Pilot study of the 5-HT2AR agonist psilocybin in the treatment of tobacco addiction

Earlier research from the 1950s–1970s suggested therapeutic potential for 5-HT2AR agonist hallucinogens in treating substance use disorders, but many of those studies lacked modern methodological rigour and the field was largely abandoned amid controversy and regulatory restrictions. More recent experimental work with psilocybin in healthy volunteers and in patients with advanced cancer has reported mystical-type experiences and enduring positive behaviour change well beyond the acute drug effect, suggesting a possible mechanism relevant to addiction treatment given associations between increased spirituality and recovery outcomes. Johnson and colleagues therefore conducted an open-label pilot study to assess the safety and feasibility of psilocybin as an adjunct to a structured smoking cessation intervention. The study aimed to establish whether a psilocybin-facilitated protocol could be delivered safely, and whether preliminary cessation rates would justify proceeding to a randomized controlled trial; no control condition was included in this first-step evaluation.

This was a 15-week, open-label pilot study in which psilocybin was administered within a manualised smoking cessation programme. Psilocybin sessions occurred in weeks 5, 7, and 13. The institutional review board approved the protocol and participants gave informed consent. Fifteen nicotine-dependent, psychiatrically healthy adults (10 male; predominantly White) were enrolled after telephone and in-person screening from a pool of 323 telephone-screened individuals. Inclusion criteria included smoking at least 10 cigarettes per day, multiple prior quit attempts, medical fitness on examination and laboratory tests, and continued desire to quit. Exclusion criteria included personal or family history of psychotic or bipolar disorder and drug dependence (other than nicotine) within the prior 5 years. Ten participants reported minimal prior hallucinogen use and five were hallucinogen-naïve; participants received no monetary compensation. The psychological intervention combined cognitive behavioural therapy (CBT) for smoking cessation with preparatory and integration work for the psilocybin sessions. Four weekly preparatory meetings preceded a Target Quit Date (TQD) set to coincide with the first psilocybin session (week 5). Sessions included a brief body-scan meditation, development of a brief motivational statement, exposure to a scented oil as a cue-management strategy, and guided imagery. Psilocybin dosing was 20 mg/70 kg for the first (moderate) session, with default higher doses (30 mg/70 kg) for the second and third sessions; participants could choose to repeat the moderate dose. Staff support included at least two team members per participant (one doctoral-level psychologist), daily phone calls for 2 weeks after TQD, and 19 in-person meetings across the programme. Participants were instructed to avoid other smoking cessation treatments during the study. Safety monitoring during sessions included periodic blood pressure and heart rate checks, continuous staff presence, a physician on call, and availability of rescue medications. Participants lay down, wore eye masks, and listened to music while staff provided non-directive interpersonal support. Measures collected throughout intake, treatment and at 6-month follow-up included breath carbon monoxide (CO), urinary cotinine, Timeline Followback (TLFB) for cigarettes per day, Fagerström Test for Cigarette Dependence, Questionnaire on Smoking Urges (QSU), Smoking Abstinence Self-Efficacy (SASE), Wisconsin Smoking Withdrawal Scale (WSWS), Mysticism Scale, States of Consciousness Questionnaire (SOCQ), Persisting Effects Questionnaire, Visual Effects Questionnaire, and a post-session headache interview (administered to the final 10 participants after an interim finding of psilocybin-associated headache in other work). Biological verification used breath CO and urine cotinine thresholds (CO ⩽6 ppm; cotinine <200 ng/mL). Analysis compared intake to 6-month follow-up: paired two-tailed t-tests assessed changes in smoking (TLFB), CO and cotinine; repeated measures ANOVA tested changes across time for QSU, SASE and WSWS; Mysticism scores were compared with paired t-tests. Persisting Effects data were summarised descriptively.

Fifteen participants completed the study; twelve completed all three psilocybin sessions and three completed two sessions but remained in follow-up. A total of 42 psilocybin sessions were conducted (16 moderate, 26 high dose). No clinically significant adverse events requiring physician intervention or pharmacologic rescue occurred. Acute cardiovascular responses were consistent with prior controlled psilocybin research: mean peak systolic blood pressure was 153 mmHg (SD 11; range 134–173) versus baseline 125 mmHg (SD 10; range 105–153); mean peak diastolic blood pressure was 87 mmHg (SD 11; range 72–105) versus baseline 71 mmHg (SD 8; range 55–89); mean peak heart rate was 87 beats/min (SD 11; range 66–120) versus baseline 68 beats/min (SD 9; range 51–89). On the SOCQ, one participant (7%) reported extreme fearful/dysphoric ratings and five others (33%) reported strong fear-related experiences at some point; these episodes resolved within the sessions with staff support. Visual Effects Questionnaire scores showed no increase in clinically significant visual disturbances at 6 months. Of the 10 participants assessed for headache, eight reported at least one post-session headache; mean duration was reported as 5.8 hours (SD 2.4; range 2.0–9.5) and mean severity was mild (mean 2.6 on a 1–6 scale). Five participants used over-the-counter analgesics after sessions. Smoking outcomes were promising: 12 of 15 participants (80%) demonstrated seven-day point prevalence abstinence at 6-month follow-up as assessed by TLFB and verified by CO and cotinine. Eleven of these 12 reported quitting on the TQD and were biologically verified as abstinent throughout the subsequent 10 weeks of active treatment; one of the 12 was unable to provide some samples due to unexpected travel but was biologically verified at attended visits. Three of the abstainers reported brief, self-corrected lapses (1, 4 and 48 cigarettes) in the period between end of treatment and 6-month follow-up. One participant relapsed after 13 weeks of abstinence, smoked an average of 5 cigarettes/day for 14 weeks (down from 19/day at intake), then returned to verified abstinence before 6 months; that person also used nicotine replacement during the relapse. Statistical analyses showed large reductions in self-reported smoking: TLFB comparisons from intake to 6-month follow-up for the entire sample yielded t14 = 11.1, p < .001. Biomarker reductions were statistically significant for breath CO (t14 = 3.8, p < .01) and urine cotinine (t14 = 2.3, p = .04) across the whole sample. Three participants who were smoking at 6 months reported reduced cigarettes/day after TQD (mean 20 before TQD to mean 14 after; t2 = 5.3, p = .03), although their CO and cotinine did not show significant changes from intake to 6 months. Measures of smoking-related psychological processes changed in directions consistent with cessation: repeated measures ANOVA found significant effects across time for SASE confidence (F2,34 = 24.9, p < .001), SASE temptation (F3,43 = 18.5, p < .001), QSU craving (F3,39 = 12.7, p < .001), and WSWS withdrawal (F4,46 = 4.0, p = .009). Linear contrasts indicated increased confidence to abstain and decreased craving and temptation across the assessment period; withdrawal symptoms peaked at 1 week post-TQD and declined significantly through 6 months. Persisting subjective effects were substantial: Lifetime Mysticism scores increased from intake to one week after the final psilocybin session (mean difference +54; t14 = 3.5, p = 0.004). On the Persisting Effects Questionnaire participants reported many more positive than negative persisting changes (mean positive subscale score 55.8% of maximum; mean negative subscale score 5.3% of maximum). Thirteen participants (87%) rated at least one psilocybin session among the 10 most meaningful experiences of their lives; 11 (73%) rated at least one session among the five most spiritually significant. When asked at three weeks post-TQD how psilocybin helped, the most commonly endorsed mechanisms were a shift in future orientation so long-term benefits outweighed immediate desires (endorsed by 73%), increased belief in ability to quit (73%), and changed life priorities making smoking less important (68%). One participant (7%) indicated psilocybin had not helped.

Johnson and colleagues interpret these findings as preliminary evidence that psilocybin can be administered safely within a structured smoking cessation programme and that the approach is feasible. Acute effects—transient increases in blood pressure and heart rate, short-lived dysphoric subjective reactions in some sessions, and post-session headaches—were manageable with non-pharmacological support and brief pharmacologic resources on call, and no serious adverse events occurred. The authors contrast the temporally confined adverse effects of psilocybin with the daily adverse-effect profiles of some approved cessation medications. Regarding efficacy, the investigators note that 80% abstinence at 6 months substantially exceeds typical 6-month abstinence rates reported for standard pharmacotherapies and behavioural programmes; they caution, however, that the open-label design, small sample size and high level of staff contact preclude causal conclusions about psilocybin per se. The study sample was relatively homogeneous (predominantly White, well educated) and this may limit generalisability. Dose-related conclusions could not be drawn: all participants who achieved abstinence did so after the first, moderate (20 mg/70 kg) session, while those who did not quit were not helped by subsequent higher-dose sessions. The authors discuss concerns about using a psychoactive substance to treat another substance dependence and argue that 5-HT2AR agonists lack features of compulsive drug-seeking and are not reliably self-administered in animal models; the observation that some participants declined additional sessions is offered as further evidence against a strong psilocybin-seeking tendency in this context. Mechanistically, participant reports and questionnaire changes suggest psychological processes such as expanded temporal perspective, greater self-efficacy, and altered life priorities may contribute to smoking cessation, and the authors recommend qualitative investigations and neuroimaging studies to probe psychological and biological mechanisms. Finally, the investigators conclude that the results justify further research, specifically a randomised controlled trial to determine efficacy and mechanisms, given the large global burden of tobacco-related mortality and the modest success of existing cessation treatments. They acknowledge key limitations of the present pilot—small N, open-label design, unrepresentative sample and inability to disentangle the contributions of intensive behavioural support versus psilocybin—and frame the study as a first step towards more definitive trials.