LSD treatment affects working memory-related brain function in patients with major depressive disorder

Classic psychedelics, including lysergic acid diethylamide (LSD), are currently under investigation for the treatment of major depressive disorder (MDD). Cognitive dysfunction is common in MDD, but effects of psychedelic treatment on cognitive brain function, measured by magnetic resonance imaging, are sparsely reported. The present study investigated effects of LSD treatment on working memory-related brain function in patients with MDD.

In a randomized, double-blind, parallel-group trial, 43 patients with MDD received either two low-dose (25 µg) or two high-dose (100 and 200 µg) oral LSD treatments 4 weeks apart. Functional magnetic resonance imaging during an n-back working memory task was conducted before and approximately 1 week after the second treatment. The treatment × time interaction effect on working memory-related brain responses (2-back and 1-back vs. 0-back) was assessed using linear mixed effects models.

We also explored associations between brain responses and depression severity, task performance, and subjective psychedelic experiences. Compared with low-dose treatment, high-dose LSD treatment significantly increased working memory-related brain responses (p = 0.007) within task-activated regions, including the dorsolateral prefrontal cortex, parietal cortex, insula, striatum, and thalamus. Working memory-related brain responses correlated with task performance and subjective perceptual changes but not treatment response.

Our observations provide the first evidence that LSD treatment in patients with MDD affects cognitive brain function. Contrary to our hypothesis, high-dose LSD significantly increased brain responses. These responses positively correlated with task performance, suggesting a potential LSD-induced boost in neural function.

The absence of an association with depressive symptoms indicates that task-related brain activity may underlie different mechanisms than changes in mood. Trial registration: Clinicaltrials.gov Identifier: NCT03866252. The authors acknowledge the assistance of Michael Arends for proofreading the manuscript.

This work was supported by the Gertrud Thalmann Fund for Depression Research (to FM), the Swiss National Science Foundation (grant no. P5R5PM_225310 to FH), and the Desirée and Niels Ydes Fond (grant 557–25 to KL). Open access funding provided by University of Basel.