Psychedelic-Assisted Therapies and Psychosocial Symptoms in Cancer: A Systematic Review and Meta-Analysis

Inclusion Criteria: This systematic review included RCTs and non-randomized interventional studies (e.g., open-label, single-arm trials) investigating PAT for psychosocial symptoms in adults with cancer or survivors. Eligible studies met the following criteria: Population: Adults (≥18 years) with active cancer (any stage) or cancer survivors (off active treatment), experiencing psychosocial symptoms (e.g., anxiety, depression, existential distress). Intervention: Psychedelic agents (psilocybin, ketamine, LSD, or MDMA) administered in therapeutic settings, with or without structured psychotherapy. Comparator: Placebo (e.g., saline, niacin) or active control (e.g., low-dose psychedelic, midazolam). Outcomes: Quantitative psychosocial outcomes (e.g., validated depression/anxiety scales) Study Design: RCTs: Prioritized for meta-analysis due to reduced bias. Non-RCTs: Open-label, single-arm, mixed-methods, or cohort studies with pre-post assessments were included in the narrative synthesis to capture feasibility, safety, and preliminary efficacy data. Language: Studies available in English.

Exclusion Criteria: Observational studies, qualitative only reports, or non-English publications.

A comprehensive search was conducted across multiple databases, including PubMed, PsycINFO, Embase, and the Cochrane Library, to identify relevant studies. The search strategy incorporated terms related to both psychedelic therapies and psychosocial symptoms in cancer patients. The initial search was performed from database inception through July 2022. To ensure the inclusion of newly published studies, the search was updated again in September 2024. Key search terms included: Terms for psychedelic therapies (e.g., “psychedelic*”, “psychedelic therapy*”, “psilocybin”, “ketamine”, “MDMA”, “LSD” and “ayahuasca”) Terms for psychosocial symptoms (e.g., “anxiety”, “depression”, “psychological distress”, “emotional distress”). Terms for cancer patients (e.g., “oncology”, “cancer survivor*”, “cancer patient*”). Example Search Strategy (specific to one database): “psychedelic*”.ab,ti. “psychedelic therapy*”.ab,ti. “hallucinogen*”.ab,ti. psilocybin.ab,ti. ketamine.ab,ti. ayahuasca.ab,ti. MDMA.ab,ti. anxiety.ab,ti. depression.ab,ti. “cancer patient*”.ab,ti. “cancer survivor*”.ab,ti. 1 OR 2 OR 3 OR 4 OR 5 OR 6 OR 7 8 OR 9 10 OR 11 12 AND 13 AND 14

A full search strategy can be reviewed in. Appendix A

The selection process involved two stages: initial screening of titles and abstracts, followed by a full-text review. Two independent reviewers assessed the studies at each stage based on the inclusion and exclusion criteria. Discrepancies were resolved through discussion, and a third reviewer was consulted if necessary.

Data were extracted independently and in duplicate by two reviewers (H.D.M.S. and C.S.) using a standardized and piloted data extraction form, developed in accordance with PRISMA 2020 guidelines [20]. Disagreements were resolved by discussion or adjudication by a third reviewer (R.M.). For each included study, the following data items were extracted:

Study Characteristics: First author, publication year, country, trial phase, study design (e.g., RCT, open-label; blinded), and journal.

For missing or unclear data, attempts were made to extract the most conservative or methodologically appropriate estimate. If outcomes were measured at multiple timepoints, the most clinically relevant post-intervention measure was selected for synthesis (e.g., first post-dose result for immediate effects, or longest available follow-up for durability assessment).

Risk of bias (RoB) was independently assessed by two reviewers (HS and CS) for all included studies. For RCTs, the Cochrane RoB 2.0 tool was used, examining five domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported results [21]. Non-randomized studies were assessed using the National Institutes of Health (NIH) Quality Assessment Tool for Before-After (Pre-Post) Studies without a control group, covering 12 domains including clarity of the study question, eligibility criteria, outcome measurement, statistical methods, and adequacy of follow-up [22]. Any discrepancies between reviewers were resolved through discussion and consensus. The comprehensive RoB assessments are detailed in Figure 1 and Table 1.

A total of 1860 records were identified through database searching. After removing 1440 duplicates, 420 unique records were screened based on titles and abstracts. Of these, 332 were excluded for failing to meet the eligibility criteria (see Section 2). Full-text review was performed on the remaining 88 articles, ultimately resulting in 15 studies being included in this systematic review. The PRISMA flow chart detailing the study selection process is presented in Figure 2. A summary of the characteristics of the studies included is provided in Table 3.

The final 15 studies encompassed randomized controlled trials (n = 11) and non-randomized experimental designs (n = 4). Publication years ranged from 2011 to 2024, with sample sizes varying from 12 to 303 participants. Majority of the studies included were conducted in the United States (n = 7), followed by China (n = 5), Switzerland (n = 2), and Canada (n = 1). Most trials focused on the use of psilocybin (five studies) and ketamine (four studies), while three studies investigated MDMA or LSD. Participant populations included individuals diagnosed with breast cancer, colorectal cancer, or other advanced malignancies, and in several cases, those with comorbid depression, anxiety, or existential distress.

The RoB for included studies is visually summarized in Figure 1 and detailed in Table 1.

RCTs: Of the 11 RCTs assessed using the Cochrane RoB 2.0 tool, most (n = 8, 73%) demonstrated low overall RoB. Two studies (18%) presented some concerns, primarily due to insufficient reporting of allocation concealment or minor deviations from intervention protocols. One study (9%) was classified as high risk of bias due to issues related to randomization processes and incomplete outcome reporting. Functional unblinding is an acknowledged limitation in PAT trials due to the noticeable psychoactive effects of substances such as psilocybin and LSD. While Schipper et al. [17] rated this as a high risk of bias in several studies, we adopted a more pragmatic approach. We rated studies as low risk in this domain when double-blinding procedures were followed, and no evidence of differential care or co-intervention was reported. Although expectancy effects are plausible, we judged bias based on observed or reported impact, rather than assumption alone, in accordance with Cochrane RoB 2.0 recommendations [21].

Non-RCTs: The four single-arm pilot trials [12,23,24,25] satisfied key methodological benchmarks—clear objectives, prespecified eligibility criteria, standardized psilocybin or ketamine dosing, and validated mood-distress outcomes—earning overall fair ratings. Their interpretability is curtailed by shared limitations: very small samples (n = 12–30), open-label conduct without blinded assessors, variable transparency in reporting how screened candidates translated into enrolled participants, and reliance on a single baseline rather than an interrupted time-series. Retention was high across studies, with all dosed participants included in endpoint analyses. Collectively, these structural constraints temper internal validity and limit the generalizability of the promising efficacy signals.

No treatment-related serious adverse events were reported across the psilocybin RCTs or non-randomized studies included in this review. In Ross et al. [9] and Griffiths et al. [41], transient increases in blood pressure, nausea, and brief anxiety episodes were documented but resolved spontaneously and were consistent with expected acute psychedelic effects. In the HOPE trial [24], participants reported mild nausea and headache; one case of sustained gastrointestinal symptoms was attributed to a viral illness rather than psilocybin. Among the ketamine studies, Rosenblat et al. [12] reported no serious adverse events. Minor side effects, e.g., dissociation, dizziness, and nausea were transient and consistent with the known pharmacological profile of ketamine. Holze et al. [36] reported a single serious adverse event during the LSD condition (acute anxiety requiring temporary withdrawal), which resolved within hours and did not require further medical intervention.

The therapeutic frameworks employed alongside psychedelic administration varied substantially across studies and may have influenced treatment outcomes. All psilocybin trials included in the meta-analysis incorporated structured psychological support, often involving preparatory and integrative sessions with trained therapists. For example, Griffiths et al. [41] and Ross et al. [9] used a manualized approach that included several hours of preparatory counseling, support during the dosing session, and post-session integration. These sessions emphasized participant safety, introspection, and meaning making, often guided by existential or spiritual psychotherapeutic principles.

In contrast, most ketamine trials did not include formal psychotherapy. Studies such as Fan et al. [32], Ren et al. [33] Liu et al. [10], and Wang et al. [34] focused on pharmacological administration in hospital or surgical settings, with no manualized therapeutic support. This distinction highlights a key divergence in how psychedelic-assisted therapy is conceptualized across agents—psilocybin is typically embedded within a psychotherapy-enhanced model, whereas ketamine is often delivered as a standalone pharmacologic intervention in medical settings.

In the exploratory trials, MDMA-assisted therapy [37] followed the Multidisciplinary Association for Psychedelic Studies (MAPS) protocol, which included structured preparation, non-directive support during the 8 h MDMA session, and follow-up integration [37]. Similarly, LSD-assisted psychotherapy [35,36] emphasized therapeutic containment, involving preparatory counseling, supported dosing sessions, and continued psychological follow-up. These models reflect a more psycholytic or experiential framework, blending pharmacologic effects with therapeutic processing.

Among the open-label studies, psilocybin was consistently administered in conjunction with psychotherapeutic support. Agrawal et al. [25], Lewis et al. [24], and Shnayder et al. [23] all utilized group-based or individual therapy sessions before, during, and after dosing. These studies emphasized safety, group cohesion, and integration of insights, typically delivered by trained clinicians in supportive environments. Similarly, Rosenblat et al. [12] administered intranasal ketamine to patients in palliative care without accompanying psychotherapy but did include supportive clinical oversight and monitoring. See Table A1 in Appendix B for a breakdown of therapeutic approaches used across included studies.

A key finding of this review is the divergent therapeutic paradigms employed across PAT trials. Psilocybin, MDMA, and LSD were consistently delivered within structured psychotherapeutic frameworks that emphasized preparation, emotional support during dosing, and post-session integration. This psychospiritual model aligns with the “set and setting” hypothesis, which posits that the therapeutic effects of psychedelics are strongly mediated by psychological and environmental context [43]. Trials using the MAPS protocol [37] or psycholytic techniques [35] reflect this experiential paradigm, aiming to catalyze enduring emotional and existential insights.

Notably, however, few psychedelic trials in oncology have adopted fully manualized psychotherapeutic frameworks, such as mindfulness-based cognitive therapy (MBCT) [44], acceptance and commitment therapy (ACT), or cognitive-behavioral therapy (CBT) that are otherwise well-established in psycho-oncology settings [45]. This may reflect both practical and philosophical factors: the dominant experiential paradigm in PAT favors individualized, non-directive support over protocolized techniques. Yet the absence of structured, reproducible therapeutic models limits opportunities for standardization, fidelity monitoring, and comparative effectiveness research.

By contrast, ketamine was typically administered as a standalone pharmacologic intervention, often in perioperative, hospital, or palliative care settings, with little or no structured psychotherapeutic support. Studies such as Fan et al. [32], Ren et al. [33], and Wang et al. [34] illustrate this biomedical orientation, prioritizing rapid symptom relief for acute psychological crises. While ketamine’s dissociative effects and NMDA antagonism contribute to its antidepressant efficacy, the absence of a therapeutic container may limit its impact on deeper existential suffering. This divergence also raises a broader definitional debate: should ketamine be considered a psychedelic at all? Unlike serotonergic psychedelics (e.g., psilocybin, LSD), ketamine operates through glutamatergic pathways and typically produces a dissociative rather than a classic hallucinogenic experience. Although some classify it as a “non-classical psychedelic” due to its capacity to induce altered states of consciousness, others argue that its mechanisms and subjective effects are fundamentally distinct [46,47]. This distinction is not merely semantic; it has implications for therapeutic delivery. The psychospiritual emphasis central to psilocybin trials may not translate directly to ketamine protocols, which often focus on neurochemical modulation over meaning-making [39].

Marguilho et al. [47] propose that the fast, albeit transient, antidepressant effects observed after ketamine infusions are mainly driven by its acute modulation of reward circuits and sub-acute increase in neuroplasticity, while its dissociative and psychedelic properties are driven by dose- and context-dependent disruption of large-scale functional networks. Wolfson and Vaid [48] emphasize that when ketamine experiences are embedded in a therapeutic relationship, they can foster personal growth, inner healing, and better relationships. However, the absence of structured psychotherapeutic support in many clinical settings may limit these benefits. Future research should continue to clarify the experiential, mechanistic, and therapeutic boundaries between ketamine and classical psychedelics to guide appropriate clinical application.

These distinct paradigms influence treatment outcomes. The sustained improvements observed in psilocybin, MDMA, and LSD trials may be partly attributable to psychotherapeutic integration [39,40]. By contrast, ketamine’s effects are often short-lived, with limited integration and ongoing concerns regarding long-term safety, including risks of dependency, cystitis, and cognitive impairment [17,49].

This distinction is particularly salient in oncology, where patients often grapple with existential concerns that extend beyond symptom control. Psilocybin’s integration with psychotherapy appears well-suited for such needs, though it remains resource-intensive and less scalable. Emerging models such as group-based psilocybin therapy [24,25], offer promising alternatives, retaining therapeutic depth while enhancing accessibility. Similarly, hybrid models like ketamine-assisted psychotherapy (KAP) may bridge the gap between pharmacologic rapidity and psychospiritual depth, warranting further exploration in cancer care [50].

Blinding remains a critical methodological challenge in psychedelic clinical trials due to the potent subjective effects of these agents. These effects, ranging from altered perception and emotional breakthroughs to dissociation, can compromise the integrity of double-blind designs, potentially inflating expectancy effects and undermining internal validity [51,52].

In psilocybin trials, active placebos such as niacin or low-dose psilocybin have been used to mimic physiological responses (e.g., flushing, tingling). However, such comparators fail to replicate the profound psychoactive and often mystical experiences of moderate-to-high dose psilocybin, making effective blinding challenging [9,31]. As Carhart-Harris and Goodwin [53] and Aday et al. [52] emphasized, the vivid and unique nature of these experiences makes them especially difficult to mask using traditional placebos, with participants and therapists correctly guessing treatment allocation in over 90% of cases in some trials [37,54].

Ketamine trials face similar limitations. While comparators such as midazolam or saline are commonly employed, these agents fail to reproduce the dissociative effects of ketamine, allowing for easy identification of the active condition [10,55]. Although midazolam offers some sedation, it does not match ketamine’s perceptual alterations, thereby introducing potential bias through expectancy [55,56,57]. Recent meta-research has shown that when stronger blinding techniques (e.g., administration under anesthesia) are used, ketamine’s effects may not differ significantly from placebo [58].

Beyond pharmacologic comparators, the integration of psychotherapy, standard in most psilocybin trials, introduces further complexity. Participants in placebo conditions still receive extensive therapeutic support, which may independently reduce symptoms and blur between-group differences [9]. This confound is particularly relevant in trials where the therapeutic container (set, setting, and therapeutic alliance) is theorized to be a key mediator of efficacy.

To address these limitations, Aday et al. [52] proposed criteria for ideal active placebos in psychedelic studies, including matched onset/duration, psychoactive and physiological effects, safety, and absence of therapeutic efficacy in the target condition. Potential candidates such as salvinorin A and dextromethorphan offer partial phenomenological overlap with psychedelics and may enhance blinding fidelity [59,60]. THC and diphenhydramine may serve as alternatives in lower-dose or microdosing trials but still present challenges, particularly among participants familiar with their effects [61,62].

While rigorous trial design is essential for establishing safety and efficacy, efforts to improve blinding raise important epistemological and cultural questions. The drive to isolate drug-specific effects through active placebos may reflect a Western scientific tendency to deconstruct holistic experiences into discrete, measurable components. Yet, in many traditional and Indigenous healing contexts, the therapeutic value of psychedelics is inseparable from their set (mental state), setting, and the broader ritual or spiritual framework in which they are used [43,63,64]. Attempting to control for these elements may inadvertently compromise the very mechanisms that contribute to healing. As the field advances, it will be important to balance methodological rigor with cultural sensitivity and openness to complexity.

In light of these issues, future trials should incorporate standardized assessments of blinding and expectancy, and consider alternative trial designs such as dose–response models, incomplete disclosure paradigms, pre-test/post-test, or pragmatic comparators (e.g., psychotherapy-only arms) [52,65]. Such refinements are essential to disentangle drug-specific effects from contextual influences and to uphold methodological rigor in this rapidly evolving field.

The global expansion of off-label ketamine use for cancer-related psychosocial distress reflects both its rapid therapeutic potential and the urgent need for oncology-specific evidence. In North America, ketamine is being used in palliative and private care settings for treatment resistant depression and MDD with suicidal ideation [66]. However, it lacks formal regulatory approval for use in cancer populations and is increasingly being applied more broadly across clinical contexts [12,67]. Its accessibility, relatively low cost, and rapid onset make it appealing in crisis contexts; yet, its use may be outpacing the evidence, raising concerns about safety, efficacy, and equity [68,69,70,71].

This pattern is echoed globally. In high-income countries such as Australia, Canada and the United States, ketamine is permitted off-label. Meanwhile, psilocybin, while demonstrating durable psychosocial and existential benefits, remains restricted to clinical trials or special access programs, limiting its scalability and integration into mainstream oncology care [19].

Canada exemplifies these tensions. While public healthcare provides universal coverage, private “psychedelic” clinics charge out-of-pocket fees (~$750 to $1400 CAD per ketamine session), creating disparities in access [72,73,74]. The absence of national oncology-specific guidelines for psychedelic use places clinicians in ethically ambiguous territory, particularly when obtaining informed consent from vulnerable patients. This underscores the need for pragmatic, equity-focused research to inform safe, scalable models of care [75].

This review provides a focused and methodologically rigorous synthesis of PAT for psychosocial symptoms in adults with cancer, a population that has often been subsumed within broader reviews of LTI. Unlike previous syntheses that pooled heterogeneous diagnoses, agents, and trial designs, we prioritized RCTs conducted specifically in oncology populations and disaggregated findings by psychedelic agent to enhance clinical relevance for psycho-oncology. However, three included RCTs [35,36,37] enrolled broader LTI populations and were retained based on partial cancer representation and because they were the only controlled trials available for LSD and MDMA. While this decision introduces some heterogeneity, it was necessary to ensure coverage of the full spectrum of psychedelic agents under investigation for cancer-related psychosocial symptoms and reflects the early stage of oncology-specific research in this field.

The Cochrane review by Schipper et al. [17] synthesized six RCTs (n = 149) assessing psilocybin, LSD, and MDMA in individuals with LTI, including cancer. While their analysis found low-certainty evidence for reductions in depression and anxiety, the inclusion of heterogeneous populations and lack of agent-specific subgroup analysis limited oncology-specific conclusions. Similarly, Schimmers et al. [19] reviewed 33 studies of PAT in terminally ill patients and reported generally positive outcomes with few adverse events; however, the review included mixed agents (e.g., psilocybin, ketamine, MDMA) and diagnoses, without a specific cancer focus.

Maia et al. [18] also reported psychological and spiritual benefits in LTI, but their inclusion of primarily non-randomized studies constrained the overall certainty of evidence. Ross et al. [76] reviewed classic psychedelics in cancer-related psychiatric distress and highlighted rapid and robust improvements, yet the absence of formal meta-analytic techniques and the combination of various study designs reduced comparability across trials.

Our review builds on these foundations by applying standardized meta-analytic methods (e.g., Hedges’ g), GRADE certainty assessments, and rigorous risk of bias evaluations using ROB 2.0 and NIH tools. We identified large effect sizes for both psilocybin and ketamine in reducing depression, anxiety, and existential distress among PLWC, but also emphasized the moderating influence of therapeutic context, particularly the presence of structured psychotherapy in psilocybin trials. Furthermore, our narrative synthesis extends these findings by integrating open-label studies [12,23,24,25] and exploratory trials involving MDMA and LSD. These studies support the feasibility and clinical promise of psychedelic interventions in cancer care, especially within group-based or psychospiritual delivery models, but highlight the need for larger RCTs and oncology-specific implementation research. In sum, this review contributes uniquely to the literature by isolating cancer-specific data, distinguishing between psychedelic agents, and contextualizing efficacy within delivery frameworks, advancing the field toward more tailored, scalable, and evidence-informed models of PAT in oncology.

This review has several limitations that should be considered when interpreting the findings. First, substantial heterogeneity was observed across included studies, particularly in the psilocybin meta-analysis. While all trials assessed changes in psychosocial symptoms using validated instruments, the selected outcomes (e.g., GRID-HAMD-17, BDI, HADS-T for psilocybin; MADRS, HAMD-17, PHQ-9, HAD-D for ketamine) varied in focus and scoring range. Timepoints for outcome collection also ranged from 24 h to several days post-treatment for ketamine, and from 2 to 7 weeks post-dose for psilocybin. These variations in outcome type and timing likely contributed to statistical heterogeneity, particularly in the psilocybin analyses, where a single study [8] had a disproportionately large effect size.

Second, the number of RCTs eligible for meta-analysis was small, especially for psilocybin, which was represented by only three RCTs totaling 101 participants. Although Griffiths et al. [8] was initially considered for exclusion due to its crossover design and extreme effect size, it was retained in the final analysis due to its methodological rigor, use of pre-crossover data, and alignment with recent Cochrane review practices. However, its inclusion substantially increased between-study heterogeneity, underscoring the need for more consistent outcome reporting and standardized trial design.

Third, non-randomized and open-label studies were excluded from the quantitative synthesis to preserve methodological rigor. These studies, while informative, particularly in exploring long-term outcomes, therapeutic frameworks, and implementation models, were synthesized narratively due to their lack of comparator groups and risk of selection and performance bias. Their exclusion from the meta-analysis limits our ability to estimate real-world effectiveness or assess the role of setting and therapy format in modulating treatment outcomes.

Fourth, only studies published in English were included, which may introduce language bias and limit the comprehensiveness of the review. Additionally, the small number of trials prevented formal tests of publication bias such as funnel plots or Egger’s test, though the risk remains a concern in emerging fields where positive findings may be more likely to be reported.

Finally, most trials evaluated short-term effects, with outcome assessments occurring within days to weeks post-treatment. While some studies included long-term follow-up [39], these were limited to small sample sizes and lacked comparator groups. As such, the durability and functional impact of PAT in oncology populations remain areas requiring further longitudinal research.

To responsibly advance the field of PAT in oncology, future research should prioritize several key directions. First, comparative effectiveness trials are needed to directly evaluate agents such as psilocybin and ketamine within cancer populations. These head-to-head studies will help clarify their respective therapeutic niches, whether for acute symptom relief, existential distress, or long-term psychosocial support. Second, the development of oncology-specific protocols is essential. This includes standardizing psychotherapy integration, optimizing dosing schedules, and harmonizing outcome measurement tools that reflect the unique psychosocial challenges of cancer care.

Third, future trials should incorporate mechanistic and biomarker endpoints, such as inflammatory markers, neuroendocrine parameters, gut microbiome, and neuroimaging correlates, to bridge subjective experiences with underlying biological processes. Fourth, equity-centered implementation research is urgently needed to ensure access and inclusion. This may include evaluating telehealth modalities, group-based formats, and culturally responsive delivery models, particularly for underserved or marginalized populations.

Finally, future research should allow for greater methodological flexibility that honors the complexity of psychedelic-assisted experiences. This may involve exploring alternative trial designs (e.g., pragmatic, hybrid, or culturally embedded models) that capture both pharmacological effects and contextual influences, especially in settings where meaning-making, ritual, and therapeutic alliance are integral to outcomes. Longitudinal studies are also critical to assess durability of effects, monitor adverse outcomes, and determine long-term impact on quality of life. Together, these priorities will support the safe, effective, and culturally attuned integration of PAT into oncology care.