Therapeutic Neuromodulation for Bipolar Disorder—The Case for Biomarker-Driven Treatment Development | Bipolar and Related Disorders | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Invited Commentary
March 12, 2021

Therapeutic Neuromodulation for Bipolar Disorder—The Case for Biomarker-Driven Treatment Development

Author Affiliations
  • 1Division of Neuropsychiatry and Neuromodulation, Massachusetts General Hospital, Harvard Medical School, Boston
JAMA Netw Open. 2021;4(3):e211055. doi:10.1001/jamanetworkopen.2021.1055

Transcranial magnetic stimulation (TMS) is a noninvasive neuromodulation technique with both diagnostic and therapeutic clinical applications. Its capacity to facilitate lasting neuroplastic changes has led to a growing number of treatment indications cleared by the US Food and Drug Administration (FDA), including major depressive disorder (MDD), migraine headaches, obsessive compulsive disorder, and smoking cessation. In 2020, the FDA granted breakthrough device designation to TMS for treating bipolar depression. This is not a formal approval, but it indicates interest, is generally supported by evidence, and provides an expedited path for FDA review.

Broadly speaking, TMS is not a therapy but rather a tool. However, when it is applied over a specific brain target using specific parameters that facilitate specific neurophysiological changes, then TMS has the potential to become a treatment. Choosing a different target or changing other stimulation parameters, however, will result in a different intervention with unique potential therapeutic indications. Indeed, although the early days of TMS treatment development overstretched the idea of dorsolateral prefrontal cortex (DLPFC) stimulation with trials exploring the benefit of this protocol for a plethora of very different neuropsychiatric conditions, all currently approved interventions include unique protocols with different anatomical targets, stimulation frequencies, and so forth, adapted to the pathophysiology of the conditions to treat.

Theta burst stimulation (TBS) was developed in the early 2000s as a safe, effective, and very efficient TMS protocol to modulate plasticity in humans.1 TBS is a translational development built on the role of hippocampal theta rhythms in learning and plasticity, in particular, the capacity to artificially induce long-term potentiation in hippocampal slice neurons using rapid bursts of electrical stimuli every 200 milliseconds (ie, 5 Hz, a theta frequency).1 The most common human TBS protocol involves rapid (50 Hz) TMS triplets applied every 200 milliseconds (5 Hz) in a continuous (cTBS) or intermittent (iTBS) fashion, leading to inhibitory or excitatory effects, respectively. A remarkable property of TBS modulation stems from the fact that with only 600 pulses requiring 40 seconds (cTBS) or 190 seconds (iTBS), measurable physiological changes in the motor cortex can last up to 60 minutes, which is substantially more than traditional repetitive TMS despite its much shorter duration.1 The short duration and longer-lasting effects of TBS have motivated a growing volume of research into its therapeutic applications. The most relevant development to date may be the finding that iTBS to the left DLPFC (3 minutes) is noninferior in efficacy and safety to the traditional 10 Hz protocol (38 minutes).2 This result led to the FDA clearance of the 3-minute iTBS protocol for the treatment of MDD.

McGirr and colleagues3 present the results of a randomized clinical trial of iTBS to the left DLPFC for the treatment of bipolar depression. The study, which was conducted across 2 Canadian centers, applied the FDA-cleared antidepressant protocol (shortening the duration from 6 to 4 weeks) to treatment-resistant bipolar patients receiving stable pharmacotherapy. The target sample size was 50 patients, but an interim futility analysis was conducted with 37 randomized participants, given the slow recruitment and low therapeutic response, and the authors concluded that the intervention was not effective and ended the study. Although the study3 has a number of limitations identified by the authors, and the investigation of iTBS to the left DLPFC in bipolar depression may warrant further work (particularly in light of other small studies with more aggressive protocols and intriguing results4), the results of this negative trial generate a number of important lessons that should help move the field forward.

Although MDD and bipolar disorder are distinct syndromes that are treated with different pharmacological strategies, it is valid to consider whether, given the phenotypical similarities of a major depressive episode in the context of MDD or bipolar disorder, the same TMS protocol may be effective for both conditions. The underlying biological assumption is that, at least at the circuit level, the 2 conditions may present with similar pathophysiological signatures that can be effectively engaged by the same therapeutic intervention. A number of previous positive trials and meta-analyses of TMS in bipolar depression, predominantly using high-frequency or low-frequency TMS over the DLPFC, may have suggested this was, in fact, true,5 but the current findings from McGirr et al3 seem to contradict this hypothesis. Although clinical trials exclusively measuring efficacy and safety (without additional biological measures) can inform such treatment-relevant mechanistic questions leading to therapeutic discoveries, a more nuanced understanding of the disease mechanisms may facilitate a faster path toward effective (mechanistically informed) treatment development. Indeed, although our understanding of the circuit-level differences between unipolar and bipolar depression is still rudimentary, pathophysiological research in this area is starting to suggest distinct patterns that differentiate 2 otherwise similar clinical phenotypes,6 suggesting disorder-specific TMS targets with translational therapeutic implications.7

Identifying the circuits and nodes differentially affected in bipolar depression is critical to defining optimal anatomical TMS targets, but we need to understand more than anatomy for successful treatment development. Physiological signatures (within the constrained anatomy of disease-relevant circuits, certainly) should be considered, leading to an exploration of optimal TMS parameters to engage such maladaptive physiology. TMS was initially developed as a tool to study human motor neurophysiology, and physiological constructs are familiar to TMS researchers. The predominant paradigm has been that of cortical and network excitability: in this framework, disorders are formulated as problems of hyperexcitability or hypoexcitability, and TMS protocols with low-frequency or high-frequency stimulation (or cTBS vs iTBS) are used to decrease or increase, respectively, local excitability. Although almost 40 years of TMS research support this canonical framework, other neurophysiological paradigms (ie, oscillations) may add critical value. High-frequency stimulation (eg, 5 Hz, 10 Hz, and 20 Hz TMS) and iTBS protocols are able to increase local excitability, but 5 Hz (theta), 10 Hz (alpha), 20 Hz (beta), and iTBS (a mix of 50 Hz gamma and 5 Hz theta) have different and very specific meaning in the brain, beyond excitability. Such oscillatory dynamics may explain why iTBS to the left DLPFC may be ineffective, whereas other excitatory protocols (10 Hz) may lead to benefit. Although both 10 Hz TMS and iTBS over the left DLPFC are therapeutic for MDD, the case may be different for bipolar depression. Hence, although research on the anatomical topography of circuits involved in bipolar depression is critical, so is an assessment of the maladaptive physiological signatures, beyond the traditional cortical excitability framework, taking into consideration oscillatory dynamics to inform the application of an oscillatory intervention.

A negative trial may be considered an undesired scientific outcome, but this study by McGirr et al3 has the potential to impact neuromodulation treatment development in bipolar depression. Clinically, it may focus the next steps in the exploration of traditional repetitive TMS protocols. Perhaps more critically, however, it emphasizes the need to deepen our characterization of circuit pathophysiology, the identification of anatomical treatment targets, and the focus on oscillatory physiological dynamics to understand both disease mechanisms and the distinct mechanisms of action (and clinical indications) of different TMS frequency interventions.

Back to top
Article Information

Published: March 12, 2021. doi:10.1001/jamanetworkopen.2021.1055

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Camprodon JA. JAMA Network Open.

Corresponding Author: Joan A. Camprodon, MD, MPH, PhD, Division of Neuropsychiatry and Neuromodulation, Massachusetts General Hospital, Harvard Medical School, 149 13th St, Ste 10.008, Boston, MA 02129 (

Conflict of Interest Disclosures: Dr Camprodon reported serving on the scientific advisory board of Hyka and Feelmore Labs and has received consultation honoraria from Neuronetics. His research is currently funded by the National Institutes of Health, the AE foundation, the Solinsky Foundation and the Gerstner foundation.

Huang  YZ, Edwards  MJ, Rounis  E, Bhatia  KP, Rothwell  JC.  Theta burst stimulation of the human motor cortex.   Neuron. 2005;45(2):201-206. doi:10.1016/j.neuron.2004.12.033PubMedGoogle ScholarCrossref
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.   Lancet. 2018;391(10131):1683-1692. doi:10.1016/S0140-6736(18)30295-2PubMedGoogle ScholarCrossref
McGirr  A, Vila-Rodriguez  F, Cole  J,  et al.  Efficacy of active vs sham intermittent theta burst transcranial magnetic stimulation for patients with bipolar depression: a randomized clinical trial.   JAMA Netw Open. 2021;4(3):e210963. doi:10.1001/jamanetworkopen.2021.0963Google Scholar
Bulteau  S, Beynel  L, Marendaz  C,  et al.  Twice-daily neuronavigated intermittent theta burst stimulation for bipolar depression: a randomized sham-controlled pilot study.   Neurophysiol Clin. 2019;49(5):371-375. doi:10.1016/j.neucli.2019.10.002PubMedGoogle ScholarCrossref
Nguyen  TD, Hieronymus  F, Lorentzen  R, McGirr  A, Østergaard  SD.  The efficacy of repetitive transcranial magnetic stimulation (rTMS) for bipolar depression: a systematic review and meta-analysis.   J Affect Disord. 2021;279:250-255. doi:10.1016/j.jad.2020.10.013PubMedGoogle ScholarCrossref
Ellard  KK, Uribe  S, Funes  CJ. Resting-state functional magnetic resonance imaging (rsfMRI) in bipolar and unipolar depression. Chapter 26. In: Martin  C, Hunter  LA, Preedy  VR, Patel  V, Rajendram  R, eds.  The Neuroscience of Depression: Genetics, Cell Biology, Neurology, Behaviour and Diet. London: Academic Press; 2021.
Ellard  KK, Zimmerman  JP, Kaur  N,  et al.  Functional connectivity between anterior insula and key nodes of frontoparietal executive control and salience networks distinguish bipolar depression from unipolar depression and healthy control subjects.   Biol Psychiatry Cogn Neurosci Neuroimaging. 2018;3(5):473-484. doi:10.1016/j.bpsc.2018.01.013PubMedGoogle Scholar
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words