Assessing the Potential Cardiovascular Risk of Microdosing the Psychedelic LSD in Mice

Psychedelic compounds such as psilocybin and LSD are increasingly used and, in some jurisdictions, decriminalised, prompting rising public interest in microdosing—the repeated ingestion of subhallucinogenic amounts of these drugs. Because many classical psychedelics are potent agonists at serotonin receptors including 5-HT2A and 5-HT2B, there is a safety concern: chronic activation of 5-HT2B receptors has previously been implicated in drug-induced valvular heart disease and pulmonary hypertension (for example, with d-fenfluramine). Although a single high dose of a psychedelic is unlikely to produce cardiac damage, the cardiovascular consequences of prolonged low-dose (microdose) use remain poorly characterised. This preclinical study set out to assess whether chronic microdosing with LSD produces structural or functional cardiac changes in mice. The investigators compared two low LSD doses against vehicle and positive controls (serotonin and d-fenfluramine) using serial echocardiography, measured behavioural evidence of psychedelic activity to define subhallucinogenic doses, characterised agonist activity at mouse and human 5-HT2B receptors in vitro, and modelled 5-HT2B receptor occupancy over time using published human pharmacokinetic data. The work aims to elucidate whether the pattern and magnitude of 5-HT2B activation with microdosing are sufficient to produce cardiotoxic effects in this mouse model.

Animals and overall design: Male and female C57BL/6J mice (ordered at 8 weeks old; habituated 1 week) were used. Treatments for the chronic microdosing paradigm were given intraperitoneally (i.p.) on a 5 days on/2 days off schedule for 8 weeks, intended to model commonly used human microdosing regimens. Echocardiography was performed at baseline, 4 weeks, and 8 weeks. Body weight and heart rate were monitored during the treatment period. Doses and controls: For LSD the study used three doses to define behavioural and chronic exposure ranges: 0.01, 0.03, and 0.1 mg/kg (i.p.). Chronic cardiovascular experiments used the 0.01 and 0.03 mg/kg doses. Positive controls included serotonin hydrochloride at 40 mg/kg (i.p.) as a cardiac pathology control, and (+)-fenfluramine at 10 mg/kg (i.p.) as a known valvulopathic comparator. For in vitro pharmacology, compounds included 5-HT, (+)-norfenfluramine, LSD and psilocin. Behavioural assay and echocardiography: The head twitch response (HTR) was recorded with high-speed video and scored by an observer blinded to treatment to establish doses that are subhallucinogenic in mice (sample sizes for HTR assays were reported per dose group). Echocardiography used a Fujifilm Visualsonics F2 system with mice anaesthetised by isoflurane (initial bolus 3%, maintenance 1.5%) and temperature maintained at 37 °C. Left ventricular inner diameter, posterior wall diameter, volumes, ejection fraction and fractional shortening were measured using standard m-mode and software-based calculations. Aortic regurgitation was assessed at 4 weeks with colour Doppler, and regurgitant jets were classified by jet length relative to left ventricular length (mild <25%, moderate 25–50%, severe >50%). In vitro pharmacology and receptor occupancy modelling: Bioluminescence resonance energy transfer (BRET) assays were used in HEK T cells to measure Gq-mediated 5-HT2B receptor dissociation (Gq and β3/γ9), evaluating potency (EC50) and efficacy (Emax) for compounds at both human and mouse 5-HT2B receptors. Concentration–response curves were analysed by nonlinear regression and normalised to 5-HT stimulation. Pharmacokinetic modelling used published human plasma concentration–time courses for LSD, the active fenfluramine metabolite d-norfenfluramine, and psilocin. The Hill equation was applied, together with the in vitro EC50/Emax values, to estimate fractional 5-HT2B receptor activation over time for single human doses (modelled doses: LSD 20 μg, psilocybin 3 mg, d-fenfluramine 30 mg). Other procedural details: Echocardiogram analyses were carried out using VevoLab software. Head twitch scoring was blinded. The extracted text does not clearly report the specific statistical tests used for group comparisons or the exact sample sizes for every treatment group in the chronic echocardiography experiments when pooled across sex; some group sizes are given (for example, several groups are described as n = 3 males/3 females).

Dose selection and head twitch response: Using the head twitch response as an index of 5-HT2A activation, LSD at 0.01 and 0.03 mg/kg produced only small increases in head twitch frequency relative to 0.1 mg/kg. The investigators therefore treated 0.01 mg/kg as a subhallucinogenic microdose and 0.03 mg/kg as a weakly hallucinogenic dose in mice. Structural and functional cardiac outcomes by echocardiography: Across the 8-week treatment period there were no treatment effects on heart rate or body weight, indicating no gross adverse health effects. Mice receiving serotonin (40 mg/kg) showed significant cardiac structural changes: reductions in left ventricular (LV) inner diameters (end diastolic and end systolic) at 4 weeks relative to baseline, significant increases in LV posterior wall diameter at 4 and 8 weeks, and decreases in LV end diastolic and systolic volumes at 4 weeks. These changes indicate ventricular thickening in the serotonin-treated group. By contrast, neither the vehicle-treated mice nor mice treated with chronic LSD at either 0.01 or 0.03 mg/kg showed significant changes in LV inner diameter, posterior wall thickness, or LV volume at 4 or 8 weeks. Measures of systolic function (ejection fraction and fractional shortening) did not show consistent treatment-related changes. No differences in ejection fraction were detected in vehicle or serotonin groups. The 0.01 mg/kg LSD group showed a statistical effect across time for both ejection fraction and fractional shortening, but post hoc comparisons found no differences from baseline at the follow-up time points. The 0.03 mg/kg LSD group showed no changes in these functional measures. Valvular function and fenfluramine control: Colour Doppler imaging after 4 weeks of dosing revealed a significant increase in aortic valve regurgitation in the d-fenfluramine (10 mg/kg) group compared with vehicle (groups reported as n = 3 males/3 females). No difference in aortic valve regurgitation was observed between the 0.03 mg/kg LSD group and vehicle. In vitro 5-HT2B pharmacology: BRET Gq dissociation assays at human and mouse 5-HT2B receptors produced potency and efficacy estimates. Reported EC50 values include norfenfluramine 6.78 nM, LSD 0.88 nM, and psilocin 3.1 nM. Relative efficacies (as a percentage of 5-HT stimulation) were higher for norfenfluramine (about 96%) compared with LSD (about 62%) and psilocin (about 52%). The extracted text does not present full statistical details for these assays, but these values were used for downstream modelling. Pharmacokinetic-based receptor occupancy modelling: Using published human plasma time–course data and the in vitro EC50/Emax parameters, the investigators estimated fractional 5-HT2B receptor activation over time for a single human microdose of LSD (20 μg), a single human dose of d-fenfluramine (30 mg), and a psilocybin microdose (3 mg). The model indicated that 5-HT2B receptor occupancy for microdoses of LSD and psilocybin peaked at roughly half the level produced by norfenfluramine and declined to negligible levels by about 10 hours after dosing. By contrast, d-norfenfluramine produced larger and more sustained plasma exposure, with estimated receptor activation remaining above 50% for 24 hours. The investigators note that daily fenfluramine dosing would therefore be expected to produce persistent and cumulative 5-HT2B activation, consistent with its known cardiotoxicity.

Effinger and colleagues interpret their findings to indicate that chronic administration of low doses of LSD, given on a 5‑days-on/2‑days-off schedule for up to 8 weeks in mice, did not produce echocardiographic evidence of ventricular remodelling or valvular dysfunction, whereas positive controls produced expected pathology: serotonin induced ventricular thickening and d-fenfluramine produced aortic valvular regurgitation. The lack of detectable ventricular or valvular changes in the LSD groups is attributed by the investigators to limited 5-HT2B receptor activation under the microdosing regimen rather than to an absence of intrinsic agonist activity at the receptor. The authors position these results in the context of earlier work linking sustained 5-HT2B activation to cardiac valvulopathy. Their in vitro assays showed that although LSD and psilocin bind 5-HT2B receptors with nanomolar affinity, their efficacy to activate Gq-dependent signalling was lower than that of norfenfluramine. Combined with pharmacokinetic modelling using published human plasma data, the researchers argue that LSD and psilocin achieve lower peak concentrations and much shorter durations of 5-HT2B receptor occupancy than d-norfenfluramine. This pharmacokinetic and pharmacodynamic profile is offered as an explanation for why chronic low-dose LSD did not reproduce the cardiotoxic effects seen with fenfluramine in this mouse model. Key limitations acknowledged by the investigators include the translational gap between mice and humans: species differences in drug absorption and metabolism may alter receptor occupancy, and although no species differences in 5-HT2B affinity or efficacy were detected in their assays, other species-specific pharmacokinetic or physiological factors may matter. The experiments were performed in presumably healthy mice, so the findings may not generalise to people with preexisting cardiovascular risk factors. The authors also note the possibility that longer treatment durations or different dosing regimens could produce different outcomes. They therefore caution that absence of observed pathology in mice should not be taken as proof of safety in humans and recommend prospective human studies to evaluate cardiac effects of microdosing in clinical populations. The discussion also briefly notes that other serotonergic drugs, for example MDMA, may elevate cardiac risk through related mechanisms.