Rapid antidepressant effects of the psychedelic... — Research Summary & Context

Rapid antidepressant effects of the psychedelic... — Research Summary & Context | Blossom

Abstract

Background

Recent open-label trials show that psychedelics, such as ayahuasca, hold promise as fast-onset antidepressants in treatment-resistant depression.

Methods

To test the antidepressant effects of ayahuasca, we conducted a parallel-arm, double-blind randomized placebo-controlled trial in 29 patients with treatment-resistant depression. Patients received a single dose of either ayahuasca or placebo. We assessed changes in depression severity with the Montgomery-Åsberg Depression Rating Scale (MADRS) and the Hamilton Depression Rating scale at baseline, and at 1 (D1), 2 (D2), and 7 (D7) days after dosing.

Results

We observed significant antidepressant effects of ayahuasca when compared with placebo at all-time points. MADRS scores were significantly lower in the ayahuasca group compared with placebo at D1 and D2 ( p = 0.04), and at D7 ( p < 0.0001). Between-group effect sizes increased from D1 to D7 (D1: Cohen's d = 0.84; D2: Cohen's d = 0.84; D7: Cohen's d = 1.49). Response rates were high for both groups at D1 and D2, and significantly higher in the ayahuasca group at D7 (64% v. 27%; p = 0.04). Remission rate showed a trend toward significance at D7 (36% v. 7%, p = 0.054).

Conclusions

To our knowledge, this is the first controlled trial to test a psychedelic substance in treatment-resistant depression. Overall, this study brings new evidence supporting the safety and therapeutic value of ayahuasca, dosed within an appropriate setting, to help treat depression. This study is registered at http://clinicaltrials.gov (NCT02914769).

Unlocked with Blossom Pro

Research Summary of 'Rapid antidepressant effects of the psychedelic ayahuasca in treatment-resistant depression: a randomized placebo-controlled trial'

Introduction

Depression affects over 300 million people worldwide, and about one-third fail to respond to at least three adequate antidepressant trials. Most licensed antidepressants act on monoamine systems and require weeks to show benefit. By contrast, ketamine, an NMDA antagonist, produces rapid antidepressant effects that peak within a day and can last about a week. Interest has grown in serotonergic psychedelics as potential fast-acting treatments: open-label trials of ayahuasca and psilocybin have reported rapid reductions in depressive symptoms, but these studies lacked placebo controls and are vulnerable to large placebo responses. Palhano-Fontes and colleagues report a parallel-arm, double-blind, randomised placebo-controlled trial designed to test whether a single dose of ayahuasca produces rapid antidepressant effects in patients with treatment-resistant major depressive disorder. The investigators also explored correlations between antidepressant outcomes and acute psychedelic effects measured during the dosing session.

Methods

This was a double-blind, parallel-arm randomised placebo-controlled trial conducted at a Brazilian university hospital. Adults aged 18–60 with unipolar major depressive disorder and treatment resistance (inadequate response to at least two antidepressants from different classes) were recruited; eligible patients had moderate-to-severe depression at screening (HAM-D ⩾ 17). Key exclusions included prior ayahuasca experience, current medical disease, pregnancy, neurological disorders, personal or family history of schizophrenia or bipolar disorder, substance abuse, and suicidal risk. The research ethics committee approved the trial and participants provided written informed consent. Randomisation used permuted blocks of size 10 to allocate patients 1:1 to ayahuasca or placebo. Blinding was supported by enrolling only ayahuasca-naïve patients and by assigning different psychiatrists for dosing-session and follow-up assessments; assignment codes were held only in the database and pharmacy. The investigators note psychiatrist blindness was not formally assessed. Participants underwent a washout of prior antidepressants (about 2 weeks on average) and were off antidepressant medication during the dosing session; benzodiazepines were permitted as needed. A single batch of ayahuasca was used and its alkaloid content was quantified by mass spectrometry (mean ± SD per ml: N,N-DMT 0.36 ± 0.01 mg; harmine 1.86 ± 0.11 mg; harmaline 0.24 ± 0.03 mg; tetrahydroharmine 1.20 ± 0.05 mg). Dosing involved a single oral administration of 1 ml/kg of the brew or placebo adjusted to contain 0.36 mg/kg of N,N-DMT; placebo was a bitter brown liquid (water, yeast, citric acid, zinc sulfate, caramel) designed to mimic taste and produce modest gastrointestinal distress. Primary and secondary outcomes were prespecified. The primary outcome was change in HAM-D from baseline to day 7 (D7). Secondary outcomes included change in MADRS from baseline to day 1 (D1), day 2 (D2), and D7, plus response (⩾50% reduction) and remission (HAM-D ⩽ 7 or MADRS ⩽ 10) rates at those time points. Acute effects and safety during the session were assessed with CADSS (dissociation), BPRS (psychopathology), YMRS (mania), and later HRS and MEQ30 for phenomenology. Assessments occurred at baseline, during the session (-10 min, +1:40 h, +2:40 h, +4:00 h), and at D1, D2, and D7. Analyses followed a modified intent-to-treat principle, including patients with baseline, dosing and D7 assessments. An initial sample size calculation had estimated 42 participants to detect a five-point HAM-D difference with 80% power. Primary statistical tests used fixed-effects linear mixed models with baseline scores as covariates and a Toeplitz covariance structure; missing data were handled by restricted maximum-likelihood estimation. Post-hoc between-group comparisons used t tests with Sidak correction. Effect sizes (Cohen's d) were reported for between- and within-group comparisons. Fisher's exact tests assessed categorical differences (responders, remitters, adverse events); odds ratios and number needed to treat (NNT) were calculated. Correlations between acute effects and MADRS change used Pearson correlations with multiple-comparisons correction by number of factors per scale. Significance was set at p < 0.05, two-tailed.

View full paper sections

INTRODUCTION

The World Health Organization estimates that more than 300 million people suffer from depression, and about one-third do not respond to appropriate courses of at least three different antidepressants. Most currently available antidepressants have a similar efficacy profile and mechanisms of action, based on the modulation of brain monoamines, and usually, take about 2 weeks to start being effective. Recent evidence, however, shows a rapid and significant antidepressant effect of ketamine, an N-methyl-D-aspartate (NMDA) antagonist frequently used in anesthesia. In recent randomized placebo-controlled trials with ketamine in treatment-resistant depression, the antidepressant effects peaked 1 day after dosing and remained significant for about 7 days. Additionally, research with serotonergic psychedelics has gained momentum. A few centers around the world are currently exploring how these substances affect the brain, and also probing their potential in treating different psychiatric conditions, including mood disorders. For instance, recent open-label trials show that psychedelics, such as ayahuasca and psilocybin, hold promise as fast-onset antidepressants in treatment-resistant patients. Ayahuasca is a brew traditionally used for healing and spiritual purposes by indigenous populations of the Amazon Basin. In the 1930s, it began to be used in religious settings of Brazilian small urban centers, reaching large cities in the 1980s and expanding since then to several other parts of the world. In Brazil, ayahuasca has a legal status for ritual use since 1987. Ayahuasca is most often prepared by decoction of two plants: Psychotria viridis that contains the psychedelic N, N-dimethyltryptamine (N,N-DMT), a serotonin and sigma-1 receptors agonist, and Banisteriopsis caapi, rich in reversible monoamine oxidase inhibitors (MAOi) such as harmine, harmaline, and tetrahydroharmine. The acute psychological effects of ayahuasca last around 4 h and include intense perceptual, cognitive, emotional, and affective changes. Although nausea, vomiting, and diarrhea are often reported, mounting evidence points to a positive safety profile of ayahuasca. For instance, ayahuasca is not addictive and has not been associated with psychopathological, personality, or cognitive deterioration, and it promotes only moderate sympathomimetic effects. In a recent open-label trial, 17 patients with major depressive disorder attended a single dosing session with ayahuasca. Depression severity was assessed before, during and after dosing, using the Hamilton Depression Rating scale (HAM-D) and the Montgomery-Åsberg Depression Rating Scale (MADRS). Significant reduction in depression severity was found already in the first hours after dosing, an effect that remained significant for 21 days. Although promising, these studies have not controlled for the placebo effect, which can be remarkably high in clinical trials for depression, reaching 30-40% of the patients. To address this issue, and to further test the antidepressant effects of ayahuasca, we conducted a randomized placebo-controlled trial in patients with treatment-resistant depression. Additionally, we explored for correlations between the antidepressant and the acute effects of ayahuasca.

Full Text PDF

Full Paper PDF

Create a free account to open full-text PDFs.

Create Free Account Already have an account? Sign in

Study Details

Study Type
individual
Population
humans
Characteristics
randomizedplacebo controlledparallel groupdouble blind
Journal
Psychological Medicine
Compound
Ayahuasca
Topics
Anxiety Disorders Depressive Disorders

Results

Between January 2014 and June 2016, 218 patients were screened, 35 met inclusion criteria, and six were excluded before dosing (five no longer met depression criteria on dosing day; one withdrew). Data from 29 patients were analysed: 14 received ayahuasca and 15 received placebo. The sample was predominantly female (72%), mean age 42.0 ± 11.7 years, and of low socioeconomic status (41% with <8 years education; 41% earning <2 minimum wages). The extracted text reports mean baseline HAM-D = 21.83 ± 5.35; the baseline MADRS mean is not clearly reported in the extracted text. Primary outcome (HAM-D change to D7): a significant between-group difference was reported at D7 (F1 = 6.31; p = 0.019), with the ayahuasca group showing greater reduction than placebo. Between-group effect size at D7 was large (Cohen's d = 0.98; 95% CI 0.21–1.75). Within-group effect sizes were large for ayahuasca (Cohen's d = 2.22; 95% CI 1.28–3.17) and medium for placebo (Cohen's d = 0.46; 95% CI −0.27–1.18). HAM-D response rate at D7 was 57% for ayahuasca versus 20% for placebo (p = 0.04; NNT = 2.69). HAM-D remission showed a trend favouring ayahuasca at D7 (43% v. 13%; OR 4.87; 95% CI 0.77–26.73; p = 0.07; NNT = 3.39). MADRS outcomes (D1, D2, D7): the linear mixed model showed significant main effects of time (F2,34.4 = 3.96; p = 0.028) and treatment (F1,27.7 = 10.52; p = 0.003), with no treatment-by-time interaction. MADRS scores were significantly lower in the ayahuasca group at D1 (F1,49.7 = 4.58; p = 0.04), D2 (F1,50.3 = 4.67; p = 0.04), and D7 (F1,47 = 14.81; p < 0.0001). Between-group Cohen's d values were 0.84 (D1; 95% CI 0.05–1.62), 0.84 (D2; 95% CI 0.05–1.63), and 1.49 (D7; 95% CI 0.67–2.32). Within-group effect sizes for ayahuasca were large at all time points (D1 d = 2.78; D2 d = 3.05; D7 d = 2.90). MADRS response rates: at D1 response was 50% (ayahuasca) v. 46% (placebo) (p = 0.87); at D2 77% v. 64% (p = 0.43); at D7 64% v. 27% (p = 0.04; NNT = 2.66). MADRS remission rates did not differ at D1 or D2; at D7 remission trended higher in the ayahuasca group (36% v. 7%; OR 7.78; 95% CI 0.81–77.48; p = 0.054; NNT = 3.44). The authors report that all patients in the ayahuasca group improved by D7, while four placebo patients worsened. Acute session effects and safety: transient increases on CADSS and BPRS+ occurred at +1:40 h after ayahuasca (34.8% BPRS+; 21.6% CADSS), with CADSS showing a trend (p = 0.052); BPRS+ changes did not reach significance and neither CADSS nor BPRS+ correlated with antidepressant improvement. YMRS scores did not increase significantly (p = 0.26). Adverse events more frequent with ayahuasca included nausea (71% v. 26%; p = 0.027) and vomiting (57% v. 0%; p = 0.0007); transient anxiety, restlessness and headache were reported but differences were not statistically significant. No serious adverse events were reported. Phenomenology scales: HRS scores were higher with ayahuasca than placebo on five subscales (perception, somaesthesia, cognition, intensity, volition; all p < 0.001 to p = 0.0003); affect did not differ. MEQ30 scores were higher in the ayahuasca group for mystical experience, transcendence of time and space, ineffability, and total score; positive mood did not differ. The MEQ30 was added mid-trial and only 15 patients completed it (8 ayahuasca, 7 placebo). Correlations between acute phenomenology and antidepressant outcome were selective: among ayahuasca responders, the HRS perception subscale correlated positively with MADRS change at D7 (r = 0.90; p = 0.002), and in the ayahuasca group the MEQ30 transcendence factor correlated negatively with MADRS change (r = −0.84; p = 0.009). The extracted text notes small subgroup sizes for these correlations.

Discussion

Palhano-Fontes and colleagues interpret their findings as evidence that a single dosing session with ayahuasca produces a rapid antidepressant effect in treatment-resistant depression, with improvements greater than placebo at D1, D2 and D7 and increasing between-group effect sizes through day 7. They compare the time-course with ketamine trials, noting that ketamine tends to show the largest between-group effect at D1 that attenuates by D7, whereas in this study the ayahuasca effect was largest at D7. Response rates were high early in both groups but diverged by D7 with significantly more responders in the ayahuasca arm. The authors acknowledge a notably high placebo response in their sample and propose several contributing factors: participants' low socioeconomic status and the provision of a comfortable, supportive study environment may have produced a strong 'care effect'; high rates of comorbid personality disorder (76%, mainly cluster B) could also elevate placebo responsiveness. Mechanistic hypotheses discussed include N,N-DMT activation of sigma-1 receptors (σ1R) and the presence of monoamine oxidase inhibitors (harmine and related alkaloids) in the brew; preclinical evidence cited links harmine and σ1R agonists to antidepressant-like effects, BDNF modulation and neurogenesis. The investigators also point to prior imaging work showing ayahuasca-related increases in blood flow to mood-related brain regions and reductions in Default Mode Network activity, and to observational studies associating regular ayahuasca use with preserved cognition and reduced anxiety and depressive symptoms. The authors note relationships between aspects of the subjective psychedelic experience and clinical outcome: higher MEQ30 subscales and HRS dimensions were observed with ayahuasca, and selected subscales (perception; transcendence of time and space) correlated with symptom change in small subsamples. Safety considerations are summarised: no serious adverse events occurred, though many patients experienced psychological distress, nausea and vomiting during the session. Key limitations acknowledged include the modest sample size, the restriction to treatment-resistant patients with long illness duration and high comorbidity (limiting generalisability), and the intrinsic difficulty of maintaining double blind in psychedelic research despite measures taken (naïve patients, separated raters, an active-tasting placebo). The authors report that some patients misclassified placebo as ayahuasca, which they argue supports preserved blinding. They conclude that, while larger trials are needed, this study is the first randomised placebo-controlled trial to test a psychedelic in treatment-resistant depression and provides preliminary evidence for the safety and therapeutic potential of ayahuasca when administered in an appropriate setting.

PROCEDURES

We used the MADRS and the HAM-Dto access depression severity. MADRS assessments were at baseline (one day before dosing), and at 1 (D1), 2 (D2), and 7 (D7) days after dosing. HAM-D was applied only at baseline and D7, as it was designed to access depression symptoms present in the last week. The liquid used as placebo was designed to simulate organoleptic properties (taste and color) of ayahuasca, such as a bitter and sour taste, and a brownish color. It contained water, yeast, citric acid, zinc sulfate and caramel colorant. The presence of zinc sulfate also produced low to modest gastrointestinal distress. A single ayahuasca batch was used throughout the study, which was prepared and provided free of charge by a branch of the Barquinha church based at Ji-Paraná-RO, Brazil. To assess alkaloids concentrations and stability of the batch, samples of ayahuasca were quantified at two different time points by mass spectroscopy analysis. On average, the ayahuasca used contained (mean ± S.D.): 0.36 ± 0.01 mg/ml of N, N-DMT, 1.86 ± 0.11 mg/ml of harmine, 0.24 ± 0.03 mg/ml of harmaline, and 1.20 ± 0.05 mg/ml of tetrahydroharmine (online Supplementary Table). After screening, patients underwent a washout period of 2 weeks on average and adjusted to the half-life time of the antidepressant medication in use. During dosing session, patients were not under any antidepressant medication, and a new treatment scheme was introduced only 7 days after dosing. If needed, benzodiazepines were allowed as a supporting hypnotic and/or anxiolytic agents (online Supplementary Tablefor demographic and clinical characteristics). Dosing sessions lasted approximately 8 h, from 8:00 a.m. to 4:00 p.m., and intake usually occurred at 10:00 a.m. After a light breakfast, patients were reminded about the effects they could experience, and strategies to help alleviating eventual difficulties. Patients were also told that they could receive ayahuasca and feel nothing, or placebo and feel something. Sessions took place in a quiet and comfortable living room-like environment, with a bed, a recliner, controlled temperature, natural, and dimmed light. Patients received a single dose of 1 ml/kg of placebo or ayahuasca adjusted to contain 0.36 mg/kg of N, N-DMT. They were asked to remain quiet, with their eyes closed, while focusing on their body, thoughts, and emotions. They were also allowed to listen to a predefined music playlist. Patients received support throughout the session from at least two investigators who remained in a room next door, offering assistance when needed. Acute effects were assessed with the Clinician-Administered Dissociative States Scale (CADSS), the Brief Psychiatric Rating Scale (BPRS), and the Young Mania Rating Scale (YMRS), applied at -10 min, +1:40 h, +2:40 h, and +4:00 h after intake. When the acute psychedelic effects ceased, patients had a last psychiatric evaluation, debriefed their experience, and responded to the Hallucinogenic Rating Scale (HRS)and Mystical Experience Questionnaire (MEQ30). Around 4:00 p.m. they could go home accompanied by a relative or friend. Only four patients presenting a more delicate condition remained as inpatients in the hospital ward for an entire week. Patients were asked to return for follow-up assessments at 1, 2, and 7 days after dosing.

STATISTICAL ANALYSIS

Analyses adhered to a modified intent-to-treat principle, including all patients who completed assessments at baseline, dosing and D7. An estimated sample size of 42 patients was estimated in G*Power software to provide 80% power to detect a five-point HAM-D difference (standardized effect size = 0.9) between baseline and D7 with two-sided 5% significance. The initial estimation was based on our previous open-label trial with ayahuasca in treatment-resistant depression. A fixed-effects linear mixed model, with baseline scores as covariate, examined changes in HAM-D at D7, and MADRS at D1, D2, and D7. A Toeplitz covariance structure was the best fit to the data according to Akaike's information criterion. Missing data were estimated using restricted maximum-likelihood estimation. Main effects and treatment v. time interaction were evaluated. Post-hoc t tests were performed for between-groups comparisons at all time points, and Sidak's test was used to control for multiple comparisons. Cohen's d effect sizes were obtained for between and within group comparisons. Between-group effect sizes were calculated using the estimated means of each group at each time point. For within-group comparisons, effect sizes of each treatment were calculated separately, using the differences between a time point and baseline values. Differences in the proportion of responders/non-responders and remitters/nonremitters were estimated using Fisher's exact test. Odds ratio (OR) and number needed to treat (NNT) were also calculated. Data from patients whose HAM-D or MADRS scores were reduced by 50% or more between washout onset and baseline, or that were in remission at dosing, were not considered for statistical analysis. Fisher's exact test was used to assess differences in the proportion of adverse events between the two treatments. We used the Mann-Whitney test to evaluate between-group differences in BPRS+, CADSS, HRS, and MEQ30. We calculated Pearson correlations between changes in MADRS scores from baseline to D7, and the acute effects during dosing assessed by BPRS, CADSS, MEQ30, and HRS. Multiple comparisons correction was based on the number of factors of each scale (N = 6, for the HRS; N = 5, for the MEQ30). Significance was set at p < 0.05, two-tailed. We used IBM SPSS Statistic 20 and Prism 7 to run the analyses.

RESULTS

From January 2014 to June 2016, we assessed 218 patients for eligibility, and 35 met criteria for the trial. Six subjects had to be excluded: five no longer met criteria for depression in the day of dosing, and one dropped out before dosing. Data from 29 patients were included in the analysis: 14 in the ayahuasca group and 15 in the placebo group. Figureshows the trial profile. On average, patients met criteria for moderate-to-severe depression (mean ± S.D.): HAM-D = 21.83 ± 5.35; MADRS =). All patients were Brazilian, most female (72%), adults (42.03 ± 11.66 yo) from low socioeconomic status backgrounds: low educated (41% with <8 years of formal education) and living in low household income (41% earn <2 minimum wages). Figureshows changes in HAM-D scores from baseline to 7 days after dosing. We observed a significant between-groups difference at D7 (F 1 = 6.31; p = 0.019), and patients treated with ayahuasca showed significantly reduced severity when compared with patients treated with placebo (online Supplementary Fig.for individual HAM-D scores). Between-group effect size was large at D7 (Cohen's d = 0.98; 95% CI 0.21-1.75). Within-group effect size (online Supplementary Table) was large for the ayahuasca group (Cohen's d = 2.22; 95% CI 1.28-3.17), and medium for the placebo group (Cohen's d = 0.46; 95% CI -0.27 to 1.18). Figureshows mean MADRS scores as a function of time. Linear mixed model showed a significant effect for time (F 2,34.4 = 3.96; p = 0.028), treatment (F 1,27.7 = 10.52; p = 0.003), but no treatment v. time interaction (F 2,34.4 = 1.77; p = 0.185). We observed significant decreased depression severity already 1 day after dosing with ayahuasca compared with placebo (F 1,49.7 = 4.58; p = 0.04). Depression severity persisted lower in the ayahuasca group at both D2 (F 1,50.3 = 4.67; p = 0.04) and D7 (F 1,47 = 14.81; p < 0.0001). Between-groups effect size was large at D1 (Cohen's d = 0.84; 95% CI 0.05-1.62) and D2 (Cohen's d = 0.84; 95% CI 0.05-1.63) and largest at D7 (Cohen's d = 1.49; 95% CI 0.67-2.32). Within-group effect sizes (online Supplementary Table) were large for the ayahuasca at all time points: Cohen's d = 2.78 at D1 (95% CI 1.74-3.82), d = 3.05 at D2 (95% CI 1.94-4.16), and d = 2.90 at D7 (95% CI 1.84-3.97). HAM-D response rate was significantly different betweengroups at D7, with 57% of responders in the ayahuasca group against 20% in the placebo group ; p = 0.04; NNT = 2.69]. HAM-D remission rate showed a trend toward significance at D7: 43% in ayahuasca v. 13% in placebo [OR 4.87 (95% CI 0.77-26.73); p = 0.07; NNT = 3.39]. Figureshows MADRS response rates as a function of time. At D1, response rates were high for both groups: 50% in the ayahuasca group, and 46% in the placebo group [OR 1.17 (95% CI 0.26-5.48); p = 0.87; NNT = 26]. At D2, they remained high in both groups: 77% in the ayahuasca group and 64% in the placebo [OR 1.85 (95% CI 0.29-8.40); p = 0.43; NNT = 7.91]. Response rate was statistically different at D7: 64% of responders in the ayahuasca group, and 27% in the placebo ); p = 0.04; NNT = 2.66]. Figureshows the MADRS remission rates as a function of time. At D1, the remission rate was of 42% in the ayahuasca group and 46% in the placebo group ( p = 0.86), at D2, 31% in the ayahuasca group and 50% in the placebo group ( p = 0.31). At D7 MADRS remission rate showed a trend toward significance: 36% of remitters in the ayahuasca group and 7% in the placebo [OR 7.78 (95% CI 0.81-77.48); p = 0.054; NNT = 3.44]. Online Supplementary Figs.andshow individual MADRS scores %-changes from baseline, at all time points. Although individual variance was high, we found improvement in depression severity for all patients in the ayahuasca group 7 days after dosing, while four patients in the placebo group have worsened their symptoms. Patients exhibited transient acute changes in CADSS and BPRS+ scales, with slightly increased scores at +1:40 h after ayahuasca intake: 34.8% (BPRS+) and 21.6% (CADSS) (online Supplementary Table). There was a trend increased toward significance observed in CADSS scores at 1:40 h after ayahuasca intake ( p = 0.052). There were no significant changes in BPRS+ scores at any time point (online Supplementary Table). Changes in BPRS+ and CADSS scores did not correlate with improvements in depression symptoms (online Supplementary Fig.). We did not observe significant increased manic symptoms as measured by the YMRS at 1:40 h after ayahuasca intake ( p = 0.26). We also observed transient nausea (Aya = 71%, Pla = 26%; p = 0.027), vomiting (Aya = 57%, Pla = 0%; p = 0.0007), transient anxiety (Aya = 50%, Pla = 73%; p = 0.263), restlessness (Aya = 50%, Pla = 20%; p = 0.128), and transient headache (Aya = 42%, Pla = 53%; p = 0.715) (online Supplementary Table). The average and standard error of mean (S.E.M.) for each factor of both scales (HRS and MEQ30) are presented in online Supplementary Table, and in online Supplementary Figs.and. Two patients did not respond to the HRS, leaving 27 respondents: 13 in the ayahuasca group, 14 in the placebo. The MEQ30 was added to the study with the trial already ongoing, and only 15 patients responded to it: eight in the ayahuasca group, seven in the placebo. Figureshows the HRS average score for all six subscales. We found significant differences groups in five of them. The ayahuasca group scored higher than the placebo group in perception ( p < 0.0001), somaesthesia ( p < 0.0001), cognition ( p < 0.0001), intensity ( p < 0.0001), and volition ( p = 0.0003). Only affect was not significantly different between groups ( p = 0.38). Figureshows scores for all factors of the MEQ30. We found significant between-groups differences in mystical ( p = 0.049), transcendence of time and space ( p = 0.0008), ineffability ( p = 0.003), and in total MEQ score ( p = 0.004). For all of these, the ayahuasca group scored higher than the placebo group. The only positive mood was not significantly different between groups ( p = 0.32). Correlations between HRS and MADRS changes from baseline to D7 were not statistically significant when assessing each group separately, ayahuasca, or placebo (online Supplementary Fig.). However, we observed a positive significant correlation between MADRS changes at D7 with the HRS subscale 'perception' (r = 0.90, p = 0.002), when considering the subgroup of ayahuasca responders only (online Supplementary Fig.). Despite the small number of patients, we found a negative correlation between changes in MADRS scores and the MEQ30 factor transcendence of time and space in patients in the ayahuasca group (r = -0.84, p = 0.009). The remaining three factors (ineffability, mystical, and positive mood) and MEQ30 total score were not significantly correlated with MADRS score changes (online Supplementary Fig.).

DISCUSSION

We found evidence of rapid antidepressant effect after a single dosing session with ayahuasca when compared with placebo. Depression severity changed significantly but differently for the ayahuasca and placebo groups. Improvements in the psychiatric scales in the ayahuasca group were significantly higher than those of the placebo group at all time points after dosing, with increasing between-group effect sizes from D1 to D7. Response rates were high for both groups at D1 and D2 and were significantly higher in the ayahuasca group at D7. Between-groups remission rate showed a trend toward significance at D7. The within-group effect size found for ayahuasca at D7 (Cohen's d = 2.22) is compatible with our earlier open-label study (Cohen's d at D7 = 1.83), and compatible with the one found in a recent open-label trial with psilocybin for depression (Hedges' g = 3.1). Our results are comparable with randomized controlled trials that used ketamine in treatment-resistant depression. Although both ketamine and ayahuasca are associated with rapid antidepressant effects, their response time-courses and mechanisms of action seem to differ. Previous studies with ketamine have found the largest between-group effect size at D1 (Cohen's d = 0.89), reducing toward D7 (Cohen's d = 0.41). In contrast, the effect sizes observed herein were large, but smallest, at D1 (Cohen's d = 0.84), and largest at D7 (Cohen's d = 1.49). These differences are also reflected in the response rates. At D1, the response rate to ketamine lies between 37 and 70%, whereas in our study 50% of the patients responded to ayahuasca. At D7, the ketamine response rate ranges between 7 and 35%, while in our study 64% responded to ayahuasca. The placebo effect was high in our study, and higher than most studies with. While we find a response rate to placebo of 46% at D1, and 26% at D7, ketamine trials have found a placebo effect of the order of 0-6% at D1, and 0-11% at D7. Several factors may account for the high placebo effect observed herein. First, the higher placebo effect has been found in patients with low socioeconomic status, which was the case of our study. Most patients were living under significant psychosocial stressors, and during our trial, they stayed at a 'very comfortable and very supportive environment', as reported by the patients themselves. Therefore, part of the increased placebo effects found in our study might be due to this 'care effect'. Second, patients with comorbid personality disorders present higher placebo responses, and in our study, most patients (76%) also suffered from personality disorders, most of them in cluster B. A growing body of evidence gives support to the observed rapid antidepressant effects of ayahuasca. For instance, sigma-1 receptors (σ1R) have been implicated in depression, and it was reported to be activated by N, N-DMT. Moreover, it has been shown that the administration of σ1R agonists results in antidepressant-like effects, which are blocked by σ1R antagonism. Furthermore, σ1R upregulates neurotrophic factors such as brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), proteins whose regulation and expression seem to be involved in the pathophysiology of depression. Nevertheless, it is worth mentioning that antidepressants with σ1R agonist profile do not present clinically significant antidepressant effect. For instance, the antidepressant fluvoxamine, which has a high affinity for σ1R do not present response rates compatible to that was found herein. The effects observed might be in part due to the presence of MAOi in the brew. In fact, studies in animal models reported that chronic administration of harmine reduces immobility time, increases climbing and swimming time, reverses anhedonia, increases adrenal gland weight, and increases BDNF levels in the hippocampus. All of these are compatible with antidepressant effects. Likewise, harmine seems to stimulate neurogenesis of human neural progenitor cells, derived from pluripotent stem cells, and progenitor cells from adult mice brains, a mechanism also observed in rodents following antidepressant treatment. In addition, a recent study in rodents found that a single ayahuasca dose increases swimming time in a forced-swim test. Brain circuits modulated by psychedelics show great overlap with those involved in mood disorders. We recently found that a single ayahuasca session in patients with depression increases blood flow in brain regions consistently implicated in the regulation of mood and emotions, such as the left nucleus accumbens, right insula and left subgenual area. Moreover, we have shown that ayahuasca reduces the activity of the Default Mode Network, a brain network found to be hyperactive in depression. Over the last two decades, mental health evaluations of regular ayahuasca consumers have shown preserved cognitive function, increased well-being, reduction of anxiety, and depressive symptoms when compared to non-ayahuasca consumers. Moreover, a recent study observed that a single dose of ayahuasca enhanced mindfulness-related capacities, and meditation practices have been associated with antidepressant effects. Prior studies suggest that elements of the psychedelic experience, such as experiences of mystical-type, account for the therapeutic benefit. We found significant increased MEQ30 scores during the effects of ayahuasca. We also observed an inverse correlation between MADRS score changes at D7 with 'transcendence of time and space' MEQ30 factor. Furthermore, HRS dimensions seem important to the clinical outcome, particularly 'perception', a subscale that comprehends changes in visual, auditory, and body sensations. Visions are common during the effects of ayahuasca, and are most frequent with the eyes closed. It has been suggested that visions may play an important role in the therapeutic effect of ayahuasca, as they may help bringing clarity to introspective events. It is interesting to observe that changes in perception taken alone are not sufficient to predict the positive clinical outcome, as for instance, we find that some patients presented increased scores in 'perception' without significant clinical response. No serious adverse events were observed during or after dosing. Although 100% of the patients reported feeling safe, the ayahuasca session was not necessarily a pleasant experience. In fact, some patients reported the opposite, as the experience was accompanied by much psychological distress. Most patients reported nausea, and about 57% have vomited, although vomiting is traditionally not considered a side effect of ayahuasca, but rather part of a purging process. Although promising, this study has some caveats and limitations worth mentioning. The number of participants is modest, and therefore randomized trials in larger populations are necessary. The study was limited to patients with treatment-resistant depression, with a long course of illness, and high comorbid personality disorder, which altogether precludes a simple extension of these results to other classes of depression. Another challenge of the research with psychedelics is maintaining double blindness, as the effects of psychedelics are unique. We were particularly keen to ensure blindness throughout the entire experiment, and to that end, we adopted a series of additional measures to preserve blindness. All patients were naïve to ayahuasca, with no previous experience with any other psychedelic substance. Clinical evaluations involved a team of five psychiatrists. For every patient, one psychiatrist was responsible for clinical evaluation during the dosing session and a different one for the follow-up assessments. The substance used as placebo increased anxiety and induced nausea. In fact, five patients misclassified placebo as ayahuasca, and two of them showed a response at D7 (online Supplementary Table). Therefore, we believe blindness was adequately preserved in our study. Since the prohibition of psychedelics in the late 1960s, research with these substances has almost come to a halt. Before research restrictions, psychedelics were at early stage testing for many psychiatric conditions, including obsessive-compulsive disorder and alcohol dependence. By mid-1960s, over 40.000 subjects had participated in clinical research with psychedelics, most of them in uncontrolled settings. To our knowledge, this is the first randomized placebo-controlled trial to investigate the antidepressant potential of a psychedelic in a population of patients with treatment-resistant depression. Overall, this study brings new evidence supporting the safety and therapeutic value of psychedelics, dosed within an appropriate setting, to help treat depression.

Company

Product

Legal

Rapid antidepressant effects of the psychedelic ayahuasca in treatment-resistant depression: a randomized placebo-controlled trial

Authors

Published

Links

Related Clinical Trial

Antidepressant Effects of Ayahuasca: a Randomized Placebo Controlled Trial in Treatment Resistant Depression

References (22)