Drug-drug interactions involving classic psychedelics: A systematic review

Classic psychedelics discussed in this review include LSD, psilocybin (psilocin as the active metabolite), mescaline, DMT and 5-MeO-DMT. Earlier pharmacology and clinical literature indicate these compounds act primarily at serotonergic receptors—most notably 5-HT2A—while also engaging other receptor systems (for example 5-HT1A and dopamine receptors). Metabolic pathways vary between agents: LSD is metabolised by multiple cytochrome P450 (CYP) enzymes (notably CYP1A2 and CYP3A4), psilocybin is converted to psilocin and handled partly by monoamine oxidase (MAO) and UDP-glucuronosyltransferases, DMT is rapidly deaminated by MAO-A unless co‑administered with MAO inhibitors (as in ayahuasca), and 5-MeO-DMT is metabolised primarily by MAO-A with a minor CYP2D6 pathway to bufotenine. These differing pharmacokinetic and pharmacodynamic features imply varied potential for drug–drug interactions (DDIs) when classic psychedelics are combined with other medications or substances. Halman and colleagues set out to fill a gap in the literature by systematically identifying and synthesising human data on DDIs involving classic psychedelics. The review aimed to capture both pharmacodynamic and pharmacokinetic interactions and to report physiological and subjective outcomes, providing a comprehensive overview to inform clinicians, researchers and policymakers as interest in therapeutic and recreational use of these agents resurges.

The review was prospectively registered in PROSPERO (CRD42022336092) and followed PRISMA 2020 guidance. Searches were run in multiple sources across two time points: an initial search on 5 June 2022 in PubMed, PsycINFO and Web of Science (no date limits), and an update on 2 September 2023 that added recently published articles and a focused search for 5-MeO-DMT. The authors also searched the MAPS "Psychedelic Bibliography" (downloaded and filtered) and ClinicalTrials.gov, and checked reference lists of included reports. Full search terms were reported in supplemental materials (not reproduced here). Screening and data extraction used the Catchii systematic review application. Duplicate detection was applied and manually verified. Records were screened in three phases (scientific databases, MAPS bibliography, and 5‑MeO‑DMT search). Two reviewers independently conducted title/abstract screening while blinded to each other's decisions; disagreements were resolved by discussion to consensus. After duplicate removal, 7,102 records underwent title/abstract screening; 77 reports proceeded to full‑text assessment. Inclusion criteria required: (1) human data; (2) administration of a classic psychedelic together with another drug; (3) reporting of physiological and/or psychological outcomes (including absence of interaction); (4) study designs comprising randomised controlled trials, observational designs or case reports/studies; and (5) publication in English. Outcome measures extracted were physiological and psychological effects, and data from eligible records were corroborated among the review authors. The final selection yielded 36 reports describing 52 studies; further methodological details and study‑level data were presented in tables and supplemental materials.

The review included 52 studies reported across 36 reports, encompassing 32 studies on LSD, 10 on psilocybin, 4 on mescaline, 3 on DMT, 2 on 5‑MeO‑DMT and 1 on ayahuasca. Studies spanned experimental trials, older clinical research from the mid‑20th century and case reports; outcomes reported were a mix of subjective measures (for example 5‑dimensional altered states of consciousness, 5D‑ASC), physiological measures (blood pressure, heart rate, temperature) and plasma pharmacokinetics where available. Interactions that attenuated psychedelic effects: Multiple studies showed that 5‑HT2A receptor antagonists attenuate or abolish subjective and some physiological effects of LSD and psilocybin. Ketanserin consistently blocked or reduced subjective effects (for example reducing 5D‑ASC domains such as anxious ego dissolution, depersonalisation, euphoria and hallucination) and lowered some physiological responses. Risperidone, which antagonises 5‑HT2A and D2 receptors, attenuated psilocybin effects in a dose‑dependent manner. Trazodone, also a 5‑HT2A antagonist, reduced LSD effects in a case report. Findings for chlorpromazine were inconsistent across older studies, with some reports of attenuation and others of enhancement; differences in dosing and timing were noted as possible explanations. 5‑HT1A agonists produced mixed effects: Pretreatment with buspirone, a 5‑HT1A partial agonist, markedly reduced psilocybin‑induced visual hallucinations. By contrast, oral ergotamine did not alter psilocybin subjective effects in a controlled trial, which the authors attributed to lower efficacy and bioavailability of ergotamine relative to buspirone. Dopamine antagonists: Haloperidol did not reduce psilocybin‑induced hallucinations in one report but diminished feelings of oceanic boundlessness and derealisation while increasing anxiety in another, suggesting a complex or indirect modulatory role of dopaminergic systems. Antidepressants and related agents: Several SSRIs (fluoxetine, sertraline, paroxetine) were reported to reduce LSD effects in at least one study, with fluoxetine also delaying onset in about half of participants and a case report noting marked decreased sensitivity to LSD. Escitalopram did not alter positive mood or most mind‑altering effects of psilocybin in a controlled setting but reduced ego disintegration and anxiety after 14 days of treatment. The extracted text also discusses a small open study (not included in the primary results) of 19 patients with treatment‑resistant depression on chronic SSRIs who received synthetic psilocybin (COMP360): no serious treatment‑emergent adverse events were reported, and 42.1% showed improvement in depression severity at week 3; however, the authors emphasised limited reporting of which specific SSRIs participants were taking and small sample size. Monoamine oxidase inhibitors (MAOIs) and DMT/ayahuasca: Non‑selective MAOIs (phenelzine, isocarboxazid, nialamide) attenuated or blocked LSD effects in several older studies, whereas iproniazid did not alter LSD effects but did reduce DMT effects in experiments. The authors hypothesise that increased brain serotonin after MAOI administration may blunt some psychedelic responses and that oral DMT becomes bioavailable when combined with MAO‑inhibiting β‑carbolines in ayahuasca. Case reports raised safety concerns: one individual on fluoxetine developed transient serotonin‑toxicity‑like symptoms after ayahuasca (resolved within 4 hours), and two case reports described severe adverse events involving 5‑MeO‑DMT consumed with β‑carbolines (harmaline/harmine), including one death; these reports were critiqued within the literature for limited detail and uncertain product composition. Other blocking agents: High‑dose niacin in older studies delayed LSD onset and, when given post‑LSD, rapidly attenuated perceptual effects; mechanisms proposed include niacin‑induced serotonin release. Phenoxybenzamine pretreatment did not alter LSD effects in another older report. Drugs that potentiated psychedelic effects: Tricyclic antidepressants (desipramine, imipramine, clomipramine) were reported in a small study to potentiate LSD effects. Chronic lithium use reportedly potentiated and hastened LSD effects in one series, although lithium has also been linked to seizures when combined with psychedelics in the literature. Reserpine pretreatment was associated with intensified and prolonged negative LSD experiences and tremors in multiple older reports; mechanistic explanations include neurotransmitter depletion and inhibition of CYP2C19/CYP2D6 and P‑glycoprotein, potentially prolonging LSD exposure. Drug combinations with mixed/no effects: Azacyclonol produced inconsistent findings across studies for LSD and mescaline, with some reports of attenuation and others of no effect. Alcohol appeared to have its subjective effects antagonised by LSD (and to a lesser degree by psilocybin), while the psychedelics' own effects were generally unchanged. Co‑administration of LSD with MDMA in a placebo‑controlled study produced longer‑lasting subjective effects (about 1.5 hours longer) and higher plasma concentrations and extended elimination half‑life of LSD; this was attributed to MDMA‑mediated inhibition of CYP2D6, an enzyme involved in LSD metabolism. Safety and adverse events: Apart from the discussed case reports, the included experimental and clinical studies did not report widespread serious adverse drug events attributable to DDIs. The authors highlighted that many included studies were old and sometimes poorly reported, and that case reports—while informative—are anecdotal and limited in generalisability.

Halman and colleagues interpret the assembled literature as evidence that classic psychedelics engage in both pharmacodynamic and pharmacokinetic drug–drug interactions that can either attenuate or potentiate subjective and physiological effects. They emphasise consistent support for a central role of 5‑HT2A receptor antagonism in blocking psychedelic effects, as demonstrated by ketanserin and several antipsychotics, while noting that other serotonergic subtypes (for example 5‑HT1A and 5‑HT1B/1C) and dopaminergic mechanisms may also modulate responses in compound‑specific ways. Differences in metabolic pathways across psychedelics appear to drive divergent interaction profiles: for example, MAO inhibition strongly affects orally administered DMT (ayahuasca) but has variable effects on LSD, and CYP‑mediated metabolism (notably CYP2D6) can be inhibited by co‑administered drugs such as MDMA, prolonging LSD exposure. The authors acknowledge substantial uncertainties. Much of the existing literature is old (1950s–1970s), sometimes lacks detailed methods or dosing information, and includes small experimental samples and case reports that limit causal inference and generalisability. They also note that studies of chronic concomitant treatments (for example long‑term SSRI therapy) are sparse; where available, evidence is mixed—some SSRIs appeared to attenuate psychedelic effects in older reports, whereas a small contemporary series found therapeutic responses to psilocybin despite ongoing SSRI treatment. Safety concerns arising from case reports—particularly combining MAOIs (or β‑carbolines in ayahuasca) with serotonergic antidepressants, and combining 5‑MeO‑DMT with MAO‑inhibiting products—are highlighted as areas requiring caution, but the authors stress the need for better‑documented evidence. Implications articulated by the authors include the need for systematic pharmacokinetic and pharmacodynamic investigation of clinically relevant combinations (for example psychedelics with commonly prescribed antidepressants, antipsychotics and enzyme inhibitors), more rigorous reporting in contemporary trials of participants' concomitant medications, and targeted surveillance for potentially serious interactions. Given the resurgence in clinical research and therapeutic interest in psychedelics, the authors conclude that improved understanding of DDIs is necessary to guide safe clinical practice and future research.

The review found evidence of both pharmacodynamic and likely pharmacokinetic DDIs involving classic psychedelics that can reduce, potentiate or prolong psychological and physiological effects. While 5‑HT2A receptor antagonism is a dominant mechanism for attenuating psychedelic effects, other receptor systems and metabolic interactions also contribute, and interaction profiles vary between psychedelics. Given the limited and heterogeneous evidence base—particularly the reliance on small, older studies and case reports—the authors call for further research to characterise clinically important interactions and to identify potentially serious adverse reactions not yet well described.