Ketamine's effect on inflammation and kynurenine pathway in depression: A systematic review

Major depressive disorder and bipolar depression remain major causes of disability, with a substantial proportion of patients not responding adequately to conventional monoaminergic antidepressants and with slow onset of therapeutic effect. Ketamine, an NMDA receptor antagonist first used as an anaesthetic, has emerged since 2000 as a rapid-acting antidepressant with high response rates reported in treatment-resistant samples and effects that can appear within hours. Interest has grown in ketamine's anti-inflammatory properties because inflammation is implicated in depression and may act via the tryptophan (TRP)–kynurenine (KYN) pathway: pro-inflammatory cytokines activate indoleamine 2,3-dioxygenase (IDO) and KMO, shifting TRP metabolism towards neurotoxic metabolites such as quinolinic acid (QUIN) and 3-hydroxykynurenine (3-HK), while kynurenic acid (KynA) is associated with neuroprotection. Kopra and colleagues set out to systematically review the evidence that ketamine alters inflammatory markers and TRP–KYN pathway metabolites in patients with unipolar or bipolar depression and in animal models of depression. The review aimed to synthesise clinical and preclinical findings to clarify whether ketamine produces anti-inflammatory and neuroprotective metabolic effects that could help explain its antidepressant action and to identify gaps for future research.

The investigators conducted a systematic search on 5 October 2020 using MEDLINE (via Ovid), Embase and APA PsycInfo with keyword combinations focused on ketamine, depression, inflammation, cytokines, CRP, TNF, interferons and KYN-pathway terms (for example kynuren*, KYNA, quinolinic, QUIN). Results were restricted to English language publications and no date limits were applied; reference lists of included papers and reviews were hand-searched for additional studies. Inclusion criteria for human studies required participants to have a current DSM or ICD diagnosis of major depressive disorder (MDD) or bipolar disorder (BD) and to be in a depressive episode at the time of study. Animal studies had to employ established models of depression (for example LPS- or stress-induced paradigms). All included studies required in vivo administration of ketamine; adjunct pharmacological or psychological treatments were grounds for exclusion unless ketamine was given as a stand-alone pharmacological exposure (participants could be maintained on their usual medications in some clinical studies). Human studies needed baseline and at least one post-treatment biomarker measurement with within-subject change data or sufficient data to interpret change; animal studies required comparison of post-treatment biomarkers between ketamine and control groups. Quality assessment was performed separately for human and animal studies. Human studies were scored against five parameters (blinding, absence of adjunct medication, number of outcome timepoints, appropriateness of statistical analyses, and completeness of data/reporting) with scores 0–2 per item for a maximum of 10. Animal studies were scored across four parameters (depression model, protocol clarity and similar treatment of groups, statistical analyses, and completeness of data/reporting) with a maximum of 8. During screening and selection the authors imported references into RefWorks, screened titles and abstracts, and assessed 62 full texts, retaining 31 studies for qualitative synthesis (nine human studies and 22 rodent studies).

The review included nine human studies published 2015–2020 and 22 rodent studies published 2013–2020. Across the clinical studies, the number of patients receiving ketamine per study ranged from 16 to 84, with a combined sample of 429. Most human trials used a 0.5 mg/kg intravenous ketamine infusion (one study also tested 0.2 mg/kg); six studies enrolled treatment-resistant patients as defined by at least two failed antidepressant trials, and some studies included medication-free samples while others allowed maintenance medication or mood stabilisers. Human studies used between 2 and 7 post-infusion sampling points, from immediately post-dose up to 2 weeks. Quality scores for human studies ranged from 4 to 8 out of 10. Clinical findings were heterogeneous but showed a tendency towards reduced peripheral inflammation in many studies. Five out of six human studies that measured inflammatory proteins reported decreases in at least one marker. Specifically, reductions were observed in IL-1β (2 of 3 studies) and IL-6 (3 of 6 studies), and TNF-α was reduced in 2 of 5 studies; single-study decreases were reported for several other cytokines and soluble receptors. Some studies observed short-lived changes lasting up to 240 minutes, while a few reported reductions persisting for 1–3 days; one study reported reductions in 14 proteins 2 weeks after a sixth infusion. Evidence on TRP–KYN metabolites in patients was mixed: two studies reported changes consistent with reduced neurotoxic KYN-pathway activation (for example increased KynA and alterations in KYN/KynA and IDO measures), other studies found no significant changes in KYN or TRP, and some reported differential effects when compared against placebo or when stratifying by antidepressant responder status. Notably, two studies that stratified by treatment response detected changes only among responders. Animal studies provided stronger and more consistent evidence of anti-inflammatory effects. Eighteen out of 21 animal studies reported reductions in one or more pro-inflammatory markers after ketamine. Decreases in IL-1β were reported in 14 of 19 studies, IL-6 in 13 of 18 studies, and TNF-α in 9 of 14 studies; a minority of studies reported increases in anti-inflammatory cytokines such as IL-10 or unexpected surges in certain markers. Regarding TRP–KYN metabolites, 4 of 6 animal studies that measured them found shifts indicating decreased inflammation: examples include increased KynA and decreased QUIN and 3-HK, and IDO protein or activity was decreased in 3 of 4 studies. Ketamine doses in rodents were most commonly 5–20 mg/kg administered intraperitoneally; there was indication of dose-dependency, with low doses (5–6 mg/kg) often producing null results while higher doses produced more robust effects. Sample source (brain versus blood) did not show marked systematic differences in outcomes across studies, and some studies that measured both compartments reported correlated changes, though analyses and sampling timing varied. Animal-study quality scores ranged from 4 to 8 out of 8. Across studies the authors did not perform a meta-analysis because of high heterogeneity in populations, biomarkers measured, sampling timepoints, and methodologies.

Kopra and colleagues interpret the aggregated evidence as showing strong preclinical support for ketamine's anti-inflammatory and neuroprotective effects, while clinical evidence is fewer in number and more heterogeneous but nevertheless suggests ketamine can reduce peripheral markers of inflammation and shift KYN metabolism toward the neuroprotective branch in at least a subset of patients. The cytokines most consistently affected were IL-1β, IL-6 and TNF-α; these cytokines are mechanistically linked to IDO and KMO activation and thus to production of neurotoxic KYN metabolites such as QUIN and 3-HK. The authors note that decreases in IDO were observed in most studies measuring it, which aligns with a model in which ketamine reduces pro-inflammatory signalling and thereby downregulates IDO activity. The investigators highlight several important uncertainties. Many clinical studies showed marker changes that were transient (often not sustained past 24 hours), and it remains unclear whether such short-term reductions can trigger downstream cascades (for example changes in BDNF, synaptogenesis or lasting shifts in KYN metabolites) that underlie ketamine's sustained antidepressant effects. Heterogeneity in clinical samples—diagnosis, treatment-resistance status, medication status, baseline inflammation—and methodological variability likely contribute to inconsistent clinical findings; for example, medicated or non-treatment-resistant patients, who often have lower baseline inflammation, were less likely to show anti-inflammatory changes. The possibility that ketamine exerts anti-inflammatory effects mainly when baseline inflammation is elevated is consistent with surgical and preclinical data and with findings that some marker changes occur only in clinical responders. Methodological limitations of the existing literature are emphasised. The majority of human studies measured circulating markers only, which may not reflect central nervous system levels because certain KYN metabolites (KynA, QUIN) are produced in the brain and cross the blood–brain barrier imperfectly. The authors therefore call for studies measuring CSF markers or using brain imaging (for example PET) and for direct measurement of IDO rather than relying on indirect proxies such as the KYN/TRP ratio. Further research priorities identified include investigation of dose-dependency and effects of repeated infusions, examination of clinical relevance of inflammatory changes (including whether baseline inflammation predicts antidepressant response), assessment of specific molecular mechanisms of immunomodulation, and inclusion of key markers such as IFN-γ and CRP which were infrequently measured in the current literature. The review's own limitations, as noted by the authors, include the small number of clinical studies to date, the decision not to search grey literature, and inability to meta-analyse because of heterogeneity between studies.

The review concludes that ketamine exhibits anti-inflammatory effects in both depressed humans and rodent models, and that it appears to shift TRP–KYN metabolism toward a more neuroprotective profile in several studies. Kopra and colleagues argue that elucidating ketamine's immunomodulatory mechanisms is important for precision medicine efforts, for improving understanding of depression neurobiology, and for identifying targets for novel antidepressants with fewer side effects. They recommend further research into the molecular basis of ketamine's effects, the clinical significance of inflammatory changes, and measurement of central nervous system markers.