Pharmacokinetics and concentration-effect relationship of oral LSD in humans

Dolder and colleagues frame lysergic acid diethylamide (LSD) as a prototypical hallucinogen with continuing recreational use and renewed interest for psychiatric research. The authors note that despite this interest, the pharmacokinetics (PK) of orally administered LSD in humans remain poorly characterised: prior data are limited to small intravenous studies and sparse oral sampling that did not permit full PK parameter estimation. This study therefore set out to define the single-dose pharmacokinetic profile of oral LSD (200 µg) in healthy men and women, to relate plasma concentrations to acute subjective and physiological effects (PK-pharmacodynamic relationship), and to quantify urinary excretion of LSD and its main metabolite 2-oxo-3-hydroxy-LSD (O-H-LSD). The aim was to provide data useful for clinical research and forensic assessment of LSD intoxication.

The investigators used a double-blind, placebo-controlled, cross-over design with two experimental sessions per subject and at least a 7-day washout between sessions. Sixteen healthy volunteers (8 men, 8 women; mean age 28.6 ± 6.2 years) participated after screening that excluded major psychiatric history, recent illicit drug use, pregnancy and other standard criteria. Subjects provided written informed consent and the study was approved by local regulatory and ethics bodies. A single oral dose of LSD (200 µg administered as two 100 µg gelatin capsules) or matched placebo was given at 09:00 in each session. Blood samples were taken before dosing and repeatedly up to 24 hours; the extracted text does not clearly report all individual post-dose blood-sampling time points. Urine (total volume) was collected in three intervals: 0–8, 8–16, and 16–24 hours after dosing. Vital signs, tympanic body temperature and repeated subjective ratings using 100-mm visual analogue scales (VAS) were recorded at multiple time points; pupil diameter was measured at several specified times with an infrared pupillometer. LSD and O-H-LSD in plasma and urine were quantified by a validated liquid-chromatography-tandem mass-spectrometry method with a lower limit of quantification of 0.1 ng/mL. Noncompartmental PK analysis was performed using Phoenix WinNonlin 6.4. Cmax and Tmax were taken from observed data; AUC0–24 was calculated by a linear-up log-down trapezoidal rule; the terminal elimination rate constant (λz) and terminal half-life were estimated from the terminal log-linear portion of the concentration–time curve and AUC∞ was extrapolated accordingly. Because the elimination rate changed after about 12 hours in many subjects, the authors separately estimated a half-life for the interval from Tmax to 12 hours using at least three data points in that phase. Renal clearance was computed as urinary recovery over AUC0–24. Pharmacodynamic analyses were descriptive. The primary PK–PD hypothesis was that no acute pharmacological tolerance would be apparent (no clockwise hysteresis). Effects after LSD were expressed as within-subject differences from placebo at corresponding time points to control for circadian changes. Hysteresis loops were constructed by plotting effect versus plasma concentration across time and the area within the hysteresis loop (AH) was calculated; AH < 0 indicates counterclockwise hysteresis (lag), AH > 0 indicates clockwise hysteresis (tolerance). Sigmoidal concentration–response (variable slope) models were fitted using data from Cmax to 24 hours to estimate EC50 for selected subjective effects because insufficient absorption-phase pairs were available for a full indirect link model. Statistical reporting was descriptive using geometric means and 90% confidence intervals; the study was noted to be underpowered (power 52%) to exclude sex differences in PK.

Plasma LSD concentrations above the 0.1 ng/mL quantification limit were measurable in all subjects up to 12 hours, in 14 subjects up to 16 hours and in 11 subjects up to 24 hours after oral administration. The observed mean Cmax was 4.5 ± 1.4 ng/mL, reached at a median Tmax of 1.5 hours (range 0.5–4 hours). From Tmax to 12 hours, concentrations declined with apparent first-order kinetics and a mean half-life of 3.6 ± 0.9 hours. In a subset of participants a slower terminal decline in concentration was observed after 12 hours, yielding a longer terminal half-life estimate of 8.9 ± 5.9 hours when those late points were included. O-H-LSD (the principal metabolite) was detectable in plasma only in about half the subjects; metabolite concentrations were generally low. Urinary recovery over 24 hours amounted to 13% of the administered dose as O-H-LSD (28.3 µg) and only 1% as unchanged LSD (2.1 µg). Of unchanged LSD recovered in urine, 56% appeared in the first 8 hours; 45% of O-H-LSD urinary recovery occurred in the 8–16 hour interval. Renal clearance of LSD was 1.32 ± 0.6 mL/min, representing roughly 1.6% of apparent total clearance after oral dosing if an oral bioavailability of 71% is assumed. No statistically significant differences between male and female subjects were observed in plasma PK parameters or overall 0–24 hour urine recoveries, although O-H-LSD was above the limit of detection in only 8 subjects and the study was underpowered to exclude sex differences. Concentration–effect analyses showed a close temporal relationship between plasma LSD levels and many dynamic effects. No clockwise hysteresis (which would indicate acute tolerance) was observed for heart rate, blood pressure or “bad drug effect”; the 95% CIs for the hysteresis-area metric (AH) for those measures overlapped zero (examples: heart rate AH mean 4.4 beats × ng/min × mL, 95% CI −13 to +22). Counterclockwise hysteresis (negative AH) was seen early for body temperature, pupil size, “any drug effect” and “good drug effect”, consistent with a lag during the absorption/distribution phase. Estimated EC50 values (mean ± SD) derived from data from Cmax to 24 hours were 1.3 ± 0.7 ng/mL for “any drug effect” and 1.0 ± 0.5 ng/mL for “good drug effect”. Heart rate, blood pressure, body temperature and “bad drug effect” increased approximately linearly with plasma concentration and did not show an apparent Emax within the observed concentration range. Adverse events were mostly mild to moderate and included difficulty concentrating, headache, exhaustion and dizziness lasting up to 24 hours. No severe adverse events were reported.

Dolder and colleagues interpret their findings as providing the first detailed single-dose PK characterisation of oral LSD in humans. They emphasise that peak plasma concentrations occurred at a median of 1.5 hours and that LSD concentrations declined with first-order kinetics up to 12 hours, with a mean half-life of about 3.6 hours in that interval; a slower and inconsistent terminal decline was seen in some subjects after 12 hours, which the authors attribute either to redistribution from tissues or to measurement uncertainty close to the lower limit of quantification. The authors note that only a small fraction of the administered dose is excreted unchanged in urine (1%), whereas 13% was recovered as O-H-LSD within 24 hours. O-H-LSD concentrations in urine were substantially higher than unchanged LSD, consistent with forensic findings. Metabolism to O-H-LSD is attributed to cytochrome P450 enzymes but the specific isoforms remain unspecified; it is also unknown whether O-H-LSD is pharmacologically active. By comparing their oral AUC data with previously published intravenous AUCs, the investigators estimate a crude oral bioavailability of approximately 71%, but caution that this estimate is approximate and based on small-sample IV data. Relating concentrations to effects, the authors report EC50 estimates for subjective effects around 1–1.3 ng/mL, values they note are comparable to LSD binding affinities at the 5-HT2A receptor and therefore biologically plausible. Pupil dilation occurred at low concentrations and persisted longer than some cardiovascular changes; heart rate, blood pressure and body temperature showed roughly linear concentration–effect relationships with no observed ceiling within the studied range, suggesting these measures could increase further at higher doses. Importantly, no acute pharmacological tolerance was observed (no clockwise hysteresis), in contrast to previous observations with MDMA using similar methodology. The authors contrast LSD’s likely direct 5-HT2A agonism with MDMA’s indirect monoamine release as a mechanistic rationale for this difference. Key limitations acknowledged in the paper include the small sample size and limited power to exclude sex differences in PK, the incomplete detection of O-H-LSD in plasma for half the sample, and uncertainty about the slower terminal phase beyond 12 hours. The authors also note that potential food effects on LSD absorption remain inadequately characterised and warrant further study. They conclude that the reported PK and PK–PD data are valuable for clinical research and forensic assessment of LSD intoxication.