Acute and Sustained Reductions in Loss of Meaning and Suicidal Ideation Following Psilocybin-Assisted Psychotherapy for Psychiatric and Existential Distress in Life-Threatening Cancer

A cancer diagnosis increases risk of suicidal ideation (SI), desire for hastened death (DHD), and completed suicide, particularly in the context of advanced illness, uncontrolled pain, major depression and existential distress such as hopelessness and loss of meaning (LoM). Demoralization syndrome, of which LoM is a core component, occurs in up to one-third of patients with advanced cancer and correlates with DHD. Existing pharmacologic treatments show limited efficacy for suicidality in this population, creating a need for rapidly acting, mechanism-informed interventions that target existential distress as well as depressive symptoms. Ross and colleagues note that single-dose psilocybin-assisted psychotherapy (PAP), delivered with existentially oriented psychotherapeutic support, has produced rapid and sustained improvements in depression, anxiety and existential distress in prior randomised trials of patients with life‑threatening cancer. Drawing on epidemiologic signals and early clinical reports suggesting possible antisuicidal effects of classic psychedelics, the study reported here presents post hoc analyses of data from a previously completed double-blind, randomised, crossover trial at NYU/Bellevue. The investigators aimed to determine whether a single moderate-to-high dose of psilocybin acutely and sustainably reduces SI and LoM, and to examine relationships between SI, other depressive symptoms, demoralization, hopelessness and spiritual well-being as potential mediators of any antisuicidal effect.

The parent study was a double-blind, randomised, crossover trial comparing a single oral dose of psilocybin (0.3 mg/kg) with an active control (niacin 250 mg), each administered alongside dyadic psychotherapy designed to address anxiety, depression and existential distress in patients with life‑threatening cancer. Some participants also took part in two long-term follow-ups at mean intervals of about 3.2 years and 4.5 years after psilocybin dosing. The original trial excluded participants with high-risk suicidal behaviour and required at least a one-year projected life expectancy; psychiatric eligibility included DSM‑IV diagnoses such as adjustment disorder with anxious and/or depressive features or generalized anxiety disorder. Of 108 prescreened, 42 consented, 31 were randomised, 29 received study medication and 11 participants met the criterion for inclusion in the present secondary analyses because they had detectable SI at baseline (composite SI score > 0). Primary and secondary outcomes were constructed from validated instruments. Suicidal ideation was operationalised as a composite score using item 9 from the Beck Depression Inventory‑II (BDI‑II) and item 9 from the Brief Symptom Inventory (BSI); because the two items used different scales, each was converted to a Z‑score, summed and transformed into a T‑score (0–100) with higher values indicating greater SI. Loss of Meaning (LoM) was taken from the LoM factor of the Demoralization Scale (five items). Other measures included the remaining BDI‑II items (non‑SI depressive symptoms), the Hopelessness Assessment in Illness (HAI), and the FACIT‑Sp‑12 measure of spiritual well‑being. SI was assessed at baseline, 8 hours after dose 1, 2 weeks after dose 1, 7 weeks after dose 1 (one day prior to dose 2) and 6.5 months after dose 2; long-term LoM and other scales were also assessed at the multi‑year follow-ups. Statistical analyses pooled data from the parent trial and long-term follow-up for post hoc testing. Mixed‑effects models with repeated measures (MMRMs) were fitted in SPSS using an AR(1) covariance structure with fixed effects for group and time; planned within‑group (change from baseline) and between‑group (psilocybin‑first vs niacin‑first) comparisons were specified a priori and Tukey post hoc tests applied. After crossover, dose‑sequence groups were combined for within‑subject long‑term analyses to increase power. The investigators conducted post hoc power calculations indicating sufficient power to detect large effects (f > 0.4) for outcomes with four repeated measures but limited power for small-to-moderate effects; effect sizes were reported tentatively as Cohen's d. Correlational analyses examined associations between SI and LoM across time points and between change in SI and changes in other depressive symptoms, demoralization, hopelessness and spiritual well‑being from baseline to 2 weeks post dose 1.

Eleven participants with detectable baseline SI comprised the subsample analysed. Mean age was 60.3 years (SD = 7.1); 63.6% were female and 90.9% were non‑Hispanic White. Gynecological cancers accounted for 54.6% of cases and 72.8% had stage III–IV disease. Prior psychedelic use was reported by 36.4% of participants. Most met DSM‑IV criteria for cancer‑related adjustment disorder with anxious and/or depressed features. Acute (precrossover) effects: there was no significant main effect of treatment group on SI (p = 0.614) and the group-by-time interaction for SI approached but did not reach significance (p = 0.065). By contrast, LoM showed a significant main effect of group [F(1,11) = 6.49, p = 0.027], with planned pairwise comparisons indicating substantially lower LoM in the psilocybin‑first versus niacin‑first groups at 2 weeks after dose 1 (p = 0.021). Reported effect sizes for statistically significant comparisons were large (Cohen's d = 1.20–3.50). Long-term effects: following the crossover and when analysed within subjects, reductions from baseline remained significant at 6.5 months post dose 2 for both SI (p < 0.001) and LoM (p < 0.001). LoM reductions persisted at the multi‑year follow-ups (3.2 and 4.5 years; p < 0.001 for both). Effect sizes for these significant long‑term changes were again large (Cohen's d ≈ 1.51–2.56). Correlational findings: across all time points there was a moderate positive correlation between LoM scores and SI scores (r = 0.41, p = 0.02). Change in SI from baseline to 2 weeks post dose 1 correlated strongly with change in other depressive symptoms over the same interval (r = 0.75, p = 0.008). Reductions in SI were also positively correlated with reductions in hopelessness (HAI: r = 0.62, p = 0.059) and with reductions in overall demoralization (DS: r = 0.57, p = 0.067); these latter correlations were large in magnitude but did not reach conventional statistical significance in this underpowered sample. Change in spiritual well‑being (FACIT‑Sp‑12) showed no association with change in SI (r = 0.015, p = 0.966).

Ross and colleagues interpret these secondary analyses as indicating that a single moderate‑to‑high dose of psilocybin, administered with psychotherapy, was associated with both rapid and enduring reductions in suicidal ideation and loss of meaning among patients with life‑threatening cancer. The investigators observed substantial within‑group improvements from baseline in the psilocybin‑first condition across precrossover time points and a between‑group difference in LoM at 2 weeks post treatment. They caution that the trial was underpowered for small-to-moderate effects and that baseline SI levels in this subsample were generally low, but suggest that the magnitude of observed effects may still be clinically meaningful and could potentially protect against progression to more active suicidality. The authors note a positive relationship between decreases in LoM and decreases in SI, consistent with the idea that meaning‑making is a plausible psychological mediator of antisuicidal effects in advanced cancer. They also found associations between reductions in SI and reductions in broader demoralization and hopelessness, whereas increases in spiritual well‑being did not correlate with SI change. Tentative mechanistic considerations are offered: potential neurobiological and psycho‑spiritual pathways (for example, changes in functional connectivity, neuroinflammatory markers and neuroplasticity) and psychological processes (mystical experience, cognitive flexibility, emotional breakthrough) might underpin durable clinical benefit. To test such mechanisms rigorously, the investigators recommend prospective trials with a parallel design, recruitment focused on clinically relevant suicidality and DHD, larger sample sizes adequate for mediation analyses, targeted measures of psychological change mechanisms, inclusion of participants with cancer pain, and integration of neuroimaging and biomarker assessments. Limitations emphasised by the study team include that the parent trial was not designed to assess suicidality as a primary outcome, exclusion of participants with more serious suicidality, underpowered statistical tests for small-to-moderate effects, the crossover design limiting long‑term controlled comparisons, a demographically homogeneous sample (e.g. 64% female, 91% Caucasian) limiting generalisability, and reliance on a post hoc composite SI score rather than a dedicated suicidality instrument. The authors conclude that while preliminary findings are promising, replication in appropriately powered, prospectively designed trials is required to establish efficacy and clarify mechanisms before clinical translation.