Major depressive disorder affects roughly 280 million people worldwide. Fewer than half of patients respond to their first prescribed antidepressant. That figure applies to SSRIs and SNRIs, the drugs that have dominated treatment for four decades. The arrival of psychedelic therapies changes the clinical landscape, but it does not solve the prediction problem. No validated biomarker predicts who will respond to psilocybin. What psychedelics may offer, though, is something the field did not expect: a new way to find the biological signals it has been looking for all along.
Three biomarker candidates have accumulated serious evidence. None is ready for the clinic.
Inflammatory markers, principally CRP and IL-6, are measured with a simple blood test. Elevated levels identify patients unlikely to respond to SSRIs but more likely to benefit from drugs that work on dopamine, norepinephrine, or glutamate rather than serotonin (Arteaga-Henríquez et al., 2019). This is a real and useful finding. But it only describes one subgroup. It says nothing about psilocybin, ketamine, or most other emerging treatments.
Neuroimaging, particularly fMRI, has produced rich mechanistic data through large studies like EMBARC, a multisite trial comparing sertraline to placebo with brain scans collected throughout (Trivedi et al., 2016). The circuits linked to treatment response are becoming clearer. The problem is the scanner. Expensive, slow, and inaccessible outside major academic centers, fMRI cannot realistically guide prescribing at scale.
EEG is the most practical candidate. It is fast, cheap, and widely available. The iSPOT-D trial enrolled 1,008 patients across 22 sites in five countries and identified three baseline EEG features that predicted which patients would respond to which antidepressant (Williams et al., 2011). A 2026 retrospective study found that patients treated according to an EEG based recommendation responded at nearly double the rate of those who were not (Provaznikova et al., 2026). Promising. But this evidence was built on SSRIs and SNRIs. No one has validated it for psilocybin.
SSRI biomarker research has a structural limitation that is rarely discussed. These drugs work slowly. The brain changes that matter, synapse formation, network reorganization, recovery of plasticity, accumulate over weeks. By the time you measure an outcome, the biology behind it is buried in time. Psilocybin is different. A single dose produces large, rapid, and measurable changes in brain activity within hours. A neuroplastic window follows that lasts days to weeks. That speed is not just clinically useful. It gives scientists something they rarely have: a biological event clear enough and fast enough to study properly.
Early findings support this. In a 2023 study, Skosnik and colleagues measured EEG theta power, a marker of synaptic plasticity, before and after a single dose of psilocybin in MDD patients. Theta power doubled two weeks after psilocybin but not after placebo. The increase correlated directly with symptom improvement (Skosnik et al., 2023). A 2024 study found that baseline patterns of brain connectivity, specifically the flexibility of functional states at rest, predicted who would respond to psilocybin in treatment resistant depression (Vohryzek et al., 2024). Meanwhile, a 2024 meta-analysis of 29 studies found that peripheral BDNF, the blood protein most commonly proposed as a marker of neuroplasticity, does not reliably change after psilocybin or any other psychoplastogen in humans (Calder, et al., 2024). The implication is direct: if you want to track what psilocybin does to the brain, you need to measure the brain, not the blood.
There is also a more fundamental point. Every patient given psilocybin shows a different acute neural response to the same dose. Under SSRIs, that individual variability takes weeks to surface. Under psilocybin, it is visible in real time, during the session itself. That difference may be exactly what biomarker research needs. The session becomes a stress test. How a brain responds to psilocybin may reveal more about its underlying state than months of conventional treatment ever could.
We are not there yet. We can describe what psilocybin does to the brain. We cannot yet predict, before a session begins, who will benefit. But the tools are in place and the mechanism, for the first time, gives the field a reason to think the answer is findable. Forty years of SSRI pharmacology built a biomarker literature on a slow biological signal. Psychedelics may finally give the field a fast one.
References
- Arteaga-Henríquez G, Simon MS, Burger B, et al. Low-grade inflammation as a predictor of antidepressant and anti-inflammatory therapy response in MDD patients. Frontiers in Psychiatry. 2019;10:458.
- Calder AE, Hase A, Hasler G. Effects of psychoplastogens on blood levels of brain-derived neurotrophic factor (BDNF) in humans: a systematic review and meta-analysis. Molecular Psychiatry. 2024.
- Provaznikova B, de Bardeci M, Altamiranda E, et al. Predictive value of EEG/ECG biomarkers for treatment response in depression. medRxiv. 2026.
- Skosnik PD, Sloshower J, Safi-Aghdam H, et al. Sub-acute effects of psilocybin on EEG correlates of neural plasticity in major depression: relationship to symptoms. Journal of Psychopharmacology. 2023;37:687-697.
- Trivedi MH, McGrath PJ, Fava M, et al. Establishing moderators and biosignatures of antidepressant response in clinical care (EMBARC): Rationale and design. Psychological Medicine. 2016;46:3047-3063.
- Vohryzek J, Cabral J, Lord LD, et al. Brain dynamics predictive of response to psilocybin for treatment-resistant depression. Brain Communications. 2024;6(2):fcae049.
- Williams LM, Rush AJ, Koslow SH, et al. International Study to Predict Optimized Treatment for Depression (iSPOT-D), a randomized clinical trial: rationale and protocol. Trials. 2011;12:4.
