Neuroimaging in psychedelic drug development: Past, present, and future

The paper is a perspective synthesis rather than a systematic review or new empirical study; the extracted text does not report a formal search strategy or selection criteria. Instead, the authors review published neuroimaging findings from acute challenge studies and clinical trials, identify methodological limitations in the literature, and propose study designs and imaging strategies that could address key knowledge gaps. Technical approaches highlighted as central to future investigations include functional MRI (resting-state and task paradigms), arterial spin labelling for cerebral blood flow, magnetic resonance spectroscopy (MRS), diffusion imaging, EEG/MEG as ancillary electrophysiological measures, and molecular PET imaging. Specific PET ligands discussed are the 5-HT2A agonist [11C]Cimbi-36 for target occupancy and functional relevance, [11C]UCB-J for synaptic vesicle glycoprotein 2A (SV2A) as a marker of synaptic density, and [18F]BCPP-EF for mitochondrial complex 1 (MC1) as a marker related to mitochondrial density and potentially postsynaptic change. In terms of study design, the authors propose early-phase (Phase I/IIa-like) experiments in healthy volunteers that would combine acute dosing (ideally using integrated PET/MR scanners for simultaneous acquisition), adaptive PET occupancy studies to establish plasma concentration–occupancy relationships, and longitudinal follow-up scans to assess longer-term functional and molecular changes. Ancillary data recommended for inclusion comprise plasma pharmacokinetics, standardised subjective measures, genotyping for relevant polymorphisms (for example in the 5-HT2A receptor), and peripheral biomarkers such as BDNF and inflammatory markers. The authors also advocate application of the established ‘‘three pillars’’ translational framework—tissue exposure, target engagement, and pharmacologic activity—when evaluating novel 5-HT2A-active compounds.

From acute challenge studies using primarily fMRI, a consistent finding is that classic serotonergic psychedelics (psilocybin, LSD, DMT) profoundly disrupt large-scale functional network organisation. Resting-state work shows network disintegration and altered connectivity patterns that underpin contemporary mechanistic models such as REBUS and CSTC. Task-based fMRI studies have revealed drug-related modulation of emotional and social processing: for example, LSD alters hedonic responses to music, psilocybin affects social/emotional processing, and MDMA influences recall of emotional memories. Molecular PET has been used less frequently, largely due to cost, logistical complexity, and until recently a lack of suitable agonist radioligands. Notwithstanding, [11C]Cimbi-36 PET studies have linked psilocin (the active metabolite of psilocybin) binding at 5-HT2A receptors to acute subjective effects, and other PET work has probed downstream changes such as dopamine release ([11C]raclopride) and cerebral metabolism ([18F]FDG). PET investigations have also assessed long-term serotonin markers following recreational use of MDMA and psychedelics. Neuroimaging adjuncts in clinical trials (mostly pre/post (f)MRI) have reported several associations between brain changes and clinical outcomes. In treatment-resistant depression cohorts, decreases in amygdala cerebral blood flow correlated with reductions in depression scores, and alterations in medial prefrontal–hippocampal functional connectivity were observed after psilocybin. Increased amygdala reactivity to emotional stimuli and heightened brain responses to music following treatment have also been described. Two groups reported reductions in brain modularity after psilocybin, and these modularity changes related to clinical improvements, whereas escitalopram did not produce similar modularity effects. An open-label study found increased cognitive and neural flexibility up to four weeks post-treatment and MRS evidence of decreased glutamate and N‑acetylaspartate in the anterior cingulate cortex. Peripheral evidence for a neuroplasticity signal includes reports of elevated serum BDNF following low-dose LSD and ayahuasca, though these are non-specific, peripheral measures. The extracted text emphasises heterogeneity and limitation across this evidence base: many studies are small, methods and analyses vary widely, over half of published resting-state work derives from two datasets, and PET data remain scarce. Consequently, the authors note that a reliable imaging biomarker of clinical response has not yet been established.

Wall and colleagues interpret the accumulated neuroimaging findings as providing a valuable, objective foundation for psychedelic therapy development while recognising substantial gaps in mechanistic understanding and the evidence base. They argue that multimodal neuroimaging, particularly combined PET/MR using modern PET ligands, offers a realistic route to link molecular target engagement at the 5-HT2A receptor with acute functional brain effects and longer-term clinical outcomes. The authors position their three priority questions—acute versus long-term brain effects, explicit characterisation of 5-HT2A pharmacology and dose–occupancy relationships, and demonstration of neuroplasticity in humans—as central to translating psychedelic treatments into robust, reproducible clinical practice. Suggested means to address these include adaptive PET occupancy studies, simultaneous PET/MR acquisition during acute dosing, longitudinal PET imaging with SV2A and MC1 ligands to probe synaptic and mitochondrial changes, complementary MRI measures (DTI, MRS, resting and task fMRI), and inclusion of ancillary data (pharmacokinetics, genotyping, EEG/MEG, peripheral biomarkers). They also discuss broader development issues. A translational ‘‘three pillars’’ framework (tissue exposure, target engagement, pharmacologic activity) is advocated for evaluating new 5-HT2A agonists and second-generation compounds. The authors caution against ‘‘psychedelic exceptionalism’’ and stress that novel compounds should follow established CNS drug-development paradigms. The role of psychotherapy is identified as an unresolved but important variable; disentangling drug effects from psychotherapeutic contributions will require large, multi-arm trials with careful control of experiential and procedural factors. Resource and feasibility constraints are acknowledged—combined PET/MR scanners, novel radioligands and large samples are costly and limited in availability—but the authors consider the proposed studies tractable and likely to materially improve the basic-science foundation for clinical development. Finally, they note the limitations introduced by the 50‑year research hiatus: compared with other psychiatric medicines, the basic-science evidence base for psychedelics is relatively sparse, increasing the need for rigorous, standardised, and replicable neuroimaging work.

Neuroimaging has been integral to the modern re-emergence of psychedelic research and is well placed to answer foundational questions needed for clinical development. The authors conclude that strategically designed, multimodal imaging studies—combining PET ligands that probe 5-HT2A engagement and synaptic/mitochondrial markers with advanced MRI, electrophysiology and ancillary biomarker measures—can build a ‘‘molecular–functional–clinical bridge’’. Although ambitious and resource-intensive, such studies are feasible with current technology and would provide a robust basic-science platform to support further clinical development and responsible evaluation of psychedelic therapies.