Why is deep brain stimulation for treatment-resistant depression a needed treatment option?

Treatment-resistant depression (TRD) is an ongoing area of concern in public health, with increasing interest in the psychiatric scientific community, given its great personal and societal costs. Its prevalence is high, and up to a third of patients do not respond to four consecutive antidepressants.1 Patients with TRD experience significant loss in quality of life, high costs, and hospitalizations, and are estimated to be twice as likely to attempt suicide at least once during their lifetime than non-resistant depressed patients, and 15 times more likely than the general population.2-5

Despite multiple antidepressant treatment options, patients with TRD encounter growing difficulties to maintain a relief of symptoms with subsequent episodes. This is true as well for newer treatments such as ketamine and esketamine.6,7 Electroconvulsive therapy (ECT) is very effective in depression, with remission rates of 60–90% reported in clinical trials, but relapse rates are high, and long courses of ECT have cumulative cognitive side effects that many times become intolerable for patients.8 Non-pharmacological treatments besides ECT have been available for years, with the main example being repetitive transcranial magnetic stimulation (rTMS); multiple meta-analysis have demonstrated efficacy of rTMS by stimulation of the dorsolateral prefrontal cortex (DLPFC).9,10 However, rTMS is a burdensome treatment (daily 40-minute sessions for up to 6 weeks), and patients with high levels of treatment resistance are less likely to respond.11 Vagus nerve stimulation (VNS) was approved for use in TRD in 2005. The device delivers low-frequency, chronic, intermittent-pulsed electrical signals to the left cervical vagus nerve. Studies have described a slow but sustained clinical response, mostly shown by long-term naturalistic follow-up of patients rather than in the primary end points of the clinical trials.12 Sadly, insurance coverage by third-party payers has been limited; therefore, the number of patients benefitting from this treatment option is low despite 15 years of commercial availability.

Deep brain stimulation (DBS) has the potential to provide a new treatment for TRD once other strategies have ceased to work. DBS may provide faster relief than VNS, as well as a sustained response for extended periods of time. Besides VNS, no other treatment option has been studied in long-term results (over 1 year). The conceptualization of psychiatric disorders as circuit-based, and the formulation of depression as the manifestation of dysfunctional brain networks, with support from neuroimaging, enabled the introduction of DBS in depression, modeled after its success in movement disorders. DBS is the most invasive of the neuromodulatory approaches, requiring neurosurgical implantation of bilateral electrodes in the selected area of interest. However, it provides a unique opportunity to achieve sustained control of symptoms of depression. Since the first report of DBS in depression in 2005, multiple targets have been investigated, with promising results.13 The largest clinical samples have studied the subcallosal cingulate white matter (SCC), nucleus accumbens (NAc), ventral capsule/ventral striatum (VC/VS), and medial forebrain bundle (MFB).14 Different open-label case series have reported response in around 40–70% of patients. The enthusiasm of these reports led to large randomized clinical trials (RCTs) in the SCC and VC/VS, which unfortunately did not meet their primary clinical endpoints.15,16 These trials were terminated early after interim analyses determined low likelihood of a positive result with completion of the desired recruitment goals.

Many opinions were expressed regarding potential causes for these failures. One main concern has been directed at the trial design: primary endpoints were reportedly too early to identify a difference between active stimulation and placebo. Supporting this hypothesis, the results from a different trial showed that discontinuation of stimulation after a period of optimization was better suited to identify a difference between active and sham stimulation. Bergfeld et al. described a 40% overall response rate in 25 patients during the open-label phase; then, a number of participants entered a randomized crossover period, in which all responders experienced return of symptoms within less than 2 weeks once stimulation was discontinued.17 Other groups identified that the possible reason for failure of larger RCTs was the surgical protocol to determine the ideal region for implantation. Initially through retrospective analysis of white matter connectivity in patients who responded to SCC DBS, and then through prospective identification of the target using diffusion tractography, Riva-Posse et al. reported that, albeit in a single center, results from DBS can improve with accurate targeting from 41% to 73% with 6 months of stimulation.18,19 This is the same approach used for target selection in MFB, as it necessarily requires identification of the white-matter bundle in which DBS leads are to be implanted.20,21 DBS in the MFB has been reported to yield rapid and effective results in small open-label trials from two separate centers, with around 70% of patients responding.22,23 Different targets (but within the same mood network) then hint at the possibility that adequate DBS requires accurate implantation. A third explanation that could improve the outcome of future trials, along with protocol design and precise targeting, lies at the problem of heterogeneity in the depressive syndrome. If, as proposed above, DBS is a specific intervention on a predefined circuit, then outcome measurement should address the expected results of modulation of that particular circuit.24 Consequently, work ahead should focus on trying to identify clinical, imaging, or physiological characteristics of patients that may respond to DBS, or changes that are exerted by the electrical modulation of the target circuit.25,26 This refinement of patient selection, and biomarker engagement with therapy, has evolved with incremental success in the field of Parkinson disease, even aiding in the determination of target selection depending on clinical characteristics.27,28

There are encouraging results in the field of DBS for depression. Patients have experienced sustained antidepressant response for years after surgery with DBS of the SCC, VC/VS, and MFB.29-31 The keys for success in the near future of DBS for depression will rely on the integration of advances in imaging, neurophysiology, and clinical expertise to plan new multicenter trials that will replicate, on a larger scale, the observations of different research groups, thus ensuring a safe and long-lasting treatment option for the TRD population.32 The stakes are high, but for clinicians who have had the privilege of observing a life-changing procedure treat depression so effectively when all else has failed, there is no other option.

Disclosure

The author has received consulting fees from Janssen Pharmaceutical.