A primer on the treatment of schizophrenia through repetitive transcranial magnetic stimulation

Keywords::

• dorsolateral prefrontal cortex
• motor cortex
• plasticity
• rTMS
• schizophrenia

Transcranial magnetic stimulation (TMS) was developed approximately 30 years ago by Barker et al. [1]. TMS has been applied as both a method to evaluate several neurophysiological processes and as a treatment for resistant psychiatric and neurological disorders. As a treatment, repetitive TMS (rTMS) has extensive evidence for efficacy in mood disorders [2,3] through the modulation of cortical excitability, inhibition and plasticity. In recent years, several studies have also explored the utility of rTMS as a treatment tool in schizophrenia with promising results. The heterogeneity of schizophrenia symptoms appears to be one of the main obstacles in developing treatments for this debilitating disorder. It may be prudent, therefore, for schizophrenia brain stimulation treatment studies to focus on specific symptom domains. In fact, some research has shown that rTMS appears more effective in the treatment of positive symptoms than negative symptoms, although one hypothesis is that this is due to fewer studies exploring the effects of rTMS on negative symptoms [4,5], or due to the scales used to measure efficacy [4,6]. Moreover, rTMS appears to have significant efficacy for auditory hallucinations [7]. rTMS can be applied as high (5–20 Hz) or low (1 Hz) frequency, with the former being usually excitatory and the latter being inhibitory [8,9]. Such selective modulation can have advantages when examined through the lens of fMRI characterization of hallucinations and negative symptoms. Simply put, the inhibitory stimulation frequency of 1 Hz has been used in studies examining treatment of positive symptoms and some of the more promising effects in negative symptoms used the excitatory stimulation frequencies of 10 or 15 Hz [10]. Below, we will focus on evidence for the treatment of auditory hallucinations and more pervasive negative symptoms, before addressing future directions for the exploration of rTMS in the treatment of schizophrenia.

Auditory hallucinations

Auditory hallucinations, as a primary symptom of schizophrenia [11], can be a main target of antipsychotic treatment. However, up to 40% of patients have only a partial response to medications [12]. As such, there has been extensive exploration of the neurocognitive basis of auditory hallucinations. Neuroanatomical and imaging studies in these ‘treatment-resistant’ patients have demonstrated hyperactivity in the left temporo-parietal cortex [13,14]. By exploiting its inhibitory role, when applied at low frequencies, an rTMS protocol delivered at 1 Hz to the posterior superior temporal gyrus (STG) [15]. Hoffman et al. were able to demonstrate a reduction in auditory hallucinations after application to this speech processing cortex. Moreover, a subsequent study by the same group demonstrated sustained improvement 15 weeks after treatment in many of the subjects [16].

Other groups targeted different anatomical areas likely to be involved in the genesis of auditory hallucinations. Unfortunately, results of treatment studies have not been robust [17]. It should be noted that the majority of rTMS treatment studies have centered on subjects with medication-resistant auditory hallucinations. Very little is known about the treatment effect rTMS would have for antipsychotic-responsive subjects.

Lastly, a recent meta-analysis by Slotema et al. examined 7 rTMS randomized controlled trials (RCTs) with a total of 189 schizophrenia subjects who were experiencing pervasive auditory hallucinations [7]. One hundred and five subjects received active rTMS treatment and there was a moderate effect size of 0.54, with only an 8.6% occurrence of side effects. While this was a somewhat smaller effect size than a previous meta-analysis [18], the clear recommendation remains for rTMS as treatment of auditory hallucinations in schizophrenia: specifically in treatment-resistant patients.

Negative & cognitive symptoms

The neuroanatomical basis of pervasive negative symptoms in schizophrenia has been localized, in part, to the medial frontal areas [19], anterior cingulate [20] and medial temporal lobe [21]. The correlation of reduced frontal activation with more severe negative schizophrenia symptoms has been corroborated by several functional neuroimaging studies [22–24]. Given the excitatory effects of high frequency rTMS (5–20 Hz) [25], several studies have explored targeting the dorsolateral prefrontal cortex (DLPFC) to reduce negative symptoms. The common hypothesis has been that high-frequency rTMS would decrease negative schizophrenia symptoms by modulating perfusion, cerebral metabolism and neuronal excitability in the prefrontal cortex [26–28]. A statistically significant, albeit small, clinical decrease in negative symptom intensity was reported by Cohen et al. in a study of six subjects [29]. Neuroimaging studies demonstrated disrupted activation of the right PFC that correlated with negative schizophrenia results [30], with similar results in bilateral prefrontal cortices [31]. Fitzgerald et al. designed a study to verify the efficacy of bilateral stimulation of the PFC in treatment of negative symptoms [32]. The results reported that there was no statistically significant difference between real and placebo stimulations. A further study, by Barr et al. focused on stable, medicated schizophrenia patients with prominent negative symptoms [33]. The efficacy of 4 weeks of daily 20 Hz bilateral stimulation targeted to the DLPFC was determined by examining changes in the scale for the assessment of negative symptoms (SANS) and positive and negative syndrome scale (PANSS). There was no difference in negative symptoms with either measure with either active or sham rTMS [33]. While the most effective rTMS parameters for the treatment of negative symptoms have yet to be defined, several studies have explored different stimulation parameters in an effort to explore treatment of negative symptoms with rTMS. With respect to stimulation frequency, 10 Hz has shown evidence to be optimal in minimizing negative symptoms of schizophrenia [34,35], and the DLPFC appears to be an effective target.

Researchers have also targeted the DLPFC in rTMS treatment studies in hopes of treating the cognitive deficits of schizophrenia. Like negative symptoms, cognitive deficits in schizophrenia are not effectively treated with antipsychotic medication [36–38]. As one of the identified cognitive deficits of schizophrenia, working memory [39] has been connected with altered DLPFC activity in schizophrenia [40–42] in neuroimaging and neurophysiological studies. Targeted rTMS to DLPFC has been found to improve working memory function in healthy subjects when applied bilaterally [43] and to the right DLPFC [44]. Through a combination of working memory tasks and neurophysiological measures, Barr et al. published a pilot study assessing the effects of high-frequency rTMS to bilateral DLPFC regions, as targeted by MRI, on working memory performance [45]. They compared 27 subjects with schizophrenia with 13 age- and sex-matched controls in a 4-week randomized double-blind sham-controlled treatment trial. Barr et al. were able to demonstrate that bilateral 20-Hz rTMS targeted to DLPFC significantly improved working memory compared with sham [45]. Statistical analysis revealed that subjects with schizophrenia performed significantly worse on working memory tasks than healthy subjects prior to rTMS administration, although this improved to a similar level as the healthy subjects after the rTMS course [45]. These results provided compelling evidence of an improvement in working memory performance in schizophrenia patients after rTMS.

Conclusion

The results discussed above must be considered with the view that most rTMS studies include subjects with very treatment-resistant forms of schizophrenia. Most studies include patients with medication-resistant auditory verbal hallucination, indicating a group with intractable symptomatology. Despite this, there is some evidence that patients with lower PANSS scores at baseline may respond better to rTMS in comparison with more severely ill patients [46]. Previous research has linked schizophrenia with a failure to integrate activity in distributed neural circuits [47–50]. rTMS has the potential to induce changes in remote, functionally connected brain areas [51,52] in addition to local effects. The application of sequential stimulation, aimed at the areas of the PFC and temporoparietal cortex, seems to be an interesting approach, with some evidence for reduction in not only negative but also positive schizophrenia symptoms and potentially improved working memory [53]. In this manner, targeting other brain areas, perhaps simultaneously or sequentially, may prove effective in targeting resistant symptoms of schizophrenia.

Although treatment with rTMS is generally well tolerated, it is quite difficult to recruit patients with prominent negative symptoms of schizophrenia to rTMS studies. Patients with negative symptoms are not motivated to undergo rTMS treatment, probably owing to the core negative symptoms such as anhedonia or loss of interest. Other patients are likely unaware of the minimal side-effect profile of rTMS, and associate brain stimulation with frightening, archaic protocols. Owing to low adherence to medications, as well as other factors, there is a high degree of pharmacoresistance associated with the disease, as well as clear deterioration of the socioeconomic status of patients, as schizophrenia progresses. There is, therefore, a medical imperative to develop new or augment existing treatments for schizophrenia and help the often intractable symptoms these patients deal with on a daily basis.

Financial & competing interests disclosure

In the last 5 years, ZJ Daskalakis received external funding through Brainsway Inc. and a travel allowance through Pfizer and Merck. ZJ Daskalakis has also received speaker funding through Sepracor Inc. and AstraZeneca, and has served on the advisory board for Hoffmann-La Roche Ltd. This work was supported by the Ontario Mental Health Foundation, the Canadian Institutes of Health Research, the Brain and Behaviour Research Foundation, the Temerty Family, the Grant Family, the Centre for Addiction and Mental Health Foundation and the Campbell Institute. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript.

No writing assistance was utilized in the production of this manuscript.