Introduction

Depressive disorders and available treatments

Depression is a common pathological condition affecting approximately 280 million people globally, constituting the second leading cause of disability worldwide due to its substantial occupational and economic impacts. It may clinically manifest through severe symptoms affecting various functional areas and can be particularly acute in recurrent or severe cases, potentially leading to suicide, especially among youths aged 15-29 [1]. The Diagnostic and Statistical Manual of Mental Disorders-Fifth edition Text Revision (DSM-5-TR) [2] removed the broad category of mood disorders and classified depressive disorders separately from bipolar disorders. Nonetheless, from a phenomenological point of view, depression can present with i) a single episode depressive disorder; ii) a recurrent depressive disorder; and iii) bipolar disorder. Furthermore, other related conditions are: a depressed mood lasting at least two years in persistent depressive disorder (or dysthymia); postpartum depression and psychotic depression, occurring in cases of severe depression, associated with psychotic features like disturbing false fixed beliefs (delusions) hallucinations [2]. This underscores the enormous eterogeneity of depressive disorders, which are probably related to different neurobiological underpinnings.

First-line non-pharmacological treatments for mild cases include psychoeducation, self-management, and psychological therapies, such as cognitive behavioural therapy or interpersonal psychotherapy [3].

Pharmacological treatments should be considered for mild depression, considering the patient’s preference, previous response to antidepressants, or lack of response to non-pharmacological interventions [3]. Other factors include the patient’s clinical features and dimensions, any comorbid conditions, and intrinsic drug-related factors (e.g., comparative efficacy/tolerability and interactions with other medications). Currently, the first-line recommendations for pharmacotherapy for Major Depressive Disorder (MDD) include: selective Serotonin Reuptake Inhibitors (SSRIs); Serotonin and Noradrenaline Reuptake Inhibitors (SNRIs) like venlafaxine, and duloxetine; the Noradrenaline and Dopamine Reuptake Inhibitor (NDRI) bupropion; the Noradrenergic and Specific Serotonergic Antidepressant (NaSSA) mirtazapine; and the Serotonin Modulator and Stimulator (SMS) vortioxetine [3]. Recommended second-line agents include Tricyclic Antidepressants (TCAs) and the Serotonin Antagonist and the Reuptake Inhibitor Antidepressant (SARI) trazodone, which show all more side effects compared to first-line treatments [3]. Finally, due to the higher side-effect burden and potentially severe drug and dietary interactions, third-line recommendations include monoamine oxidase inhibitors (MAOIs) [3]. ‘Adjunctive strategies’ refers to the addition of a second medication to the initial medication, i.e., adding a second antidepressant to the first or adding another medication that is not an antidepressant (i.e., aripiprazole, lithium, or quetiapine, etc.), taking into consideration both efficacy and tolerability [4].

Despite the availability of various therapeutic options, approximately 30-50% of patients with depression do not achieve complete symptom remission even after multiple treatment steps [5]. This has led to a significant ‘revolution’ in the field of antidepressant therapies over recent years, with the exploration of new, rapid-acting treatments based on pharmacodynamic mechanisms that differ from conventional monoaminergic treatments.

The “antidepressant revolution”: ketamine/esketamine and psychedelics renaissance

Despite structural and pharmacological differences, arylcyclohexamines (i.e. ketamine/esketamine) and tryptamines (e.g., LSD, psilocybin, ayahuasca, and dimethyltryptamine/DMT) have high evidence for treating depressive disorders. Their respective pharmacodynamic mechanisms goes beyond the mono-aminergic theory of conventional antidepressants, involving either glutamatergic, opioidergic, dopaminergic and serotoninergic systems [6, 7], but resulting in a shared mechanism of increase brain plasticity via BDNF/mTOR pathways [8].

The non-competitive N-methyl-D-aspartate receptor (NMDAr) antagonist ketamine, which has long been used as a general anaesthetic, produces remarkable results in patients with treatment-resistant depression (TRD), post-traumatic stress disorder (PTSD), bipolar disorder, obsessive-compulsive disorders [9,10,11]. An intranasal formulation of esketamine, currently branded as Spravato®, has been approved by the Food and Drug Administration (FDA) in the United States in March 2019 and by the European Medical Agency (EMA) in December 2019 for TRD [12, 13], and seems to be effective also in complicated populations with multiple comorbidities [14,15,16]. Despite the broad efficacy of ketamine/esketamine in clinically heterogeneous samples, there remains an absence of definitive indicators regarding clinical and biological markers of efficacy [17], with only preliminary data available from retrospective studies [18].

Classical psychedelic drugs, including psilocybin, lysergic acid diethylamide (LSD), and mescaline, have been extensively used in psychiatry before being banned as compounds listed in Schedule I of the United Nations Convention on Drugs in 1967 [19,20,21]. In the current resurgence of research into psychedelics, these substances are re-emerging as potential clinical therapies for treating various disorders. They present a model of single-dose, rapid-effect interventions that demonstrate substantial efficacy in treatment-resistant mental disorders, offering a distinct benefit as a potential monotherapy for mental illnesses [22]. Since 2006, there have been several trials using psychedelics (especially psilocybin) in non-psychotic psychiatric disorders, such as depression, anxiety, PTSD, and addiction, showing initial evidence of safety and efficacy, especially in the context of extensive psychological support, i.e., assisted psychotherapy [20, 23,24,25].

Some studies have suggested that psilocybin may induce profound psychological and spiritual effects, referred to as ‘mystical experiences’, such as high empathy, a sense of connection with the universe, and temporary alterations of thought patterns, mood, and time/space perception. These effects are considered potential facilitators of the antidepressant effects of psychedelics like psilocybin [26]. Nonetheless, this approach would be limited by several factors, including i) personal patients’ expectancies which might condition primary outcomes of drug response; ii) generalisability; iii) legal and regulatory barriers; iv) long-term effects of psychedelics; and v) ethical issues [19, 20].

Macroscopic and microscopic structural changes in MDD

Magnetic resonance imaging (MRI) and neuroimaging in general, have been used to investigate the neurophysiological changes associated with MDD. These approaches have the advantage of being non-invasive and allow repeated measures. Gray matter changes have been identified in several areas involved in depressive disorders (i.e., increased cortical thickness of frontal and parietal lobes, as well as decreased thalamic, caudate, pallidum, and putamina volumes) (see Table 1 for details) [27]. Furthermore, neuroplasticity, i.e., the capacity of the nervous system to adapt its activity in response to intrinsic and extrinsic stimuli, resulting in structural and functional reorganization of its connections, is thought to be involved in the pathophysiology of depression. The enhancement of neuroplasticity in the hippocampus and prefrontal cortex and the modulation of glutamate transmission are two antidepressant mechanisms activated by rapidly-acting compounds like ketamine, esketamine, and tryptamines [28].

Table 1 Brain region involved in Major Depressive Disorder (MDD).
Full size table

Brain networks involved in depressive disorders

A growing body of evidence supports the notion that depressive disorders are associated with widespread network dysconnectivity rather than aberrant alterations of individual brain regions [29] (Table 2). Different studies have coupled neuroimaging data on functional brain connectivity with specific clinical features of MDD to phenotype depressed patients based on clinical and neurobiological markers [30, 31].

Table 2 Brain networks involved in Major Depressive Disorder (MDD) and their clinical correlates.
Full size table

Four core networks have been frequently implicated in depressive disorders: the Affective Network (AN), the Reward Network (RN) - both part of the so-called Salience Network of the Triple Network Model - (Menon, 2020), the Default Mode Network (DMN), and the Central Executive Network (CEN) [29]. The role of these networks in depressive disorders is depicted in Table 2. Overall, aberrant connectivity of the AN has been liked to negative affection and dysphoria [32, 33], while enhanced DMN activity has been related to self-referential thoughts and depressive ruminations [34], frequently related to a prior history of trauma exposure [35]. The hypoactivation of the RN has been associated with anhedonia, lack of interests and reduced motivation [36, 37]. Finally, diminished brain connectivity of the CEN seems to underlie inadequate cognitive control, particularly regarding negative thoughts and emotions [38, 39].

Aim of the study

Overall, there is still a lack of evidence on the use of tryptamines and ketamine/esketamine as therapy concerning changes in functional connectivity (FC) and brain networks, their safety, outcomes, and duration of treatment. Therefore, this systematic review investigates changes in neuroimaging features in MDD, TRD, or bipolar disorder patients treated with tryptamines (e.g., LSD, psilocybin, ayahuasca, and dimethyltryptamine/DMT) or arylcyclohexamines (ketamine/esketamine).

Finally, the study analyses existing protocols, treatment outcomes, and the safety of these treatments, describing any recorded efficacy/tolerability problems.

Materials and methods

Data extraction

The review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines of 2015 [40, 41]. A literature search was performed using Pubmed (via Medline), Web of Science and Scopus databases on March 1st, 2022, without any time restrictions. We used the following search string: (fMRI or MRI or functional magnetic resonance or neuroimaging or magnetic resonance or PET or positron emission tomography or SPECT or single photon emission computed tomography) and (psychedelics or psilocybin or ayahuasca or LSD or lysergic acid diethylamide or dimethyltryptamine or DMT or tryptamines or ketamine or esketamine) and (treatment-resistant depression or depression or MDD or depressive symptoms) not review not meta-analysis not animal not in vitro. only original articles reporting data in patients affected from depression (i.e. MDD, Bipolar Depression and TRD) and treated with tryptamines (i.e., LSD, psilocybin, ayahuasca, and DMT) or ketamine/esketamine written in English were selected. In addition, only studies evaluating brain circuit alterations through specific neuroimaging techniques (i.e., Positron Emission Tomography/PET, Single Photon Emission Computed Tomography/SPECT, MRI, functional magnetic resonance imaging/fMRI) were included. Experimental and observational studies, post-marketing surveillance reports, case reports, case series, and fatality reports were included. the exclusion criteria included non-original research (e.g., reviews, commentaries, editorials, book chapters, and letters to the editor), non-full-text articles (e.g., meeting abstracts), works in a language other than English, and studies involving animal/in vitro experiments. Furthermore, articles not assessing post-treatment variation through PET/SPECT/MRI investigation were excluded. Although letters to the editor, conference proceedings, and book chapters were excluded from the literature review, they were still considered to retrieve further secondary references. We also performed secondary searches using the reference lists of all eligible papers.

Data synthesis strategy

The search for results was carried out individually by six investigators (L.C., F.M.S., R.T., F.M., D.D.B., and A.M.) and supervised by S.C., G.D.A, F.D.C., A.MI., and M.P., while G.M. and M.D.G. discussed unclear cases. The selection and eligibility phases of the articles were carried out independently by the two selected members, and afterward, the papers were subjected to a final cross-check. Any questions not solved by the team related to understanding the topic covered in the articles were requested directly from the authors if they were contactable. Data were collected in a word table containing the first names and year of publication of the study, the design, demographic variables (e.g., gender, age, and psychiatric history), details on the intervention drug (dosage and route of administration), and any other drug in combination, as well as the drugs’ antidepressant effect (i.e. clinical outcome), neuroimaging findings, as well as any adverse event recorded. Data synthesis was carried out by five independent members of the team (F.M.S., R.T., F.M., D.D.B., and A.M.) and compared at the end of the extraction process. All titles/abstracts were examined in the first selection phase, and full texts of potentially relevant papers were obtained and evaluated. Relevant works were selected to obtain a full representation of the available literature data on the chosen topic.

From an initial list of 890 studies (PubMed = 369; Scopus = 252; WOS = 269), duplicates (n = 256) were eliminated, as were papers unrelated to the topic (n = 411) and those that did not meet the inclusion criteria (e.g., non-original articles, n = 131, non-English papers n = 11). This yielded records relevant for screening. Studies deemed irrelevant to the topic considering the title and abstract (e.g., animal/in vitro studies; articles that did not deal with diagnosis of depression, specific neuroimaging techniques, e.g. PET, SPECT, fMRI, MRI, and the therapeutic use of tryptamines or ketamine/esketamine). Of the 81 full-text articles assessed for eligibility, 28 articles did not meet the inclusion criteria, and 4 were unavailable. Therefore, 49 articles were eventually considered and analysed (Fig. 1).

Fig. 1
figure 1

Prisma Flow Diagram.

Full size image

Results

Findings are reported according to the class of substance, type of study, and alphabetical order (see Table 3). Of the selected articles, the majority were related to ketamine/esketamine (n = 44), while fewer (n = 5) were related to psychedelic tryptamines, specifically one to ayahuasca and four to psilocybin. From the total 49 studies, 9 were randomized-controlled trials (RCT), 25 were open-label studies, 4 were double-blind trials, 8 were observational studies, and 3 cross-over studies.

Table 3 Main findings of retrieved studies.
Full size table

Specifically, the sample consisted of 687 patients suffering from MDD, 598 from TRD and 95 from bipolar disorder. All the study cohorts included adult subjects (mean age ranged from 30.2 to 51.13 years) except for one study, which included an adolescent cohort [42].

Most studies (n = 24) employed fMRI to investigate ketamine-related neural changes, while a few reported by magnetic resonance spectroscopy data [43,44,45,46,47], SPECT [48], or PET [49,50,51,52,53,54,55,56,57]. Seven studies only evaluated structural brain changes (i.e., brain volumes, cortical thickness, grey- or white-matter changes) [58,59,60,61,62,63].

Clinical measures included the Montgomery-Asberg Depression Rating Scale (MADRS) in n = 21 studies, the Hamilton Depression Scale (HAM-D) in n = 19 studies, the BDI in n = 8 studies, the Quick Inventory of Depressive Symptomatology-self-report (QIDS-SR) in n = 4 studies, and the Snaith-Hamilton Pleasure Scale (SHAPS) in n = 6 studies.

Clinical outcomes

In the only study evaluating the use of ayahuasca in 17 MDD subjects, depressive symptoms, assessed using the HAM-D, the MADRS, and the Brief Psychiatric Rating Scale (BPRS), significantly decreased from 80 min to day 21 [48].

In the four studies concerning psilocybin [64,65,66,67], a significant reduction in depressive symptoms (BDI, HAM-D, QIDS-SR) compared to baseline was described 4 and 5 weeks after treatment. Besides, all four studies reported mystical experiences associated with psilocybin administration [64,65,66,67]. However, a single study [64] indicates a significant correlation between these experiences and neuroimaging changes in individuals with treatment-resistant depression.

Ketamine administration reduced BDI [43, 45, 52, 68, 69], HAM-D and MADRS [47, 50, 52, 59, 62, 68,69,70,71,72,73,74] scores compared to placebo. Furthermore, several studies highlighted the anti-anhedonic action of ketamine, with a significant reduction of SHAPS scores after multiple ketamine infusions [53, 54, 62, 75, 76].

Safety and tolerability

In relation to the selected studies, vomiting (47%) and dissociative symptoms were recorded after the administration of ayahuasca (from 40 to 80 min) [48]. Psilocybin has never been associated with adverse events, whereas in the case of ketamine, fluctuating sensations were reported, dissociation, dizziness/nausea, chest tightness, etc [47, 55, 57]. Moreover, one suicide, one active suicidal ideation, and one antidepressant abuse were recorded upon ketamine treatment [46]. Finally, other ketamine-related symptoms such as fatigue, inattention, sedation, light-headedness, restlessness, and palpitations were reported [74]. Other symptoms included transiently impaired vigilance and increased blood pressure (which did not require pharmacologic intervention) [47, 52, 69]. Those effects were not related to the use of concomitant oral treatments.

Neuroimaging findings

Ayahuasca

In a single open-label study [48], SPECT was used to assess the effects of ayahuasca on brain connectivity in individuals with recurrent depression. The study found, after ayahuasca intake, increased perfusion in the left nucleus accumbens, right insula, and left subgenual area, all regions involved in regulating mood and emotion.

Psilocybin

Studies on psilocybin found significant changes upon drug intake, mainly regarding DMN areas and – among subcortical structures- the amygdala. For instance, a significant relationship between decreased cerebral blood flow in the left amygdala and reduction of depressive symptoms was shown at the fMRI, thus confirming a significant role for this structure in the improvement of emotion processing and mood balance, even upon pharmacological modulation. Interestingly, the reduction of the resting-state FC between this structure and DMN-related cortical regions was also linked to mystical experiences, possibly due to the role of the amygdala in the induction of mindful states [64, 67, 77]. The centrality of the amygdala in psilocybin-driven neural and clinical changes was also observed with a face/emotion perception fMRI task [66]. These neuroimaging findings showed decreased FC between the ventromedial prefrontal cortex (VMPFC) and the right amygdala correlating with lower rumination levels, while increased FC between the amygdala and occipital regions correlated with a reduction of depressive symptoms at one week. Furthermore, increased right amygdala FC specifically during the processing of fearful faces task, significantly correlated with the therapeutic response and remission of BDI scores [66]. Further studies should focus on the predictive role of face/emotion fMRI tasks on treatment response before clinical changes occur.

From a cortical point of view, an increased resting-state FC between the VMPFC and the bilateral inferior lateral parietal cortices predicted treatment response at one week, while decreased post-treatment FC between para hippocampal and prefrontal cortices predicted 5-weeks responses [64]. This finding is consistent with the role of VMPFC in emotion processing – an activity strictly coupled with the amygdala. Finally, FC between the dorsal anterior cingulate cortex (ACC) and the posterior cingulate cortex (PCC) was investigated concerning psilocybin-driven changes in cognitive flexibility. However, the two changes appear not correlated [65].

Ketamine/esketamine

Several reports on fMRI changes after ketamine/esketamine intake highlighted the role of the lateral and medial prefrontal cortices, whose alterations have been typically linked to MDD cognitive – but also affective – dysfunctions [29]. Following ketamine administration, increased global brain connectivity regression (GBCr) in dorsal prefrontal cortices (PFC), both lateral and medial (DLPFC, DMPFC), was documented in TRD patients - compared to healthy controls [43, 78]. This finding indirectly indicates a restoring of physiological functioning of prefrontal areas, since decreased PFC GBCr has been regarded as a graph-theory-based signature of conditions related to chronic stress, including PTSD and MDD [79], possibly related to impaired glutamatergic transmission [78].

A position paper suggested the existence of a “Cognitive dyscontrol” biotype of depression, where a hypoconnectivity between the ventral ACC and the DLPFC, along with hyperconnectivity between the dorsal part of the ACC and the DLPFC may promote inhibition deficits linked to complex attention dysfunctions and the incapacity of suppressing DMN-related processes like rumination [80]. After ketamine infusion, a reduced FC between the left dorsal and right ACC and between the right DLPFC and right frontal pole was shown [70], thus suggesting a normalization of the hyperconnectivity between the dorsal ACC and the DLPFC in MDD [80]. The studies by Sahib at al. confirmed the modulation of the brain areas involved in response inhibition, like the DLPFC and the dorsal ACC, documenting their altered FC with several other networks, like the DMN, the FPN, the DAN, and the SN within the right hemisphere, after ketamine administration [60]. Similarly, increased FC between the subgenual ACC and several cortico-subcortical brain regions, like the right anterior PFC in ketamine responders [81], or with caudal anterior cingulate nuclei and anterior insulae, was frequently observed upon ketamine administration [74].

The ventrolateral prefrontal cortex (VLPFC) is also involved in cognitive tasks and emotion control, participating in object-related working memory functions (as opposed to spatial-related ones, which are mostly controlled by the DLPFC) [82]. VLPFC atrophy can be observed in depressed patients, and the activation of the left VLPFC has been linked to abnormal processing of negative affective content [29]. In this context, increased connectivity after ketamine infusions between prefrontal areas (as VLPFC) and subcortical striatal regions (i.e., caudate) has been reported in several fMRI studies [63, 76, 78, 83]. Increases in FC between dorsal caudate-right VLPFC immediately after ketamine infusion, or 10 days after increased dorsal caudate-perigenual ACC connectivity significantly correlated with clinical improvement of anhedonic symptoms (SHAPS score) [76]. Furthermore, other studies [63, 78, 83] indicated global FC increases between right caudate, nucleus accumbens and prefrontal areas, which were significantly related to antidepressant response. Additionally, activity in other subcortical regions associated with fronto-striatal networks and reward-related activity (e.g., ventral striatum; ventral tegmental area) were enhanced in response to reward stimuli after ketamine infusion in MDD subjects [69].

The amygdala participates in several functions, including emotion processing, cognitive control, and elaboration of pain stimuli, through highly flexible interactions with the CEN, SN, and DMN [29]. Disrupted FC of the amygdala has been observed in several conditions, including generalized anxiety and depressive disorders. Thus, the observation of significant changes in the amygdala’s FC upon ketamine administration, mostly involving the CEN, is in line with the literature. In particular, increased FC was observed between the right amygdala FC and the CEN, while decreased FC with the left CEN was documented at the subsequent follow-up. Decreased FC was also observed between the left amygdala and SN regions and predicted anxiety improvements [62].

From a structural point of view, increased volumes of the left amygdala and the left cornu ammonis 4 body were documented after six ketamine infusions, as well as a positive correlation between the pre-treatment volumes of the right thalamus and left subiculum head of the hippocampus, and a subsequent improvement of depressive symptoms [84]. The same group described a significant improvement in depressive symptoms and a small increase in right hippocampal volume following ketamine infusions [85]. These findings were confirmed in a study highlighting the critical roles of the amygdala and hippocampal subfields in producing antidepressant effects after repeated ketamine treatment [63].

Changes in cerebral blood flow (CBF) after ketamine infusion have also detected in several fMRI studies. Specifically, increased CBF was documented in the thalamus, while decreased CBF was found in the lateral occipital cortex [71] Interestingly, a 24-h decrease in ventral basal ganglia’s CBF values was detected in non-responders, and changes in depression symptom scores negatively correlated with these values [71]. On the contrary, responders showed a correlation between symptom relief and a 6-h-decreased CBF within the medial prefrontal cortex (MPFC) [71]. Increased global CBF has been observed after the first ketamine infusion, with normalization – after the fourth infusion – within the cingulate and primary and higher-order visual association regions. Finally, regional CBF reduction was documented in hippocampi and the right insula after serial infusion [86].

On fewer occasions compared to fMRI, fluorodeoxyglucose PET (18FFDG-PET) has also been employed to investigate the effects of ketamine. The significant effect on the ACC region was also confirmed, since a dose-dependent increased activation of the dorsal ACC and the supplementary motor area (SMA), was found to negatively correlate with depressive symptom severity after ketamine infusion, compared to placebo [51]. Increased cerebral metabolic rates were also documented in the dorsal ACC and the ventral striatum (VST), linked to the anti-anhedonic effect of ketamine in bipolar depression [53, 54]. Similar correlations were also documented for depressive symptoms. Increased metabolic rates in the right VST correlated to the improvement in MADRS scores, and lower metabolism in the left hippocampus was directly associated with depressive symptom severity [56]. Moreover, reduced metabolic rate in the right habenula and the extended medial and orbital prefrontal networks was linked to clinical response to ketamine [50]; whereas, increased regional cerebral metabolic rate in bilateral occipital, right sensorimotor, left parahippocampal, and left inferior parietal cortices post infusion, was linked to the illusory phenomena experienced with ketamine [50]. Higher tracer uptake in the dorsal ACC, along with the prefrontal cortex, SMA, and precentral gyrus, than placebo also predicted treatment response, according to Li and colleagues [55].

Increased infralimbic cerebral metabolic rates positively correlated with baseline suicidal ideation in TRD patients. Accordingly, ketamine administration decreased the infralimbic metabolic rate while increasing it in a large posterior cluster encompassing regions from the left lingual gyrus to the left superior cerebellum and the lateral occipital cortex [49].

In addition to 18FDG, other tracers have been used by some research groups to investigate the various effects of ketamine through PET imaging. Esterline et al. found a significant ketamine-induced decrease in metabotropic glutamatergic receptor (mGluR5) availability through PET with [11 C]ABP688 in MDD and healthy subjects. Based on these findings, they hypothesized a functional link with the immediate glutamatergic surge that occurs after ketamine infusion. Furthermore, they reported a positive correlation between the antidepressant effects of ketamine and the decrease in mGluR5 availability [52].

MRI spectroscopy (MRS) techniques were only used in a few studies, whose results converged upon a dose-dependent effect of ketamine on glutamine-glutamate cycling. Reduction in MPFC glutamine levels after ketamine infusion was documented, possibly in relation to the compound’s antidepressant effect [46]. Accordingly, ketamine increased prefrontal glutamate-glutamine cycling compared to placebo, and the ratio of 13 C glutamate/glutamine enrichment, a putative measure of neurotransmission strength, correlated with Clinician-Administered Dissociative States Scale (CADSS) scores [43]. These findings were not observed in the occipital cortices, where baseline levels of glutamate, gamma-aminobutyric acid (GABA), and glutamine were not associated with significant changes in HAM-D scores [47].

Discussion

This is the first systematic review of the literature focusing on neural changes in depressed patients treated with tryptamines or ketamine/esketamine. Most of the data assessed by this review derive from open-label studies, a substantial number of RCT, observational studies, double-blind trials, and cross-over studies were also included. In the following sections, we discuss the results of the review and how tryptamines and ketamine/esketamine can modulate brain networks to counteract depressive symptoms.

Tryptamines

Concerning the antidepressant effect of ayahuasca and its neuroimaging correlates, a single, open-label study on the subject does not permit us to reach meaningful conclusions [48]. Multiple studies support the potential antidepressant efficacy of ayahuasca, with recent data also generated from placebo-controlled RCT [87] that indicate safety and low risk of side effects. Therefore, the therapeutic use of ayahuasca in mood disorders is promising but more studies are needed to evaluate the neural correlates of antidepressant responses.

Psilocybin antidepressant mechanism relies upon the modulation of intra- and inter-network connectivity of the DMN. This compound can restore the DMN physiological activity and produce a transient ‘hyperconnectivity state’ that allows a brain resetting and the departure from rigid and negative network states that produce depressive disorders [88, 89]. In fact, DMN hyperactivity, a trait feature of depressive disorders [90, 91], is, therefore, a primary effect of psilocybin that determines an overall reduction of relapses and inhibition of recurrent depressive episodes. This hypothesis is in line with studies indicating a long-term and sustainable antidepressant action of this compound [92], lasting several months after the first administration. In support, a growing body of evidence indicates that hyperactivity of the DMN is associated with rumination, mind-wandering, and self-referential thoughts in MDD [34, 80]. These features are considered core elements related to the response to psilocybin [93, 94], being suggestive of an effectiveness of psilocybin in MDD patient subtypes, such as those with ruminative depression.

Despite the presence of numerous studies on the acute effects of psychedelics, little is known on the relation between the character of the acute psychedelic experience (i.e., the way the experience goes, not simply the intensity) and the persisting changes in mood, behaviour, and personality [26, 95]. Concerning the ‘psychedelic state’, psychedelics appear to dysregulate cortical activity, producing an ‘entropic’ brain state characterized by compromised modular but enhanced global connectivity [88, 89]. These effects have been found to correlate with significant aspects of the ‘psychedelic experience’, including ‘ego-dissolution’, and were predictive of post-acute changes in the personality domain of ‘openness’. Such experiences, known as “Complete Mystical Experiences” (CME), are hypothesized to be instrumental in producing long-lasting positive effects of psychedelics [96]. The magnitude of psilocybin-induced mystical-type experience positively correlates with improvements in subjective life quality, meaning in life, and mood in patients suffering from anxiety and depression and persisting positive effects 12 months after psilocybin [97].

Finally, psilocybin intake can impact cognitive processes, improving cognitive flexibility and possibly relieving the cognitive disorders that frequently complicate MDD.

Ketamine

With a prominent effect on prefrontal areas (i.e., VMPFC, DLPFC, ACC), confirmed by multi-modal neuroimaging studies, ketamine can normalize the MDD-relate cortico-subcortical dysconnectivity by enhancing long-distance connectivity and restoring the central role of the prefrontal cortices in cognitive and emotional processing. The subcortical structures which show a more prominent response to ketamine modulation are the amygdala and the caudate nucleus. The modulation of PFC and caudate correlates with treatment success. Post-treatment FC changes in these regions show more central and balanced features in responders. Prefrontal and striatal structures play a critical role in higher cognitive control, particularly in exploration and goal-directed behaviour. Thus, the enhanced engagement of these frontostriatal regions could underlie the behavioural shift from depression, withdrawal, and rumination to exploratory and externally focused behaviour following recovery [78]. Furthermore, the modulation of glutamatergic cycling, suggested by MRS studies, indicates that ketamine administration can restore glutamatergic signalling involved in key top-down circuits.

Indeed, top-down circuits such as frontostriatal networks appear to be related to specific clinical features, such as anhedonia, a core and difficult-to-treat dimension of depressive disorders [98]. Historically, frontostriatal neural networks associated with hedonic functioning (i.e. RN) have been linked to the activation of dopaminergic pathways [99]. Nevertheless, recent research has suggested that glutamate neurotransmission can exert indirect control over these pathways through a top-down action exerted on striatal areas [100]. In this context, several studies here analysed [53, 54, 62, 63, 69, 76, 78, 83] indicated that multiple ketamine infusions can boost connectivity between striatal areas (e.g., VST, nucleus accumbens and right caudate nucleus) and frontal cortical regions (e.g., VLPFC, ACC, DLPFC), thus leading to a decrease in anhedonic symptoms [53, 54, 62, 69, 76]. Further investigations are needed to elucidate the precise mode of action of ketamine and its derivatives on hedonic system-related circuits, as well as the neurobiological correlates of their rapid anti-suicidal effect, which has been demonstrated in several clinical studies on this matter [101, 102]. This research could be highly beneficial in determining patient populations most likely to respond to ketamine and esketamine.

Network-wise, several studies point out that ketamine modulates the FC between limbic and resting state networks implicated in MDD, as well as between limbic regions and the CEN [62, 63, 72, 74, 75, 103]. These observations indicate that ketamine plays a pivotal role in restoring top-down control of emotion processing, possibly through restoring physiological and flexible cortico-subcortical connections involving the amygdala. This activity is related to the ketamine-induced increase of cognitive control over emotional stimuli, a process that reduces negative affect mood stages [60, 62].

In summary, compared to psychedelics, ketamine has a more distinct impact on networks that involve prefrontal areas (i.e., CEN, RN, AN), thereby indirectly affecting the DMN and inter-network connectivity (see Fig. 2). Its specific action on prefrontal-striatal glutamatergic pathways may explain ketamine’s efficacy as an anti-anhedonic treatment [53, 54, 62, 69, 76].

Fig. 2: Effects of antidepressant treatments with ketamine and psylocibin on brain networks.
figure 2

While Ketamine seems to act directly on prefrontal areas and related circuits (e.g., CEN, RN, and AN), psylocibin seems to operate directly on DMN areas, with a specific action on restoring DMN activity. Am, Amigdala; AN, Affective Network; ANG, Gyrus Angular; Cau, Caudate; CEN, Central Executive Network; dACC, dorsal Anterior Cingulate Cortex; DLPFC, Dorsolateral Prefrontal Cortex; DMN, Default-Mode-Network; HIP, Hippocampus; INS, Insula; mPFC, medial Prefrontal Cortex; NAc: Nucleo Accumbens; OFC, Orbito Frontal Cortex; PCC, Posterior Cingulate Cortex; PCUN: precuneus; RN, Reward Network; vACC, Ventral Anterior Cingulate Cortex.

Full size image

In general, the application of novel therapies is justified only if the risks do not outweigh the benefits. Concerns are still being expressed when considering the therapeutic applications of psychedelics. Of note, no addictive potential was recorded in the present review. Nevertheless, significant concerns persist regarding the potential for abuse of these compounds [104, 105]. Future studies should rigorously address this issue, particularly by evaluating subjects with depressive disorders and comorbid substance use disorders.

Furthermore, there is a slight risk for certain patients to experience sporadic adverse psychological effects, including the risk of developing psychotic symptoms. Thus, to minimize risks, clinical trials have excluded participants with a family history of first-degree relatives affected by psychiatric disorders like psychosis and schizophrenia [89]. However, future studies should include patients with psychotics features to better inform if those treatments are safe and effective in this subpopulation [106].

Study limitations

The present review is the first systematic assessment of scientific literature on the use of ketamine, esketamine, and tryptamines for depressive disorders. The first limitation of the current review concerns the predominance of heterogeneous studies, small sample sizes, and the high rate of descriptive studies. Given this heterogeneity and the scarcity of RCTs or double-blind studies, it was impossible to precisely assess the studies’ quality or carry out a meta-analysis. Moreover, the evaluated studies had a limited duration of follow-up. Therefore, estimating the long-term benefits and/or potential long-term side effects produced by the reviewed compounds was impossible. Lastly, this review only included studies published in English.

Conclusions

The main structural and functional changes involved prefrontal regions (e.g., DLPFC, VLPFC, ACC), as well as subcortical structures (i.e., amygdala and caudate nucleus). Accordingly, the effects of these compounds mostly targeted key hubs of large-scale networks, like the CEN, the AN, the RN, and more directly (tryptamines), or indirectly (ketamine), the DMN. In particular, psylocibin’s mechanism of action leads to a ‘reset’ of DMN activity, whereas ketamine shows instead a more specific action on the networks encompassing prefrontal areas, exerting only an indirect modulation of the DMN. The overall evidence provides a neurophysiological basis for ketamine’s efficacy as anti-anhedonic treatment, and its critical role in increasing cognitive control over emotional stimuli, thus reducing negative affect mood stages.