanism.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: Functional abnormality of the gastrointestinal tract
    term:
      id: HP:0012719
      label: Functional abnormality of the gastrointestinal tract
  evidence:
  - reference: PMID:37474417
    reference_title: "Prevalence of gastrointestinal symptoms in autism spectrum disorder: A meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      The prevalence of GI symptoms ranged between 0% and 69%, with an estimated general prevalence of 33% (95% CI, 13%-57%), higher than that reported by a previous meta-analysis for the general paediatric population.
    explanation: >-
      The pooled estimate supports FREQUENT while the range and interval document
      substantial heterogeneity.
- category: Co-occurring
  name: Seizure
  description: >-
    Epilepsy occurs in a minority, with higher prevalence in older groups and
    people with intellectual disability. Seizures require separate diagnosis
    and management and must not be used to infer one ASD mechanism.
  frequency: OCCASIONAL
  phenotype_term:
    preferred_term: Seizure
    term:
      id: HP:0001250
      label: Seizure
  evidence:
  - reference: PMID:34510916
    reference_title: "Prevalence of epilepsy in autism spectrum disorders: A systematic review and meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      About 1/10 autistic individuals also had epilepsy, which was common in the clinical setting, adolescents, adults, females, or patients with intellectual disability and less common in the country with high human development index.
    explanation: >-
      The meta-analysis supports an occasional overall frequency and identifies
      important modifiers rather than a uniform risk.
- category: Co-occurring
  name: Attention Deficit Hyperactivity Disorder
  description: >-
    ADHD commonly co-occurs and can materially affect learning, daily function,
    and social participation. It is a separate co-occurring diagnosis, not a
    defining ASD manifestation.
  phenotype_term:
    preferred_term: Attention deficit hyperactivity disorder
    term:
      id: HP:0007018
      label: Attention deficit hyperactivity disorder
  evidence:
  - reference: PMID:37913872
    reference_title: "Prevalence of co-occurring conditions in children and adults with autism spectrum disorder: A systematic review and meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Point prevalence of developmental coordination disorder, sleep-wake problem, gastrointestinal problem, ADHD, anxiety disorder, overweight/obesity, feeding and eating disorder, elimination disorder, disruptive behavior, and somatic symptoms and related disorder were the most frequent CCs.
    explanation: The systematic review supports ADHD as a frequent co-occurring condition.
genetic:
- name: CHD8
  gene_term:
    preferred_term: CHD8
    term:
      id: hgnc:20153
      label: CHD8
  association: Risk Factor
  notes: >-
    Illustrative pleiotropic ASD risk gene supported by rare-variant sequencing;
    inclusion does not imply that CHD8 variation is necessary or sufficient for
    broad ASD, or that every carrier has the same phenotype.
  evidence:
  - reference: PMID:24387789
    reference_title: "A de novo convergence of autism genetics and molecular neuroscience."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and catenin (cadherin-associated protein), beta 1, 88 kDa (CTNNB1, beta-catenin).
    explanation: >-
      The review identifies CHD8 among rare-variant loci supported by targeted
      large-scale resequencing; the record is therefore typed as a risk factor,
      not a sufficient cause of broad ASD.
- name: SCN2A
  gene_term:
    preferred_term: SCN2A
    term:
      id: hgnc:10588
      label: SCN2A
  association: Risk Factor
  notes: >-
    Illustrative pleiotropic ASD risk gene supported by rare-variant sequencing;
    variant class and functional direction matter, and SCN2A-associated
    neurodevelopmental conditions are not interchangeable with broad ASD.
  evidence:
  - reference: PMID:24387789
    reference_title: "A de novo convergence of autism genetics and molecular neuroscience."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and catenin (cadherin-associated protein), beta 1, 88 kDa (CTNNB1, beta-catenin).
    explanation: >-
      The review identifies SCN2A among rare-variant loci supported by targeted
      large-scale resequencing; the record does not generalize across variant
      mechanisms or treat the gene as a sufficient cause.
- name: DYRK1A
  gene_term:
    preferred_term: DYRK1A
    term:
      id: hgnc:3091
      label: DYRK1A
  association: Risk Factor
  notes: >-
    Illustrative pleiotropic ASD risk gene supported by rare-variant sequencing;
    this bounded record is not a claim that DYRK1A explains ASD generally.
  evidence:
  - reference: PMID:24387789
    reference_title: "A de novo convergence of autism genetics and molecular neuroscience."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and catenin (cadherin-associated protein), beta 1, 88 kDa (CTNNB1, beta-catenin).
    explanation: >-
      The review identifies DYRK1A among rare-variant loci supported by targeted
      large-scale resequencing and supports risk-factor, rather than universal
      causal, typing.
- name: CTNNB1
  gene_term:
    preferred_term: CTNNB1
    term:
      id: hgnc:2514
      label: CTNNB1
  association: Risk Factor
  notes: >-
    Illustrative pleiotropic ASD risk gene supported by rare-variant sequencing;
    CTNNB1-related syndromic disease remains distinct from the broad ASD entry.
  evidence:
  - reference: PMID:24387789
    reference_title: "A de novo convergence of autism genetics and molecular neuroscience."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Targeted large-scale resequencing studies have confirmed the significance of specific loci, including chromodomain helicase DNA binding protein 8 (CHD8), sodium channel, voltage-gated, type II, alpha subunit (SCN2A), dual specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A), and catenin (cadherin-associated protein), beta 1, 88 kDa (CTNNB1, beta-catenin).
    explanation: >-
      The review identifies CTNNB1 among rare-variant loci supported by targeted
      large-scale resequencing; it does not make CTNNB1 a sufficient or
      spectrum-wide cause.
environmental:
- name: Prenatal and perinatal epidemiologic associations
  description: >-
    Advanced parental age, maternal metabolic conditions, short interpregnancy
    interval, and some proxies of perinatal hypoxia have been associated with
    ASD in observational studies. Residual confounding, genetics, and correlated
    exposures limit causal inference; these are population risk associations,
    not explanations for an individual diagnosis. Vaccination is not associated
    with ASD.
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      considered causal, but could be reactive, independent or contributory
    explanation: >-
      The primer explicitly cautions against promoting observational risk
      associations to established causes.
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      including clear evidence that ASD is not associated with vaccination
    explanation: Directly records the well-established negative association.
- name: Prenatal valproate exposure
  exposure_term:
    preferred_term: gestational maternal valproate exposure
    term:
      id: XCO:0001598
      label: gestational maternal exposure to valproate
  influences_mechanisms:
  - target: Heterogeneous neurodevelopmental molecular effects
    environmental_effect: PREDISPOSES
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      The best-evidenced environmental association in this entry: a population
      cohort reporting increased risk after adjustment for maternal epilepsy,
      which is the confounder that would otherwise explain it. Recorded as
      predisposing and never as triggering, because this entry is careful
      throughout that these are population-level risk associations and not
      explanations for an individual diagnosis, and because the exposure's own
      description warns against abrupt medication changes in pregnancy. No
      cited sentence follows valproate to any molecular step, so the
      intermediates are unknown despite the node being molecular.
    evidence:
    - reference: PMID:23613074
      reference_title: "Prenatal valproate exposure and risk of autism spectrum disorders and childhood autism."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: "Maternal use of valproate during pregnancy was associated with a significantly increased risk of autism spectrum disorder and childhood autism in the offspring, even after adjusting for maternal epilepsy."
      explanation: >-
        Danish population cohort finding maternal valproate use in pregnancy
        associated with significantly increased offspring risk even after
        adjusting for maternal epilepsy. An adjusted population association.
  description: >-
    In a Danish population cohort, maternal valproate use during pregnancy was
    associated with increased offspring ASD risk after adjustment for maternal
    epilepsy. This is a population-level medication-exposure association, not a
    deterministic explanation for an individual diagnosis; pregnancy treatment
    decisions require individualized assessment of maternal seizure and fetal
    risks rather than abrupt medication changes.
  evidence:
  - reference: PMID:23613074
    reference_title: Prenatal valproate exposure and risk of autism spectrum disorders and childhood autism.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Maternal use of valproate during pregnancy was associated with a significantly increased risk of autism spectrum disorder and childhood autism in the offspring, even after adjusting for maternal epilepsy.
    explanation: >-
      The population cohort supports a specific adjusted association while the
      wording avoids treating exposure as sufficient or inevitable causation.
- name: Prenatal and early-childhood heavy metal exposure
  influences_mechanisms:
  - target: Heterogeneous neurodevelopmental molecular effects
    environmental_effect: PREDISPOSES
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Graded below the valproate link and carrying its own contradiction on
      purpose. The same review that reports a comparatively consistent
      adjusted association for lead and mercury reports inconsistent
      associations for arsenic and cadmium, and bounds its own conclusion to
      certain populations and exposure conditions. All three sentences are
      carried here, so the reader of the pathograph meets the metal-by-metal
      disagreement rather than a uniform risk factor. The candidate mechanisms
      named in this exposure's notes are recorded there as candidates and are
      not asserted by any link.
    evidence:
    - reference: PMID:42540380
      reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
      supports: SUPPORT
      evidence_source: OTHER
      snippet: "In contrast, elevated prenatal and early childhood Pb and Hg concentrations in blood and hair samples showed a significant consistent association with both increased ASD risk and symptom severity, even after adjustment for key demographic and environmental confounders."
      explanation: >-
        Reports a significant consistent adjusted association for lead and
        mercury with risk and symptom severity. A narrative synthesis rather
        than a pooled estimate.
    - reference: PMID:42540380
      reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
      supports: SUPPORT
      evidence_source: OTHER
      snippet: "We revealed inconsistent associations between prenatal and childhood urinary, blood, and hair, As and Cd exposure, and ASD outcomes."
      explanation: >-
        The same review recording inconsistent associations for arsenic and
        cadmium. Carried so this exposure class is not read as uniform across
        metals.
    - reference: PMID:42540380
      reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
      supports: SUPPORT
      evidence_source: OTHER
      snippet: "Rather, lead and mercury may contribute to ASD risk in certain populations or under specific exposure conditions, but these findings require cautious interpretation due to methodological differences and potential biases."
      explanation: >-
        The authors' own bound: lead and mercury may contribute in certain
        populations or under specific exposure conditions, with cautious
        interpretation required.
  description: >-
    Prenatal and early-childhood exposure to lead, mercury, cadmium, and arsenic
    has been examined as a possible contributor to ASD risk. The reported
    associations are metal specific and inconsistent: reviews of human
    epidemiological studies report a more consistent association for lead and
    mercury than for arsenic and cadmium, while acknowledging residual
    confounding and heterogeneous exposure-assessment methods. This is a
    contested population-level exposure association, not an established cause of
    an individual diagnosis, and the reviewed evidence does not support a
    blanket claim that heavy metals cause ASD.
  notes: >-
    Candidate mechanisms, recorded as candidates only. Three routes are
    repeatedly proposed for lead and mercury developmental neurotoxicity:
    oxidative stress with mitochondrial dysfunction and lipid peroxidation;
    impaired synaptic plasticity, including altered NMDA receptor subunit
    composition and reduced synaptic protein expression; and microglial and
    astrocytic activation with neuroinflammatory signaling. A zebrafish arm of
    one birth-cohort study additionally nominates PPAR-linked lipid and steroid
    metabolic disruption. Every one of these is characterized in rodent,
    zebrafish, or cell models; none has been demonstrated in human
    neurodevelopment, at the cord-blood concentrations measured in human
    cohorts, or as a mediator of autistic traits. They are therefore documented
    here rather than curated as pathophysiology nodes, and the open
    translational question is tracked in the discussion
    mismatch_asd_heavy_metal_model_mechanism.
  chemicals:
  - lead
  - mercury
  - cadmium
  - arsenic
  exposure_term:
    preferred_term: exposure to heavy metal
    term:
      id: ECTO:9002163
      label: exposure to heavy metal
  evidence:
  - reference: PMID:42540380
    reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      In contrast, elevated prenatal and early childhood Pb and Hg concentrations in blood and hair samples showed a significant consistent association with both increased ASD risk and symptom severity, even after adjustment for key demographic and environmental confounders.
    explanation: >-
      A critical review of 36 human epidemiological studies reports a
      comparatively consistent adjusted association for lead and mercury. It is
      a narrative synthesis rather than a pooled estimate, so it supports a
      population-level association only.
  - reference: PMID:42540380
    reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      We revealed inconsistent associations between prenatal and childhood urinary, blood, and hair, As and Cd exposure, and ASD outcomes.
    explanation: >-
      The same review records that arsenic and cadmium associations are
      inconsistent, so this exposure class must not be curated as a uniform risk
      factor across metals.
  - reference: PMID:42540380
    reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Rather, lead and mercury may contribute to ASD risk in certain populations or under specific exposure conditions, but these findings require cautious interpretation due to methodological differences and potential biases.
    explanation: >-
      The authors' own summary bounds the claim to specific populations and
      exposure conditions and requires cautious interpretation.
  - reference: PMID:41110787
    reference_title: "Prenatal exposure to environmental metal mixtures and social development in early childhood: Evidence from a birth cohort and mechanistic study."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Prenatal exposure to the mixture significantly impaired motor and social functions, with Pb contributing most to motor deficits and Cd to social dysfunction.
    explanation: >-
      A prospective birth cohort with cord-blood metal measurement reports a
      mixture association with early social and motor development. The outcome
      is a developmental measure rather than an ASD diagnosis, so it supports
      the exposure association and not an ASD-specific causal pathway.
  - reference: PMID:42541535
    reference_title: "Molecular and cellular mechanisms of lead-induced neurotoxicity: comparative insights from rodent and zebrafish models."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      Once in the brain, lead induces oxidative stress through excessive reactive oxygen species generation, mitochondrial dysfunction, lipid peroxidation, DNA damage, and depletion of antioxidant defenses.
    explanation: >-
      Documents the oxidative-stress candidate mechanism. The finding is from
      rodent and zebrafish models and is recorded as a candidate route, not as a
      demonstrated human ASD mechanism.
  - reference: PMID:42541535
    reference_title: "Molecular and cellular mechanisms of lead-induced neurotoxicity: comparative insights from rodent and zebrafish models."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      Lead also impairs synaptic plasticity by altering NMDA receptor subunit composition, reducing synaptic protein expression, and dysregulating genes involved in neurodevelopment.
    explanation: >-
      Documents the synaptic candidate mechanism, again in animal models. It is
      deliberately not linked to this entry's excitation-inhibition node, which
      is curated from genotype-defined human subgroup work.
  - reference: PMID:42541535
    reference_title: "Molecular and cellular mechanisms of lead-induced neurotoxicity: comparative insights from rodent and zebrafish models."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      In parallel, it activates both intrinsic and extrinsic apoptotic pathways and enhances neuroinflammatory signaling through microglial and astrocytic activation, further contributing to neuronal injury.
    explanation: >-
      Documents the glial-activation candidate mechanism. Demonstrating this in
      human tissue at cohort-relevant exposure levels is the condition for
      curating any heavy-metal neuroinflammation mechanism node.
treatments:
- name: Individualized developmental and behavioral intervention
  action_category: THERAPEUTIC
  description: >-
    Goal-directed developmental and behavioral programs can improve selected
    communication, adaptive, cognitive, play, or social-engagement outcomes in
    some young children. Program model, intensity, acceptability, access,
    outcomes, and response vary. Evidence does not support a promise of
    normalization or one mandatory program for every child; goals should be
    selected collaboratively and monitored for benefit and burden.
  treatment_term:
    preferred_term: behavioral intervention
    term:
      id: NCIT:C15184
      label: Behavioral Intervention
  target_phenotypes:
  - preferred_term: Abnormal nonverbal communicative behavior
    term:
      id: HP:0000758
      label: Abnormal nonverbal communicative behavior
  - preferred_term: Delayed speech and language development
    term:
      id: HP:0000750
      label: Delayed speech and language development
  evidence:
  - reference: PMID:41502379
    reference_title: "Clinically Significant Outcomes of Early Intensive Behavioral Intervention for Children With Autism Spectrum Disorders: An Individual Participant Data Meta-Analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Across the 17 identified studies, we obtained participant data from 15 studies: 341 children received EIBI and 280 were in comparison-groups. All studies had a serious risk of bias due to the lack of random assignment.
    explanation: >-
      The individual-participant meta-analysis supports possible benefit while
      requiring explicit acknowledgment of serious bias.
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Psychosocial interventions in children can improve specific behaviours, such as joint attention, language and social engagement, that may affect further development and could reduce symptom severity.
    explanation: Supports domain-specific, non-universal benefit.
- name: Communication, educational, environmental, and family supports
  action_category: THERAPEUTIC
  description: >-
    Speech-language therapy, augmentative and alternative communication,
    occupational and educational supports, sensory or environmental
    accommodations, caregiver education, transition planning, and assistance
    with community participation, housing, transport, employment, and benefits
    should be individualized to communication, function, preferences, and
    support needs across the lifespan.
  treatment_term:
    preferred_term: supportive care
    term:
      id: NCIT:C15747
      label: Supportive Care
  evidence:
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Children and youth with ASD have service needs in behavioral, educational, health, leisure, family support, and other areas.
    explanation: The AAP report supports broad, coordinated pediatric service needs.
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Families are often the major source of support for people with autism throughout much of life and need to be considered, along with the perspectives of autistic individuals, in both research and practice.
    explanation: Supports lifespan family support and autistic-person participation.
- name: Pharmacotherapy for severe irritability and co-occurring conditions
  action_category: THERAPEUTIC
  therapeutic_modality: SMALL_MOLECULE
  description: >-
    Medication does not treat autism as a whole. Risperidone or aripiprazole can
    reduce severe irritability, aggression, self-injury, or tantrums in selected
    children and adolescents, with sedation, weight/metabolic effects, and
    movement-disorder risk requiring monitoring. ADHD, anxiety, depression,
    epilepsy, sleep, and gastrointestinal disorders should be assessed and
    treated on their own evidence base with autism-relevant tolerability and
    communication considerations.
  treatment_term:
    preferred_term: pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: risperidone
      term:
        id: CHEBI:8871
        label: risperidone
    - preferred_term: aripiprazole
      term:
        id: CHEBI:31236
        label: aripiprazole
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      All medications that have evidence of benefit for ASD treat the associated symptoms
    explanation: Directly constrains medication claims to associated symptoms and diagnoses.
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Risperidone and aripiprazole (both of which are atypical antipsychotics) are approved in the USA to treat
    explanation: >-
      The cached line fragment identifies the two agents and their US approval
      context; the source context and bounded treatment description supply the
      associated-symptom target without claiming core-ASD treatment.
- name: Melatonin for co-occurring sleep disturbance
  action_category: THERAPEUTIC
  therapeutic_modality: SMALL_MOLECULE
  description: >-
    Melatonin can improve total sleep time, sleep latency, or sleep efficiency
    for some autistic children with sleep disturbance after behavioral and
    environmental contributors are assessed. Trial heterogeneity is substantial;
    benefit is symptom-targeted rather than treatment of autism as a whole, and
    dose, formulation, interactions, and adverse effects require clinical review.
  treatment_term:
    preferred_term: pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: melatonin
      term:
        id: CHEBI:16796
        label: melatonin
  target_phenotypes:
  - preferred_term: Sleep disturbance
    term:
      id: HP:0002360
      label: Sleep disturbance
  evidence:
  - reference: PMID:36584862
    reference_title: "Melatonin for sleep disorders in people with autism: Systematic review and meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Melatonin has possible efficacy over total time, latency, and efficiency sleep parameters.
    explanation: >-
      The meta-analysis supports possible sleep-domain benefit, while the
      treatment description records its substantial between-study heterogeneity
      and symptom-specific scope.
clinical_trials:
- name: NCT03504917
  phase: PHASE_III
  status: TERMINATED
  description: >-
    Industry-sponsored randomized double-blind placebo-controlled Phase III
    trial of oral balovaptan in 322 autistic adults. The registry reports that
    it was terminated after a futility analysis found the predefined primary
    objective highly unlikely to be met; no new safety concern was identified.
    This negative program illustrates why social-neuropeptide hypotheses cannot
    be promoted to established treatment mechanisms.
  target_phenotypes:
  - preferred_term: Abnormal nonverbal communicative behavior
    term:
      id: HP:0000758
      label: Abnormal nonverbal communicative behavior
  evidence:
  - reference: clinicaltrials:NCT03504917
    reference_title: A Phase III, Randomized, Double-Blind, Placebo-Controlled, Efficacy, and Safety Study of Balovaptan in Adults With Autism Spectrum Disorder With a 2-Year Open-Label Extension
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      This study will evaluate the efficacy, safety, and pharmacokinetics of 10 mg of oral administration balovaptan once a day (QD) compared with matching placebo in adults (18 years and older) with autism spectrum disorder (ASD).
    explanation: >-
      The registry supports the intervention, comparator, population, and
      purpose; phase, enrollment, termination, and futility reason were
      rechecked in the live registry on 2026-08-05.
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      balovaptan, a vasopressin AVPR1A antagonist in adults with ASD showed negative results on its primary
    explanation: >-
      The disease primer independently reports the negative primary outcome;
      the live registry supplies the later termination and futility status.
experimental_models:
- name: Genotype-defined patient iPSC-derived neuronal networks
  experimental_model_type: IPSC_DERIVED_MODEL
  cell_source: Patient-derived induced pluripotent stem cells
  description: >-
    Neuronal networks derived from autistic participants with defined variants
    can test cellular excitability, synchrony, connectivity, and rescue. They
    retain donor/genotype relevance but lack whole-brain development, behavior,
    immune and environmental context, and population representativeness. The
    TSC2 result is a genotype-specific mechanistic model, not a model of all ASD.
  evidence:
  - reference: PMID:33076974
    reference_title: Pharmacological intervention to restore connectivity deficits of neuronal networks derived from ASD patient iPSC with a TSC2 mutation.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      We find that ASD patient-derived neurons with a functional loss of TSC2, in addition to possessing neuronal hyperactivity, develop a dysfunctional neuronal network with reduced synchronisation of neuronal bursting and lower spatial connectivity.
    explanation: Defines the model finding and its genotype-limited scope.
- name: Multi-genotype human cortical organoid and neural-progenitor panel
  experimental_model_type: ORGANOID
  cell_source: Induced pluripotent stem cell-derived, including isogenic perturbations
  description: >-
    A panel spanning selected ASD-associated mutations and idiopathic donor
    lines can compare mutation-specific and convergent transcriptional effects
    across neural progenitors and cortical organoids. Immaturity, variable cell
    composition, absent long-range inputs and vasculature, and selected genotypes
    limit inference to the human brain or the entire spectrum.
  evidence:
  - reference: PMID:41611887
    reference_title: Developmental convergence and divergence in human stem cell models of autism.
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: >-
      Early time points harboured the largest mutation-specific changes, but different mutations converged on shared transcriptional changes as development progressed.
    explanation: Supports the panel's central mutation-specific and convergent readout.
animal_models:
- name: Cntnap2 knockout mouse with maternal immune activation
  species: Mouse
  genotype: Cntnap2 knockout with maternal immune activation exposure
  category: Gene-environment interaction model
  description: >-
    Combines a defined genetic perturbation, prenatal immune challenge, and sex
    to test interaction effects on selected mouse behaviors. "Autistic-like"
    behavior is not equivalent to the human diagnosis, and results cannot
    establish maternal immune activation as a human ASD cause.
  evidence:
  - reference: PMID:41898431
    reference_title: Analysis of Gene, Environment, and Sex Interaction in the Development of Autistic-like Phenotype in Mice.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      The three-hit theory states that the vulnerability of an individual to develop ASD is modulated by the interplay between genetic predisposition, sex, and environmental insults.
    explanation: Supports the interaction model while not validating it as a universal human pathway.
- name: Prenatal interferon-alpha exposed rat
  species: Rat
  genotype: Prenatal interferon-alpha exposure
  category: Prenatal exposure model
  description: >-
    Tests how prenatal cytokine exposure affects neurotransmitter measures,
    histology, and selected offspring behaviors. It is an exposure/toxicity
    model and does not reproduce ASD's clinical definition or heterogeneous
    human genetic architecture.
  evidence:
  - reference: PMID:41484215
    reference_title: Prenatal Interferon-Alpha Exposure Induces Autism-Like Neurobehavioral and Neurochemical Alterations in Male Offspring.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      IFN-α exposure resulted in significant reductions in GABA, 5-HIAA, and GAD-67 levels, particularly in male offspring, indicating neurotransmitter dysregulation.
    explanation: Documents the model readout without equating it to human ASD.
diagnosis:
- name: Multidisciplinary clinical developmental assessment
  presence: >-
    Diagnosis is clinical and integrates developmental history, direct
    observation, current social communication and restricted/repetitive and
    sensory behavior, language, cognition, adaptive function, medical and
    psychiatric assessment, and information across settings. Hearing, vision,
    language, motor, learning, trauma, and co-occurring conditions are evaluated
    as appropriate. No laboratory or imaging test establishes broad ASD.
  diagnosis_term:
    preferred_term: clinical assessment
    term:
      id: NCIT:C124351
      label: Clinical Evaluation
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Diagnosis of ASD is made on the basis of
    explanation: >-
      The cached PDF line fragment identifies that diagnosis has an assessment
      basis but is truncated by line-number extraction; the complete case
      definition and adjunct-tool evidence above supply the substantive clinical
      diagnostic claim.
- name: Standardized diagnostic instruments as adjuncts
  presence: >-
    Instruments such as ADOS-2, ADI-R, CARS, and structured rating scales can
    organize observation and history, but thresholds do not replace expert
    clinical synthesis. Accuracy varies by tool, age, comparator, setting, and
    study quality; cultural, sex/gender, language, intellectual, and masking
    effects require consideration.
  diagnosis_term:
    preferred_term: clinical assessment
    term:
      id: NCIT:C124351
      label: Clinical Evaluation
  evidence:
  - reference: PMID:40274203
    reference_title: "Autism diagnosis in children and adolescents: A systematic review and meta-analysis of test accuracy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Diagnostic tools should be regarded as adjunctive aids rather than comprehensive substitutes for diagnosis.
    explanation: Directly states the intended use and limitation of standardized tools.
- name: Etiologic and co-occurring-condition evaluation
  presence: >-
    Genetic testing and targeted metabolic, neurologic, sleep, gastrointestinal,
    hearing, vision, or other evaluation can identify an etiology or separate
    condition in selected people. A pathogenic finding may change counseling or
    management but does not serve as a general ASD diagnostic biomarker, and a
    negative genetic test does not exclude ASD.
  diagnosis_term:
    preferred_term: clinical assessment
    term:
      id: NCIT:C124351
      label: Clinical Evaluation
  evidence:
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Primary care providers should be familiar with the diagnostic criteria for ASD, appropriate etiologic evaluation, and co-occurring medical and behavioral conditions
    explanation: Supports etiologic and co-occurring-condition evaluation as distinct from the clinical diagnosis.
differential_diagnoses:
- name: Social pragmatic communication disorder
  description: >-
    Social-communication impairment without the required history or presence of
    restricted/repetitive behavior supports social pragmatic communication
    disorder rather than ASD; the newer category has a less mature validation
    base.
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      with social communication problems but not restricted and repetitive behaviours who would previously
    explanation: The primer directly states the distinguishing absent domain.
- name: Intellectual disability or global developmental delay without ASD
  description: >-
    Developmental level can explain delayed communication and social behavior;
    ASD requires social-communication differences beyond those expected for
    developmental level plus the restricted/repetitive domain. Intellectual
    disability and ASD can also co-occur.
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Not better explained by intellectual disability or global developmental delay
    explanation: The reproduced diagnostic criteria state this exclusion boundary.
- name: Attention Deficit-Hyperactivity Disorder
  description: >-
    Inattention, impulsivity, social difficulty, or dysregulation can resemble
    parts of ASD, but ADHD does not require the two ASD diagnostic domains. The
    diagnoses frequently co-occur and one should not be used to erase the other.
  disease_term:
    preferred_term: attention deficit-hyperactivity disorder
    term:
      id: MONDO:0007743
      label: attention deficit-hyperactivity disorder
  evidence:
  - reference: PMID:31843864
    reference_title: Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      Primary care providers should be familiar with the diagnostic criteria for ASD, appropriate etiologic evaluation, and co-occurring medical and behavioral conditions (such as disorders of sleep and feeding, gastrointestinal tract symptoms, obesity, seizures, attention-deficit/hyperactivity disorder, anxiety, and wandering) that affect the child's function and quality of life.
    explanation: Supports ADHD as a distinct, potentially co-occurring condition.
- name: Genetic and neurologic syndromes with autistic features
  description: >-
    Rett syndrome, fragile X syndrome, tuberous sclerosis complex, and other
    genetic or neurologic conditions can include autistic features or a
    co-occurring ASD diagnosis. Regression pattern, examination, seizures,
    dysmorphology, family history, and targeted genetic evaluation can identify
    an additional etiology; the syndromic diagnosis does not automatically
    establish or exclude ASD.
  evidence:
  - reference: PMID:31949163
    reference_title: Autism spectrum disorder.
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      other disorders including ADHD, intellectual disability, language delay and genetic syndromes.
    explanation: Supports dual diagnosis and cautions against treating genetic syndromes as interchangeable with broad ASD.
discussions:
- discussion_id: gap_asd_validated_biological_stratification
  prompt: >-
    Which reproducible combinations of genotype, development, physiology, and
    phenotype define treatment-relevant ASD subgroups without collapsing
    heterogeneous people into a universal biomarker signature?
  kind: KNOWLEDGE_GAP
  status: OPEN
  attaches_to:
  - pathophysiology#Polygenic neurodevelopmental liability
  - pathophysiology#Rare coding and copy-number variant liability
  - pathophysiology#Distributed developmental circuit differences
  rationale: >-
    Imaging, EEG, immune, metabolomic, microbiome, and molecular differences are
    generally group associations with limited replication or specificity. No
    validated biological test diagnoses broad ASD or selects a general
    core-symptom treatment. Large, diverse, longitudinal, externally replicated
    studies with prespecified individual-level performance are needed.
  evidence:
  - reference: PMID:40274203
    reference_title: "Autism diagnosis in children and adolescents: A systematic review and meta-analysis of test accuracy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Overall, certainty of the evidence was low and very low except for ADOS-2 (moderate).
    explanation: Even established behavioral tools have important evidence limitations.
- discussion_id: gap_asd_lifespan_natural_history_and_supports
  prompt: >-
    Which supports improve self-defined quality of life, health, communication,
    autonomy, safety, participation, housing, and employment across diverse
    autistic adults and people with high communication or support needs?
  kind: KNOWLEDGE_GAP
  status: OPEN
  attaches_to:
  - progression#Adulthood and later life
  rationale: >-
    Adult and later-life cohorts, people with intellectual disability or
    limited speech, and people outside high-income settings remain
    underrepresented. Outcomes should extend beyond symptom scores and be
    selected with autistic people and families.
  evidence:
  - reference: PMID:41926193
    reference_title: "Interventions for autistic adults: A meta-analysis."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: >-
      Thirty-five studies were included (N = 1,631 total participants).
    explanation: >-
      The modest adult evidence base motivates broader, more representative
      lifespan research.
- discussion_id: gap_asd_causal_status_of_immune_and_microbiome_associations
  prompt: >-
    Do immune or microbiome changes causally modify defined symptoms or
    co-occurring conditions in reproducible ASD subgroups, or are they
    consequences of diet, medication, gastrointestinal disease, stress,
    sampling, and other confounding?
  kind: CONTROVERSY
  status: OPEN
  attaches_to:
  - pathophysiology#Immune and neuroinflammatory associations in studied groups
  - pathophysiology#Gut microbiome associations in studied groups
  rationale: >-
    Current reviews supply biological plausibility and group associations but
    not a general causal ASD pathway. Mechanistic trials require prespecified
    subgroups, appropriate non-autistic and gastrointestinal controls,
    longitudinal sampling, and clinically meaningful outcomes.
  evidence:
  - reference: PMID:41550027
    reference_title: "The gatekeepers breached: claudin dysregulation in psychiatric disorders."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      In autism and ADHD, altered tight junction protein profiles imply more subtle or context-dependent barrier dysfunction.
    explanation: The context-dependent wording illustrates the unresolved causal status.
- discussion_id: gap_asd_ibs_hpa_gri_brain_gut_coupling
  prompt: >-
    Does hypothalamic-pituitary-adrenal (HPA) axis dysregulation with impaired
    glucocorticoid-responsive immune (GRI) signaling in peripheral immune cells
    (monocytes, natural killer cells, B cells; reported core genes LRFN1,
    NUAK2, TMEM154, GAPT) causally couple the immune and
    microbiome associations reported in autism spectrum disorder to those
    reported in the frequently co-occurring irritable bowel syndrome, or is the
    shared GRI signature a common downstream stress response in both conditions?
  kind: KNOWLEDGE_GAP
  status: OPEN
  attaches_to:
  - pathophysiology#Immune and neuroinflammatory associations in studied groups
  - pathophysiology#Gut microbiome associations in studied groups
  - Irritable_Bowel_Syndrome:pathophysiology#Gut-Brain Axis Dysfunction
  - Irritable_Bowel_Syndrome:pathophysiology#Immune Activation and Mast Cell Degranulation
  rationale: >-
    Functional gastrointestinal symptoms co-occur with ASD, and a shared
    glucocorticoid-responsive immune transcriptional signature in peripheral
    blood mononuclear cells has been proposed as a molecular link between ASD
    and IBS. The available evidence, however, is an integrative computational
    reanalysis of existing transcriptomes (ssGSEA, WGCNA, machine-learning
    feature selection, deconvolution, Connectivity Map inference). It identifies
    candidate regulators and a correlated signature but establishes neither
    causal ordering nor a mechanism coupling peripheral GRI dysregulation to
    the group-level immune and microbiome associations recorded in this entry.
    This gap is narrower than
    gap_asd_causal_status_of_immune_and_microbiome_associations, which asks
    whether those associations are causal at all: it asks specifically whether
    HPA/GRI signaling is the substrate of the ASD-IBS co-occurrence rather than
    a bystander stress response shared by both. It deliberately makes no claim
    about an enteric-nervous-system causal chain, which the 2026-08-05
    publication-readiness review removed from this entry as unsupported.
  proposed_experiments:
  - experiment_id: exp_asd_ibs_gri_stress_challenge_multiomics
    name: Stress-challenge PBMC GRI multi-omics cohort in ASD with and without IBS
    description: >-
      Enroll ASD participants stratified by presence or absence of IBS-type
      gastrointestinal symptoms plus matched non-autistic and IBS-only controls;
      apply a standardized HPA-axis challenge (for example dexamethasone
      suppression or an acute psychosocial stressor with serial cortisol); and
      measure, before and after challenge, single-cell PBMC transcriptomes
      (monocytes, NK cells, B cells) for the reported GRI core genes LRFN1,
      NUAK2, TMEM154, and GAPT, circulating
      cytokines, intestinal-permeability biomarkers, and stool microbiome
      composition, relating each to prespecified social-communication and
      gastrointestinal symptom measures.
    experiment_type:
      preferred_term: longitudinal stress-challenge multi-omics cohort
    model_systems:
    - name: Human ASD/IBS stress-challenge cohort
      description: >-
        Patient cohort designed to temporally order HPA activation, peripheral
        GRI gene dysregulation, systemic immune signaling, gut-barrier and
        microbiome change, and behavioral readouts across the ASD-IBS
        co-occurrence.
      experimental_model_type: OTHER
      organism:
        preferred_term: human
        term:
          id: NCBITaxon:9606
          label: Homo sapiens
    decision_criterion: >-
      HPA/GRI signaling is supported as an upstream driver of the ASD-IBS
      co-occurrence if post-challenge GRI dysregulation in monocytes, NK cells,
      and B cells precedes and quantitatively mediates gut-barrier, microbiome,
      and immune change and tracks with gastrointestinal symptom severity; a
      shared-bystander interpretation is favored if GRI changes neither precede
      nor mediate those organ-level readouts.
    would_support:
    - pathophysiology#Immune and neuroinflammatory associations in studied groups
    - pathophysiology#Gut microbiome associations in studied groups
  evidence:
  - reference: PMID:42424309
    reference_title: "Dysregulated glucocorticoid-responsive immune genes in peripheral blood mononuclear cells as a shared molecular signature of autism spectrum disorder and irritable bowel syndrome."
    supports: SUPPORT
    evidence_source: COMPUTATIONAL
    snippet: "While dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis and impaired glucocorticoid-responsive immune (GRI) signaling are proposed links between these disorders, the precise molecular mechanisms remain poorly understood."
    explanation: >-
      The source study explicitly frames the HPA/GRI brain-gut link between ASD
      and IBS as proposed but mechanistically unresolved, defining the gap.
  - reference: PMID:42424309
    reference_title: "Dysregulated glucocorticoid-responsive immune genes in peripheral blood mononuclear cells as a shared molecular signature of autism spectrum disorder and irritable bowel syndrome."
    supports: SUPPORT
    evidence_source: COMPUTATIONAL
    snippet: "These findings define a shared GRI-associated molecular signature linking systemic stress adaptation to immune dysregulation along the brain-gut axis."
    explanation: >-
      A computational reanalysis of ASD and IBS PBMC transcriptomes reports a
      shared GRI signature along the brain-gut axis, motivating the causal
      question posed by this gap.
- discussion_id: gap_asd_heavy_metal_exposure_causal_status
  prompt: >-
    Are the reported associations between prenatal or early-childhood lead and
    mercury exposure and ASD causal, or do they reflect residual confounding,
    reverse causation, and heterogeneous exposure-assessment methods that also
    explain why arsenic and cadmium findings are inconsistent?
  kind: CONTROVERSY
  status: OPEN
  attaches_to:
  - environmental#Prenatal and early-childhood heavy metal exposure
  rationale: >-
    Existing studies differ in biomarker matrix (blood, hair, urine), exposure
    window, timing relative to diagnosis, ascertainment, and confounder
    adjustment, and most measure a single metal rather than the correlated
    mixtures children are actually exposed to. Lead and mercury show the most
    consistent signal and arsenic and cadmium the least, but the source review
    declines to conclude that heavy metals cause ASD and calls for large
    prospective cohorts. This item concerns whether the human association is
    real and causal at all; the separate question of whether animal-model
    mechanisms apply to human neurodevelopment is recorded in
    mismatch_asd_heavy_metal_model_mechanism.
  proposed_experiments:
  - experiment_id: exp_asd_prospective_metal_mixture_birth_cohort
    name: Prospective mother-child cohort with repeated metal biomarkers and mediation analysis
    description: >-
      Follow a large, diverse pregnancy cohort with repeated maternal and child
      metal biomarkers across defined prenatal and early postnatal windows,
      prespecified mixture models that separate correlated metals from each
      other and from co-exposures, blinded standardized ASD ascertainment, and
      genetic and socioeconomic confounder control. Sibling or
      negative-control-exposure comparisons and biomarker path analysis through
      oxidative-stress and inflammatory intermediates test whether any
      exposure-to-outcome association survives and is mediated.
    experiment_type:
      preferred_term: prospective pregnancy and birth cohort with mixture and mediation analysis
    model_systems:
    - name: Diverse prospective pregnancy and birth cohort
      description: >-
        Human mother-child cohort with repeated exposure biomarkers, blinded
        diagnostic ascertainment, and sibling comparisons for unmeasured
        familial confounding.
      experimental_model_type: OTHER
      organism:
        preferred_term: human
        term:
          id: NCBITaxon:9606
          label: Homo sapiens
    perturbations:
    - name: Measured prenatal and early-childhood metal exposure
      target: environmental#Prenatal and early-childhood heavy metal exposure
      exposure_term:
        preferred_term: exposure to heavy metal
        term:
          id: ECTO:9002163
          label: exposure to heavy metal
      effect: >-
        Observed rather than assigned exposure, measured repeatedly by window
        and modeled as a correlated mixture rather than one metal at a time.
    readouts:
    - name: Blinded ASD ascertainment and exposure-response gradient
      target: environmental#Prenatal and early-childhood heavy metal exposure
      interpretation: >-
        A window-specific exposure-response gradient that survives mixture,
        sibling, and negative-control analyses is the minimum needed to move
        this exposure beyond an observational association.
    decision_criterion: >-
      A causal contribution is supported if window-specific lead or mercury
      biomarkers predict ASD outcomes with exposure-response gradients that
      persist in mixture models, sibling comparisons, and negative-control
      analyses; a confounding interpretation is favored if associations
      attenuate to the null once familial and socioeconomic factors and
      correlated co-exposures are accounted for.
    would_support:
    - environmental#Prenatal and early-childhood heavy metal exposure
  evidence:
  - reference: PMID:42540380
    reference_title: "Association Between Heavy Metal Exposure and Autism Spectrum Disorders: Discrepancies, Research Gaps, and Future Priorities of Human Epidemiological Studies."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      The findings are inconsistent across metals and studies, and should be interpreted with caution due to potential residual confounding and heterogeneity in exposure assessment methods. Large prospective cohort studies are needed to clarify causal relationships.
    explanation: >-
      The source review states the unresolved causal status and the study design
      needed to resolve it, defining this controversy.
- discussion_id: mismatch_asd_heavy_metal_model_mechanism
  prompt: >-
    Do the oxidative-stress, synaptic, and glial-activation mechanisms
    characterized for lead and mercury in rodent, zebrafish, and cell models
    operate at human-relevant developmental exposure levels, and do they mediate
    autistic traits in people, or are they model-system toxicology that does not
    transfer to human ASD?
  kind: HUMAN_MODEL_MISMATCH
  status: OPEN
  attaches_to:
  - environmental#Prenatal and early-childhood heavy metal exposure
  - pathophysiology#Immune and neuroinflammatory associations in studied groups
  rationale: >-
    Metal neurotoxicity mechanisms are described in detail in experimental
    animals, including reactive oxygen species generation, impaired synaptic
    plasticity, microglial and astrocytic activation, and lipid and steroid
    metabolic disruption, and a zebrafish arm of one birth-cohort study
    nominates a PPAR-linked pathway. None of this has been demonstrated in human
    neurodevelopment, at the cord-blood concentrations measured in cohorts, or
    as a mediator of autistic traits, and reviews of developmental mercury
    toxicity state that no single process explains the observed effects. That is
    why this entry records the exposure as an environmental association and adds
    no heavy-metal pathophysiology node: promoting these model findings to a
    human ASD mechanism node would repeat the animal-only
    maternal-immune-activation claim that the 2026-08-05 publication-readiness
    review removed. The prior question of whether the human epidemiological
    association is causal at all is recorded in
    gap_asd_heavy_metal_exposure_causal_status.
  proposed_experiments:
  - experiment_id: exp_asd_human_cortical_organoid_metal_exposure
    name: Human cortical organoid and microglia-containing assembloid exposure at cohort-relevant metal concentrations
    description: >-
      Expose human iPSC-derived cortical organoids and microglia-containing
      assembloids, including lines carrying defined ASD-associated variants and
      isogenic controls, to lead and methylmercury across a dose range anchored
      to measured human cord-blood concentrations. Read out oxidative stress,
      microglial activation state, synaptic density and network activity, and
      transcriptomic and lipidomic changes including the PPAR pathway nominated
      by the zebrafish work, and test whether effects appear at human-relevant
      rather than only supraphysiological doses.
    experiment_type:
      preferred_term: dose-anchored human organoid exposure experiment
    model_systems:
    - name: Human iPSC-derived cortical organoid and microglia-containing assembloid panel
      description: >-
        Human cortical organoids with and without integrated microglia, spanning
        ASD-associated genotypes and isogenic controls, used to test whether
        animal-model metal neurotoxicity mechanisms are reproduced in human
        neural tissue at cohort-relevant exposure levels.
      experimental_model_type: ORGANOID
      organism:
        preferred_term: human
        term:
          id: NCBITaxon:9606
          label: Homo sapiens
      cell_types:
      - preferred_term: neural progenitor cell
        term:
          id: CL:0011020
          label: neural progenitor cell
      - preferred_term: microglial cell
        term:
          id: CL:0000129
          label: microglial cell
    perturbations:
    - name: Cohort-anchored lead exposure
      target: environmental#Prenatal and early-childhood heavy metal exposure
      exposure_term:
        preferred_term: exposure to lead
        term:
          id: ECTO:9000945
          label: exposure to lead
      effect: >-
        Applies lead across a dose range spanning measured human cord-blood
        concentrations, testing human-relevant rather than only high-dose effects.
    - name: Cohort-anchored mercury exposure
      target: environmental#Prenatal and early-childhood heavy metal exposure
      exposure_term:
        preferred_term: exposure to mercury
        term:
          id: ECTO:0001571
          label: exposure to mercury
      effect: >-
        Applies methylmercury across a comparably anchored dose range.
    readouts:
    - name: Oxidative stress and glial activation
      target: pathophysiology#Immune and neuroinflammatory associations in studied groups
      biological_processes:
      - preferred_term: response to oxidative stress
        term:
          id: GO:0006979
          label: response to oxidative stress
      - preferred_term: microglial cell activation
        term:
          id: GO:0001774
          label: microglial cell activation
      interpretation: >-
        Detects whether the oxidative and glial responses reported in animal
        models occur in human neural tissue at cohort-relevant doses.
    - name: Synaptic and network development
      target: pathophysiology#Excitation-inhibition and synaptic-signaling differences in studied subgroups
      biological_processes:
      - preferred_term: regulation of synaptic plasticity
        term:
          id: GO:0048167
          label: regulation of synaptic plasticity
      interpretation: >-
        Tests whether metal exposure reproduces the synaptic and network
        measures already curated for genotype-defined subgroups.
    decision_criterion: >-
      Translational relevance is supported if oxidative, glial, and synaptic
      effects appear at concentrations within the measured human cord-blood
      range and scale with dose; the model-system interpretation stands if
      effects require concentrations well above human exposure or fail to
      reproduce in human neural tissue.
    would_support:
    - pathophysiology#Immune and neuroinflammatory associations in studied groups
  evidence:
  - reference: PMID:42541535
    reference_title: "Molecular and cellular mechanisms of lead-induced neurotoxicity: comparative insights from rodent and zebrafish models."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      Once in the brain, lead induces oxidative stress through excessive reactive oxygen species generation, mitochondrial dysfunction, lipid peroxidation, DNA damage, and depletion of antioxidant defenses.
    explanation: >-
      Establishes that a detailed mechanism exists in animal models, which is
      the model side of the mismatch. The review covers rodent and zebrafish
      data and makes no human ASD claim.
  - reference: PMID:42541535
    reference_title: "Molecular and cellular mechanisms of lead-induced neurotoxicity: comparative insights from rodent and zebrafish models."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      In parallel, it activates both intrinsic and extrinsic apoptotic pathways and enhances neuroinflammatory signaling through microglial and astrocytic activation, further contributing to neuronal injury.
    explanation: >-
      Names the glial-activation route that would have to be demonstrated in
      human tissue before a heavy-metal neuroinflammation node could be curated.
  - reference: PMID:41110787
    reference_title: "Prenatal exposure to environmental metal mixtures and social development in early childhood: Evidence from a birth cohort and mechanistic study."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: >-
      Integrated omics analyses identified substantial disruption of lipid and steroid metabolic pathways, particularly involving the peroxisome proliferator-activated receptor (PPAR) signaling cascade.
    explanation: >-
      The candidate PPAR mechanism comes from the zebrafish arm of the study,
      not from the human cohort arm, so it is a nominated hypothesis awaiting
      human confirmation.
  - reference: PMID:26987277
    reference_title: "Methylmercury and brain development: A review of recent literature."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: >-
      The mechanism of toxicity, however, is not fully understood. No single process can explain the multitude of effects observed in MeHg-induced neurotoxicity.
    explanation: >-
      States directly that developmental methylmercury neurotoxicity has no
      single settled mechanism, so no specific mechanism can be curated as the
      route from exposure to ASD.
review_notes: |-
  Publication-readiness review (2026-08-05): the OpenScientist deep-research
  report and every publication cache cited by the prior entry were checked
  against the source text. The report is keyed to MONDO:0005260 rather than this
  entry's MONDO:0005258 and was used only as a lead list. Genetic pathway
  convergence and genotype-defined cellular findings were retained with scope
  limits. E/I, immune, blood-brain-barrier, and microbiome claims were downgraded
  to provisional or hypothetical subgroup findings. The asserted ENS causal
  chain, animal-only maternal-immune-activation environmental risk claim,
  Female Protective Effect pseudo-gene, unsupported universal phenotype
  frequencies, and preclinical CRISPRa "treatment" were removed.

  D2P audit dispositions (all 22 recommendations from the pre-edit audit):
  1. HP:0011024 gastrointestinal abnormality — REPLACE the overly broad source
     suggestion with HP:0012719 functional gastrointestinal abnormality, backed
     by a 2023 symptom meta-analysis (33%, 95% CI 13%-57%); intentionally leave
     unlinked because association does not prove a gut-to-core-ASD causal path.
  2. HP:0007018 ADHD — ADD with systematic-review evidence; intentionally leave
     unlinked as a separate co-occurring diagnosis.
  3. HP:0000733 motor stereotypy — RETAIN/ADD exact evidence as a variable
     restricted/repetitive manifestation; do not duplicate the parent-domain
     pathograph edge.
  4. HP:0001250 seizure — ADD with meta-analysis evidence and OCCASIONAL overall
     frequency (~1/10); do not adopt the OMIM "frequent" label across broad ASD
     and do not invent a universal ASD-to-epilepsy mechanism.
  5. HP:5200046 sensory behavioral abnormality — RETAIN with exact definitional
     evidence and remove the unsupported 70%-95% figure; leave unlinked because
     no general causal route is established.
  6. HP:0000758 abnormal nonverbal communicative behavior — ADD as a core
     diagnostic phenotype and link from the explicitly provisional distributed-
     circuit node with PARTIAL evidence.
  7. HP:0000717 Autism — DO NOT ADD; it restates the disease as its own
     phenotype and adds no atomic manifestation.
  8. HP:0000750 delayed speech/language — ADD as a variable associated feature,
     explicitly not required for diagnosis; leave unlinked rather than assert a
     universal language mechanism.
  9. HP:0002353 EEG abnormality — DO NOT ADD as a top-level ASD phenotype; EEG
     findings are heterogeneous group readouts or epilepsy-context findings,
     not a defining or universal manifestation.
  10. HP:0003144 increased serum serotonin — DO NOT ADD; hyperserotonemia is a
      nonspecific group association without validated individual diagnostic or
      mechanistic status.
  11. HP:0000732 inflexible adherence to routines — ADD with exact parent-domain
      evidence; do not duplicate the linked restricted/repetitive parent node.
  12. HP:0001249 intellectual disability — ADD as a co-occurring feature with a
      39.6% ADDM estimate among children with cognitive data; leave unlinked
      because it is neither required nor explained by one broad-ASD mechanism.
  13. HP:0000721 lack of spontaneous play — DO NOT ADD; this narrow historical
      presentation is not required and is not supported as a general phenotype
      by the reviewed contemporary sources.
  14. HP:0000723 restrictive behavior — ADD as the core restricted/repetitive
      domain and link from the provisional distributed-circuit node.
  15. HP:0011024 local-unlinked recommendation — REVIEWED; replace with the
      narrower HP:0012719 and retain unlinked for the causal reason in
      disposition 1.
  16. HP:0007018 local-unlinked recommendation — REVIEWED; retain unlinked for
      the co-occurring-diagnosis reason in disposition 2.
  17. HP:0000733 local-unlinked recommendation — REVIEWED; retain unlinked as a
      narrow child of the already linked restricted/repetitive domain.
  18. HP:0012760 reduced social responsiveness — REPLACE the overly broad local
      term with HP:0000758 plus source-exact HP:0000728, preserving atomic
      social-communication and relationship manifestations.
  19. HP:0001250 local-unlinked recommendation — REVIEWED; retain unlinked for
      the subgroup and causal-heterogeneity reason in disposition 4.
  20. HP:5200046 local-unlinked recommendation — REVIEWED; retain unlinked for
      the causal-evidence reason in disposition 5.
  21. HP:0002360 sleep disturbance — RETAIN with systematic-review evidence but
      remove the unsupported 50%-80% and immune-pineal causal generalization;
      leave unlinked as a heterogeneous co-occurring condition.
  22. HP:0000728 reduced ability to form peer relationships — ADD the source-
      exact term and evidence; leave it as an atomic subdomain rather than
      duplicate the parent social-domain edge.
datasets:
- accession: geo:GSE107878
  title: Disruption of Autism Spectrum Disorder-Susceptibility Genes Predominantly Reduces Functional Connectivity of Isogenic Human Neurons
  description: Autism Spectrum Disorder (ASD) is phenotypically and genetically heterogeneous, but genomic analyses have identified candidate susceptibility genes. We present a CRISPR gene editing strategy to insert a protein tag and premature termination sites creating an induced pluripotent stem cell (iPSC) knockout resource for functional studies of 10 ASD-relevant genes (AFF2/FMR2, ANOS1, ASTN2, ATRX, CACNA1C, CHD8, DLGAP2, KCNQ2, SCN2A, TENM1). Neurogenin 2 (NEUROG2)-directed differentiation of iPSCs allowed production of cortical excitatory neurons, and mutant proteins were not detectable.
  organism:
    preferred_term: human
    term:
      id: NCBITaxon:9606
      label: Homo sapiens
  data_type: BULK_RNA_SEQ
  sample_count: 86
  publication: PMID:30392976
  notes: Identified by GEO DataSets index search for Autism Spectrum Disorder (scripts/discover_datasets.py); accession and metadata verified against NCBI E-utilities on 2026-07-31. Title, sample count, and organism are GEO's own values.
- accession: geo:GSE328363
  title: Developmental dynamics of the cortical cellular and molecular landscapes in autism spectrum disorder models
  description: Recent research has identified over 100 causal genes in autism spectrum disorder (ASD), raising the question of how mutations in genes with diverse functions result in similar clinical presentations. Here, we profiled 251 samples from eleven monogenic ASD mouse models using single-nucleus multi-omic sequencing across three developmental stages, both sexes, and two brain regions. We discovered that, despite wide genetic heterogeneity, ASD-linked mutations converged on perturbations of the radial glial cell lineage. This converging alteration primarily reflects a transient developmental delay rather than a lasting lineage misspecification and resolves by postnatal stages.
  organism:
    preferred_term: mouse
    term:
      id: NCBITaxon:10090
      label: Mus musculus
  data_type: SINGLE_CELL_RNA_SEQ
  sample_count: 135
  publication: PMID:42310454
  notes: Identified by GEO DataSets index search for Autism Spectrum Disorder (scripts/discover_datasets.py); accession and metadata verified against NCBI E-utilities on 2026-07-31. Title, sample count, and organism are GEO's own values.
- accession: geo:GSE286067
  title: Autism spectrum disorder associated chromatin modifiers converge on transcription with sex-specific regulatory signatures
  description: We sought to understand the transcriptional disruptions and functional impacts of the loss of 9 autism spectrum disorder (ASD) risk genes that encode transcriptional regulators in neurons. In addition to understanding how these signature converge or diverge, we aimed to study how sex could modulate these consequences.
  organism:
    preferred_term: mouse
    term:
      id: NCBITaxon:10090
      label: Mus musculus
  data_type: BULK_RNA_SEQ
  sample_count: 66
  publication: PMID:40196547
  notes: Identified by GEO DataSets index search for Autism Spectrum Disorder (scripts/discover_datasets.py); accession and metadata verified against NCBI E-utilities on 2026-07-31. Title, sample count, and organism are GEO's own values.
- accession: ega:EGAS00001000556
  title: Whole genome sequencing of autism spectrum disorder
  description: Autism Spectrum Disorder (ASD) demonstrates high heritability and familial clustering, yet the genetic causes remain only partially understood as a result of extensive clinical and genomic heterogeneity. Whole-genome sequencing (WGS) shows promise as a tool for identifying ASD risk genes as well as unreported mutations in known loci, but an assessment of its full utility in an ASD group has not been performed. We used WGS to examine 32 families with ASD to detect de novo or rare inherited genetic variants predicted to be deleterious (loss-of-function and damaging missense mutations).
  organism:
    preferred_term: human
    term:
      id: NCBITaxon:9606
      label: Homo sapiens
  publication: PMID:23849776
  notes: 'European Genome-phenome Archive study, matched because the disease is named in the study''s own title ("Autism Spectrum Disorder"); description-level mentions were not accepted. EGA study_type: Other. Controlled access -- data require a Data Access Agreement. EGA metadata retrieved 2026-08-01.'
- accession: ega:EGAS00001000829
  title: Detection of Clinically Relevant Genetic Variants in Autism spectrum Disorder by Whole-Genome Sequencing
  description: Autism Spectrum Disorder (ASD) demonstrates high heritability and familial clustering, yet the genetic causes remain only partially understood as a result of extensive clinical and genomic heterogeneity. Whole-genome sequencing (WGS) shows promise as a tool for identifying ASD risk genes as well as unreported mutations in known loci, but an assessment of its full utility in an ASD group has not been performed. We used WGS to examine 32 families with ASD to detect de novo or rare inherited genetic variants predicted to be deleterious (loss-of-function and damaging missense mutations).
  organism:
    preferred_term: human
    term:
      id: NCBITaxon:9606
      label: Homo sapiens
  notes: 'European Genome-phenome Archive study, matched because the disease is named in the study''s own title ("Autism Spectrum Disorder"); description-level mentions were not accepted. EGA study_type: Other. Controlled access -- data require a Data Access Agreement. EGA metadata retrieved 2026-08-01.'
- accession: ega:EGAS00001001943
  title: Histone Acetylome-wide Association Study of Autism Spectrum Disorder
  description: 'H3K27ac ChIP-seq were performed on postmortem samples from autism spectrum disorder and matched control brains. Tissues were chosen from three brain regions: prefrontal cortex, temporal cortex and cerebellum.'
  organism:
    preferred_term: human
    term:
      id: NCBITaxon:9606
      label: Homo sapiens
  notes: 'European Genome-phenome Archive study, matched because the disease is named in the study''s own title ("Autism Spectrum Disorder"); description-level mentions were not accepted. EGA study_type: Other. Controlled access -- data require a Data Access Agreement. EGA metadata retrieved 2026-08-01.'
- accession: metabolomics_workbench:ST002332
  title: Plasma metabolomic profiling of individuals with autism spectrum disorder and their family members.
  notes: Located via OmicsDI, which aggregates across omics repositories; this record comes from metabolomics_workbench. Only repositories with no other discovery route in this project and with a working accession resolver are curated from OmicsDI -- GEO, ArrayExpress, PRIDE, MetaboLights and EGA hits are excluded as duplicates of dedicated passes. Matched because the disease is named in the dataset's own title ("Autism Spectrum Disorder"). Retrieved 2026-08-02.
- accession: massive:MSV000085232
  title: Metaproteomics investigation on the gut microbiota of children affected by Autism Spectrum Disorder
  description: Metaproteomics investigation of the Gut Microbiota in young subjects with Autism Spectrum Disorders and their relatives
  notes: Located via OmicsDI, which aggregates across omics repositories; this record comes from massive. Only repositories with no other discovery route in this project and with a working accession resolver are curated from OmicsDI -- GEO, ArrayExpress, PRIDE, MetaboLights and EGA hits are excluded as duplicates of dedicated passes. Matched because the disease is named in the dataset's own title ("Autism Spectrum Disorder"). Retrieved 2026-08-02.

📚

References & Deep Research

References

3

PMID:31949163

Autism spectrum disorder.

No top-level findings curated for this source.

PMID:31843864

Identification, Evaluation, and Management of Children With Autism Spectrum Disorder.

No top-level findings curated for this source.

PMID:40274203

Autism diagnosis in children and adolescents: A systematic review and meta-analysis of test accuracy.

No top-level findings curated for this source.

Deep Research

1

Executive Summary ▸

Executive Summary

Autism Spectrum Disorder (ASD) is a highly heritable (~80%), clinically heterogeneous neurodevelopmental condition now affecting approximately 1–2.8% of children globally, with a consistent male-to-female diagnostic ratio of approximately 3.6:1. This comprehensive characterization, synthesizing evidence from 79 peer-reviewed publications, reveals that ASD pathophysiology converges on three core mechanistic axes: (1) excitatory/inhibitory (E/I) neural imbalance driven by synaptic, chromatin remodeling, and transcriptional pathway disruptions; (2) neuroinflammation and immune dysregulation, including microglial activation, altered cytokine profiles, and blood–brain barrier compromise; and (3) gut–brain axis dysbiosis involving altered short-chain fatty acid and tryptophan metabolism. These biological processes are shaped by a complex interplay of hundreds of genetic risk variants and prenatal environmental exposures—particularly maternal immune activation (MIA)—operating through epigenetic mechanisms with sex-specific vulnerability patterns.

The most critical unmet clinical need in ASD is the absence of any approved pharmacotherapy targeting core symptoms of social communication deficit and restricted/repetitive behaviors. Current evidence-based interventions, principally Applied Behavior Analysis (ABA), produce medium-sized improvements in IQ (9–15 points) and adaptive behavior but do not significantly improve core ASD symptomatology. Emerging frontiers—including CRISPR-based gene activation for haploinsufficient genes, multimodal early biomarker detection (neonatal movement analysis, eye-tracking, neuroimaging), and microbiome-targeted therapies—offer the most transformative potential for future diagnosis and treatment. The disorder carries a substantial societal burden, with increased all-cause mortality (particularly from epilepsy, comorbidities, and injury), catastrophic family healthcare expenditures in low- and middle-income countries, and estimated annual societal costs of €28 billion in France and $74 billion in the United States.

Key Findings

Finding 1: ASD Genetic Architecture — De Novo Mutations Converge on Synaptic, Chromatin, and Transcriptional Pathways

The genetic architecture of ASD is characterized by a convergence of rare, high-impact de novo mutations onto a limited number of biological pathways. Landmark exome sequencing of 3,871 autism cases and 9,937 ancestry-matched controls implicated 22 autosomal genes at FDR < 0.05 and a broader set of 107 genes at FDR < 0.30, with de novo loss-of-function mutations present in over 5% of autistic subjects (PMID: 25363760). Targeted resequencing confirmed specific high-confidence loci including CHD8 (chromatin remodeling), SCN2A (sodium channel), DYRK1A (kinase signaling), CTNNB1 (Wnt/β-catenin signaling), SHANK3 and NRXN1 (synaptic scaffolding) (PMID: 24387789). These genes organize into three convergent functional pathways:

Pathway	Representative Genes	Function
Chromatin remodeling	CHD8, ARID1B, ASH1L	Epigenetic regulation of gene expression
Wnt signaling	CTNNB1, DYRK1A	Cell proliferation, neuronal differentiation
Synaptic function	SHANK3, NRXN1, SCN2A	Synaptic transmission and plasticity

An Indian cohort study (n=101 trios) corroborated these findings, with whole exome sequencing yielding a 30% diagnostic rate—predominantly de novo variants in synaptic formation, transcription regulation, and chromatin remodeling genes, with MECP2 as the most recurrently mutated gene (PMID: 37543562).

Finding 2: Neuroinflammation and Immune Dysregulation Are Core Pathophysiological Features

Multiple independent lines of evidence establish neuroinflammation as a central, not peripheral, feature of ASD pathophysiology. Postmortem studies consistently reveal microglial and astroglial activation, while peripheral biomarker studies demonstrate altered cytokine profiles including elevated IL-1β, IL-6, and TNF-α, alongside markers of oxidative stress such as glutathione imbalance and lipid peroxidation (PMID: 41947852). This neuroinflammatory state has functional consequences: ASD individuals show higher nocturnal salivary TNF levels, with sleep breathing dysfunction positively correlated with TNF (r = 0.42, P < 0.01) and inversely correlated with melatonin metabolite aMT6s (r = −0.31, P < 0.05) (PMID: 33421193).

Blood–brain barrier integrity is also compromised. Claudin-5, the predominant tight junction protein of the BBB, shows altered expression in ASD, potentially permitting peripheral inflammatory mediators to access brain tissue and sustain a pro-inflammatory cycle (PMID: 41550027). A study of aggressive behavior in ASD males (n=42) found elevated plasma TNF-α, IL-6, IL-8, IL-13, IFN-γ, vasopressin, and EGF in the aggressive subgroup, with spatial transcriptomics revealing pro-inflammatory gene overexpression in fronto-limbic regions involved in emotional regulation (PMID: 40721173).

Finding 3: Gut–Brain Axis Dysregulation as an Emerging Mechanism

Gastrointestinal symptoms affect 40–70% of ASD individuals, and an accelerating body of research (1,391 articles published 1999–2024) now positions the gut–brain axis as a mechanistic contributor to ASD rather than a mere epiphenomenon. Microbial metabolites—including short-chain fatty acids (SCFAs), tryptophan metabolites, and neurotransmitter precursors—directly influence brain development and behavior, with ASD-associated dysbiosis impacting neuroinflammatory processes (PMID: 39733842). The enteric nervous system (ENS) itself is now recognized as an active driver: disruptions in ENS neurotransmission, gut microbiota balance, and local neuroinflammation contribute to disease pathogenesis across neurodevelopmental disorders (PMID: 40088964).

Emerging therapeutic approaches targeting this axis include fecal microbiota transplantation, probiotics, and dietary modifications, though clinical benefit remains variable and standardization of protocols is needed.

Finding 4: Prevalence, Sex Ratio, and the Female Protective Effect

ASD prevalence estimates vary by geography and methodology but consistently demonstrate rising rates and pronounced male bias:

Population	Prevalence	Male:Female Ratio	Source
China (meta-analysis, 21 studies)	0.7% (95% CI: 0.006–0.008)	OR = 3.198 (95% CI: 2.489–4.109)	PMID: 38811881
Istanbul (n=25,839 screened)	0.9%	3.6:1	PMID: 39049996
United States (CDC, 2023)	~2.78% (1 in 36)	~4:1	CDC surveillance

The male bias is partially explained by the female protective effect: females diagnosed with ASD carry a significantly higher burden of putative functional de novo mutations (loss-of-function and predicted deleterious missense) than males, indicating that females require a higher genetic load to reach the diagnostic threshold (PMID: 32066658). Mechanistically, ASD candidate genes are significantly more frequently co-expressed in female brains than in male brains, suggesting greater compensation capacity. Gene prioritization identified 60 shared, 91 male-specific, and 23 female-specific candidate genes, reinforcing the concept of sex-differential genetic architecture.

Finding 5: Excitatory/Inhibitory Imbalance as Central Pathophysiology

The E/I imbalance hypothesis is supported by convergent evidence across multiple model systems:

mGluR5 dysfunction: 25-fold peripheral GRM5 downregulation in ASD patients, with specific variants (rs905646, rs762724) showing biased paternal transmission and association with increased symptom severity (PMID: 41653294).
15q11-13 duplication model: Facilitated LTP of glutamate synapses onto layer 5 pyramidal neurons due to decreased inhibitory synapses and altered serotonergic modulation of fast-spiking interneurons (PMID: 31901366).
TSC2-mutant iPSC neurons: Neuronal hyperactivity, reduced network synchronization, and elevated expression of GABA and glutamate signaling genes (PMID: 33076974).
Gliotransmission: Astrocytic release of glutamate, D-serine, and GABA dysregulates tonic E/I balance, contributing to sensory, cognitive, and social impairments (PMID: 40122634).

Finding 6: ABA Interventions Show Medium Effects on IQ but Not Core Symptoms

The most rigorously evaluated behavioral interventions for ASD are those based on Applied Behavior Analysis. A meta-analysis of 11 RCTs (n=632) found:

Outcome	SMD (95% CI)	Significance
Intellectual functioning	0.51 (0.09–0.92)	Significant
Adaptive behavior	0.37 (0.03–0.70)	Significant
Language abilities	—	Not significant vs. controls
Symptom severity	—	Not significant vs. controls
Parental stress	—	Not significant vs. controls

(PMID: 36864429)

A broader narrative review confirmed IQ gains of 9–15 points with early intensive behavioral interventions (EIBIs) and naturalistic developmental behavioral interventions (NDBIs), but effects on core autism symptoms were described as "more variable" (PMID: 41080225). High-intensity interventions showed notably greater effects on language skills (SMD = 0.72) compared to low-intensity (SMD = 0.34) (PMID: 41454358). This evidence gap—robust improvement in cognitive/adaptive domains but not in core social-communication deficits or repetitive behaviors—represents the most critical unmet therapeutic need.

Finding 7: Emerging Biomarkers Enable Detection as Early as the Neonatal Period

A multimodal portfolio of early biomarkers is advancing toward clinical translation:

Modality	Finding	Age	Source
Neonatal movement	Sleep-state spontaneous movement features predict ASD risk at 18 months	Neonatal	PMID: 37620366
Eye-tracking	ASD toddlers (n=57) show fewer/shorter fixations on eyes/mouth vs. TD	Toddler	PMID: 37410255
Brain MRI	Smaller nucleus accumbens, larger ventricles in pre-diagnostic ASD (n=81)	<3 years	PMID: 34455432
Mobile app gaze	Computer vision on smartphone distinguished 40 ASD from TD toddlers (AUC=0.90)	Toddler	PMID: 33900383
Hair cortisol	HCC inversely associated with ASD trait severity and ADHD comorbidity	2–17 years	PMID: 41610559

These converging biomarker modalities suggest that scalable, objective early screening tools are within reach, potentially reducing the current diagnostic delay that defers intervention past the critical neurodevelopmental window.

Finding 8: High Comorbidity Burden

ASD is rarely an isolated condition. Convergent prevalence estimates indicate:

Comorbidity	Estimated Prevalence
Sensory processing issues	70–95%
Sleep disturbances	50–80%
GI symptoms	40–70%
Anxiety	30–50%
ADHD	30–60%
Intellectual disability	25–40%
Epilepsy	10–30%
Depression	15–40%

A Japanese pediatric claims database (n=21,145 hypnotic prescriptions) confirmed ASD as the most common comorbidity (32.5%), followed by depression (23.4%), ADHD (19.8%), and anxiety (18.2%) (PMID: 41800554). Mechanistic links between ASD genetics and anxiety comorbidity have been demonstrated: Neuroligin-3 R451C knock-in mice exhibit heightened anxiety susceptibility through CCK upregulation in medial prefrontal cortex (PMID: 41699722). Higher behavioral comorbidity—particularly attention and thought problems—is strongly associated with poorer social functioning in ASD children (n=225) (PMID: 41604128).

Finding 9: Maternal Immune Activation as Key Environmental Risk Factor

Three distinct MIA pathways have been characterized in preclinical models, all converging on ASD-relevant neurodevelopmental disruption:

G-CSF pathway: Poly(I:C) MIA increases G-CSF in maternal plasma and embryonic tissue, causing increased dendritic spine density with immature spines in mPFC and altered social preference (PMID: 41825653).
Kynurenine pathway: MIA activates IDO enzyme, increasing kynurenine metabolism in fetal tissue and reducing NMDA receptor subunit expression selectively in male fetal brains (IDO-knockout prevents this effect) (PMID: 41577052).
IFN-α pathway: Prenatal IFN-α exposure causes reductions in GABA, 5-HIAA, and GAD-67, neuronal loss in hippocampal and cerebellar regions, elevated TNF-α, and reduced sociability—with heightened male vulnerability (PMID: 41484215).

These findings support a three-hit model of ASD vulnerability: genetic predisposition × biological sex × environmental insult, with the prenatal period as a critical window mediated by epigenetic mechanisms including DNA methylation, histone modifications, and non-coding RNA regulation (PMID: 41898431).

Finding 10: Gene Editing and Precision Medicine as Emerging Frontiers

No FDA-approved drugs target core ASD symptoms. Current treatments remain primarily symptomatic. The most promising emerging approaches include:

CRISPRa gene activation for haploinsufficient NDD genes, which could restore expression of ASD risk genes carrying loss-of-function variants (PMID: 41278953).
Precision biomedical treatments guided by pre-treatment biomarkers: folinic acid for patients with folate receptor autoantibodies, methylcobalamin for impaired methylation, mitochondrial cofactors for mitochondrial dysfunction (PMID: 31801452).
Microbiome-targeted therapies including fecal microbiota transplantation and psychobiotics (PMID: 40076598).

Finding 11: Increased Mortality and Substantial Economic Burden

A systematic review of 15 studies (n=216,045) found significantly elevated all-cause mortality in autistic individuals, with key causes being epilepsy, medical comorbidities, and injury—highest risk in those with co-occurring intellectual disability (PMID: 37042154). Suicide risk is also elevated, with autistic college students showing OR = 2.06 for suicidal ideation and OR = 2.39 for attempts (PMID: 39382895).

The economic burden is profound:

Region	Annual Cost	Details
France	~€28 billion/year	All NDDs combined (PMID: 39956665)
United States	~$74 billion/year	ASD-related costs
Sweden	~€50,000/year per child	Additional societal cost; parents spend ~1,000 extra hours/year caregiving (PMID: 17942458)
India	71.25% of families exceed catastrophic expenditure	>10% monthly income on healthcare (PMID: 41841515)

Mechanistic Model

The evidence synthesized across 12 findings supports an integrated mechanistic model of ASD:

GENETIC SUSCEPTIBILITY                    ENVIRONMENTAL EXPOSURES
(De novo mutations in                     (Maternal immune activation,
 synaptic/chromatin/Wnt genes;             infections, toxicants,
 common polygenic risk;                    epigenetic insults)
 ~80% heritability)                              |
 |                                       |
 v                                       v
    ┌─────────────────────────────────────────────────┐
    │          PRENATAL NEURODEVELOPMENT               │
    │  Epigenetic reprogramming (DNA methylation,      │
    │  histone mods, ncRNA) → placental mediation      │
    │  → MODULATED BY FETAL SEX                        │
    │  (Female protective effect: higher co-expression │
    │   compensation in female brains)                 │
    └──────────────────────┬──────────────────────────┘
           │
           v
    ┌─────────────────────────────────────────────────┐
    │         THREE CORE PATHOPHYSIOLOGICAL AXES       │
    │                                                  │
    │  1. E/I IMBALANCE                                │
    │     - mGluR5 dysfunction                         │
    │     - Reduced inhibitory synapses                │
    │     - Altered gliotransmission                   │
    │                                                  │
    │  2. NEUROINFLAMMATION                            │
    │     - Microglial/astroglial activation           │
    │     - Elevated IL-1β, IL-6, TNF-α               │
    │     - BBB claudin dysregulation                  │
    │     - Oxidative stress                           │
    │                                                  │
    │  3. GUT-BRAIN AXIS DYSBIOSIS                     │
    │     - Altered SCFAs, tryptophan metabolism        │
    │     - ENS dysfunction                            │
    │     - Disrupted gut barrier                      │
    └──────────────────────┬──────────────────────────┘
           │
           v
    ┌─────────────────────────────────────────────────┐
    │              CLINICAL PHENOTYPE                   │
    │  Core: Social communication deficits,            │
    │        restricted/repetitive behaviors           │
    │  Comorbid: Sensory (70-95%), Sleep (50-80%),     │
    │           GI (40-70%), Anxiety (30-50%),         │
    │           ADHD (30-60%), Epilepsy (10-30%)       │
    │  Outcomes: Increased mortality, economic burden   │
    └─────────────────────────────────────────────────┘

Key mechanistic insights:

Bidirectional reinforcement: Neuroinflammation and E/I imbalance are not independent—TNF-α and IL-6 directly modulate synaptic function, while E/I imbalance can trigger inflammatory cascades. Gut dysbiosis feeds into both through SCFA-mediated immune modulation and tryptophan-serotonin pathway disruption.
Sex as a biological variable: The female protective effect operates at the genetic (higher mutation burden required), transcriptomic (greater co-expression compensation), and immunological (sex-specific HLA allele patterns) levels. MIA effects show consistent male vulnerability across G-CSF, kynurenine, and IFN-α pathways.
Critical developmental windows: Prenatal MIA at specific gestational timepoints produces distinct synaptic, neurochemical, and behavioral outcomes, suggesting that timing of environmental insult interacts with the maturational state of developing neural circuits.

Evidence Base

This characterization draws on 79 peer-reviewed publications. The most critical evidence supporting each mechanistic domain is summarized below:

Genetic Architecture

PMID: 25363760 — Landmark exome sequencing study (3,871 cases) establishing convergence on synaptic, chromatin, and transcriptional pathways
PMID: 24387789 — Confirmation of CHD8, SCN2A, DYRK1A, CTNNB1 as high-confidence ASD genes
PMID: 37543562 — Indian cohort validating de novo variant architecture and WES diagnostic yield

Neuroinflammation

PMID: 41947852 — Comprehensive review linking microglial activation and cytokine alterations to functional connectivity changes
PMID: 33421193 — Direct evidence of TNF–melatonin–sleep disruption axis in ASD
PMID: 41550027 — Claudin-5 dysregulation and BBB permeability in psychiatric disorders including ASD

E/I Imbalance

PMID: 41653294 — Multi-level analysis of mGluR5 dysfunction in ASD
PMID: 33076974 — TSC2-mutant iPSC neurons demonstrating network synchronization deficits
PMID: 31901366 — 15q11-13 duplication model showing serotonergic modulation of E/I balance in PFC

Environmental Risk and Epigenetics

PMID: 41825653 — G-CSF as novel MIA mediator
PMID: 41577052 — Kynurenine pathway mediating sex-specific MIA effects on NMDA receptors
PMID: 41484215 — IFN-α prenatal exposure producing ASD-like changes with male vulnerability
PMID: 41898431 — Three-hit model: gene × sex × environment interaction in ASD mouse models

Sex Differences and Female Protective Effect

PMID: 32066658 — Genetic evidence for higher de novo mutation burden in ASD females and greater gene co-expression compensation in female brains
PMID: 41480043 — Systematic review of sex differences in additive genetic variants in autism
PMID: 39337366 — Sex-related HLA allele risk/protective patterns in Italian ASD cohort

Treatment and Intervention

PMID: 36864429 — Definitive meta-analysis of ABA interventions (11 RCTs, n=632)
PMID: 41080225 — Narrative review of EIBI and NDBI effects on IQ and core symptoms
PMID: 41454358 — Mixed-methods systematic review showing intensity-dependent language improvements
PMID: 41278953 — CRISPRa as emerging gene therapy for haploinsufficient NDD genes
PMID: 31801452 — Targeted biomedical treatments guided by biomarker-defined ASD subgroups

Biomarkers

PMID: 37620366 — Neonatal movement patterns predicting ASD risk
PMID: 37410255 — Eye-tracking gaze patterns as early diagnostic biomarker
PMID: 34455432 — Pre-diagnostic neuroimaging biomarkers in infants
PMID: 33900383 — Mobile app computer vision gaze tracking (AUC=0.90)

Epidemiology and Burden

PMID: 38811881 — China meta-analysis quantifying prevalence and sex ratio
PMID: 39049996 — Istanbul community-based screening confirming 3.6:1 sex ratio
PMID: 41841515 — Catastrophic financial burden in Indian families
PMID: 17942458 — Per-child societal cost quantification in Sweden
PMID: 37042154 — Systematic review of mortality in autistic individuals

Limitations and Knowledge Gaps

Genetic heterogeneity: Despite identification of >100 high-confidence risk genes, the majority of ASD genetic risk remains unexplained. Common variant contributions are poorly characterized, and gene–gene interactions are largely unexplored.
Biomarker validation: While multiple early biomarker modalities show promise, none have been validated in large, prospective, population-level screening studies with sufficient sensitivity and specificity for clinical deployment.
Treatment evidence quality: The evidence base for ABA and other interventions is limited by small sample sizes, high risk of bias, variable outcome measures, and short follow-up periods. Most RCTs are graded as low to very low quality of evidence.
Gut–brain axis causality: The majority of gut microbiome studies in ASD are cross-sectional and correlational. Longitudinal studies establishing directionality and interventional trials demonstrating symptom modification through microbiome manipulation are lacking.
Sex-specific research gaps: Most ASD research cohorts are heavily male-dominated. The female phenotype, diagnostic criteria sensitivity for females, and female-specific genetic architecture remain understudied, as highlighted by systematic reviews finding inconclusive evidence in sex-stratified analyses (PMID: 41480043).
Translational gap: Preclinical MIA and genetic models, while informative, may not fully recapitulate the polygenic, multi-hit nature of human ASD. The pathway from mechanistic insight to therapeutic target remains long and uncertain.
Geographic representation: Prevalence data, genetic studies, and intervention trials are predominantly from high-income countries. ASD characterization in low- and middle-income settings is limited, despite evidence of catastrophic economic burden.

Proposed Follow-up Experiments and Actions

Near-term (1–3 years)

Prospective biomarker validation study: Combine neonatal movement analysis, eye-tracking, and structural MRI in a large birth cohort (n > 5,000) to develop a composite early detection algorithm with target sensitivity > 80% and specificity > 90%.
Sex-stratified GWAS mega-analysis: Pool existing ASD GWAS datasets with enforced female enrichment to power the detection of female-specific common variant associations and refine the female protective effect model.
Longitudinal microbiome study: Follow infants at high familial risk (n > 500) from birth through age 5 with serial stool microbiome, metabolomics, and behavioral assessments to establish temporal relationships between gut dysbiosis and ASD symptom emergence.
Core symptom intervention trials: Design adequately powered RCTs specifically targeting social communication outcomes (not IQ or adaptive behavior as primary endpoints) for existing and novel interventions, including oxytocin, bumetanide, and microbiome-targeted approaches.

Medium-term (3–7 years)

CRISPRa preclinical pipeline: Advance CRISPRa approaches for the top 10 haploinsufficient ASD genes (CHD8, SHANK3, SCN2A, etc.) through systematic in vitro and in vivo studies assessing efficacy, specificity, and safety.
Precision medicine framework: Develop a clinical decision algorithm that matches biomarker profiles (folate receptor antibodies, mitochondrial markers, inflammatory panels) to targeted treatments, and evaluate in a pragmatic clinical trial.
Neuroinflammation-targeted therapeutics: Test anti-inflammatory agents (selective cytokine inhibitors, microglial modulators) in ASD subgroups with documented elevated inflammatory biomarkers, measuring both biological and behavioral outcomes.

Long-term (7+ years)

Gene therapy clinical trials: Translate the most promising CRISPRa or antisense oligonucleotide approaches into Phase I/II trials for monogenic or oligogenic ASD subtypes with clear loss-of-function mechanisms.
Global ASD burden study: Conduct population-based prevalence and economic burden studies across diverse LMIC settings to inform global health policy and resource allocation.

Report generated from systematic analysis of 79 publications across 5 investigation iterations, encompassing ASD genetics, pathophysiology, epidemiology, treatment, and societal burden.

Ask OpenScientist

OpenScientist Report

Classifications

Definitions

Mechanistic Hypotheses

Discussions and Knowledge Gaps

Pathophysiology

Pathograph

Phenotypes

Genetic Associations

Medical Actions

Environmental Factors

Diagnosis

Progression

Prevalence

Differential Diagnoses

Related Datasets

Clinical Trials

Experimental Models

Animal Models

Source YAML

References & Deep Research

References

Deep Research

Executive Summary

Key Findings

Finding 1: ASD Genetic Architecture — De Novo Mutations Converge on Synaptic, Chromatin, and Transcriptional Pathways

Finding 2: Neuroinflammation and Immune Dysregulation Are Core Pathophysiological Features

Finding 3: Gut–Brain Axis Dysregulation as an Emerging Mechanism

Finding 4: Prevalence, Sex Ratio, and the Female Protective Effect

Finding 5: Excitatory/Inhibitory Imbalance as Central Pathophysiology

Finding 6: ABA Interventions Show Medium Effects on IQ but Not Core Symptoms

Finding 7: Emerging Biomarkers Enable Detection as Early as the Neonatal Period

Finding 8: High Comorbidity Burden

Finding 9: Maternal Immune Activation as Key Environmental Risk Factor

Finding 10: Gene Editing and Precision Medicine as Emerging Frontiers

Finding 11: Increased Mortality and Substantial Economic Burden

Mechanistic Model

Evidence Base

Genetic Architecture

Neuroinflammation

E/I Imbalance

Environmental Risk and Epigenetics

Sex Differences and Female Protective Effect

Treatment and Intervention

Biomarkers

Epidemiology and Burden

Limitations and Knowledge Gaps

Proposed Follow-up Experiments and Actions

Near-term (1–3 years)

Medium-term (3–7 years)

Long-term (7+ years)