Abstract by Kristian Larsen
Research on classical psychedelic drugs is advancing rapidly, as compounds such as psilocybin and LSD have entered phase-3 clinical trials for major depressive disorder (MDD) and generalised anxiety disorder. Psychedelic neuroimaging is trying to uncover both the underlying neural mechanisms behind the therapeutic potential and the intriguing subjective effects that these drugs produce in humans. The present thesis evaluates the neurovascular, cognitive, and functional effects of classical psychedelics using multimodal neuroimaging methods. These methods include positron emission tomography (PET),
blood-oxygen-level-dependent (BOLD) functional magnetic resonance imaging (fMRI), arterial spin labelling (ASL), time-of-flight angiography (TOF), and phase contrast mapping (PCM).
Papers 1 and 2 reveal a surprising finding in which psilocybin and LSD produced opposite effects on brain haemodynamics, despite these drugs sharing agonism and the serotonin 2A receptor (5-HT2AR). In Paper 1, psilocybin reduced whole-brain cerebral blood flow (CBF) by 11.7% at peak plasma levels and caused acute internal carotid artery (ICA) vasoconstriction (10.5% diameter reduction), whereas the 5-HT2AR antagonist ketanserin did not affect CBF. By contrast, Paper 2 demonstrated that LSD increased whole-brain CBF by 19.3%, ICA flow rate by 28% without vasoconstriction, and produced neocortical 5-HT2AR occupancy ranging from 42-85% across doses of 25-200 μg. Combined, these findings suggest that 5-HT2AR engagement alone is insufficient to explain the haemodynamic effects and highlight the polypharmacological complexity of these drugs. Notably, both LSD and psilocybin showed a negative correlation between CBF and global connectivity (GCOR) despite opposite effects on each measure separately. This finding further highlights that haemodynamic changes may influence BOLD-derived measures in pharmacological studies involving vasomodulatory drugs.
Paper 3 evaluated LSD effects on working memory-related brain function in patients with MDD, who received either two separate low-dose (2×25 μg) or high-dose (100 followed by 200 μg) treatments. Compared to the low-dose treatment, high-dose LSD significantly increased BOLD activation in working memory brain regions, but these effects were not associated with the therapeutic outcomes, suggesting that cognitive and therapeutic effects of LSD might operate through distinct neural systems.
Paper 4 provides the first direct comparison of the functional brain effects of classical psychedelics (LSD, psilocybin, mescaline) versus non-psychedelic psychostimulants (MDMA, d-Amphetamine) in the largest psychedelic neuroimaging dataset to date (79 participants, 255 sessions). Compared to psychostimulants, psychedelics significantly increased brain complexity and entropy in three of six measures evaluated: meta-state complexity; multiscale sample entropy (MSSE, scale 1); and dynamic conditional correlation entropy, and decreased MSSE at scale 5. The significant effects were primarily located in the brain's transmodal networks. These findings refine a popular hypothesis in psychedelic neuroimaging, the Entropic Brain Hypothesis, by demonstrating that psychedelic-specific effects involve rapid increases in temporal variability, which may coexist with more stable, slower fluctuations, and that some brain complexity and entropy measures are sensitive to psychedelic effects, but do not distinguish them from the effects produced by psychostimulants.
Collectively, the multimodal imaging work presented here contributes to our understanding of psychedelic drug action by (i) demonstrating that psilocybin and LSD produce opposite haemodynamic responses, and that the BOLD signal might be sensitive to these changes, (ii) that cognitive and therapeutic effects of LSD co-occur in patients with MDD, but independently, and (iii) some, but not all, complexity- and entropy-based measures capture psychedelic-specific effects when compared to psychostimulant effects.