Copenhagen Magnetic Personalized Accelerated Brain Circuit Therapy (CoMPACT) trial: study protocol for a randomized, double-blinded, sham-controlled, three-armed, parallel-group trial of transcranial magnetic stimulation for treatment-resistant depression

TomSo apparently this happened: Copenhagen Magnetic Personalized Accelerated Brain Circuit Therapy (CoMPACT) trial: study protocol for a randomized, double-blinded, sham-controlled, three-armed, parallel-group trial of transcranial magnetic stimulation for treatment-resistant depression. I need your honest reaction to it.

CarolThree arms sounds like overkill when half the people stuck in treatment-resistant depression can barely afford a single consultation. You are designing a lab experiment for folks whose lives are already hanging by a thread, and the name alone feels less like medicine and more like a marketing pitch.

TomThe fancy name aside, randomizing patients into a group that might get zero benefit is exactly how we figure out if TMS actually works better than doing nothing. You want results without the messy data? That’s just guessing.

CarolRandomization solves the science problem, but it ignores the ethical one: you cannot justify withholding proven care from a subset of participants while testing an unproven device on another. The protocol needs to specify exactly what happens to those assigned to the sham arm if their condition deteriorates mid-trial, because right now that gap reads like an afterthought.

TomWithholding proven care isn't happening here since TRD means they're already failing standard meds anyway, so the sham arm just delays starting TMS rather than denying help entirely. What matters is whether the protocol lets clinicians pull anyone out early once symptoms spike past a certain threshold, because keeping someone in a useless control group during a crisis is where the real danger lies.

CarolEarly exit clauses sound good on paper until you realize they let researchers judge who counts as "worse off" based on their own bias. For a population already struggling to trust providers, that subjective call could mean some patients get kicked out of potential treatment while others stay trapped in the placebo group longer than they should. We need objective neurological markers, not just a doctor's gut feeling, to decide when the risk outweighs the data collection.

TomObjective markers are great for peer review, but they don’t tell a patient staring down a bad week that their brain activity hasn’t crossed a line yet. In a relationship, you don’t wait for a blood test to know things aren’t working; you talk it out and adjust. The protocol has to allow for human judgment alongside the hard data, otherwise it treats these people like sensors rather than humans.

CarolComparing clinical protocols to fixing a marriage misses the point that one side holds all the power and the other has no legal recourse. A relationship coach leaning on that analogy suggests the patient’s discomfort is just a communication failure rather than a safety violation, which completely bypasses the actual harm being done. You need independent oversight boards with binding authority, not a clinician trying to read the room while simultaneously collecting data.

TomYou’re building a fortress against the very patients you hired to fill it. An independent board sounds safe, but it adds layers of bureaucracy that slow down response times precisely when a person is spiraling, turning a quick check-in into a committee meeting. That setup protects the institution’s liability more than it keeps anyone alive.

CarolBureaucracy slows things down, sure, but so does a researcher who is too busy checking his watch to notice a patient slipping into a dangerous state. The board exists specifically to catch the blind spots that a busy clinician will inevitably miss. Without that external check, the only accountability left is the researcher’s own ego.

TomEgo is a fair word, but the bigger issue is that this trial targets depression that meds can’t fix, so the baseline pain is already through the roof before day one. Adding a waiting board doesn’t lower that pain, it just stretches the time between suffering and relief until the data looks clean.

CarolStretching the timeline might make the charts look neat, but it leaves the participant holding the bag with no support until the next appointment. The protocol has to define specific interim support measures for the sham group, not just promise eventual treatment, because waiting for the data to settle is not a substitute for basic care.

TomInterim support sounds reasonable, but it turns the sham group into a managed care plan where the goal is comfort rather than testing efficacy. If you pad the control arm with extra counseling to keep them stable, you muddy the results and can’t tell if iTBS actually works better than standard maintenance. The trial needs a clear definition of what “support” excludes so the comparison stays honest.

CarolDefining exclusions is easy until you ask who gets to enforce them without contaminating the blinding process. If the sham group receives targeted help only after a crisis hits, you’ve already introduced a variable that distorts whether TMS is effective or just reactive triage. The protocol must specify proactive monitoring thresholds that trigger intervention without breaking the double-blind design.

TomProactive thresholds sound rigorous, but they turn every symptom into a potential breach that breaks the blind before the last session ends. You cannot measure a magic pill if half the participants suspect exactly why their headaches stopped. Keep the monitoring passive and let the real test happen at the end, not during the scan.

CarolPassive monitoring ignores the reality that some symptoms in TRD are immediate threats, not just data points to be logged later. You can’t ethically wait for the end of the trial to address acute distress in the sham group; the protocol has to build in rapid unblinding pathways for severe cases without compromising the overall efficacy analysis. This isn’t about perfecting the charts, it’s about ensuring no participant suffers unnecessarily under the guise of scientific rigor.

TomRapid unblinding solves the ethical problem but kills the statistical integrity, leaving you with a dataset full of missing values that proves nothing about iTBS versus sham. Until you figure out how to separate safety triggers from efficacy endpoints without contaminating the blinding, this CoMPACT protocol is just expensive guesswork dressed up in lab coats.