Evaluating the Potential Use of Serotonergic Psychedelics in Autism Spectrum Disorder

The review organises existing clinical and preclinical findings across behavioural, neurobiological and historical clinical-trial domains. Behaviourally, recent studies in neurotypical adults indicate that serotonergic psychedelics can produce prosocial and anxiolytic effects. Examples cited include two psilocybin therapy sessions that increased extraversion and openness for up to 3 months, single-dose LSD rising sociability and desire for social contact, and LSD acutely increasing emotional empathy and circulating oxytocin. Preclinical rodent work reported both acute and repeated LSD-enhanced social behaviour. With respect to mood and anxiety, LSD and psilocybin have reduced symptoms in patients with life-threatening illness; the anxiolytic effects of LSD were noticeable after two therapy sessions and reported to last up to 12 months in one study, while psilocybin effects have been observed up to 6 months after a single administration. The authors also note a recent double-blind trial in which psilocybin's antidepressant effect was not statistically different from escitalopram, highlighting limits in the current evidence base. On synaptic function, the review summarises genetic and cellular findings linking ASD to synaptic pathology: mutations or epigenetic alterations in genes such as SHANK3, CNTNAP2 and FMR1 produce dendritic and synaptic deficits in animal models, and induced pluripotent stem cell-derived neurons from people with ASD show reduced excitatory synaptic activity and impaired NMDA receptor function. In this context, psychedelics have been reported to enhance synaptic plasticity: a single psilocybin dose produced AMPA receptor-mediated synaptic strengthening in mouse hippocampal slices, and LSD and DMT promoted dendritic arbor growth, spine formation and synaptogenesis in neuronal cultures and invertebrate models. Those plasticity effects were attributed to signalling through mTOR and 5-HT2A pathways. However, the authors caution that dysregulated mTOR signalling (for example hyperactive mTOR in some ASD samples) and reduced 5-HT2A expression or binding in individuals with ASD may attenuate or alter these synaptic effects. Regarding serotonergic signalling, the review reports that elevated blood serotonin is a replicated peripheral finding in ASD, with a meta-analysis indicating 28.3% of individuals with ASD show elevated 5-HT levels. Brain serotonin measures—mostly proxy markers such as PET binding or cerebrospinal fluid metabolites—generally point to lower central serotonergic indices in ASD, and reduced 5-HT1A and 5-HT2A receptor binding has been reported in limbic and neocortical regions. Because psychedelics are 5-HT1A/5-HT2A agonists and can increase brain serotonin, they could theoretically restore altered signalling; yet trials of serotonin-releasing agents such as fenfluramine produced at best modest benefits, suggesting that simply increasing serotonin is unlikely to be sufficient. The prefrontal cortex (PFC), and especially the medial PFC (mPFC), is reviewed as a key node in social cognition that is both altered in ASD and strongly modulated by 5-HT2A agonists. Animal experiments cited by the authors found that photoinhibition of mPFC excitatory neurons reduces sociability and blocks LSD's prosocial effects, consistent with human imaging results showing that psilocybin and LSD alter mPFC activity in ways correlated with subjective and social outcomes. The thalamocortical system is discussed next: large imaging studies in ASD show two partly consistent trends—hyperconnectivity between thalamus and sensorimotor cortex and hypoconnectivity between thalamus and multimodal association cortices. Psychedelic effects on thalamocortical connectivity are heterogeneous in the literature, with some reports of general 5-HT2A-mediated increases in thalamocortical coupling after LSD and others showing region-specific increases to sensory cortices and decreases to associative regions. The authors also note their own observation that LSD increases firing in the mediodorsal thalamic nucleus (MDT), a structure with strong mPFC connections and a role in sociability. Finally, the review summarises historical clinical trials in children carried out prior to Schedule I controls. Most trials used LSD at medium to high doses (25–400 µg), administered from single doses to daily dosing for up to 18 months. The reviewers report that greater improvements in mood, sociability, affectionate behaviour, speech, eye contact, playfulness and reduced repetitive or aggressive behaviours were observed with extended dosing regimens and when therapists engaged actively in psychotherapeutic interactions. Behaviour tended to regress after drug discontinuation, though not always to the previous baseline. Adverse events documented in those early studies included rapid mood swings, ataxia, moderate-to-severe anxiety, panic-like states, seizures (at least one reported), increased biting and pinching, persistent aggression in some cases, sleep disturbance, and occurrences of anxious, aggressive or self-harming behaviour following combined LSD/psilocybin. The authors emphasise that these early reports had major methodological and diagnostic limitations—subjects were often classified as "autistic schizophrenic" or "severely emotionally disturbed," which does not correspond to contemporary DSM-V ASD criteria—reducing the reliability and generalisability of the findings.

Markopoulos and colleagues interpret the assembled evidence as indicating a plausible, but unproven, therapeutic rationale for serotonergic psychedelics in some ASD-related problems. They propose that the compounds' capacity to enhance synaptic plasticity, modulate serotonergic signalling, alter mPFC function and influence thalamocortical dynamics could underlie observed prosocial and anxiolytic effects in non-ASD populations and in preclinical models. At the same time, the authors repeatedly flag reasons why those mechanisms might not translate straightforwardly to people with ASD: ASD is neurobiologically heterogeneous, some individuals show reduced 5-HT2A receptor expression or altered mTOR signalling, and structural or functional thalamocortical alterations may change circuit responses to psychedelics. The review highlights substantial limitations in the evidence base. There is a paucity of systematic, double-blind, placebo-controlled clinical trials of classical psychedelics in either neurotypical individuals or people with ASD. Early paediatric trials from the mid-twentieth century suffer from diagnostic ambiguity, ethical and methodological shortcomings, and variable dosing regimens, making their results difficult to interpret by modern standards. Safety concerns are central to the authors' discussion: documented adverse effects include anxiety, aggression, seizures and, given the elevated comorbidity of psychotic disorders in some people with ASD, a potential risk of precipitating psychosis in vulnerable individuals. The practical and ethical challenges of conducting trials in children and in people with intellectual disability—particularly obtaining informed consent—are emphasised. In terms of implications, the authors call for cautious, rigorously designed contemporary research rather than clinical application at present. They recommend targeting trials to well-characterised subgroups within the ASD spectrum, identifying optimal dosing and psychotherapy support models that mitigate risks, and assessing whether the net risk-to-benefit ratio justifies further development. The review also suggests that mechanistic studies that better define how psychedelics interact with ASD-specific neurobiology (for example 5-HT2A and mTOR status) would be important before large-scale clinical testing.

The authors conclude that, given the limited treatment options for core social and communicative features of ASD, serotonergic psychedelics warrant further investigation but only with caution. Clinical and preclinical data suggest these compounds can enhance social behaviour and reduce co-occurring anxiety and depression by acting on synaptic plasticity, serotonin signalling, prefrontal circuits and thalamocortical networks. Nevertheless, the historical record of adverse effects, together with ASD heterogeneity and potential biological limitations (such as altered 5-HT2A expression and mTOR signalling), mean that only some subsets of individuals with ASD might benefit. Markopoulos and colleagues stress that future studies must proceed under contemporary scientific and ethical standards, rigorously assess risk versus benefit, and address consent challenges in paediatric and intellectually disabled populations before validating the early positive findings from "first wave" research.