In a double‑blind placebo‑controlled trial of treatment‑resistant depression patients and healthy controls, ayahuasca produced a significant 48‑hour reduction in C‑reactive protein (CRP) and, in patients, larger CRP decreases correlated with greater reductions in depressive symptoms. No significant effects were seen for interleukin‑6 or BDNF, and these biomarkers did not predict antidepressant response or remission.
Papers cited by this study that are also in Blossom
Galvão-Coelho, N. L., de Almeida, R. N., de Menezes Galvão, A. C. et al. · Frontiers in Psychology (2019)
Background
Ayahuasca is a traditional Amazon brew and its potential antidepressant properties have recently been explored in scientific settings. We conducted a double-blind placebo-controlled trial of ayahuasca with treatment-resistant depression patients ( n = 28) and healthy controls ( n = 45).
Aims
We are evaluating the blood inflammatory biomarkers: C-reactive protein and interleukin 6, as a potential consequence of ayahuasca intake and their correlation with serum cortisol and brain-derived neurotrophic factor levels. Blood samples were collected at pre-treatment and 48 hours after substance ingestion to assess the concentration of inflammatory biomarkers, together with administration of the Montgomery-Åsberg Depression Rating Scale.
Results
At pre-treatment, patients showed higher C-reactive protein levels than healthy controls and a significant negative correlation between C-reactive protein and serum cortisol levels was revealed ( rho = –0.40, n = 14). C-reactive protein in those patients was not correlated with Montgomery-Åsberg Depression Rating Scale scores. We observed a significant reduction of C-reactive protein levels across time in both patients and controls treated with ayahuasca, but not with placebo. Patients treated with ayahuasca showed a significant correlation ( rho = + 0.57) between larger reductions of C-reactive protein and lower depressive symptoms at 48 hours after substance ingestion (Montgomery-Åsberg Depression Rating Scale). No significant result with respect to interleukin 6 and brain-derived neurotrophic factor was found. Furthermore, these biomarkers did not predict the antidepressant response or remission rates observed.
Conclusions
These findings enhance the understanding of the biological mechanisms behind the observed antidepressant effects of ayahuasca and encourage further clinical trials in adults with depression.
Major depressive disorder (MDD) is heterogeneous and many patients do not achieve sustained remission with standard pharmacological treatments, leaving a subset with treatment-resistant depression. Earlier research has identified raised circulating inflammatory markers in some people with depression, notably interleukin 6 (IL-6) and C-reactive protein (CRP), and has linked chronic stress, reduced brain-derived neurotrophic factor (BDNF), and inflammation to depressive pathophysiology. Findings on whether conventional antidepressants reduce inflammatory markers are inconsistent, and baseline inflammation may influence treatment effects. Classical serotonergic psychedelics have attracted renewed interest as potentially fast-acting antidepressants, and some preclinical and mechanistic work has proposed direct anti-inflammatory actions mediated via 5-HT2A and sigma-1 (σ1R) receptors. Leite Galvão‑Coelho and colleagues set out to examine whether a single dose of ayahuasca alters blood inflammatory biomarkers and whether such changes relate to antidepressant effects. Using data from a randomised, double-blind, placebo-controlled trial in adults with treatment-resistant depression and healthy controls, the investigators measured plasma CRP and serum IL-6 before treatment and 48 hours after dosing, and tested correlations with serum cortisol and BDNF as well as associations with clinical change on the Montgomery-Åsberg Depression Rating Scale (MADRS). They hypothesised that changes in inflammatory markers would correlate with symptom improvement and might predict antidepressant response or remission at 48 hours.
The study was embedded in a randomised, double-blind, placebo-controlled, parallel-arm trial of ayahuasca versus placebo conducted at the Federal University of Rio Grande do Norte (University Hospital Onofre Lopes) in Brazil. Ethical approval and trial registration were in place and all volunteers gave written informed consent. In total 73 adult, psychedelic‑naïve volunteers were enrolled: 28 patients with treatment-resistant depression (patient group, PG) and 45 healthy controls (control group, CG). Treatment resistance was defined as failure to remit after at least two antidepressants from different classes. Patients were in a current moderate-to-severe depressive episode diagnosed using the Structured Clinical Interview for DSM-IV and had baseline MADRS scores >18; MADRS was the primary clinical outcome. All patients completed a washout period adjusted to prior medication half-lives and were not taking antidepressants during the study, although benzodiazepines or hypnotics were permitted if required. Participants were randomised 1:1 to receive a single oral liquid dose of ayahuasca (1 mL/kg) or an inert placebo. A single, characterised batch of ayahuasca was used (mean concentrations reported for N,N-dimethyltryptamine, harmine, harmaline and tetrahydroharmine). Placebo was formulated to mimic some gastrointestinal effects (water, yeast, citric acid, caramel colourant, zinc sulphate). Dosing sessions lasted about six hours in a comfortable hospital room; participants remained seated with eyes closed and listened to curated playlists. No formal psychotherapeutic protocol was provided beyond brief preparation and in-session support. Blood samples were collected at 07:00 following an 8-hour fast immediately before dosing (D0) and 48 hours after dosing (D2). Measured analytes were plasma CRP (immunoturbidimetry), serum IL-6, cortisol and BDNF (enzyme-linked immunosorbent assay), and liver enzymes GOT and GPT to screen for hepatic causes of altered CRP. Laboratory assays were performed blinded to treatment allocation. Statistical analyses treated group (CG, PG), treatment (ayahuasca, placebo) and phase (D0, D2) as categorical independent factors. MADRS and biomarker levels at D0 and change scores (Δ) to D2 were treated as quantitative outcomes. Body mass index (BMI) was included as a covariate. The authors used t-tests, Chi-square, Mann–Whitney tests, general linear models (GLM) with interaction terms, Spearman correlations, multiple regression and binary logistic regression to explore prediction of response (≥50% reduction in MADRS) and remission (MADRS ≤10) at D2. Analyses were performed in IBM SPSS Statistics 25 with bootstrap enabled and significance set at p < 0.05.
Groups (CG and PG), treatment (ayahuasca and placebo), and phases (D0 and D2) were considered categorical independent variables. MADRS, CRP, IL-6, cortisol, and BDNF levels at D0 and their changes between D2 and D0 (ΔMADRS, ΔCRP, ΔIL-6, Δcortisol, and ΔBDNF) were considered quantitative dependent variables. Body mass index (BMI) was used as a co-variable in multivariable analyses due to comorbid between treatmentresistant depression and obesity, which has been associated with inflammation. At D0, the student t-test was used to compare GOT and GPT, and Chi-square test to compare gender, both between groups. Mann-Whitney test was applied to compare BMI and age between groups and to analyze CRP and IL-6 with respect to comorbidities, anxiety, and personality disorders in the patient group. One patient with a personality disorder in cluster A was excluded from the personality disorder statistical analysis. Multiple regression was used to test if the volunteer's gender and age modulated baseline depressive symptoms (MADRS) and if the number of unsuccessful previous antidepressant treatments predicts CRP and IL-6 levels. For this, we clustered patients into three groups: 2-3, 4-5 and 6-9 treatments, with respect to the number of previous different antidepressants used. The general linear model (GLM) was used to investigate CRP and IL-6 for group, treatment, and phase. Spearman correlation test was applied to quantitative dependent variables for each group and treatment separately. Multiple regression and binary logistic regression were applied to predict the antidepressant response and remission rates of patients treated with ayahuasca at D2. CRP, IL-6, ΔCRP, ΔIL-6, cortisol, BDNF, and BMI were used as potential predictor variables. Response criteria was defined as a reduction of 50% or more in baseline scores and remission rates were defined as MADRS scores ⩽10 at D2. All statistical analyses were performed using the IBM SPSS statistics 25. The bootstrap function remained enabled for statistical testing. The significance level considered was p < 0.050.
In the pre-treatment phase we found treatment-resistant depressive patients had higher CRP levels than healthy controls. Moreover, for patients, CRP was negatively correlated with serum cortisol levels. Then 48 hours after treatment we observed a reduction of CRP levels in both patients and controls treated with ayahuasca, but not in placebo-treated individuals. Moreover, patients treated with ayahuasca, and not with placebo, showed a significant correlation between CRP reductions and improvements in depressive symptoms. The pre-treatment CRP levels observed here in patients were characterized as a low-grade inflammatory profile (CRP level >3 mg/L; healthy adult population <1 mg /L). Patients and controls showed similar levels of IL-6 in the pretreatment phase. Similar findings have been identified the CRP, as a non-specific etiological factor for MDD. However, opposite of our results, high levels of IL-6 also have been reported in MDDet al., 2019). In part these contradictory findings are due to the involvement of IL-6 in the pro-inflammatory and anti-inflammatory response, which have been described by trans-signaling pathway of tissue repair. Therefore, to distinguish if the IL-6 change is a pro-inflammatory or anti-inflammatory response, it is necessary to explore downstream molecules, such as its receptors. Although traditionally IL-6 and CRP are physiologically linked, some studies, including in depressive patients, have shown increases in CRP independently of IL-6. However, the mechanisms behind such changes are not completely understood, some hypotheses have been raised: (a) other inflammatory factors than IL-6, such as IL-1β, could stimulate CRP synthesis and release it from the liver; (b) IL-6 increases are not simultaneous to CRP, but precede CRP by 1-2 days; and (c) the activation of the IL-6 trans-signaling pathway reduces IL-6 levels in the blood. The inflammatory biomarkers of the pre-treatment phase were not correlated with the severity of depressive symptoms. This finding is contrary to our initial hypothesis, but corroborates other studies. In fact, different correlations (positive, negative, or non-significant) are seen between inflammation and clinical characteristics of depression. Furthermore, there was no association between the number of previous antidepressant treatments and the levels of CRP and IL-6 in the pre-treatment phase. Patients' mental comorbidities also did not modulate inflammatory biomarker levels and were not related to autoimmune, infectious, or liver diseases. In this study ayahuasca, but not placebo, decreased CRP levels in both patients and controls 48 hours after ingestion. In addition, in patients this reduction was moderately correlated with improvement in depressive symptoms. Other studies with depressive patients have also reported significant decreases in CRP levels after treatment; however, differences exist between classes of antidepressants. Because some studies have shown an association between the presence of low-grade inflammatory profiles and treatment resistance in depressive patients, these results point to a relevant role of ayahuasca, which shows concomitant antidepressant and anti-inflammatory action. We did not find changes in IL-6 after treatment (placebo or ayahuasca). The literature reports conflicting responses of IL-6 after treatment with antidepressants, possibly due to the heterogeneous study designand IL-6 complex physiological mechanisms. It is proposed that classic serotonergic psychedelics have antiinflammatory actions through 5-HT2A and σ1R. Harmine, present in ayahuasca, also shows anti-inflammatory activity by suppression of nuclear factor-κB signaling. Anti-inflammatory properties have also been proposed for ketamine, an anesthetic with psychedelic and antidepressant actions; however, such effects remain controversial. Pro-inflammatory biomarkers did not predict the response or remission rates at D2. This finding may be due to suboptimal measurement points: the highest response and remission rate was observed 7 days after ayahuasca intake, but measurement of the inflammatory biomarkers was only performed at D2. However, other publications have reported no correlations between inflammation and remission. As expected, we observed a negative correlation between CRP and cortisol levels in patients. In a previous study, we showed that this depression group had lower cortisol levels (hypocortisolemia) compared to controls prior to ayahuasca intake. Therefore, we can argue that the inflammatory basal profile observed in patients could be, at least in part, due to low cortisol levels, because cortisol is a potent anti-inflammatory hormone. A similar relation between inflammation and cortisol was observed in depressive patients. However, this correlation at baseline was not observed in the control group, probably because they presented eucortisolemia. We also considered possible correlations between inflammatory, BDNF, and cortisol responses to treatment, as interaction between them has been describedand taking into account that ayahuasca increased BDNF levels at D2 in volunteers of this trial. However, no correlation was observed between them. The absence of expected correlations can be due to some factors, such as: 1. these biomarkers are modulated by multiple routes and they show circadian oscillation, then, a single blood collection may not be enough to show them. 2. as patients showed alterations in these biomarkers, the correlation between them probably could be absent. Therefore, it is also important to point out the limitations of this study. Besides a single blood sample per volunteer, we only measured total BDNF levels instead of pro-and mature BDNF molecules, which are more specific biomarkers of apoptosis and neuroplasticity, respectively. With respect to the sample, it was limited to a modest number of patients with treatmentresistant depression and comorbid personality and anxiety disorder. Therefore, trials in larger populations of depressive patients are needed, with sequential blood measurements throughout the day and on different days. Moreover, studies with multiple sessions of ayahuasca may provide a stronger anti-inflammatory effect. As inflammation has an important role in various features of major depression such as reward processing and reactivity to negative information, there is a clear rationale for including reduction in inflammation as a goal of treatment. Therefore, our findings support the use of ayahuasca as an alternative treatment with anti-inflammatory effects for MDDand reinforce the utility of biomarkers in depression.
Create a free account to open full-text PDFs.
Araújo, A. M., Carvalho, F. M., Carvalho, M. et al. · Archives of Toxicology (2015)
De Araujo, D. B., Ribeiro, S., Cecchi, G. A. et al. · Human Brain Mapping (2011)
Flanagan, T. W., Nichols, C. D. · International Review of Psychiatry (2018)
Osório, F. L., Sanches, R. F., Macedo, L. et al. · brazilian Journal of Psychiatry (2015)
Palhano-Fontes, F., Andrade, K. C., Tófoli, L.F. et al. · PLOS ONE (2015)
Palhano-Fontes, F., Barreto, D., Onias, H. et al. · Psychological Medicine (2018)
Ross, S., Bossis, A. P., Guss, J. et al. · Journal of Psychopharmacology (2016)
Sanches, R. F., Osório, F. L., Dos Santos, R. G. et al. · Journal of Clinical Psychopharmacology (2016)
Szabo, A., Kovacs, A., Frecska, E. et al. · PLOS ONE (2014)
Papers in Blossom that reference this study
Thuery, G., Crossen, F., Mc Loone, D. et al. · Therapeutic Advances in Psychopharmacology (2026)
Garcia-Romeu, A., Naudé, G. P., Rebman, A. W. et al. · Scientific Reports (2026)
Silva-Costa, N., Pessoa, J. A., Andrade, K. C. et al. · Journal of Psychopharmacology (2025)
Caspani, G., Ruffell, S. G. D., Tsang, WF. et al. · Pharmacological Research (2024)
Mason, N. L., Szabo, A., Kuypers, K. P. C. et al. · Brain Behavior and Immunity - Health (2023)
Burmester, D., Madsen, M. K., Szabo, A. et al. · Comprehensive Psychoneuroendocrinology (2023)
Galvão-Coelho, N. L., De Meiroz Grilo, M. L. P., de Sousa, G. M. et al. · Frontiers in Behavioural Neuroscience (2022)
Rossi, G. N., Hallak, J. E., Baker, G. et al. · European Archives of Psychiatry and Clinical Neuroscience (2022)
Almeida, C. A. F., Pereira, B. P., Pereira Junior, A. A. et al. · Behavioural Brain Research (2022)
Dursun, S. M., Kelly, J. R., Gillan, C. M. et al. · Frontiers in Psychiatry (2021)
Garcia-Romeu, A., Darcy, S., Jackson, H. et al. · Current Topics in Behavioral Neurosciences (2021)
Kuiperes, Z., Schreiber, R. · Neurobiology of Learning and Memory (2021)
Rush, B., Marcus, O., García, S. et al. · Frontiers in Pharmacology (2021)
Rocha, J. M., Rossi, G. N., de Lima Osório, F. et al. · Journal of Clinical Psychopharmacology (2021)
Baseline characteristics and pre-treatment biomarkers: The sample comprised 28 patients (seven men, 21 women) and 45 controls (20 men, 25 women). Patients were older than controls and had higher BMI; liver enzymes (GOT, GPT) were within reference ranges and did not differ significantly between groups. A large majority of patients had comorbid personality disorder (71%) and 43% had an anxiety disorder. Adjusting for BMI, patients had higher plasma CRP than controls (F = 30.60, df = 1, p = 0.001, Cohen's d = 0.74). IL-6 levels did not differ between patients and controls (F = 0.35, p = 0.550). In patients, baseline CRP showed a significant inverse correlation with serum cortisol (rho = -0.40, p = 0.033); no correlation was found between CRP and BDNF, and IL-6 was not correlated with the other measured variables. The number of prior unsuccessful antidepressant trials did not predict baseline CRP or IL-6. Effects of ayahuasca versus placebo at 48 hours (D2): Across both patients and controls, individuals who received ayahuasca (n = 37) showed a reduction in CRP at D2 compared with D0 after adjustment for BMI, whereas placebo recipients (n = 36) did not (GLM treatment*phase interaction: F = 8.20, df = 1, p = 0.005). Post hoc testing for the ayahuasca group indicated a significant D0–D2 decrease (p-adjusted Bonferroni = 0.040, Cohen's d = -1.37), while the placebo group showed no significant change. Quantitatively, patients treated with ayahuasca had a 26.6% reduction in CRP (controls reduced 32.39%); patients on placebo had a 20.4% increase (controls on placebo reduced 2.45%). During the dosing session, 90% of patients reported nausea and 57% vomited. Relation to clinical outcome: Among patients who received ayahuasca (n = 14), greater reductions in CRP (ΔCRP) correlated with lower MADRS scores at D2 (Spearman rho = 0.57, p ≤ 0.050). In this subgroup MADRS mean decreased from 28.28 (± SEM 3.01) at D0 to 14.23 (± SEM 3.05) at D2 (t = 3.26, df = 25, p = 0.003, Cohen's d = -1.25). No such correlation was observed in placebo-treated patients. Changes in CRP were not correlated with changes in cortisol or BDNF, and neither baseline CRP nor ΔCRP (alone or combined with other variables) predicted response or remission at D2. IL-6 showed no significant main effects or interactions by group, treatment or phase (GLM p = 0.43) and ΔIL-6 did not correlate with clinical change nor predict response/remission at D2.
Leite Galvão‑Coelho and colleagues interpret their findings as evidence that people with treatment-resistant depression exhibited a low-grade inflammatory profile at baseline, reflected by higher CRP, and that a single dose of ayahuasca reduced CRP levels 48 hours later in both patients and healthy controls. They emphasise the observed negative correlation between baseline CRP and cortisol in patients, consistent with the notion that hypocortisolaemia may contribute to elevated inflammation because cortisol is an anti-inflammatory hormone. The authors highlight that, in patients, larger reductions in CRP after ayahuasca correlated with greater improvements in depressive symptoms at D2, whereas IL-6 did not change and was not associated with clinical outcomes. The discussion situates these results within inconsistent prior literature on IL-6 and CRP in depression and on inflammatory responses to antidepressant treatment. The investigators note the biological complexity of IL-6, including its pro- and anti-inflammatory roles and timing differences that could decouple IL-6 and CRP measures. Possible mechanisms for an anti-inflammatory action of ayahuasca are referenced, including 5-HT2A and σ1R-mediated effects of classical psychedelics and harmine-mediated suppression of nuclear factor-κB signalling. The authors acknowledge that pro-inflammatory biomarkers did not predict response or remission at D2 and suggest this could reflect suboptimal timing of biomarker measurements because the largest clinical effects were observed at 7 days in the parent trial while biomarker sampling was limited to D2. Key limitations noted by the investigators include the modest sample size, the single time-point blood collections (one morning draw at each phase) which may miss circadian or delayed changes, the measurement of total rather than pro- and mature BDNF isoforms, and the presence of comorbid personality and anxiety disorders in the patient sample. They recommend larger trials with sequential sampling across the day and over multiple days, and studies assessing repeated ayahuasca sessions to evaluate whether more sustained anti-inflammatory effects occur. In their interpretation the findings support ayahuasca’s concurrent antidepressant and anti-inflammatory actions and underscore the potential value of biomarkers in depression research.
Dos Santos, R. G., Hallak, J. E., Baker, G. et al. · Journal of Psychopharmacology (2021)
da Silva, M. G., Daros, G. C., de Bitencourt, R. M. · Behavioural Brain Research (2021)
Galvão-Coelho, N. L., Marx, W., Sinclair, J. et al. · Psychopharmacology (2021)
Nichols, D. E., Walter, H. · Pharmacopsychiatry (2020)
Maia, L. O., Daldegan-Bueno, D., Tófoli, L.F. · Journal of Psychoactive Drugs (2020)
Kadriu, B., Greenwald, M., Ba et al. · International Journal of Neuropsychopharmacology (2020)
Thompson, C., Szabo, A. · Immunology Letters (2020)