For two decades, the working assumption in treatment-resistant depression was that ketamine and electroconvulsive therapy worked through entirely different mechanisms: one a glutamate disruptor that reorganizes synaptic plasticity within hours, the other a broad neuroelectric intervention whose antidepressant effects take weeks to accumulate. That assumption just cracked. A multi-centre Bayesian re-analysis of PET data published in Molecular Psychiatry found that both treatments produce convergent increases in serotonin 1B receptor (5-HT1B) binding, a shared mechanistic fingerprint that changes how we should be thinking about rapid-acting antidepressant development.
The finding is not incremental. When two structurally unrelated interventions, one pharmacological and one electrical, leave the same receptor signature across independent datasets, you are looking at a mechanism that matters to the antidepressant response itself, not a side effect of how either treatment works in isolation.
What the Serotonin 1B Signal Actually Means
The 5-HT1B receptor has been a background player in antidepressant neuropharmacology for years, largely overshadowed by the serotonin transporter as the primary target of SSRIs. A PET study published in the International Journal of Neuropsychopharmacology had already shown that escitalopram increases 5-HT1B binding in patients with major depressive disorder, correlating those changes with antidepressant treatment effect. That finding established a precedent: the 5-HT1B receptor is not just a bystander in the serotonin system but a downstream participant in therapeutic response. What the new Molecular Psychiatry analysis adds is that two treatments with no shared primary mechanism both converge on this receptor, which strongly suggests 5-HT1B upregulation is a common downstream feature of successful rapid antidepressant action rather than a mechanism-specific artifact.
Having treated more than 30,000 patients with guided ketamine therapy across Mindbloom’s network, I have watched the clinical response patterns long enough to know that ketamine’s speed is not its only unusual feature. The durability question is the one that keeps me up at night: why do some patients sustain remission for months after a six-session series while others begin to plateau within weeks? If 5-HT1B upregulation is part of the mechanism producing that durable signal, then patients who fail to maintain that receptor upregulation may represent a distinct biological subgroup. We have not been measuring that. No trial I know of has enrolled with 5-HT1B binding status as a stratification variable.
The Bayesian methodology used in this re-analysis also deserves recognition on its own terms. The FDA’s draft guidance on Bayesian methods in clinical trials, flagged by regulators as a tool to improve efficiency and reduce development time, is finally being applied to neuroimaging datasets in psychiatry at multi-centre scale. That matters because single-site PET studies in depression have always been underpowered for the kind of receptor-level precision this field requires. Pooling datasets with a Bayesian framework, rather than the frequentist approaches that reward large homogeneous samples, is exactly how you extract mechanistic signal from the heterogeneous, multi-site populations that reflect real TRD patients.
The Trial Design Question Nobody Is Asking
Treatment-resistant depression is a massive unmet need. Approximately 21 million adults in the United States are anticipated to be affected by TRD, and the market is valued at $3.29 billion as of 2024. Yet trial designs in this space continue to treat TRD as a category rather than a spectrum of biologically distinct non-responders. The ENLIGHTEN trial, a Phase 2b study of liafensine in TRD, was the first prospectively biomarker-guided depression trial in psychiatry, enrolling 189 patients selected by ANK3 gene status. That is a proof of concept for what selective enrollment looks like. The 5-HT1B convergence finding opens a parallel question: can PET-confirmed receptor binding status at baseline predict who will respond to rapid-acting interventions and who will need a combination approach?
In my practice running multi-site trials, the hardest recruitment problem in TRD research is not finding treatment-resistant patients. It is finding patients whose treatment resistance has a defined biological substrate. Most protocols define TRD by treatment history: two failed adequate antidepressant trials. The KADS study, which demonstrated meaningful remission rates with subcutaneous ketamine in 179 participants with treatment-resistant depression, is still using that historical definition. Meanwhile, ECT, which a decade-long dataset covering 2,074 patients showed produces a remission rate of 51% and a response rate of 73% for moderate to severe depression, is routinely the comparator of last resort rather than a mechanistic anchor for understanding what biological features predict rapid antidepressant response.
The convergent 5-HT1B signal should prompt sponsors designing next-generation TRD trials to ask whether 5-HT1B binding at baseline or after an index treatment could serve as a stratification biomarker or a surrogate endpoint, not just a mechanistic curiosity. Multi-centre PET harmonization now has a global credentialing framework behind it, as a recent international agreement to standardize PET/CT and PET/MR imaging for clinical trials demonstrates, meaning the logistical barrier to incorporating receptor imaging into Phase 2 enrichment strategies is lower than it was five years ago. The field has the tools. The question is whether trial designers will build the endpoint structures to use them.
What Comes Next
A shared mechanistic endpoint between ketamine and ECT also raises a more provocative combination therapy question: if both interventions drive 5-HT1B upregulation through different upstream routes, could sequential dosing or co-administration produce additive or synergistic receptor effects in patients who plateau on either treatment alone? That is the clinical hypothesis I will be watching for at the next major neuroimaging and mood disorders conferences. If a Phase 2 sponsor is designing a ketamine-plus-ECT augmentation trial with receptor imaging endpoints, that readout will tell us more about rapid antidepressant biology in a single study than the last decade of endpoint-agnostic TRD trials combined.
References
- Molecular Psychiatry: “Convergent increases in serotonin 1B receptor binding following ketamine and electroconvulsive therapy: a multi-centre bayesian re-analysis of PET data”
- International Journal of Neuropsychopharmacology, “Escitalopram and 5-HT1B receptor binding in MDD”
- Applied Clinical Trials Online: “FDA Issues Draft Guidance to Advance Bayesian Methods in Clinical Trials”
- Research and Markets: “Treatment-Resistant Depression Market Size, 2024”
- Psychiatric Times: “New Biomarker-Guided Clinical Trial of Liafensine for Treatment-Resistant Depression (ENLIGHTEN)”
- Psychedelic Health: “KADS Study: Ketamine for Treatment-Resistant Depression, Positive Results”
- PubMed: “Electroconvulsive therapy response and remission rates, 2009–2018, 2074 patients”
- ANIF: “Global Agreement Signed to Standardise PET Imaging for Clinical Trials and Collaboration”
Dr. Leonardo Vando, MD is Principal Investigator of Psychiatry at Vando Medical Services.
