After Two Failed Antidepressants, a Dutch Trial Pits rTMS Against the Next Drug
- Entry threshold: patients had already failed at least two adequately dosed antidepressant trials in the current episode (HDRS-17 above 16, resistance quantified via DM-TRD and ATHF).
- N = 89 randomized (48 rTMS, 41 medication switch) across seven Dutch mental health centers, both arms combined with weekly CBT.
- At 8 weeks: response 37.5% (rTMS) vs 14.6% (medication); remission 27.1% vs 4.9% – secondary measures, alongside the trial's primary continuous HDRS-17 outcome.
- A separate 12-month economic analysis (2025) found the rTMS-first pathway cost 2,280 euros less per patient while adding 0.066 quality-adjusted life years.
Roughly two-thirds of people treated for major depressive disorder fail to remit on their first antidepressant trial, and guidance for what to do after a second failure is thin. A pragmatic trial across seven Dutch mental health centers put an actual number on that fork: patients who had already failed two adequately dosed antidepressant trials were randomized either to the next step in the national medication algorithm or to left-prefrontal rTMS, both alongside weekly CBT.
Where the Trial Drew the Line
Eligibility was not a symptom checklist but a resistance threshold: HDRS-17 above 16, and at least two prior antidepressant trials confirmed as adequately dosed by the Antidepressant Treatment History Form, quantified overall by the Dutch Measure for quantification of Treatment Resistance in Depression (DM-TRD). Patients meeting that bar were randomized 1:1, in practice 48 to rTMS and 41 to medication, stratified by center, number of prior episodes, and baseline severity. The rTMS arm received 10 Hz stimulation of the left dorsolateral prefrontal cortex, located with the BeamF3 scalp-based method, at 120% of resting motor threshold, 3,000 pulses per session, up to 25 sessions across 8 weeks on a tapering schedule of four sessions a week down to two, with early stopping once HDRS-17 fell below 8 on two consecutive assessments. The comparison arm received no placebo stimulation; it received the next step specified by the Dutch national depression treatment algorithm - a switch to a tricyclic antidepressant, or augmentation of the current antidepressant with lithium or a second-generation antipsychotic. Both arms attended weekly cognitive behavioral therapy throughout.
At 8 weeks, response, a HDRS-17 drop of at least half, was reached by 37.5% of the rTMS group against 14.6% of the medication group; remission, HDRS-17 below 8, by 27.1% against 4.9% – both secondary outcomes alongside the trial's primary continuous HDRS-17 measure. Anxiety and anhedonia improved more in the rTMS group as well, though the trial was not powered to treat either as a primary endpoint. A separate 12-month economic analysis, published in 2025, followed a modified version of the same two pathways - medication-arm patients who had not improved by week 8 were themselves offered rTMS - and found the rTMS-first pathway cost 2,280 euros less per patient while producing 0.066 more quality-adjusted life years, with response and remission rates converging, 27.1% versus 24.4%, and 25.0% versus 17.1%, once that crossover is factored in. The trial was open-label: patients and treating clinicians knew which arm they were in, and its comparator, its cost model, and its reimbursement math all describe the Dutch system specifically, not a universal treatment ladder.
At the same threshold of treatment resistance, 27.1% of patients assigned to rTMS reached remission at 8 weeks, versus 4.9% assigned to the next step in the medication algorithm.
Open-label design, no blinded raters; the medication comparator and the cost-effectiveness conclusions describe the Dutch national treatment algorithm and reimbursement system specifically, not a universal standard of care.