Psychedelic Research Recap: March 2026

March was busy and unusually wide-ranging. Clinical studies again dominated, especially in depression, PTSD, OCD and smoking cessation, but the month also brought useful work on biomarkers, brain mechanisms, therapeutic context, implementation, and safety. Across the papers, a familiar pattern held: the acute experience still matters, but so do dose, setting, support, and the practical realities of care. Several papers also pushed the field to look harder at what is actually being measured, from blinding and expectancy to subjective effects, biomarker change, and the lived experience of psychedelics versus psychosis.

Clinical Trials In Real Patients

This section brings together the papers closest to patient care. It starts with larger and more direct trials, then moves into real-world cohorts, biomarker work, and post hoc analyses from individual treatment studies. Read together, these papers show where the clinical signal is strongest and where the evidence still depends on small samples, open-label designs, or secondary analyses.

In treatment-resistant depression, the EPISODE Trial found a clear split between its endpoints. The primary binary response outcome at week 6 was not significant, but continuous measures of depression improved with psilocybin (25mg) plus psychotherapy, and the trial also reported acute safety signals, including suicidal ideation on dosing days and one case of persisting perceptual disturbance. It is a careful, clinically important paper because it shows both promise and the limits of overconfident interpretation.

The GH001 trial was more straightforwardly positive. In adults with treatment-resistant depression, GH001 produced a large reduction in MADRS scores by day 8, with remission in over half of those treated and no severe or serious adverse events during the placebo-controlled period. The open-label extension suggests that the benefit may persist with occasional retreatment, though the unblinding problem (see below) remains hard to ignore.

For smoking cessation, a pilot randomised trial found that a single dose of psilocybin combined with cognitive behavioural therapy outperformed nicotine patches plus the same therapy at 6 months. The abstinence signal was strong, adverse events were not serious, and the authors framed the result as a promising step towards a more durable cessation treatment.

In a small randomised trial in obsessive-compulsive disorder (OCD), repeated weekly low- and high-dose psilocybin administrations were generally safe and were linked with meaningful symptom reduction over eight weeks, with benefits still visible at six months. The study was small, but it adds to the evidence that psilocybin may have value beyond depression, while also reminding us that dose, cumulative exposure and recent SSRI discontinuation may shape response.

A proof-of-concept trial of intranasal 5-MeO-DMT given alongside stable SSRI treatment found that the combination was generally well tolerated and associated with rapid, sustained symptom improvement over 12 weeks. The sample was very small, but the paper is notable because it directly addresses a question the field has often avoided: whether classic (serotonergic) psychedelics can be studied without stopping antidepressants.

A biomarker analysis nested within the low-dose oral ketamine treatment for PTSD trial found small decreases in serum BDNF and VEGF-A, with the two measures positively correlated across timepoints. The clinical picture was mixed but interesting: inflammatory subgroups were stable, cytokine changes were not consistent, and several biomarker-symptom links survived only partially after correction, making the findings feel more suggestive than definitive.

Another real-world inpatient study of intranasal esketamine found meaningful improvement in depression severity and suicidal ideation, with older age and extended induction appearing particularly helpful for some patients. The work is pragmatic rather than polished, but that is also its strength: it shows what treatment looks like in routine care, not only in trials.

In a retrospective naturalistic study of subcutaneous esketamine, depression scores fell steadily over six weeks. Benzodiazepine use was associated with higher symptom severity during treatment, whereas lithium and lamotrigine were not, suggesting that some common co-medications may matter more than others in day-to-day practice.

An open-label LSD microdosing analysis in major depressive disorder found short-term improvements in daily mood and added pharmacokinetic data for sublingual LSD across an eight-week dosing regimen. It follows earlier trial reporting from the same study, so the main caveat remains the lack of a placebo group.

A post hoc path analysis from a Phase II psilocybin trial in treatment-resistant depression found that measures of the psychedelic experience were more strongly linked to depression improvement than pre-dosing therapeutic alliance. The alliance still seemed to support the quality of the experience, which keeps both drug effects and therapy context in view.

A post hoc ESKALE analysis found that intranasal esketamine was associated with a rising response rate during the first four weeks of induction treatment for treatment-resistant depression. Early dissociation may have signalled later response, though this kind of real-world analysis is better for hypothesis generation than firm prediction.

Mechanisms And Brain Measures

These studies ask what changes during or after psychedelic and ketamine administration. Most are not treatment-efficacy papers; they use imaging, pharmacokinetics, psychometrics, surveys, or interviews to link subjective effects with brain, body, and context. The useful thread is not one mechanism, but the way different measures capture different aspects of the experience.

A PET study in treatment-resistant depression suggested that the dynamics of AMPA receptors may help explain ketamine response. Baseline AMPAR availability tracked illness severity and differed from healthy controls, while ketamine-related AMPAR changes in specific regions were linked to symptom improvement, pointing to a regionally specific and partly restorative mechanism.

In healthy men, a ketamine study using a multimodal approach found a significant increase in anterior cingulate glutamate and altered functional connectivity, but no clear group-level increase in the synaptic PET marker ([11C]-UCBJ VT). The paper is a good reminder that mechanistic signals can be uneven across levels, with neurochemistry, connectivity and synaptic markers not always moving in lockstep.

Another ketamine imaging study focused on accumbal glutamate and its link to altered states of consciousness. There was no significant group-level rise in glutamate, but individual changes were associated with anxious ego dissolution and reduced vigilance, which is a neat example of how subjective experience may be the more sensitive readout than the group mean.

The intravenous DMT study provided clear dose-response and pharmacokinetic data in healthy participants. Effects peaked within minutes, faded quickly, and levelled off at higher doses, while the open-label dose-escalation arm suggested that context and expectancy made the experience more tolerable. It is one of the month’s best demonstrations that psychedelic effects are both pharmacological and deeply shaped by administration context (set & setting).

A brain imaging analysis argued that psilocybin shapes the slow, global propagation of brain activity over the cortical layout of 5HT2A receptors. Faster wave propagation went along with narrower connectivity gradients and stronger functional connectivity, suggesting that what often looks like a connectivity effect may partly reflect a change in large-scale travelling waves.

In a mixed-methods analysis from a randomised controlled mindfulness retreat, DMT-harmine produced a richer and more intense phenomenology than placebo while still sharing some semantic themes with meditative states. The study’s blend of topic modelling and manual coding is especially useful because it shows how language can capture both the psychedelic experience itself and the retreat context around it.

The psychometric analysis of Mystical dynamics argued that mystical oneness is closely linked with renewal and luminous light, and moderately to strongly linked with ego disintegration. It is a careful re-analysis, but also a reminder that the field still lacks a perfect vocabulary for some of the core experiences it keeps trying to measure.

A large survey suggested that age moderates the relationship between lifetime psychedelic use and mental health in naturalistic settings. Classic psychedelics were associated with lower anxiety and depression in younger adults, but those associations weakened with age and, in some cases, reversed, which makes age an important variable rather than a background detail.

A survey on Therapeutic-Like Context and Relational Support During Psychedelic Use found that setting and support moderated how life stress, challenging experiences, and later psychological outcomes were linked. The paper gives a useful, grounded account of why supportive context can reduce harm without necessarily removing the difficulty of the experience itself.

In interviews with people who had lived experience of psychosis and psychedelics, most participants said the two states were not especially similar. There was overlap in altered thinking and meaning-making, but the paper argues that psychedelics are usually more sensory, more controllable, and less self-referential than psychosis.

An exploratory biomarker study of MDMA-assisted group therapy in veterans with PTSD found small, mixed changes in IL-6, TNF-alpha and CRP, with baseline inflammation linked to worse symptoms. The sample was small and the effects modest, but the paper adds to a still thin literature on immune change in psychedelic-assisted therapy.

Reviews And Commentary

These broader papers step back from single trials. They review evidence, compare trial designs, and ask how implementation, safety, and expectancy should be handled as the field matures.

A living systematic review of MDMA-assisted therapy for PTSD found better short-term symptom reduction, response, and remission than control treatment across six randomised trials, but the overall certainty of evidence was low. The public open-data resource and planned updates are as important as the pooled estimates, because the evidence base is moving quickly and still carries substantial concerns about functional unblinding and generalisability.

A meta-analysis comparing psychedelic therapy with antidepressants under equal unblinding conditions found that psychedelic-assisted therapy was not more effective than open-label traditional antidepressants for depression. The result is less a verdict against psychedelics than a warning that blinding, expectancy and comparator choice shape what the headline effect sizes really mean.

Trip killers is a practical review of drugs that might be used to reduce severe distress during a bad trip. Its main takeaway is cautious and sensible: psychological containment should come first, but when medication is needed, selective 5-HT2A antagonism and a small number of antipsychotic options look most plausible.

The NHS implementation commentary on psilocybin argues that service delivery may be possible if psychological support is planned carefully and embedded in a realistic staffing model. It is less about efficacy than about what would need to change for public services to deliver psychedelic treatment fairly and safely.

The postpartum depression review is a useful map of a very under-studied area. It is appropriately cautious, noting that serotonergic psychedelics may have promise, but that safety, breastfeeding, long-term effects and peripartum-specific safeguards still need far more work.

The exercise-and-psychedelics commentary is speculative, but it raises an interesting possibility: that physical activity and psychedelic therapy may reinforce each other biologically and behaviourally. It is best read as a prompt for future studies rather than a treatment recommendation.

The Roland R. Griffiths tribute is a reminder that the modern field still rests on decades of careful behavioural pharmacology and methodical human research. It also nicely captures how work on alcohol, nicotine, caffeine and benzodiazepines helped create the scientific credibility that later psychedelic studies needed.

Also Published This Month

The remaining paper is preclinical and therefore gets a shorter note than the human and clinical work above. It is still useful for mechanism-building, but it sits further from patient care.

A rat study of oxa-noribogaine found reduced alcohol drinking and relapse-like drinking, apparently through stronger aversion learning and changes in glutamatergic activity in the medial prefrontal cortex. It stays in the preclinical lane, but it adds mechanism detail to ibogaine-adjacent alcohol research.