Effects of psilocybin on sleep quality and brain microstructure in chronic cluster headache

Of the eleven patients with chronic cluster headache who completed baseline scanning, eight had complete follow-up microstructural data. Healthy controls were younger on average, but there were no significant differences in sex distribution or body mass index. Weekly cluster headache attacks ranged from 6 to 23, with a median of 11. Headache frequency was strongly correlated with PSQI score at baseline (r = 0.83, 95% CI 0.37-0.96) and follow-up (r = 0.87, 95% CI 0.43-0.98). At baseline, patients with chronic cluster headache had substantially worse subjective sleep quality than healthy controls. Ten of 11 patients (91%) had impaired sleep quality, defined as PSQI > 6, compared with 1 of 24 controls (4%). After adjustment for age, body mass index, scanner, PSQI, and other covariates, there were no statistically significant group differences in the individual microstructural or diffusivity measures when analysed one by one. After psilocybin treatment, patients reported fewer attacks per week, with a mean change of -6.1 attacks (SD 4.9), corresponding to a 50% reduction, and the change was statistically significant (p = 0.009). Subjective sleep quality also improved, with mean PSQI decreasing by 2.50 points (SD 2.1), a 24% improvement, and this remained significant after family-wise error correction (p = 0.015). By contrast, no individual microstructural measure changed significantly at the group level from baseline to follow-up. Nonetheless, 7 of 8 patients showed a numerical decrease in white matter intra-neurite volume fraction and extra-neurite mean diffusivity, although these trends were not statistically significant. Correlations between PSQI and grey matter diffusivity measures were statistically significant before correction but did not survive correction for multiple testing. White matter measures and intra-neurite volume fraction were not significantly related to sleep quality. Change in sleep quality did not correlate with change in any microstructural measure. Because the measures were highly intercorrelated, the researchers also performed principal component analysis. In the baseline PCA, the first two components explained 55.5% and 27.9% of the variance, and patients with chronic cluster headache were separated from controls in multivariate space, with the difference driven mainly by grey matter measures (R2 = 0.12, p = 0.01). However, the principal components were not significantly associated with sleep quality, and neither were the components derived from change scores after psilocybin.

Brendstrup-Brix and colleagues interpret the findings as showing that chronic cluster headache is associated with markedly impaired subjective sleep, and that psilocybin treatment is followed by improved sleep quality alongside fewer headache attacks. They suggest that the sleep improvement may largely track reduction in cluster headache burden, although they do not exclude a more direct effect of psilocybin on sleep. The authors place their findings in the context of previous work showing disturbed sleep in cluster headache and limited evidence on psychedelic effects on sleep. They note that earlier studies have reported higher white matter diffusivity in cluster headache, whereas their adjusted analyses did not find significant single-measure white matter differences between patients and controls. They suggest this discrepancy could relate to the use of a multi-compartment diffusion model and to accounting for subjective sleep quality. Their multivariate analyses indicated that the patient-control difference was driven mainly by grey matter, which they note may be relevant given prior links between grey matter microstructure and cognitive function. Although psilocybin did not produce statistically significant group-level changes in microstructure or diffusivity, the authors highlight numerical patterns, particularly in white matter, that were consistent with lower myelin-related measures after treatment in most patients. They also note borderline and moderate correlations between sleep quality and microstructural metrics, which they interpret as potentially meaningful despite not meeting corrected significance thresholds. They discuss this alongside earlier findings that acute sleep deprivation and chronic poor sleep may be associated with different diffusion patterns. The main limitations they acknowledge are the very small sample, missing follow-up scans in some patients, the absence of a placebo-control condition, and reliance on subjective rather than objective sleep measurement. They also note that they did not collect detailed timing of headache attacks, which limits separation of direct drug effects from headache-related effects. In addition, they caution that the diffusion model relies on assumptions and should be interpreted carefully. Overall, they describe the study as exploratory and argue that larger studies using objective sleep measures are needed to clarify how sleep, headache activity, and psilocybin-related brain changes relate to one another.

The authors conclude that low subjective sleep quality in chronic cluster headache improved after psilocybin treatment, alongside symptom improvement. They report baseline differences in brain microstructure and diffusivity between patients and healthy controls that were mainly driven by grey matter, and they note non-significant numerical changes suggesting reduced white matter myelin content after treatment. They state that their findings support further research into sleep and brain microstructure in relation to psilocybin treatment in chronic cluster headache, ideally in larger samples.