Introduction

The discovery of racemic (R, S)-ketamine (hereafter referred to as ‘ketamine’) as a rapid-acting antidepressant—and the subsequent regulatory approvals for treatment-resistant depression (TRD) and suicidal ideation of its (S)-enantiomer, esketamine, by the US Food and Drug Administration (FDA) and other major pharmaceutical regulatory agencies around the world—ushered in a new era for psychiatric therapeutic development. Ketamine, canonically an N-methyl-D-aspartate receptor (NMDAR) antagonist, was first synthesized by Parke-Davis in 1956 and developed in 1962 as a short-acting sedative to replace phencyclidine. In the ensuing decades of the 20th century, it was primarily used as an anesthetic in adult, pediatric, and veterinary settings [1]. Given the potential anesthetic and dissociative effects associated with ketamine use and abuse, its distribution is controlled in Schedule III of the Controlled Substances Act in the United States, though the World Health Organization (WHO) does not currently recommend its classification as a scheduled substance.

A surge of preclinical research into the potential antidepressant effects of MK-801, another NMDAR antagonist [2], spurred reevaluation of ketamine’s potential antidepressant effects. In a seminal clinical study, Berman and colleagues demonstrated that an intravenous (IV) ketamine infusion alleviated symptoms of depression within hours [3], a finding that was subsequently confirmed in TRD and bipolar depression [4, 5]. Since these initial findings, multiple large-scale randomized controlled trials (RCTs) have confirmed ketamine’s effectiveness as a rapid-acting antidepressant with both a clinical response and apparent neurobiological pathway distinct from traditional antidepressant pharmacology [6]. In 2019, FDA approval of Spravato, an intranasal (IN) formulation of ketamine’s (S)-enantiomer, broadened ketamine’s use as an antidepressant throughout the medical community; its relative success, despite implementation barriers, has inspired further pharmaceutical development beyond the oral “me-too” medications that have long dominated this space.

Ketamine is highly water- and lipid-soluble, allowing it to rapidly cross the blood-brain barrier [7]. Its binding site is located deep within the NMDAR channel, necessitating activation of the receptor and removal of the magnesium block before ketamine can exert its antagonism [8] through an open-channel block that prevents the movement of ions through the channel [9]. Ketamine is metabolized within minutes through nitrogen-mediated demethylation driven primarily by the action of cytochrome P450 liver enzymes [10, 11]. Norketamine, ketamine’s first major metabolite, can subsequently be metabolized to either hydroxynorketamine (HNK) or dehydroxynorketamine (DHNK). With a half-life of around two to three hours, the elimination of ketamine is around equal to that of liver blood flow (12-20 ml/min/kg) [12, 13], though some research suggests that women have an approximately 20% higher clearance rate [14]. Ketamine’s antidepressant effects last significantly longer than its NMDAR blockade and rapid clearance, suggesting the activation of downstream signaling cascades that cause lasting effects.

Initial studies of ketamine’s neuropsychiatric effects focused on IV racemic (R, S)-ketamine, which is composed of two optical enantiomers rotated around an asymmetric second carbon of a cyclohexanone radical: (S+) and (R-)-ketamine (Fig. 1). However, each enantiomer holds a distinct pharmacokinetic profile. (S)-ketamine is demethylated at a significantly higher rate than (R)-ketamine or racemic ketamine [15], lending it a wider distribution pattern than its counterparts. While the exact reason underlying this pharmacokinetic difference is unknown, research has suggested that CYP3A5 (one of the main liver enzymes responsible for the demethylation of ketamine) is able to demethylate (S)-ketamine far more rapidly than (R)-ketamine [16]. As a racemic mixture, (R)-ketamine is also able to inhibit the metabolism of (S)-ketamine [17]. Although clinical enantiomer-specific research has been limited, preclinical models have found that prophylactic effects against chronic unpredictable mild stress and lipopolysaccharide injection differ between the enantiomers, and that there are additional differences between both their rapid-acting and sustained antidepressant-like effects [18]. Despite great promise in preclinical models, (R)-ketamine has yet to demonstrate significant clinical effects in Phase 2 clinical trials [19, 20]. In contrast, the success of esketamine in preclinical models has translated into observed clinical effects to address symptoms of TRD, leading to approvals of the intranasally-administered Spravato by the FDA and European Medicines Agency (EMA). In turn, the successful translation to approved treatment has further energized mechanistic studies and the search for putative biomarkers of antidepressant response to esketamine.

Fig. 1
figure 1

Adapted from [18]

The chemical structures of (S)-ketamine and (R)-ketamine and their metabolic pathways involve stereoselective metabolism by P450 liver enzymes. Initially, both enantiomers undergo nitrogen-mediated demethylation, converting them to their norketamine (NK) forms. Further metabolism results in either dehydroxynorketamine (DHNK) or hydroxynorketamine (HNK) through hydroxylation. Each enantiomer can also be metabolized to hydroxyketamine (HK), serving as another intermediary step towards HNK.

Full size image

Although animal models have been extremely useful in elucidating the molecular mechanisms of ketamine and its enantiomers [18], a full description of the preclinical evidence is beyond the scope of this current review. Nevertheless, considerable research into esketamine’s molecular mechanisms is germane to this focused discussion. Generally, both (S)- and (R)-ketamine seem to enhance the likelihood of glutamate release post-administration, thereby increasing α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptor (AMPAR) activity [6]. This increase in AMPAR activity triggers downstream effects, such as increases in brain-derived neurotrophic factor (BDNF) release and activation of mechanistic target of rapamycin complex 1 (mTORC1) or tropomyosin receptor kinase B (TrkB) signaling pathways [21]. Despite this consistent observation, uncertainty persists about the exact mechanisms through which ketamine and its enantiomers promote this glutamate surge. One prominent hypothesis is that ketamine selectively inhibits GluN2b-containing NMDARs, which are predominantly found on gamma aminobutyric acid (GABA)-ergic interneurons. This inhibition leads to the disinhibition of cortical pyramidal neurons, resulting in a glutamate influx into the synaptic cleft, where it binds to AMPARs and initiates downstream signaling cascades [22, 23]. Interestingly, while (S)-ketamine (Ki = 465 nM) shows high affinity for NMDARs, (R)-ketamine (Ki = 1340 nM) exhibits considerably lower potency. Another potential mechanism by which ketamine might facilitate glutamate release is through blockade of extrasynaptic NMDARs, leading to the dephosphorylation of eukaryotic elongation factor 2, which in turn disinhibits BDNF release. This disinhibition results in the increased insertion of GluA1 and GluA2 into the post-synaptic membrane, thereby inducing homeostatic scaling of the synapse and enhancing synaptic AMPAR signaling [24,25,26].

Repeated intraperitoneal administration of (S)-ketamine has also been found to counteract stress-induced deficits in behavior, neuronal structure, and hippocampal long-term potentiation. This effect is achieved through Rac1-mediated synaptic plasticity, which upregulates the expression of GluA1, PSD-95, and Synapsin I [27]. Other studies have indicated that (S)-ketamine might preferentially activate mu-opioid and kappa-opioid receptors more than (R)-ketamine, potentially contributing to its risk of misuse [28, 29] as well as its antidepressant effects [30,31,32,33]. Another mechanism critical for ketamine’s antidepressant effects is suppression of NMDAR-dependent burst firing in the lateral habenula (LHb). The LHb, generally known as an “anti-reward” center, is suppressed by ketamine, which contributes to its anti-anhedonic effects [34]. A single systemic injection of racemic ketamine is capable of blocking NMDAR activity in the LHb for up to 24 h, in contrast to its much shorter actions observed in other brain regions. Through neural activity-mediated manipulation of the rate of untrapping, researchers were able to prolong or shorten ketamine’s antidepressant-like effects [35]. In this context, while preclinical investigations have yielded significant insights into the molecular mechanisms of novel therapeutics, understanding the actions of these agents in clinical populations is particularly important. For example, one clinical study found that the mTORC1 inhibitor rapamycin extended, rather than blocked, ketamine’s antidepressant effects [36]. While this does not negate other preclinical findings, it demonstrates the importance of considering clinical translation within any attempt to evaluate pharmacological actions.

Given the measured pace of neuropsychiatric drug development, knowledge of potential clinical biomarkers for future adaptive trial designs would be invaluable. This review seeks to synthesize the evidence across clinical trials in order to assess the impact of esketamine on proposed biomarkers of depression and treatment response. Because this is a nascent field, there are areas in which racemic ketamine’s effects have been extrapolated to fill the gaps in our understanding. Further work is required to elucidate the putative biomarker effects that are specific to or can be attributed to esketamine and those that cannot.

Clinical updates

Esketamine’s first clinical approval in 2019 in an IN spray formulation (Spravato, Janssen Pharmaceuticals, Inc.) for TRD heralded what seemed to be the realization of decades of anticipation; in stark contrast to previous generations of pharmacological antidepressant treatment, esketamine had a dramatically short latency to response of hours or days rather than the weeks or longer timeline that occurs with conventional antidepressants. In 2020, its approval was expanded to treat individuals with major depressive disorder (MDD) with suicidal ideation or behavior. However, in practice, several barriers remain to its widespread use, including limitations to availability, access, and affordability [37]. In the US, esketamine includes a Boxed Warning highlighting the risk of suicidal thoughts and behaviors in pediatric and young adult patients taking antidepressants. As a result, a Risk Evaluation and Mitigation Strategy (REMS) was put into place to mitigate the risks of serious adverse outcomes resulting from sedation, dissociation, respiratory depression, or potential abuse or misuse, and the medication is available to patients only through healthcare providers at REMS-certified treatment centers. Similar barriers vary by country. For example, esketamine is listed as a Schedule 2 Controlled Drug in the UK, where it is subject to the complete Controlled Drug requirements related to prescribing, safe custody, and maintenance of a register [38]. In New Zealand, esketamine is classified as a Class C Controlled Drug (moderate risk) with specific prescribing and dispensing regulations [39], while in Australia it is controlled as a Schedule 8 drug, like ketamine and many narcotics, making it available with restrictions to reduce abuse, misuse, and dependence [40]. Even across EMA member states, the status of special and restricted medical prescriptions are categorized at the member state level [41]. Additional caveats for patients are that both approved indications require concurrent treatment with an oral antidepressant medication; prescribers must monitor patients for at least two hours after administration; and the multiple device activations needed to attain the therapeutic dose may make delivery cumbersome (for instance, in order to use the currently approved device, an individual prescribed the maximum 84 mg dose would need to receive six doses total administered across three nasal spray devices).

Although expert opinion and consensus statements recommend caution and use of ketamine and esketamine only after most other pharmacological approaches have been exhausted [42, 43], the use of both agents continues to grow. According to IQVIA National Prescription Audit, total prescriptions for esketamine dispensed in the US were approximately 87,274 in 2020, 145,475 in 2021, and 235,906 in 2022 [44]. A variety of delivery approaches also continue to be explored for ketamine and related molecules, including esketamine, in the context of other indications such as postpartum depression, anorexia nervosa, and substance use disorders [45,46,47,48,49]. Ketamine is widely available via mail-order from telehealth services as sublingual troches, inviting further questions about the balance between regulatory and patient safety landscapes versus practicality, patient comfort, and access to treatment [50].

The first evidence that ketamine could be safely and effectively delivered intranasally was a randomized, crossover, double-blind ketamine study of 18 patients [51]. The potential benefits of IN delivery include greater availability in a wider range of treatment settings, the need for fewer support services, and less monitoring and potential patient discomfort. IN ketamine had already been in use to treat headache [52] and chronic pain [53, 54] in ambulatory patients. Although few direct comparisons have been done, the magnitude of esketamine’s antidepressant benefit has generally been shown to be similar to racemic ketamine infusion [55, 56]. One non-inferiority trial found that IV esketamine (0.25 mg/kg) was non-inferior to racemic ketamine (0.5 mg/kg) at 24 h. At seven days there seemed to be a slight increase in remission rates in the racemic ketamine group compared to esketamine, but this was not statistically significant [57]. A retrospective comparative analysis of 210 patients found no differences in suicidality items, or in remission and response rates, between the two agents, although the study was unrandomized and unblinded [58]. An observational study of 62 adults in real-world clinical settings also found no difference in response or remission but did report a faster time to remission with IV racemic ketamine compared to IN esketamine [59]. When pooled data from three of these studies comparing IV ketamine to IN esketamine were combined, similar response and remission rates were observed, but response time for IV ketamine was quicker [60]. A systematic review of IN esketamine versus IV racemic ketamine demonstrated a similar number-needed-to-treat (NNT) for both treatments, with NNTs of two at one day and 11 at four weeks for esketamine, and NNTs of three at one to seven days and nine at four weeks [61]. In terms of dosing, IV esketamine at 0.2 mg/kg was superior to placebo in one trial [62], while IV racemic ketamine at 0.2 mg/kg was not superior in two other randomized trials, potentially due to low-dose administration not allowing for proper metabolite concentrations [63, 64]. Collectively, this suggests that there is insufficient evidence from which to strongly recommend one treatment modality over another from a clinical efficacy standpoint.

Landmark intranasal esketamine clinical studies

Development of ketamine’s (S)-enantiomer proceeded based on its reportedly higher affinity for the NMDAR than (R)-ketamine [65]. One of the first studies to assess the efficacy and safety of IN esketamine for TRD was a Phase 2, double-blind, placebo-controlled study across multiple outpatient referral centers; 67 individuals with TRD (defined as inadequate response to two or more antidepressants) were randomized to receive placebo or esketamine at 28, 56, or 84 mg twice weekly while continuing their current antidepressant regimen [66]. Most of those who received placebo were re-randomized to the four treatment arms. All three doses of esketamine were superior to placebo in reducing Montgomery-Asberg Depression Rating Scale (MADRS) total score after one week in a dose-dependent manner; this effect was sustained for up to nine weeks after decreasing frequency during the open-label phase to biweekly.

Another double-blind, multicenter, placebo-controlled study randomized 68 participants to receive IN esketamine alongside standard-of-care treatment. With twice weekly dosing, IN esketamine reduced MADRS scores at 4 h and 24 h after the initial dose, but not at 25 days [67]. The MADRS suicidal thoughts score at 4 h was also significantly reduced, but not at later timepoints, and there was no reduction in clinician global judgment of suicide risk at any timepoint. Dissociative symptoms, which peaked at 40 min after dosing and resolved by two hours, attenuated with repeated dosing over time. Three participants required a dose reduction due to intolerance, while five had adverse events leading to early discontinuation (agitation, aggression, unpleasant taste, and ventricular extrasystoles in one participant each, and dizziness, dyspnea, and nausea in one participant).

The TRANSFORM-1 trial, a Phase 3, double-blind, multicenter study, enrolled 346 adults with moderate-to-severe depression who had not responded to at least two antidepressant trials during the current depressive episode [68]. Participants were randomized to 56 or 84 mg of IN esketamine or placebo spray twice weekly alongside a newly initiated open-label oral medication. After four weeks, the combined esketamine groups together demonstrated a clinically meaningful effect versus placebo, although the primary endpoint specifically comparing 84 mg esketamine versus placebo was not met. The positive finding was confirmed in another Phase 3 efficacy trial (n = 197), where change in MADRS score at day 28 was significantly greater in those receiving flexibly dosed IN esketamine versus placebo [69].

As mentioned previously, the IN esketamine formulation Spravato was granted FDA and EMA approval in 2019 [70]. Subsequent FDA approval in July 2020 of esketamine to treat MDD with acute suicidal ideation or behavior was based on two identical Phase 3 trials. ASPIRE I, which was conducted between June 2017 and December 2018, assessed the effects of IN esketamine (84 mg, twice weekly for four weeks) or placebo on 226 participants with MDD and active suicidal ideation with intent. IN esketamine was given alongside comprehensive standard-of-care treatment, including initial psychiatric hospitalization and optimization or initiation of oral antidepressants. The primary endpoint was MADRS score 24 h post-initial dose, and significant decreases were observed for esketamine plus standard-of-care compared to placebo plus standard-of-care. These differences were also present at four hours post-dose and sustained through day 25 of treatment. The most common adverse events included dizziness, headache, dissociation, somnolence, and nausea. Depression- and suicide-related adverse events, which were considered unrelated to esketamine treatment, included three attempted and one completed suicide in those who received esketamine during the follow-up phase, and two attempted suicides in the placebo group [71]. ASPIRE II (n = 230), which used an identical study design, similarly found significant reductions in MADRS score compared to placebo at 24 h, which continued to favor esketamine at timepoints through day 25. Both groups experienced significant reductions of severity of suicidality and were not significantly different. The adverse events profile was similar to the first trial, with the addition of dysgeusia and paresthesia amongst the common complaints for those who received esketamine [72]. The ESCAPE-TRD open-label, randomized, multi-site trial compared esketamine to extended-release quetiapine, which is an approved adjunctive treatment for TRD; both agents were additionally compared to treatment with a selective serotonin reuptake inhibitor (SSRI) or serotonin-norepinephrine reuptake inhibitor (SNRI) during an initial eight-week treatment period as well as during a subsequent 24-week maintenance phase [73]. Of the 776 randomized patients, more patients in the esketamine group achieved remission at week 8. After 32 weeks of maintenance and follow-up, esketamine was favored with regard to sustained remission rates, response rates, and change in MADRS score from baseline.

Given that FDA approval was granted in 2019, data are beginning to accumulate with regard to esketamine’s long-term maintenance effects. For instance, post-hoc results from the SUSTAIN-2 trial suggest efficacy in adjusting treatment frequency in accordance with depressive symptoms. After the initial phase of twice-weekly dosage, participants received esketamine once per week for the following four weeks; 76% maintained clinical benefit or had further reductions in depressive symptom scores. Antidepressant response continued to be maintained in a similar proportion of individuals when the esketamine dose was further reduced to every other week; furthermore, in 90% of those who worsened, re-increasing the dose to once a week improved clinical benefit or stabilized mood ratings [74]. In addition, interim results from SUSTAIN-3, a long-term (2 + years) open-label study (n = 1148), suggest that esketamine has antidepressant effects that last throughout the optimization and maintenance periods, with persistent reductions observed in MADRS scores [75]. Roughly 31% of the participants in the SUSTAIN-3 study discontinued treatment during the optimization/maintenance phase for a variety of reasons, with no differences in discontinuation rates observed across length of time in the study. In addition, a SUSTAIN-3 subgroup analysis in those who experienced a second induction and maintenance period after relapse found remission rates around 61% throughout the second optimization/maintenance period, suggesting that potential benefits are associated with a second induction period [76].

Based on these seminal studies, current guidelines recommend co-administration of a traditional oral antidepressant with esketamine. Dosing for IN ketamine is typically 56 mg on day 1, with subsequent dosing of 56 mg or 84 mg per insufflation. The recommended frequency is twice a week for four weeks, followed by once a week for four weeks, then once every one to two weeks thereafter. Monitoring is recommended for at least two hours after administration, which may require reporting of vital signs and return to baseline functioning to a centralized monitoring system.

Putative clinical biomarkers research

Blood, saliva, and other measures

To date, few clinical studies have examined biomarkers associated with esketamine; the current literature is summarized in Table 1. For example, one systematic review published in 2022 found only two previous studies of blood-based putative biomarkers examining response to esketamine [77]. A genome-wide association study (GWAS) of the participants in the SUSTAIN-2 and TRANSFORM-3 Phase 3 trials found significant associations between clinical response to esketamine and interleukin-1 receptor associated kinase-3 (IRAK3). NME/NM23 family member 7 (NME7) was also associated with change in MADRS score as determined by gene-level association analysis. NME7 is a functional component of the γ-tubulin ring complex and involved in microtubule organization and cell division in various tissues [78], though little is known about its relationship to depressive symptoms. Pathway enrichment analyses also suggested that glucocorticoid metabolic processes (p = 3.53 × 10− 5) and neuronal action potential (p = 0.0001) were associated with change in MADRS score after treatment. The genetic loading for depressive symptoms was most strongly associated with esketamine efficacy as identified by polygenic risk score analysis, though this did not reach study-wide significance after multiple testing corrections (p = 0.001, standardized coefficient β=− 3.1, SE = 0.9) [79]. In a metabolomic profile, decreases in tryptophan metabolites (indole-3-lactate and indole-3-acetate) were observed two hours post-esketamine infusion, potentially implicating kynurenine signaling and the gut microbiome in esketamine’s mechanism of action [80]. Finally, a meta-analysis of metabolomic profiles after ketamine and esketamine treatment found changes in mitochondrial function and kynurenine signaling but emphasized the importance of considering time after administration of ketamine in analyses [81].

Table 1 Biomarkers from clinical studies of esketamine
Full size table

Another meta-analysis demonstrated significant, but transient and relatively small, effects for both ketamine and esketamine on increases in systolic and diastolic blood pressure, as well as heart rate [97]; no significant differences were observed between ketamine and esketamine. In healthy volunteers, esketamine almost doubled production of saliva and plasma cortisol expression over a three-hour period in a manner consistent with circadian rhythms [98]. Future research is needed to correlate enantiomer-specific expression after racemic ketamine administration with biological and clinical measures using convenient tools such as chiral liquid chromatography/tandem mass spectrometry (LC-MS/MS) assays [99, 100].

Neuroimaging and electrophysiology

Increases in frontoparietal gamma power, functional connectivity within the prefrontal cortex (PFC), and striatum activation have all been previously associated with magnitude of antidepressant response to ketamine [101,102,103]. While most research in this area has focused on racemic ketamine, two studies that examined both ketamine and esketamine in their analyses found associations between clinical response and increased thalamic blood flow through perfusion MRI [86] and increases in resting-state functional connectivity between the PFC and subgenual cingulate [92]. Other studies found that greater baseline volume of the bilateral rostral anterior cingulate cortex (ACC) [93], greater pregenual ACC (pgACC) activity during emotional stimulation, and increased glutamate levels at 24 h [94] all significantly predicted treatment response to both ketamine and esketamine. Another study found that low-dose esketamine administered during anesthesia had no impact on electromyography measures [104].

In healthy volunteers, esketamine led to immediate upregulation of resting-state functional connectivity (rsFC) to the dorsomedial PFC, followed by a decreased connectivity of the pgACC to the parietal lobe and dorsolateral PFC (dlPFC). The immediate increases in rsFC also correlated with increased MRS glutamate levels, as estimated by a neurometabolite fitted spectral curve, in the pgACC [89]. Additional neuroimaging research in healthy volunteers found that esketamine may increase hippocampal subfield volumes [88], increase resting cerebral blood flow [91], lead to changes in slow and fast (gamma) wave frequencies [87], and alter pgACC to medial ACC connectivity [86]. pgACC cortical thickness has also been found to be negatively correlated with disembodiment ratings after esketamine administration [90]. However, results from healthy volunteers should be interpreted with caution, given that ketamine administration is also commonly used as a model for schizophrenia and neurocognitive impairment within healthy populations [105]. Further research is needed within TRD populations, and multiple clinical trials using a variety of neuroimaging techniques are currently recruiting TRD participants and healthy volunteers to further understand brain network activity after esketamine administration (NCT06012916 (K-BRAINED), NCT04587778, NCT06002100, NCT05137938).

Beyond clinical trials

Research in this area is understandably limited by what can be accomplished within human populations. Within the past decade, novel methods of assessing human neurobiology have been developed, including in vitro induced pluripotent stem cell (iPSC) models, machine learning, and network analyses. The genetically encoded biosensors iSKetSnFR1 and iSKetSnFR2 fluorescently respond to the presence of esketamine in different cellular compartments and can be used in in vitro models. In iPSC-derived dopaminergic neurons, esketamine rapidly enters the endoplasmic reticulum, suggesting that organellar ion channels, receptors, and transporters may be a potential target of esketamine’s antidepressant effects [95]. Network pharmacology analysis using the Swiss Target Prediction software predicted esketamine to uniquely interact with the GABAergic system, identifying main targets with gene ontology and KEGG enrichment analysis [96]. Additional insights into esktamine’s effects on clinical populations will likely be gained as computational modeling and in vitro models improve.

Putative biomarkers of esketamine in other indications

The use of subanesthetic esketamine in other contexts can also provide useful insights into the biological mechanisms of treatment response. For instance, prefrontal EEG revealed that subanesthetic esketamine decreased the power of slow, delta, and alpha waves while increasing the power of beta-gamma bands during sevofluorane anesthesia, though differences in cognition and emergence time were not noted in comparison to placebo [85].

In breast cancer patients, pre-treatment with esketamine before surgery significantly increased serum levels of BDNF and serotonin; these levels, in turn, were negatively correlated with significant decreases in post-operative depression compared to both ketamine and placebo [82]. Ongoing trials seek to assess the effects of an intraoperative sub-anesthetic dose of esketamine on depression in women undergoing radical mastectomies [106] and on post-operative delirium in the elderly undergoing non-cardiac thoracic surgery (NCT05242692). Another study found that low-dose esketamine (0.5 mg/kg) also effectively reduced post-operative anxiety and depression in those undergoing thoracic surgery, and that symptom decreases were associated with increased serum BDNF levels and decreased S100β and interleukin-6 levels [84]. Similarly, 0.5 mg/kg of esketamine increased serum BDNF and serotonin levels for one to three days post-hysterectomy, and these increases were associated with significantly decreased Visual Analogue Scale (VAS) and Hamilton Depression Rating Scale (HAMD-17) scores [83].

Adverse events and risk of misuse

While esketamine has demonstrated significant clinical efficacy in TRD, its clinical use is not without risks. Concomitant use of benzodiazepines may interfere with response to ketamine, leading to the general practice of withholding them at least 24 h prior to ketamine infusion, a guideline that is also usually applied to IN esketamine administration [107,108,109]. In a seminal RCT (n = 67) followed by a nine-week open-label phase, adverse events such as syncope, headache, dissociation, and ectopic pregnancy led to study discontinuation in one participant each [66]. Double-blind RCTs comparing IN esketamine to placebo administered alongside standard care found that nausea (~ 27–40%), dizziness (~ 12–22%), dissociation (~ 12–26%), unpleasant taste (14–16%), and headache (~ 20–26%) were the most commonly reported side effects [67, 68]. It should also be noted that long-term esketamine use may increase risk of urinary tract symptoms, though this has not been reported from administration following clinical guidelines [110].

Although there has also been concern regarding the impact of ketamine on cognition, most research in individuals with TRD suggests that ketamine/esketamine administration leads to overall improvements in cognition-related tasks such as working memory, processing speed, and cognitive flexibility, among others [111]. Repeated dose esketamine in 51 adolescents with MDD and suicidal ideation also led to significant improvements in processing speed and working memory after 12 days [112]. Another small study in eight patients with TRD also found long-term improvements in cognition after three months of treatment with IN esketamine [113]. Another study found that a single dose of esketamine did not impair driving performance eight hours after administration compared to placebo, in contrast to a positive control of an impairing dose of mirtazapine [114]. Longer-term studies, larger sample sizes, and the inclusion of careful controls are necessary to fully determine any potential impact of esketamine on cognition.

Although patient monitoring during esketamine administration can be burdensome, new technologies like the “MindMed Session Monitoring System” have been developed to alleviate this strain. This system, tested for use with Spravato administration, includes heart rate, motion, audio, and additional activity data that provide passive physiological monitoring of patients [115]. The use of these and similar systems, or even widespread technology such as fitness trackers, smart watches, or other wearables, will enable the future collection of additional empirical data regarding esketamine’s impact on the body.

In addition to administration-related adverse events, esketamine also carries a risk of potential misuse liability, even though delivery application devices were designed with this risk in mind. Recent preclinical studies suggest that esketamine may preferentially activate mu-opioid receptors, suggesting it may have stronger reinforcement properties than racemic or (R)-ketamine [29]. An analysis of esketamine cases extracted from the FDA Adverse Event Reporting System and EMA EudraVigilance database found cases that may reflect abuse potential, as determined by the presence of qualities such as “feeling drunk”, “hallucination”, and “derealization” as well as preferred terms such as “withdrawal syndrome”, “dependence”, and “off-label use”, among others [116]. Despite these findings, little evidence suggests that esketamine administered according to clinical guidelines leads to substance misuse [58]. However, caution is necessary in real-world settings where off-label prescribing and improper supervision could have potentially grievous consequences.

Conclusions

Esketamine is the first widely available agent to emerge from the ketamine-inspired revolution in psychiatric pharmacotherapy. The evidence demonstrates that esketamine is a novel clinical treatment option addressing TRD and suicidality and a unique tool to probe for putative biomarkers of antidepressant response and mechanisms of action; indeed, the limited existing evidence suggests that esketamine seems to have efficacy on par with that of racemic ketamine, although more direct, designed comparisons are needed to test this. However, the study of biomarkers associated with response to esketamine in clinical research is still preliminary, as few studies have sought to identify potential key components of antidepressant response. Significant associations with genes such as IRAK3 and NME7 suggest that inflammation may play a role in esketamine’s rapid-acting antidepressant effects, though future research is necessary given the mixed literature to date on ketamine’s anti- versus pro-inflammatory properties [117]. Metabolomic profiles also suggest that the gut microbiome and mitochondrial function may play a role in mediating esketamine’s effects, although that research is preliminary [80]. Transient effects on cortisol levels, blood pressure, and heart rate could also contribute to both esketamine’s side effects and treatment profile. Clinical neuroimaging and electrophysiological studies to date have been more common than blood-based biomarker studies; these seem to consistently implicate the PFC, striatum, and ACC in esketamine’s rapid and longer-lasting antidepressant effects. Novel methodologies such as iPSC models and network pharmacology may also provide new insight into esketamine’s mechanisms. Finally, clinical research into other indications has also provided useful insight into potential biological mechanisms, including demonstrating promise in perioperative settings for reducing depression and anxiety, with corresponding changes in biomarkers like BDNF and serotonin. Future clinical research should prioritize genomic, proteomic, and brain connectivity biomarkers in TRD populations to fully elucidate the mechanisms underlying esketamine’s therapeutic effects.