A patient I’ve been treating for three years — treatment-resistant MDD, two failed SSRIs, one failed SNRI, buprenorphine-maintained for co-occurring opioid use disorder — finally stabilized on a combination of esketamine nasal spray and maintenance ketamine infusions through our telehealth program. She responds. Reliably. Within hours of each session. But the response window narrows over time, and every six months I find myself asking the same question: what is actually changing in her brain, and why does it eventually stop changing enough? That question got a more precise answer this month than anything I’ve read in years.

Two companion studies published in Cell — Gongwer et al. and Johnson et al. — identify intratelencephalic (IT) neuronal projections to the anterior insula as the central plasticity substrate driving the antidepressant effects of accelerated intermittent theta burst stimulation (aiTBS). The fronto-insular circuit, specifically the prelimbic-to-anterior-insula pathway, turns out to be where the therapeutic action lives. IT neurons show increased synapse-related plasticity following aiTBS. Disrupt that cell type, and the antidepressant effect disappears. Preserve it, and you get rapid relief — not in weeks, but in the compressed timelines that have made aiTBS clinically compelling in the first place.

What the Circuit Is Telling Us

The anterior insula has long been implicated in interoception, emotional salience, and the subjective experience of distress — the felt sense of being depressed, not just the behavioral correlates. What these studies add is a mechanistic address: IT neurons are the plasticity gate. That specificity matters enormously, because for years the field has been running neuromodulation trials with endpoints calibrated to detect average symptom change across heterogeneous populations, with no way to stratify by who actually has a responsive IT-insular circuit. We’ve been treating the target region like a neighborhood when we should have been treating it like a specific apartment.

Compare this to what we know about ketamine. The 2006 Zarate et al. randomized crossover study established that a single IV ketamine infusion produced significant antidepressant effects in treatment-resistant patients within 110 minutes — a timeline that stunned the field and reframed what “rapid-acting” could mean. Ketamine’s mechanism runs primarily through NMDA receptor antagonism — preferentially on cortical GABAergic interneurons, whose disinhibition produces a glutamate surge onto pyramidal neurons — and downstream AMPA potentiation, converging on synaptic plasticity in prefrontal circuits. The IT-insula finding in aiTBS suggests a convergent but distinct plasticity node — and the question worth asking now is whether these two rapid-acting mechanisms are targeting the same downstream synapse remodeling through different upstream routes, or whether they’re genuinely independent therapeutic substrates. If they’re partially convergent, that has direct implications for combination sequencing. If they’re independent, it opens a stratification logic: some patients’ circuits may be selectively responsive to one mechanism and not the other.

The current aiTBS clinical picture is encouraging but incomplete. A systematic review published in Frontiers in Psychiatry found iTBS response and remission rates comparable to high-frequency rTMS across 474 adult participants with treatment-resistant depression — a meaningful finding, but one that tells us nothing about which patients have IT-insula plasticity profiles that make them candidates for robust response versus those who will plateau after modest gains. The Gongwer and Johnson data now give us a cellular target specific enough to build a biomarker hypothesis around.

The Trial Design Gap This Opens

Standard TMS protocols — including the MagVenture 3-minute TBS protocol cleared by FDA for MDD in August 2018 — were designed around population-level efficacy thresholds, not circuit-specific responder identification. That was appropriate for initial clearance. It is not appropriate for the next generation of aiTBS development, and the Gongwer/Johnson findings make the case for building prospective trials that pair neuroimaging or electrophysiological biomarkers of IT-insular connectivity with clinical outcome. In my experience running trials across SUD and mood disorder populations, the patients who show the most durable rapid-acting responses — to ketamine, to esketamine, and in the literature to aiTBS — tend to share a profile: high baseline interoceptive sensitivity, significant emotional dysregulation, often dual diagnosis. That phenotype may be mapping onto IT-insula circuit architecture in ways the field hasn’t formally measured.

Accelerated TMS protocols are not yet widely covered by insurance, which means reimbursement access lags behind the science considerably. The fastest path to changing that is efficacy data in precision-selected populations — exactly the kind of stratified enrollment that IT-neuron biomarkers could enable. Running another large trial that averages together IT-responsive and IT-unresponsive patients will produce a moderate effect size and another coverage denial. The field has absorbed that outcome enough times to know better.

I’ll be watching for whether any of the groups behind these Cell papers move toward translational biomarker work in human imaging — specifically whether fronto-insular functional connectivity at baseline can predict aiTBS response magnitude. That is the signal that would turn a mechanistic breakthrough into a trial design revolution.

References

  1. Cell — Gongwer et al. and Johnson et al., “This is IT: How accelerated intermittent theta burst stimulation relieves depression symptoms”
  2. PubMed — Zarate et al. (2006), “A Randomized Trial of an N-methyl-D-aspartate Antagonist in Treatment-Resistant Major Depression”
  3. Frontiers in Psychiatry — Systematic review of iTBS vs. HF-rTMS in treatment-resistant depression (474 participants)
  4. BioSpace — MagVenture FDA clearance for 3-minute TBS protocol (Express TMS®), August 2018
  5. Cognitive FX — Insurance coverage landscape for accelerated TMS depression protocols
+ posts

Dr. Leonardo Vando, MD is Principal Investigator of Psychiatry at Vando Medical Services.