Addressing blinding in classic psychedelic studies with innovative active placebos

Abstract

Classic psychedelics have shown promise in the treatment of various neuropsychiatric disorders. However, weak blinding integrity has been argued to limit the interpretability of therapeutic effects observed in psychedelic clinical trials, highlighting the need to explore alternative active placebos. Here, we aimed to describe the drawbacks of current placebo conditions used in classic psychedelic studies, propose criteria for suitable active placebos, and review interventions that may putatively fit these criteria. Considerations for the characteristics of ideal active placebos in classic psychedelic studies include (1) acute psychoactive effects, (2) acute physiological effects, (3) onset and duration of acute effects, (4) safety, and (5) lack of therapeutic effects in the target disease. We identified several pharmacological agents that may have potential as active placebos in trials involving moderate-to-high doses of certain short-acting and long-acting classic psychedelics, as well as low-dose administration and microdosing regimes. To accurately assess the safety and efficacy of classic psychedelics as therapeutics, future research should apply a thoughtful process for selecting active placebos and consider ancillary strategies to improve blinding in trials involving these substances.

INTRODUCTION

Classic psychedelics (eg, psilocybin, lysergic acid diethylamide [LSD], N,N-dimethyltryptamine [DMT]) are a group of drugs that act as full or partial serotonin 2A receptor agonists to produce dose-dependent, acute psychoactive effects.¹ These acute effects can, at certain doses, include major positively or negatively valenced alterations in the sense of self, sensory perception, and mood.² Recent studies suggest that classic psychedelics have therapeutic potential in the treatment of a number of neuropsychiatric disorders, including depressive disorders, substance use disorders, and some types of headache and pain disorders.^3,4 For example, in a recent randomized, placebo-controlled trial, a single moderate-to-high (25 mg) dose of psilocybin paired with psychological support was shown to reduce depressive symptoms compared with a placebo (ie, niacin) in patients with major depressive disorder.⁵ In another randomized, placebo-controlled trial, a pulse of low-dose (10 mg/70 kg) psilocybin administered 3 times 5 days apart was followed by a significant reduction in cluster attack frequency in patients with cluster headache.⁶ Despite these promising findings, there are several methodological challenges that need to be addressed to increase confidence in interpreting efficacy outcomes in clinical trials with classic psychedelics.^7,8 One fundamental challenge is accounting for placebo effects with adequate control conditions. In this article, we aim to briefly describe the drawbacks of current placebo conditions used in classic psychedelic studies, propose criteria for suitable active placebos, and review interventions that may fit these criteria.

Blinding and Expectancy Issues in Classic Psychedelic Research

Early modern-day studies and pilot investigations with classic psychedelics often used an open-label design, wherein participants, investigators, and those collecting data were aware of treatment assignment.^9,10 However, open-label design studies cannot definitively prove the efficacy or safety of the novel intervention because of factors such as regression to the mean and the placebo effect. The placebo effect is the measure of clinical benefit obtained from an intervention beyond treatment-specific therapeutic effects, and reflects several ancillary factors, such as belief in the efficacy of the treatment. Prior research has demonstrated that the placebo effect can produce substantial symptom improvements in patients with various neuropsychiatric disorders.^11-13 Hence, controlling for the placebo effect in novel interventions, such as classic psychedelics, is crucial to demonstrating their efficacy and safety.¹⁴

To accurately assess safety and efficacy, researchers use blinded control conditions to compare against the intervention under study. The control condition may be a known treatment with proven efficacy—or a placebo. There are 2 types of placebos that can be used in clinical research: inert and active placebos. While neither is expected to produce therapeutic effects in the target disease, inert placebos are physiologically (and psychologically as in the case of classic psychedelics) inactive, and active placebos are meant to resemble some aspects of the intervention under study.

The use of an inert placebo (eg, microcrystalline cellulose) does not usually suffice in classic psychedelic research. This is because, at certain doses, classic psychedelics produce acute robust alterations of sensory and perceptual systems,¹⁵ as well as consciousness itself.² These acute subjective effects are often obvious to study participants, study therapists, and raters, making it easy to infer treatment allocation and, therefore, leading to functional unblinding.¹⁴ Using an inert placebo raises the risk of unblinding and allows treatment expectations to obfuscate the outcomes of the study. For instance, a participant who has correctly identified that they have been assigned to the active treatment may benefit through positive expectation alone. Similarly, if a participant does not experience the expected acute classic psychedelic effects, they will infer that they did not receive the active treatment (ie, they received placebo instead), which could, in turn, lead to disappointment and symptom worsening.⁷ In addition, study therapists and other staff who infer the treatment received by participants may inadvertently engage with them based on that inference.¹⁶ These various factors complicate the interpretability of results in classic psychedelic trials that have weak blinding integrity. To minimize the likelihood of functional unblinding, it is therefore important for classic psychedelic research to use active placebos that produce acute subjective effects that are not easily distinguishable from the active treatment.

The selection of an active placebo must be considered carefully. The active placebos that have been administered in modern-day trials involving classic psychedelics typically include niacin, methylphenidate, diphenhydramine, or a low dose of the classic psychedelic under investigation.^17,18 These active placebos may produce some acute subjective effects, but they do not produce moderate-to-high-dose classic psychedelic-like effects, and blinding integrity, in the rare instances it has been systematically assessed, has been poor.¹⁹ For example, in a recent randomized, placebo-controlled trial in patients with alcohol use disorder, 2 moderate-to-high doses (25 mg/70 kg, first session; 25–40 mg/70 kg, second session) of psilocybin, paired with psychotherapy, were associated with a reduction in the percentage of heavy drinking days, compared with a placebo (ie, diphenhydramine). However, study therapists and participants accurately guessed treatment assignment more than 90% of the time,²⁰ which may limit the interpretability of the trial. This is a particularly timely issue considering that the US Food and Drug Administration’s (FDA) Psychopharmacologic Drugs Advisory Committee in June 2024 voted against Lykos Therapeutics’ New Drug Application 3,4-methylenedioxymethamphetamine (MDMA), paired with psychotherapy, as a treatment for posttraumatic stress disorder, partly because of concerns regarding functional unblinding.²¹ Notably, the application was eventually denied in August 2024, with the FDA requiring a more rigorous Phase III study. Similarly, trials of ketamine for mental health conditions have recently come under scrutiny for potential unblinding, and when strong blinding procedures have been deployed (eg, administration under general anesthesia), some studies have found benefits to be no stronger than placebo.²² Although MDMA and ketamine are not considered classic psychedelics, these concerns are similarly applicable to classic psychedelic research, which highlights the need to carefully consider approaches to adequately blind studies involving these drugs.

Ideal Active Placebo Characteristics in Classic Psychedelic Research

To identify suitable pharmacological placebo candidates, it is important to define the characteristics of an ideal active placebo for use in studies with classic psychedelics. First, the active placebo should produce acute psychological changes that are perceived as generally similar to those produced by the classic psychedelic and the dose under investigation. Second, the active placebo should produce acute somatic/physiological effects similar to that of the classic psychedelic and the dose under investigation. Third, the active placebo should have an onset and duration of acute effects similar to that of the classic psychedelic and the dose under investigation. Fourth, to minimize the risk of adverse reactions, the active placebo should be known to be generally nontoxic and safe in controlled settings with no persisting adverse effects. Fifth, the active placebo should not be expected or known to have therapeutic effects in the target disease (eg, major depressive disorder), which would otherwise make it difficult to distinguish the true effects of the classic psychedelic treatment.

If most, if not all, of the above criteria were fulfilled by a pharmacological agent, it could potentially address issues with regard to both placebo and nocebo effects in studies involving classic psychedelics. However, the active placebos that have typically been used in modern-day trials involving moderate-to-high-dose classic psychedelics do not meet all the proposed criteria of an ideal placebo. For example, oral niacin has a similar onset of acute effects to oral psilocybin, but the duration and nature of the effects (ie, 30 minutes of flushing) do not mimic those of moderate-to-high-dose psilocybin (eg, 6 hours of closed-eye visuals, time distortion). There are, however, other pharmacological agents that may more successfully maintain blinding integrity and accurately control for the placebo effect in moderate-to-high-dose classic psychedelic studies (Table 1).

Table 1.

Characteristics of placebos in classic psychedelic trials.

	Inert placebo (eg, MCC)	Niacin(oral)	Salvinorin A(IV or vaped)	Dextromethorphan(oral)	Delta-9-tetrahydrocannabinol(oral)	Diphenhydramine(oral)	Stimulant(eg, caffeine, methylphenidate)(oral)
Primary receptor target	n/a	GPCR 109A niacin receptor activation	Kappa opioid receptor agonist	NMDAR antagonist; Sigma-1 agonist	Cannabinoid CB1 receptor agonist	Histamine H1 receptor antagonist	Adenosine receptor antagonist, catecholamine reuptake blockade
Proposed for blinding	Not recommended	Limited utility	IV or vaped DMT	Oral psilocybin and LSD	Oral psilocybin and LSD	Low-dose oral psilocybin and LSD	Microdosing (chronic sub-perceptual dosing)
(1) Acute psychological changes	None	Minimal	+	+	+	+	Mild stimulation possible
(2) Acute physiological changes	None	None	+	+	+	+	Mild stimulation possible
(3a) Onset of acute effects	n/a	Rapid	Rapid	Slow	Slow	Slow	Intermediate
(3b) Duration of acute effects	n/a	Brief (30 minutes)	Brief (30 minutes)	Hours	Hours	Hours	Hours
(4) Safety	+	+	+	+	+	+	+
(5) Therapeutic target	None	None	Unlikely with limited dosing	Unlikely with limited dosing	Unlikely with limited dosing	Unlikely with limited dosing	Possible with chronic dosing

Abbreviations: DMT, N,N-dimethyltryptamine; LSD, lysergic acid diethylamide; THC, tetrahydrocannabinol.

The ideal active placebo produces acute (1) psychological and (2) physiological effects with an (3a) onset (3b) duration matching that of the classic psychedelic and dose being investigated. It should be (4) safe and (5) lack target effects at the dose and regimen being administered. An inert placebo only has safety and lack of target effects. Niacin meets some criteria but does not match the onset/duration of longer-lasting classic psychedelics such as LSD and lacks substantive psychological effects. Salvinorin A and dextromethorphan are good considerations for classic psychedelics with shorter-lasting and longer-lasting acute effects, respectively. THC is a good consideration for orally administered psilocybin or LSD, possibly better for lower doses. Diphenhydramine might meet all criteria if being used against low-dose psilocybin or LSD, but not moderate–high doses. Stimulants might be useful in microdosing studies, which unlike other study designs involve chronic and/or prolonged dosing, but depending on the outcome measure, these might also have therapeutic effects.

Potential Active Placebos for Moderate-to-High Doses of Classic Psychedelics

Salvinorin A is a selective kappa opioid receptor (KOR) agonist that produces dose-related, acute dissociative, and classic psychedelic-like effects. Previous research suggests that the phenomenological effects of salvinorin A may have moderate overlaps (eg, cognitive, perceptual, somatosensory), including potentially challenging experiences, with those induced by moderate-to-high doses of classic psychedelics.^23-25 The onset and duration of acute effects between salvinorin A and a short-acting classic psychedelic, such as moderate-to-high doses of intravenous DMT, are also comparable (ie, 10–30 minutes). Notably, both drugs can be administered via the same routes (vaporization or intravenously^26,27), thus facilitating blinding. The therapeutic effects (eg, antidepressant) of salvinorin A have not been studied in clinical trials, making it difficult to assess whether it would violate one of the most important features of an ideal active placebo (ie, not be expected or known to have therapeutic effects). In animal models, KOR agonists are aversive, anxiogenic, and produce dysphoric outcomes which can be reversed by KOR antagonists,²⁸ suggesting that salvinorin A is unlikely to have antidepressant effects. On the other hand, it is conceivable, but as of yet untested, that KOR agonists may trigger post-acute prodepressant effects in clinical populations, which would inflate the antidepressant effects of psychedelics. It should also be noted that research suggests that salvinorin A, similarly to psilocybin and LSD, may increase neural entropy and decrease default mode network (DMN) activity²⁹—hypothesized therapeutic mechanisms of action of classic psychedelics. Although it might be possible to develop a different dosing regimen of salvinorin A (eg, slower intravenous administration) that would allow a similar duration of effects to any of the long-acting classic psychedelics (eg, psilocybin, LSD), it may not be practical or safe. Other KOR agonists such as spiradoline (U-62066) that have a slower onset and longer duration of acute effects³⁰ could instead be explored as an active placebo for orally administered classic psychedelics that have slower and longer acute psychedelic effects.

With regard to studies with long-acting classic psychedelics, another drug with intriguing potential as an active placebo is dextromethorphan (DXM), an N-methyl-D-aspartate (NMDA) receptor antagonist found in many over-the-counter cold medicines. Notably, at certain doses, DXM can cause acute effects characteristic of moderate-to-high doses of certain classic psychedelics such as LSD and psilocybin.³¹ For example, in a double-blind randomized controlled trial comparing low (10 mg/70 kg), moderate (20 mg/70 kg), and high (30 mg/70 kg) oral doses of psilocybin and high-dose DXM in 20 healthy participants, Carbonaro and colleagues³² found both drugs yielded comparable time courses and increases in participant ratings of peak overall drug effect strength. The drugs had no serious adverse events reported and shared some phenomenological similarities, but moderate-to-high doses of psilocybin produced significantly greater scores on insight and mystical-type experiences than high-dose DXM—a finding that might be especially important if the therapeutic effects of classic psychedelics are driven in part by the acute subjective experience, as suggested by some³³ but not others.³⁴ If certain aspects of the psychedelic experience, such as mystical-type effects and insight, are crucial therapeutic mechanisms of action with psilocybin, DXM, and other control conditions that closely mimic the subjective effects without providing those elements may make for excellent active placebo conditions. It is also important to highlight, however, that the percentage of participants receiving high-dose DXM who have guessed that they received a classic psychedelic-type drug like psilocybin or LSD has differed across studies,^32,35,36 suggesting that extra-pharmacological factors may influence blinding integrity in classic psychedelic studies with high-dose DXM as an active placebo. Lastly, DXM is an NMDA receptor antagonist and may have therapeutic potential in depression. In fact, when combined with bupropion hydrochloride (Auvelity), it is approved for the treatment of depression.^37,38 DXM in combination with quinidine (Nuedexta) is also used in the management of pseudobulbar effect in amyotrophic lateral sclerosis and multiple sclerosis.³⁹ However, these DXM-containing treatments must be taken daily to take effect and are not expected to provide long-lasting clinical benefit after a single or few doses.

Future studies involving long-acting classic psychedelics could also investigate delta-9-tetrahydrocannabinol (THC). When administered orally, THC has a comparable onset and duration of acute effects to certain long-acting classic psychedelics (eg, psilocybin) and is known to alter perception and mood. The use of THC as an active placebo in classic psychedelic studies could potentially also be more acceptable to some participants, given the increasingly positive public perception of cannabis in countries such as the United States.⁴⁰ Earleywine et al.⁴¹ found that high doses of cannabis were associated with subjective effects comparable to findings from high-dose psilocybin trials, converging with other recent research supporting that altered states of consciousness induced by THC may overlap with classic psychedelics.⁴² Importantly, THC is also generally well tolerated and is not known to produce prolonged therapeutic benefits after a single administration for any clinical condition, fulfilling crucial criteria for an active placebo in classic psychedelic studies. Lastly, neuroimaging studies suggest that THC may have opposing effects on neural correlates thought to be therapeutically relevant to classic psychedelics, such as DMN activity and neural entropy.^43,44 Although oral THC represents a promising active placebo candidate in high-dose psychedelic studies, it should be noted that blinding integrity might be compromised in study participants with familiarity with cannabis or other THC products through recreational or medicinal use.

Potential Active Placebos for Low or Sub-Perceptual Doses of Classic Psychedelics

It is also worth considering pharmacological agents that may not be ideal as active placebos in moderate-to-high-dose classic psychedelic studies that could instead be used in studies involving low doses of classic psychedelics. The low-dose range of psilocybin and LSD that has been administered in clinical trials of some types of headache and pain disorders⁴⁵ is based in large part on the experience of patients who self-medicate with these drugs.⁴⁶ For example, in a recent migraine trial investigating the preventive effects of a single low dose (10 mg) of psilocybin compared with a placebo (diphenhydramine 25 mg), between the 2 dosing sessions participants could not identify the drug when only receiving one or the other, though they correctly identified psilocybin when receiving both drugs (Schindler et al., under review). This finding is informative and raises an additional consideration in study design related to the number of dosing sessions and whether study participants, based on the design, will have the opportunity to compare the acute subjective effects when receiving both the classic psychedelic and the active placebo. In addition to this important design consideration, other pharmacological agents that have mild psychoactive effects (eg, benzodiazepines, low doses of THC) could also be investigated as potential active placebos in low-dose classic psychedelic trials.

Although most clinical trials with classic psychedelic research have involved the single administration of a perceptual (ie, low, moderate, or high) dose of these drugs (or some with 2–3 total doses within a period of weeks or months), recent randomized, placebo-controlled trials have also investigated prolonged, regular dosing with sub-perceptual or minimally perceptual doses of classic psychedelics (ie, “microdosing”⁴⁷). These studies are taking place in part due to emerging practices of self-administration that have a variety of goals, including improved mood, well-being, cognition, and creativity. These microdosing studies have not included active placebos,⁴⁸ which limits the interpretability of findings. However, because sub-perceptual doses of classic psychedelics have limited acute psychoactive or physiological effects, the range of potential pharmacological agents that could blind participants in these studies may be greater than in those involving low, moderate, or high doses of classic psychedelics. For example, caffeine, diphenhydramine, methylphenidate, or other stimulants may be suitable as active placebos for these types of regimens, especially since sub-perceptual doses of classic psychedelics often involve stimulant-like acute effects.^49-51 Both caffeine and methylphenidate have comparable onset times to classic psychedelics, but the acute effects vary across these drugs and selection should therefore be informed by the type (ie, short- vs. long-acting) of classic psychedelic under investigation. It is also important to consider the outcomes of interest when selecting an active placebo for these studies, given that caffeine and methylphenidate may increase specific cognitive domains that overlap with reports from those who self-administer classic psychedelic microdosing regimens.⁴⁸ Similarly, repeated administration of stimulants has been shown to have therapeutic efficacy for some disorders (eg, depression,⁵² pain⁵³), underscoring the importance of considering the indication under study as well as the drug.

Other Considerations

Regardless of the pharmacological agent that is selected as the active placebo in trials involving short- or long-acting classic psychedelics, as well as low-dose administration and microdosing regimes of these drugs, additional strategies may further improve blinding in classic psychedelic studies. Such strategies could include, for example, recruiting classic psychedelic-naive participants, including placebo run-in periods, considering incomplete disclosure of the treatment arms, noting to participants significant variability in drug response, listing all study drug effects together unspecified, and measuring expectations and blinding so that they can be included as covariates in statistical analyses (see Aday et al.⁷ for a review). Another strategy that has been proposed involves combining the active treatment and the active placebo with virtual reality (VR) paradigms that have been developed to simulate the classic psychedelic experience.⁵⁴ However, there are possible issues with this approach, including potential discomfort or interruption of therapeutic processes caused by engagement with VR for the length of the classic psychedelic dosing sessions, unblinding during periods where participants are not immersed in VR (eg, interactions with study staff, restroom breaks, etc.), as well as possible therapeutic effects in the target disease by VR itself (eg, Kaup et al.⁵⁵).

CONCLUSION

In summary, classic psychedelics have shown promise as a treatment for various neuropsychiatric disorders, but weak blinding integrity in trials has limited the capacity to definitively attribute therapeutic effects to the classic psychedelic and dose under investigation, which highlights the need to explore alternative active placebos. Using active placebos with acceptable safety profiles, no known therapeutic effects on the target disease, and that approximate the time-course, subjective, and physiological effects of classic psychedelics should be a suitable strategy to improve blinding. In our review, we offered a set of criteria for the ideal active placebo to maintain blinding integrity in psychedelic studies and described various drugs that may meet some, if not all, criteria for moderate-to-high doses of certain short-acting (eg, DMT) and long-acting (eg, psilocybin) classic psychedelics, as well as low-dose administration and microdosing regimes of these drugs. However, there are additional factors in this field of research requiring scrutiny to fully understand the therapeutic effects of this drug class. Some of these factors include high expectation (eg, fueled by media hype and parties with commercial interest¹⁹), concomitant therapies that are proposed to work synergistically with classic psychedelics (eg, psychotherapy^8,56), curated settings that enhance the acute classic psychedelic experience,⁵⁷ and the role of the classic psychedelic experience itself.⁵⁸ These many unresolved facets of classic psychedelic research create a continued need for rigorous research in the field that is not obfuscated by functional unblinding or other non-drug factors.

Author contributions

Jacob Aday (Writing—original draft [equal], Writing—review & editing [lead]), Otto Simonsson (Conceptualization [equal], Writing—original draft [Equal], Writing—review & editing [equal]), Emmanuelle A.D. Schindler (Writing—original draft [equal], Writing—review & editing [equal]), and Cyril D'Souza (Writing—original draft [equal], Writing—review & editing [equal])

Funding

O.S. was supported by Olle Engvist Foundation. J.S.A’s effort in this publication was partially supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the National Institutes of Health under T32AR007080-44S1. E.A.D.S. has received research grant support administered through Yale University from Heffter Research Institute, Ceruvia Lifesciences, Wallace Research Foundation, Clusterbusters, Inc., and VA R&D unrelated to her efforts on this publication. D.C.D. has received research grant support administered through Yale University from Heffter Research Institute, Wallace Research Foundation, VA R&D, and the Vikram Sodhi Endowed Chair of Psychiatry.

Conflicts of interest

O.S. was a co-founder of Eudelics AB and has once received a small payment from Mindfully Sweden AB for educational content. E.A.D.S. serves on the Scientific Advisory Board of Ceruvia Lifesciences, Clusterbusters, and OptoSom. She is listed as an inventor on patent US20210236523A1. D.C.D. serves on the Scientific Advisory Board of Ceruvia Lifesciences and the Physicians Advisory Board of the Connecticut Medical Marijuana Program. He has served as a Consultant for Atai, Abide Therapeutics, Jazz Pharmaceuticals, and Biohaven. He is listed as an inventor on patent US20210236523A1 and serves on the Scientific Advisory Board of the Shulgin Institute.

Data availability

There is no primary data associated with this manuscript.