Major depressive disorder is a highly prevalent condition (
1), estimated to affect approximately 280 million people worldwide (
2). It is well established that a significant proportion—approximately one-third—of individuals with major depression develop treatment-resistant depression after failing to respond to first-line therapies (
3). In fact, the 12-month prevalence of treatment-resistant depression is 2.76 million in the United States alone (
4). There remains a strong need for effective treatments in this population. In this issue, Cole et al. (
5) report findings on a new treatment protocol developed to rapidly and effectively treat individuals with treatment-resistant depression.
The Cole et al. study demonstrates safety and significant efficacy of Stanford neuromodulation therapy (SNT) in a placebo-controlled design with 29 participants. In this protocol, patients are treated with repetitive transcranial magnetic stimulation (rTMS) 10 times per day over the course of a week, with 50 minutes between each brief treatment. Each treatment consists of a total of 1,800 pulses of intermittent theta burst stimulation (iTBS) rTMS over the left dorsolateral prefrontal cortex (DLPFC), localized by MRI. Stimulation is at 90% of resting motor threshold. The authors of the SNT protocol previously published findings on their accelerated rTMS treatment, identical to the treatment of focus in this article, in uncontrolled trials. They showed the potential of this rapid treatment protocol in a sample of six patients with highly treatment-resistant depression, among whom five met criteria for response and four for remission at end of treatment (
6). In a follow-up study, with a sample of 21 patients, they found response and remission rates of 90% at treatment end (
7). Their use of a placebo control in this latest study is important given the relatively high likelihood of a placebo response in treatment-resistant depression (
8) and in accelerated rTMS protocols (
9), since the treatment involves intensive patient involvement over a 5-day treatment program. The authors successfully maintained the blind throughout the study, and there were limited differences in side effects between groups, with headache in the active treatment group being the only one that was significantly greater. In terms of the primary outcomes, the authors reported an impressive 69.2% response rate and a 46.2% remission rate 4 weeks after treatment, with a very large effect size (Cohen’s d) of 1.4. They justifiably compare their outcomes to ECT, the gold-standard treatment in this patient group.
Other research groups have previously demonstrated the feasibility, safety, and efficacy of accelerated TBS. Compared with the SNT protocol, however, these studies included far fewer sessions per day and alternative treatment parameters with regard to percent resting motor threshold of treatment, time period between treatments, and laterality of treatment target (
10–
13). None of these previous studies have delivered the large remission rates demonstrated in the SNT protocol.
The Cole et al. study represents a significant therapeutic innovation for the use of rTMS as a treatment for major depressive disorder. There has long been a general understanding in the field of noninvasive brain stimulation that given the wide range of available treatment parameters that rTMS affords, there would be significant room for improvements in treatment design for major depression. The first rTMS treatment approved by the U.S. Food and Drug Administration for major depression was a 10-Hz protocol taking up to 37.5 minutes and given once a day for 4–6 weeks, which was effective but grossly inefficient as an interventional treatment. The next significant advance was approval of a condensed variant of rTMS, theta burst stimulation (TBS). The impetus for the development of TBS came from preclinical studies exploring neuroplastic effects of a form of brain stimulation that mimics the endogenous burst discharges in the theta range on EEG (
14). The 10-Hz and the iTBS protocols were found to be equally efficacious (
15), and similar to the 10-Hz protocol, iTBS was initially approved to be delivered once a day, but over a short 3-minute period, which was a significant reduction in treatment time. Ultimately, this considerable reduction in protocol time subsequently freed up researchers to experiment with accelerated treatment schedules, leading to major opportunities for pragmatic improvements in the growing field of rTMS, and leading to our discussion here.
While the SNT protocol demonstrates strong treatment effects in this study, there are some questions with regard to study design and patient inclusion, which may limit the generalizability and full adoption of this treatment innovation in the treatment-resistant depression population. First, the severity of treatment-resistant depression and level of treatment resistance in the study sample appears to be moderate, not severe, owing partially to the relatively few failed medication trials during the current episode (the mean number of adequate antidepressant trials in the current episode was two for the active treatment group and one for the sham treatment group). It remains to be seen how effective this treatment will be in patients with more severe illness in whom ECT is commonly used (e.g., inpatients or those with emergent suicidality [
16]). Nevertheless, the logistics of delivering SNT may actually be more practical for a patient who is already stationed in a hospital setting all day. It may be difficult for patients employed full-time to lose an entire work week, especially given the limited evidence on the long-term durability of response with SNT (the response and remission rates declined by week 4 in this study). It is therefore not yet clear who the optimal target population is for this treatment. The other issue with generalizability here is that the authors use an MRI-guided technique to localize treatment to the left DLPFC. Recent advances in these methods have improved the ability to optimally target specific locations within the DLPFC. What has yet to be determined is whether these methods lead to improved remission rates above standard targeting techniques (
17). Moreover, MRI-guided targeting of the DLPFC could be prohibitive for clinics running small margins, or for patients who are partially insured. Lastly, only 16% of individuals screened online were included in this study, which implies that patients were carefully selected. Even so, the overall results are both exciting and compelling, and follow-up studies with the SNT approach are currently under way.
There is a growing body of evidence to support the notion that targeting the left DLPFC with rTMS dynamically changes the connection of this brain region with the subgenual cingulate cortex, an area strongly implicated in the pathophysiology of major depression (
18,
19). This theory suggests that rTMS activates the underactive left DLPFC in depressed individuals, which in turn reduces overactive subgenual cingulate cortex activity in major depression, possibly through a long-term potentiation method at the neuronal level. In fact, there is some preclinical evidence to support the specific 50–60 minute intertreatment window in the SNT protocol as a component of the treatment that enhances clinical improvement through the process of additive metaplasticity on the cell and circuit level (
20). Metaplasticity refers to the process of change in magnitude or direction of plasticity; put more simply, it is plasticity of neuroplasticity. Thus, treatment with the SNT protocol may, in theory, accelerate neuroplasticity through metaplastic effects on the magnitude and intensity of TBS-induced neuroplasticity. This could be one reason why the SNT protocol yields such strong effects while earlier accelerated protocols with shorter intertreatment windows far shorter than 50 minutes did not. A large open-label accelerated trial currently under way has also found robust responses with this same intertreatment window, albeit with only 600 pulses per treatment (D. Blumberger et al., personal communication). It still remains to be determined whether the intertreatment interval (50 minutes) is necessary to produce the robust treatment response or whether this is simply a reflection of the uniquely very high number of stimuli delivered daily in SNT (18,000). While it has been suggested that repeated sessions of iTBS produce neuroplasticity to achieve the beneficial effects of accelerated iTBS in depression (20), this has recently been called into question (21). Hence, the high number of pulses delivered in an even shorter time window (i.e., between 1 and 2 hours) may potentially achieve comparable results, and such trials should be considered.
With the growing acceptance of accelerated TBS treatments by physicians, researchers, and patients, a new avenue of rTMS research is now open. Future studies in this area should further refine treatment parameters and investigate the role for accelerated protocols in other psychiatric diseases, such as bipolar depression and severe treatment-resistant depression in the inpatient setting. Further study of these treatment approaches may lead to improved outcomes in a substantial proportion of patients, offering tremendous hope for those who suffer with disabling conditions.