TMS for Depression: What 30+ Clinical Trials Actually Show

The evidence base in context

TMS for major depressive disorder has one of the strongest evidence bases of any non-pharmacological psychiatric treatment. More than 30 randomized controlled trials, multiple large meta-analyses, and extensive real-world data back its use in treatment-resistant depression.

But numbers without context can mislead. Response rates, remission rates, and effect sizes shift depending on the study population, protocol, and how “response” gets defined. This summary cuts through the noise. The strengths, the limitations, and what you can realistically expect.

The landmark studies

O’Reardon et al. (2007) — The pivotal trial

Citation: O’Reardon JP, Solvason HB, Janicak PG, et al. “Efficacy and Safety of Transcranial Magnetic Stimulation in the Acute Treatment of Major Depression: A Multisite Randomized Controlled Trial.” Biological Psychiatry. 2007;62(11):1208-1216.

The study that led to the first FDA clearance. A double-blind, multisite trial that randomized 301 medication-free patients with treatment-resistant MDD (n=155 active, n=146 sham) to 10 Hz left DLPFC stimulation at 120% motor threshold, 3,000 pulses per session, five days per week for 4-6 weeks.

Active TMS was significantly superior to sham on the MADRS at week 4, and on the HAMD17 and HAMD24 at weeks 4 and 6. Response rates were significantly higher with active TMS on all three depression scales. Remission rates were approximately twofold higher with active TMS at week 6, reaching significance on the MADRS and HAMD24. The dropout rate for adverse events was low (4.5%), limited to transient scalp discomfort.

Modest by later standards? Yes. But context matters. The study required a complete medication washout — that’s not how TMS is used in the real world. The treatment course was relatively conservative. And the strict research setting excluded many people who would later benefit from TMS in actual clinical practice.

Despite the modest numbers, this trial put TMS on the map as a legitimate antidepressant treatment. Everything that followed built on this foundation.

George et al. (2010) — The OPT-TMS trial

Citation: George MS, Lisanby SH, Avery D, et al. “Daily Left Prefrontal Transcranial Magnetic Stimulation Therapy for Major Depressive Disorder: A Sham-Controlled Randomized Trial.” Archives of General Psychiatry. 2010;67(5):507-516.

This NIMH-sponsored multisite trial (OPT-TMS) randomized 190 medication-free, treatment-resistant patients to active TMS or sham over three weeks, followed by a three-week open-label phase. They used 3,000 pulses per session — higher than the 1,500-2,000 common in earlier protocols.

Remission rates during the blinded phase: 14.1% active versus 5.1% sham — a roughly fourfold difference that hit the study’s primary statistical endpoint. Here’s the key finding: remission climbed to 30.5% after the open-label extension. Many people needed more than three weeks to fully respond.

The takeaway: TMS is a course of therapy, not a one-shot fix. More sessions generally mean better outcomes.

Levkovitz et al. (2015) — Deep TMS for depression

Citation: Levkovitz Y, Isserles M, Padberg F, et al. “Efficacy and Safety of Deep Transcranial Magnetic Stimulation for Major Depression: A Prospective Multicenter Randomized Controlled Trial.” World Psychiatry. 2015;14(1):64-73.

The pivotal trial for BrainsWay’s Deep TMS H1 coil. This international, double-blind, multicenter RCT randomized 212 patients with treatment-resistant MDD to active deep TMS or sham. The H1 coil reaches deeper brain structures (4-6 cm below the skull) compared to conventional figure-8 coils (1.5-2 cm).

Response rates: 38.4% active versus 21.4% sham. Remission rates: 32.6% versus 14.6%. These were notably stronger effect sizes than the original NeuroStar trial, and the results demonstrated efficacy for patients with 1-4 failed medication trials. Benefits were sustained through 16 weeks of follow-up.

Blumberger et al. (2018) — The THREE-D trial

Citation: Blumberger DM, Vila-Rodriguez F, Thorpe KE, et al. “Effectiveness of Theta Burst Versus High-Frequency Repetitive Transcranial Magnetic Stimulation in Patients With Depression (THREE-D): A Randomised Non-Inferiority Trial.” The Lancet. 2018;391(10131):1683-1692.

This one changed everything. A landmark Canadian non-inferiority trial that enrolled 414 patients with treatment-resistant depression across multiple sites, randomized to standard 10 Hz rTMS or intermittent theta burst stimulation (iTBS) delivered to the left DLPFC, five days per week for 4-6 weeks.

Both groups showed equivalent improvement on the HRSD-17, from baseline scores of ~23.5 to ~13.4. Response rates: 49% for iTBS versus 47% for 10 Hz rTMS. Remission rates: 32% for iTBS versus 27% for 10 Hz. iTBS was definitively non-inferior.

Why did this matter so much? iTBS delivers a full treatment in about 3 minutes. Standard 10 Hz takes 37.5 minutes. Same results, fraction of the time. That meant clinics could treat more people, you’d spend less time in the chair, and costs came down. The THREE-D trial accelerated the adoption of theta burst protocols worldwide and fundamentally changed how TMS gets delivered.

What the meta-analyses show

Response and remission rates

Pool all the high-quality RCT data and the numbers converge:

Response rates (50% or greater reduction in depression scores): 50-60% in clinical practice, 30-40% in sham-controlled trials.

Remission rates (depression scores falling below a clinical threshold): 30-35% in clinical practice, 15-25% in controlled trials.

Why the gap between trials and real-world numbers? Same thing that happens with antidepressants. In practice, you keep your medications during TMS (most trials require washout). Real-world treatment courses run longer — 30-36 sessions versus 20-30 in trials. And clinical populations include people with less severe treatment resistance than the heavily resistant patients enrolled in FDA studies.

The largest real-world outcomes dataset comes from Sackeim et al. (2020), published in the Journal of Affective Disorders (277:65-74). This registry study collected data from 5,010 patients across 103 clinical practice sites — the largest outcomes registry for any depression treatment. Response rates ranged from 58% to 83% and remission rates from 28% to 62% depending on whether self-report or clinician-administered measures were used. These numbers far exceeded the earlier controlled trials.

Berlim et al. (2014), in a systematic review and meta-analysis of randomized, double-blind, sham-controlled trials published in Psychological Medicine (44(2):225-239), calculated a number needed to treat (NNT) of 6 for clinical response and 8 for remission — numbers that compare favorably with antidepressant medications.

Mutz et al. (2019) published a comparative network meta-analysis in the BMJ (364:l1079) examining all non-surgical brain stimulation techniques for depression. High-frequency left DLPFC rTMS had the largest body of supporting evidence. Bilateral rTMS, deep TMS, and theta burst stimulation also showed significant efficacy over sham, though with more variable evidence.

Comparison to antidepressant medication

Head-to-head comparisons between TMS and medication are limited. But the available data suggest broadly comparable efficacy — with some telling differences.

The STAR*D trial, the largest study of antidepressant sequencing, found that after failing one medication, a second achieves remission in roughly 25% of patients. After two failures, a third medication gets about 14%. TMS is typically offered after 1-2 medication failures. That makes a 30-35% real-world remission rate look pretty good by comparison.

Where TMS clearly wins: side effects. No weight gain. No sexual dysfunction. No drowsiness. No withdrawal symptoms. If you can’t tolerate medications or want to avoid systemic pharmacological effects, TMS offers a genuinely different risk-benefit picture.

Where TMS is less convenient: time commitment. Five sessions per week for 4-6 weeks is a real logistical burden compared to swallowing a daily pill. Accelerated protocols are starting to close this gap.

Durability of response

The question everyone asks: how long does the benefit last?

Six-month data

Dunner et al. (2014), published in the Journal of Clinical Psychiatry (75(12):1394-1401), followed 257 patients who completed acute TMS treatment across multiple sites over 52 weeks. Among 120 patients who met response or remission criteria at the end of acute treatment, 75 (62.5%) continued to meet response criteria throughout long-term follow-up. Only 10 of 99 patients (10.1%) met criteria for full relapse during the first 6 months. When symptom worsening did occur, 84.2% of those patients (32 of 38) re-achieved symptomatic benefit with TMS reintroduction.

A meta-analysis of durability data found that among initial responders, 66.5% sustained response at 3 months, 52.9% at 6 months, and 46.3% at 12 months.

Twelve-month data

At one year, the picture gets more variable. Some people maintain their response indefinitely with no additional treatment. Others see gradual symptom return, typically between 6 and 12 months after the initial course. The Dunner et al. 12-month follow-up showed statistically significant reductions in depression scores were sustained throughout the full year compared to pre-TMS baseline (all P < .0001). Approximately 36% of patients received TMS reintroduction during follow-up, averaging 16.2 sessions.

Maintenance and retreatment

If symptoms come back, retreatment works. Several studies show that a second course of TMS produces response rates comparable to the first — and often works faster, requiring fewer sessions.

Some clinics now offer maintenance TMS. Periodic sessions — weekly, biweekly, or monthly — designed to prevent relapse. The evidence base is still building, but early data from multiple open-label studies look encouraging. People who get maintenance sessions appear to stay in remission longer.

Who responds best

Not everyone responds equally. Research has identified several factors tied to better outcomes:

Carpenter et al. (2012), published in Depression and Anxiety (29(7):587-596), conducted the first large naturalistic outcomes study of TMS in real-world clinical practice. Across 42 sites, 307 outpatients received an average of 28.3 TMS sessions. Clinician-assessed response rate was 58.0% and remission was 37.1%. Patient-reported response rates ranged from 41.5% to 56.4% depending on the measure used. These numbers confirmed that real-world TMS outcomes substantially exceeded the controlled trial results.

Who responds best

Not everyone responds equally. Research has identified several factors tied to better outcomes:

• Fewer failed medication trials — 1-2 failures respond better than 4+
• Shorter current episode — Chronic, multi-year episodes are harder to treat
• Less comorbid anxiety — TMS often improves anxiety too, but severe comorbid anxiety can reduce depression response rates
• Younger age — TMS works across age ranges, but some data suggest modestly better outcomes in younger adults
• Concurrent therapy — People in psychotherapy alongside TMS tend to do better

Neuroimaging biomarkers are emerging as predictors too. Fox et al. (2012), in a study published in Biological Psychiatry (72(7):595-603), demonstrated that the functional connectivity between the DLPFC stimulation site and the subgenual anterior cingulate cortex (sgACC) predicts antidepressant response. DLPFC sites with stronger anticorrelation to the sgACC produced better clinical outcomes. This finding, replicated in subsequent prospective studies, drives the neuroimaging-guided targeting approaches used in protocols like the Stanford SAINT trial. Several research groups are developing tools to use this for personalized targeting.

The bottom line

TMS for depression has a real evidence base that has matured considerably since 2008. If you have treatment-resistant MDD — particularly after failing 1-3 adequate medication trials — TMS offers response rates of 50-60% and remission rates of 30-35%, with a side effect profile that’s hard to beat and reasonable durability.

It’s not a cure-all. Roughly 40-50% of people won’t respond meaningfully. Maintaining remission often requires ongoing management. But for a treatment-resistant population that has already failed the most common interventions, those numbers represent real hope.

To explore TMS treatment options, visit our provider directory to find qualified TMS specialists in your area, or learn more about the different TMS technologies available today.