Toward Translatable Biomarkers of Psychedelic-Induced Neuroplasticity

Depression and many other neuropsychiatric diseases are extremely heterogenous, which is unsurprising given that numerous genetic and environmental factors converge to produce each patient’s unique constellation of symptoms. As a result, it can often take several attempts for a patient to find an appropriate medication or other intervention to suit their needs. Patient stratification offers a potential solution to this problem by minimizing trial and error, but identifying robust biomarkers for subtyping patients is challenging. The rapidly emerging field of psychedelic medicine could benefit greatly from patient stratification, especially given that evidence suggests that both psychological and neurobiological mechanisms may play roles in the therapeutic effects of psychedelics, and that these factors may be dissociable (1, 2).

High doses of psychedelics reliably induce mystical-type experiences and can facilitate emotional breakthrough or psychological insight, which could benefit patients (3). Unfortunately, the powerful subjective effects of psychedelics necessitate costly and complex in clinical administration, which ultimately reduces patient access (4). In addition to their psychological effects, psychedelics have been classified as psychoplastogens (5, 6) because they promote robust structural neuroplasticity in the cortex leading to rapid and sustained increases in dendritic spine density. Increased spine density in the prefrontal cortex has been shown to be causally related to the sustained antidepressant-like effects of ketamine in rodents (7), and blocking the plasticity-promoting effects of psychedelics using genetic or pharmacological tools also blocks their antidepressant-like effects (8, 9). Furthermore, evidence suggests that the psychoplastogenic effects of psychedelics can be decoupled from their subjective effects, with several non-hallucinogenic psychoplastogens being recently reported (10–13). As the rescue of cortical atrophy is a hallmark of many antidepressant interventions (4, 14), psychedelic-induced cortical structural plasticity might be sufficient to ameliorate disease symptoms for many patients. Biomarker-based strategies for determining which patients exhibit cortical atrophy could streamline treatment by identifying the patients more likely to benefit from take-home, non-hallucinogenic psychoplastogens and those that might require an in clinic psychedelic experience.

Human imaging, postmortem analysis, and preclinical models all demonstrate that cortical atrophy is common in subjects exhibiting depressive phenotypes (14). Moreover, preclinical work in rodents has shown that chronic administration of selective-serotonin reuptake inhibitors (SSRIs) or a single administration of ketamine, psychedelics, or nonhallucinogenic psychoplastogens can induce cortical plasticity (15–17). However, given the challenges associated with measuring structural neuroplasticity non-invasively in a human brain, it has been difficult to find a translatable biomarker capable of detecting changes at the synaptic level. Recent advances in positron emission tomography (PET) imaging of synaptic vesicle protein 2A (SV2A) provide tantalizing opportunities for psychedelic-related drug discovery.

The radioligand [¹¹C]UCB-J has been used to selectively image SV2A in the human brain, with this measure serving as a proxy for synaptic density (18). Using this relatively new tool, decreased cortical SV2A density has been observed in a number of conditions including depression (19), cocaine use disorder (20), schizophrenia (21), and Alzheimer’s disease (22). Importantly, cortical and global SV2A densities were inversely correlated with the severity of depression (19) and cognitive performance (23) in patients with major depressive disorder and Alzheimer’s disease, respectively. These results are largely consistent with those from animal models, including models of depression that utilize chronic stress paradigms (4, 14).

Preclinical testing has demonstrated that several pharmacological agents can rescue cortical neuron atrophy and synapse loss, but the potential translatability of these findings to humans is still under active investigation, with several notable advances occurring in the past few years. First, the Knudsen Lab recently demonstrated that SV2A density increased over time in healthy people undergoing daily administration of the SSRI escitalopram. This time course is consistent with the fact that SSRIs require chronic, daily administration to demonstrate clinical efficacy. In contrast, psychoplastogens increase cortical spine density rapidly following a single administration (5). To assess the effects of the psychoplastogen ketamine on SV2A density, Esterlis and colleagues performed [¹¹C]UCB-J PET imaging 24 hours after drug administration to major depressive disorder patients with mild to moderate symptoms. When analyzed collectively, there was no statistical difference in cortical SV2A density between patients who received ketamine and those who received placebo (24). However, when the patients were stratified into two groups based on their baseline SV2A densities, they found that ketamine increased cortical SV2A density in the patients with SV2A deficits at baseline, and that the percent change correlated with reduction in disease symptoms. To date, no PET studies have been published describing the effects of psychedelics on human SV2A density. However, the Knudsen Lab has demonstrated that a single dose of psilocybin rapidly increased SV2A density in pigs, and this effect lasted long after the drug was cleared from the body (25).

Determining the effects of psychedelics on SV2A density in living humans has obvious implications for stratifying patient populations, but it could also drastically accelerate drug discovery efforts to develop the next generation of psychoplastogenic compounds. Psychedelics have previously been shown to increase the expression of presynaptic proteins in cortical cultures (6), and if they have similar effects on SV2A density, SV2A could serve as a truly translatable biomarker of psychoplastogen efficacy providing a direct link between experiments performed in vitro, in animals, and in humans. Currently, high-throughput methods for measuring psychedelic-induced structural plasticity are lacking, and most studies have focused on quantifying changes in postsynaptic structures such as dendrites and dendritic spines. In contrast to dendritic branching and spine quantification, measuring SV2A density is relatively straightforward and can be accomplished using standard high-content imaging data analysis software. Importantly, if SV2A is truly a translatable biomarker of psychedelic-induced neuroplasticity, it could be used to conduct high-throughput, concentration-response studies in cortical cultures, enabling the calculation of potency and efficacy parameters that could be used to prioritize novel compounds for further development. These in vitro studies could easily be confirmed in animals by administering a psychoplastogen prior to measuring SV2A density at a later time point via immunohistochemistry.

By focusing on phenotypic changes using a biomarker of structural neuroplasticity, the psychoplastogen field could embrace polypharmacology rather than prioritizing compounds based on efficacy at a single target. There are multiple receptors that lead to activation of cortical neuron growth pathways (16, 17), and simultaneous targeting could have additive or synergistic effects. Moreover, a translatable biomarker of structural neuroplasticity like SV2A could be used to optimize drug dose and dosing frequency in humans. In rodents, psilocybin can increase cortical spine density for over a month (26), but durability is likely to be compound specific. Optimizing dosing frequency is very important given that a single dose of the psychedelic DMT leads to robust cortical neuron growth (6) while chronic intermittent administration may begin to engage homeostatic mechanisms leading to either no change in spine density or slight neuronal atrophy (27). While we found that chronic intermittent administration of DMT still produced antidepressant and anxiolytic effects in rodent assays in the absence of obvious increases in spine density (27), we hypothesize that more frequent or longer dosing might become detrimental in much the same way that chronic stress results in negative outcomes. Finally, having an objective biomarker of structural neuroplasticity could enable psychoplastogens to more easily move into conditions beyond mood disorders such as schizophrenia and neurodegenerative diseases where cortical atrophy is a hallmark.

While early evidence suggests that SV2A may be a translatable biomarker of psychedelic-induced structural neuroplasticity, it has several drawbacks. First, loss of presynaptic markers might only be obvious in the most severely ill patients (19). Postsynaptic structures might be more sensitive to both disease state and the effects of drugs. Given that the majority of psychedelic-induced structural neuroplasticity studies have focused on changes in postsynaptic structures like dendritic spines, identifying a robust biomarker of postsynaptic changes will be useful. Second, [¹¹C]UCB-J PET imaging is very expensive, making it impractical to use as a widespread diagnostic tool. Thus, it will be important to determine if easily measurable clinical phenotypes correlate with synaptic deficits or find alternative biomarkers of psychedelic-induced neuroplasticity that are cheaper and more easily implemented on a large scale such as electroencephalographic (EEG) correlates of neuroplasticity (28). While an EEG-based biomarker would have significant advantages for routine clinical use, a molecular marker of psychedelic-induced neuroplasticity that can be measured in cell culture, in rodents, and in humans could increase confidence across all stages of the drug discovery process and drastically accelerate the development of novel psychoplastogens.