Large-scale brain connectivity changes following the administration of lysergic acid diethylamide, d-amphetamine, and 3,4-methylenedioxyamphetamine

Abstract

Psychedelics have recently attracted significant attention for their potential to mitigate symptoms associated with various psychiatric disorders. However, the precise neurobiological mechanisms responsible for these effects remain incompletely understood. A valuable approach to gaining insights into the specific mechanisms of action involves comparing psychedelics with substances that have partially overlapping neurophysiological effects, i.e., modulating the same neurotransmitter systems. Imaging data were obtained from the clinical trial NCT03019822, which explored the acute effects of lysergic acid diethylamide (LSD), d-amphetamine, and 3,4-methylenedioxymethamphetamine (MDMA) in 28 healthy volunteers. The clinical trial employed a double-blind, placebo-controlled, crossover design. Herein, various resting-state connectivity measures were examined, including within-network connectivity (integrity), between-network connectivity (segregation), seed-based connectivity of resting-state networks, and global connectivity. Differences between placebo and the active conditions were assessed using repeated-measures ANOVA, followed by post-hoc pairwise t-tests. Changes in voxel-wise seed-based connectivity were correlated with serotonin 2 A receptor density maps. Compared to placebo, all substances reduced integrity in several networks, indicating both common and unique effects. While LSD uniquely reduced integrity in the default-mode network (DMN), the amphetamines, in contrast to our expectations, reduced integrity in more networks than LSD. However, LSD exhibited more pronounced segregation effects, characterized solely by decreases, in contrast to the amphetamines, which also induced increases. Across all substances, seed-based connectivity mostly increased between networks, with LSD demonstrating more pronounced effects than both amphetamines. Finally, while all substances decreased global connectivity in visual areas, compared to placebo, LSD specifically increased global connectivity in the basal ganglia and thalamus. These findings advance our understanding of the distinctive neurobiological effects of psychedelics, prompting further exploration of their therapeutic potential.

Fig. 1. Substance-induced changes in network integrity.

Depicted are changes in network integrity (i.e., within-network connectivity) induced by LSD, d-amphetamine, and MDMA, compared to placebo. All substances induced reductions in network integrity in the VIS and FPN. Amphetamines induced reduced integrity in SAL and ASM, whereas LSD uniquely reduced integrity in the DMN. Repeated-measures ANOVAs were computed on dual regression-derived parameter estimates and corrected for multiple comparisons (P_FDR < 0.05). The ANOVA was not significant for SAL and DAN. * - depicts significant differences. Abbreviations: VIS visual network, SAL salience network, FPN frontoparietal network, DMN default mode network, ASM auditory-sensorimotor network.

Fig. 2. Substance-induced changes in network segregation.

Depicted are changes in network segregation (i.e., between-network connectivity) induced by LSD, d-amphetamine, and MDMA, compared to placebo. LSD induced more extensive increases in connectivity. Compared to LSD, the amphetamines also induced increased segregation vs. placebo. Repeated-measures ANOVAs were computed on Pearson correlation coefficients of dual regression-derived time series of pairs of networks, followed by r-to-z transformations, and corrected for multiple comparisons (P_FDR < 0.05). * - depicts significant differences. Abbreviations: VIS visual network, SAL salience network, FPN frontoparietal network, DMN default mode network, ASM auditory-sensorimotor network.

Fig. 3. Substance-induced changes in seed-based connectivity.

Depicted are voxel-wise repeated-measures analysis of variance (ANOVA) parametric maps reflecting contrasts between active substances and placebo for RSN whole-brain connectivity. The substances mainly increased connectivity between the seed network and other areas of the brain (shown in yellow/red) but reduced connectivity within the seed network (shown in blue). As no voxel-wise differences were observed between MDMA and placebo conditions for DAN iFC no image is depicted. The analyses were computed in SPM12 (P < 0.001, cluster-level familywise error–corrected P < 0.05); x, y, and z indicate Montreal Neurological Institute coordinates.

Fig. 4. Substance-induced changes in degree centrality.

Depicted are voxel-wise repeated-measures analysis of variance (ANOVA) parametric maps reflecting contrasts between active substances and placebo for global connectivity (A) and between the active conditions only (B). Compared to placebo, substances differentially increased global connectivity (i.e., degree centrality) in the brain (shown in yellow/red) but similarly induced reductions in visual areas (shown in blue). Compared to the amphetamines directly, LSD mainly increased global connectivity in sensorimotor areas. As no voxel-wise differences were observed between d-amphetamine and MDMA, no image is depicted. The analyses were computed in SPM12 (P < 0.001, cluster-level familywise error–corrected P < 0.05); x, y, and z indicate Montreal Neurological Institute coordinates.

Fig. 5. Associations between altered connectivity and 5-HT2AR density.

Depicted are comparisons of Pearson’s correlation coefficients between the iFC of RSNs and 5-HT2AR density in the same regions for all conditions. Notably, LSD showed a stronger positive association with the 5-HT2AR density across all networks, except for VIS, compared to both placebo and amphetamines. Whiskers represents the minimum and maximum correlation coefficients values within 1.5 times the interquartile range. Repeated-measures ANOVAs were computed on the Pearson coefficients and corrected for multiple comparisons (P_FDR < 0.05). * - depicts significant differences. Abbreviations: VIS visual network, SAL salience network, FPN frontoparietal network, DMN default mode network, ASM auditory-sensorimotor network.