Globally, the 12-month prevalence of substance use disorder (SUD) is 2.2%, with alcohol use disorder as the most prevalent (1.5%) of all SUDs (excluding tobacco use disorder). The 12-month prevalence of SUDs is even higher among high-income countries (1). Alcohol use accounts for 4.2% of all disability-adjusted life-years (DALYs), while drug use accounts for 1.3% of all DALYs worldwide (2). In addition to the detrimental health consequences, there are also enormous economic consequences of SUD, with an estimated cost of $249 billion from alcohol, $300 billion from tobacco, and $193 billion from other drugs a year in the United States of America (USA) alone. Despite these adverse health and economic consequences of SUDs, therapeutic options are still limited. Food- and Drug Administration (FDA) approved medications are available for alcohol, tobacco, and opioid use disorders. Still, for other SUDs, such as cocaine and cannabis use disorders, no approved medications are available (3). Relapse rates remain high across all SUDs, with >75% of patients with primary cannabis and cocaine use disorder and >65% of primary alcohol and opioid use disorder showing relapse throughout 12 months after evidence-based treatment (4). Arguably, relapse rates are even higher in non-substance-related disorders, as >90% of patients with gambling disorder who had recently quit gambling showed relapse throughout 12 months (5). This means, there is an unmet need for effective treatments of SUD and non-substance-related disorders.

Psychedelics [e.g., with lysergic acid diethylamide (LSD)] have shown promising results as treatment of SUDs. In the majority of studies, some form of psychotherapy was given in combination with the psychedelic treatment, but not in all, which is usually referred to as psychedelic-assisted therapy (6). Psychedelic-assisted therapy comprises three stages: (1) preparation; (2) the psychedelic session; and (3) integration. Preparation is thought to be vital for maximizing the potential benefit from the psychedelic session, while integration is believed to be important for prolonging the improvements (7). Currently, interest in the efficacy of psychedelic treatment for SUDs is increasing (8). The term “psychedelic” was first proposed by Oswald in 1957 and comes from the Greek words “psyche” (i.e., mind) and “deloun” (i.e., manifesting) and refers to the subjective effects of these agents. A categorization of psychedelics can be made based on their chemical structure into three different classes: (1) tryptamines (e.g., psilocybin); (2) ergolines (e.g., LSD); and (3) phenethylamines (e.g., mescaline) (9). All these agents are closely related to the endogenous neurotransmitter serotonin [i.e., 5-hydroxytryptamine (5-HT)] and induce their effect via activation of the 5-HT_2A-receptors (10). Sometimes, a more broad definition of psychedelics is used, and dissociates, and deliriants (e.g., ketamine and ibogaine) are included as well, which have a different mechanism of action (9). Swiss chemist dr. Albert Hoffman first synthesized LSD in 1938, (11) while he first identified psilocybin and its active metabolite psilocin as the psychoactive compounds of the psilocybe mushrooms in 1958. In the subsequent year, he synthesized psilocybin, which was later marketed as Indocybin by the pharmaceutical company Sandoz (12). Throughout the 1950s and 1960s, considerable research was conducted investigating the efficacy of LSD and psilocybin for various conditions, including anxiety, depression, and existential distress in patients with terminal cancer, in patients with opioid use disorder, and in patients with alcohol use disorder (6, 13). Possible therapeutic mechanisms of psychedelics might be at the biochemical, neural, and psychological level (14).

Krebs and Johansen (15) conducted a meta-analysis of randomized controlled trials (RCTs) evaluating the efficacy of LSD for alcohol use disorder. They identified k = 6 eligible trials, including n = 536 patients, and found a beneficial effect of LSD on alcohol use disorder with comparable efficacy to disulfiram (15). Savage and McCabe (16) also found a beneficial effect of LSD on heroin use disorder in n = 78 inmates of correctional institutions (16). However, in 1970, the Controlled Substances Act was signed into law by former USA President Richard Nixon placing psychedelics such as LSD and psilocybin in Schedule 1 (i.e., has a high potential for abuse, has no currently accepted medical use in treatment, and has a lack of safety for use under medical supervision), which hindered conducting psychedelic research (17). Other factors, such as negative publicity about psychedelics, doubts regarding their efficacy, and allocation of funding to related research fields (e.g., schizophrenia), increasingly discouraged research on psychedelics (18).

Recent years have seen a resurgence in psychedelic research, a development sometimes referred to as the psychedelic renaissance. Since 2016, three systematic reviews have been conducted evaluating the efficacy of psychedelics (i.e., psilocybin, but also other classic serotonergic psychedelics such as LSD were included in these systematic reviews) as a treatment for psychiatric disorders (i.e., SUD, but also other psychiatric disorders such as depression and anxiety were included in these systematic reviews) (19–21). Although these reviews provide an excellent overview of clinical trials published in the last 25 years, including multiple psychedelics and conditions, they may have missed clinical trials evaluating the efficacy of psychedelics that were conducted before the 1980s (19–21). As shown by Fuentes et al. (6), many trials have been conducted assessing the efficacy of LSD in SUD in these early decades, which may have been an important era for psilocybin research as well (6). In addition, many trials evaluating the efficacy of psilocybin in SUDs have been started since 2016, some of which may already have been published and not yet included in previous systematic reviews. Therefore, we performed a systematic review without restrictions on publication date. Our aim was to assess the efficacy of psilocybin in patients with SUD and non-substance-related disorders. In addition, a quality assessment was done of the included clinical trials.

We retrieved a total of k = 6832 unique records through our systematic search in various electronic databases (Figure 1). After screening titles and abstracts, k = 36 full-text articles were assessed for eligibility, and k = 6 articles were finally included in this systematic review. K = 2 articles were long-term follow-up results from the same clinical trial. All included clinical trials were conducted either in the USA or Poland.

A total of k = 130 registered clinical trials were identified through our systematic search in the two trial databases (Supplementary Figure 1). After screening, k = 11 ongoing trials on September 2, 2022, were included in this systematic review.

In this systematic review, four clinical trials (n = 151 patients) assessing the efficacy of psilocybin in SUD were included, of which three clinical trials were in patients with alcohol use disorder and one clinical trial was in patients with tobacco use disorder. One of the trials in patients with alcohol use disorder was a double-blind, placebo-controlled RCT; the other three trials were single-arm pilot studies with a substantial risk of bias. All four clinical trials, which combined psilocybin with some form of psychotherapy, provided evidence for a significant beneficial effect of psilocybin-assisted therapy in patients with either alcohol or tobacco use disorder. These encouraging results have resulted in several (double-blind, placebo-controlled) RCTs currently ongoing in alcohol, tobacco, cocaine, opioid, and methamphetamine use disorders. Notable is the lack of trials evaluating the efficacy of psilocybin in patients with cannabis use disorder, despite the high prevalence and the lack of any approved medications for this condition.

Fuentes et al. (6) showed in their systematic review that eight eligible RCTs evaluating the efficacy of LSD in patients with alcohol use disorder and one eligible RCT in patients with opioid use disorder were conducted in the 1950s, 1960s, and 1970s (6). Krebs and Johansen (15) identified 33 clinical trials, of which only six met modern-day criteria of randomization, blinding, and outcome assessment (15). This is in stark contrast with the number of clinical trials evaluating the efficacy of psilocybin in patients with SUD before the 1980s, which was only one, and was contrary to our expectations. One trial evaluating psilocybin in patients with alcohol use disorder conducted in the 1960s was excluded from this systematic review. It did not include an outcome assessing severity of substance use or abstinence, but only evaluated treatment-related adverse effects (31). Three trials included in this systematic review were conducted in the 21st century in the era after the Controlled Substances Act was signed in 1970, while no trial has been conducted in the past four decades evaluating the efficacy of LSD in SUD (6). Although it has been claimed that >2000 publications have been published evaluating psychedelics in conditions such as SUD, it seems that the majority used LSD as intervention, not psilocybin (32, 33). Both LSD and psilocybin primarily act via stimulation of the 5-HT_2A-receptors, but LSD exhibits additional affinity for dopaminergic and α-adrenergic receptors (34). The acute effects of LSD (100 and 200 μg) were directly compared to psilocybin (15 and 30 mg) in a double-blind, placebo-controlled RCT in healthy volunteers. The two psychedelics showed qualitatively, and quantitatively very similar subjective effects and differences in effect are dose-dependent rather than compound-dependent. The onset of effects is significantly earlier, and the duration is substantially longer in LSD compared to psilocybin (35). A dose of 30 mg psilocybin, which is comparable to the dosages used by Bogenschutz et al. (25), Bogenschutz et al. (26), and Johnson et al. (29), (25, 26, 29) was found similar to 100 and 200 μg LSD (35). However, in the meta-analysis by Krebs and Johansen (15), which showed a beneficial effect of LSD on alcohol use disorder, 83% (5/6) of included trials used a single fixed dose of ≥450 μg LSD with a maximum of 800 μg LSD (15). It is unclear whether higher and/or multiple dosages translate to a more favorable effect than lower and/or single dosages of psilocybin or LSD. In addition, all the included studies in this systematic review adjusted psilocybin dose to body weight, (25–30) although there is no clear benefit for this adjustment and fixed dosing is more practical in clinical trials (35, 36).

Proposed working mechanisms of psychedelics, including psilocybin, for improving psychiatric symptoms such as depression and SUD are both biological (e.g., by inducing brain neuroplasticity through elevating Brain-Derived Neurotrophic Factor (BDNF) levels, which are diminished in psychiatric conditions) (34) and psychological (e.g., the degree to which a patient has a mystical-type experience during the psilocybin session seems an important mediator for the enduring effect of the psilocybin) (25, 37). The latter proposed working mechanism seems especially important in SUD as the mystical-type experience seems to induce behavioral change in a patient with SUD. This working mechanism would also explain why psychedelics seem effective in different SUDs as opposed to the currently approved pharmacological treatments for SUD, which have a different working mechanism and are, in most cases, only effective in a specific SUD (e.g., disulfiram in alcohol use disorder).

In the design of clinical trials (with psilocybin treatment), several important issues need to be taken into account, including an adequate control group, randomization, blinding, and generalizability of the results. Regulatory agencies even recognize single-arm trials with external historical controls to assess promising treatments for specific indications. However, three trials in this systematic review were non-randomized, had no adequate control group, and were therefore prone to bias. Preferably, future trials are double-blind, placebo-controlled RCTs to account for a potential placebo effect. Most tobacco use disorder patients had a past use of psychedelics, which could have influenced the expectancy of these patients regarding the efficacy of psilocybin and have thus an effect on the results (29). Pretreatment-positive expectations about psychedelic microdosing predicted improvements in psychiatric symptoms (38). A meta-analysis by Wilkinson et al. (39) evaluating the efficacy of ketamine for major depressive disorder or bipolar disorder showed a dramatic reduction in effect size when comparing the impact of an inactive (saline) to the active placebo midazolam (effect size d = 1.8, 95% CI = 1.4–2.2 vs. d = 0.7, 95% CI = 0.4–0.9) (39). Despite comparing psilocybin to an active placebo, Bogenschutz et al. (26) still observed a medium effect size for the efficacy of psilocybin. However, even in a double-blind, placebo-controlled RCT, this problem with expectancy effects, a key contributor to the placebo response, is not easily accounted for given >90% of patients and study therapists correctly guessed treatment assignment to either psilocybin or the active placebo diphenhydramine (26). Maintaining successful masking remains a major challenge in psychedelic clinical trials for which there is not yet an adequate solution. Another challenging problem is the generalizability of the results. Johnson et al. (29) screened n = 323 individuals, and only n = 15 (5%) patients were eventually included. (29) Bogenschutz et al. (25) screened n = 70 individuals, and only n = 10 (14%) patients were eventually included (25). Bogenschutz et al. (26) screened n = 569 individuals, and only n = 95 (17%) were eventually randomized (26). From a safety point of view strict in- and exclusion criteria makes sense. Still, a risk is that external validity is relatively low and that diminished effects will be observed with psilocybin in a therapeutic setting (33). Coadministration of an extensive psychological treatment (motivational enhancement therapy and cognitive behavioral therapy) during the clinical trials makes interpretation of the treatment effect of psilocybin on alcohol and tobacco use disorder even more difficult (25, 26, 29). Ideally, an RCT would compare four different treatment arms as in the RCT by Grabski et al. (40) evaluating the efficacy of adjunctive ketamine in patients with alcohol use disorder to separate the different treatment effects. In this double-blind, placebo-controlled RCT, the four treatment arms were: (1) ketamine + psychotherapy; (2) placebo + psychotherapy; (3) ketamine + alcohol education (therapy control); and (4) placebo + alcohol education. The greatest reduction in alcohol use was seen in the ketamine + psychotherapy group vs. placebo + alcohol education (mean difference of 15.9, 95% CI = 3.8–28.1) (40). It seems the eventual treatment strategy will likely consist of psilocybin combined with psychotherapy, given the latter seems to be needed to derive meaning from the psilocybin sessions and incorporate insights from the psilocybin sessions into daily life and change current substance use behavior. However, alternative hypotheses assume more direct pharmacological working mechanism of psychedelics, and thus microdosing would be a potential treatment eliminating the need for a psychedelic experience. However, so far, results with psychedelic microdosing are less promising (41).

Currently, > 10 clinical trials are registered evaluating the efficacy of psilocybin in SUD, most are in alcohol use disorder. Other patient populations include tobacco, cocaine, opioids, and methamphetamine use disorder. Conducting clinical trials in alcohol and tobacco use disorder can be understood from their high prevalence worldwide. It is less clear why no clinical trials in cannabis use disorder are currently conducted as it is the most common drug use disorder worldwide, especially among youth (42). Other interesting patient populations to study the efficacy of psilocybin would be benzodiazepines or hypnotics use disorder and gambling disorder, for which there is a pressing need for effective treatments. Almost all registered trials have a control group and are randomized. The risk of confounding, which was serious to critical in the included non-randomized studies in this systemic review, is minimized. The registered RCT by Johnson (ClinicalTrials.gov Identifier: NCT01943994) is of special interest, evaluating the efficacy of psilocybin combined with psychotherapy versus an approved pharmacological treatment (i.e., nicotine replacement therapy) combined with psychotherapy. Here, psilocybin will be compared for the first time with an approved active pharmacological treatment (in patients with SUD). Maintaining masking in double-blind, placebo-controlled RCTs evaluating psychedelics is usually unsuccessful, which may result in an increased effect size for the efficacy of psilocybin. It is essential to know how psilocybin compares to approved pharmacological treatments and its effect size in comparative efficacy trials. Given that psychological treatments have high relapse rates in the long term, the limited number of available approved pharmacological interventions for SUD, and the high adverse health and economic consequences of SUDs, alternative treatment options are highly warranted for this population. Psilocybin has shown promising first results in SUD treatment that should be replicated and established in larger trials. Questions whether psychotherapy is a necessary treatment component for psilocybin to be effective and if so, what form of psychotherapy is indicated, and whether higher and/or multiple dosages translate to a more favorable effect compared to lower and/or single dosages of psilocybin are not addressed in the currently registered clinical trials. The results of ongoing and future studies assessing the efficacy of psychedelics in SUD will reveal their true potential.

In this systematic review, we identified only one double-blind, placebo-controlled RCT, and three small clinical trials, of which three were conducted in the 21st century assessing the efficacy of psilocybin in patients with alcohol and tobacco use disorder. All four studies combined psilocybin with some form of psychotherapy and showed a beneficial effect of psilocybin-assisted therapy on SUD, but the risk of bias ranged from some concerns to critical. Future (double-blind, placebo-controlled) RCTs in patients with SUD need to evaluate whether psilocybin-assisted therapy is effective in this population.

The original contributions presented in this study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

PM and AB designed the study. PM and JF reviewed the abstracts and the manuscripts, obtained the data from the selected manuscripts, and performed the quality assessment of the included manuscripts. PM wrote the first and successive versions of the manuscript. PM, JF, AK, JS, MW, AG, KK, AS, MS, MB, and AB contributed to the interpretation of the results, critical revisions to the drafts of the manuscript, and approved the final version. All authors contributed to the article and approved the submitted version.