If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Evidence suggests that schizophrenia constitutes a neurodevelopmental disorder, characterized by a gradual emergence of behavioral and neurobiological abnormalities over time. Therefore, applying early interventions to prevent later manifestation of symptoms is appealing.
Objective
This review focuses on the use of cortical neuromodulation in schizophrenia and its potential as a preventive treatment approach. We present clinical and preclinical findings investigating the use of neuromodulation in schizophrenia, including the current research focusing on cortical non-invasive stimulation and its possibility as a future preventive treatment.
Methods
We performed a search in Medline (PubMed) in September 2020 using a combination of relevant medical subject headings (MeSH) and text words. The search included human and preclinical trials as well as existing systematic reviews and meta-analysis. There were no restrictions on language or the date of publication.
Results
Neurodevelopmental animal models may be used to investigate how the disease progresses and thus which brain areas ideally should be targeted at a given time point. Here, abnormalities of the prefrontal cortex have been often identified as an early and persistent impairment in schizophrenia. Currently there is insufficient evidence to either support or refute the use of neuromodulation to the cortex in adult patients with already manifested symptoms. However, preclinical results show that early non-invasive neuromodulation to the prefrontal cortex of adolescent animals, sufficiently prevents later psychosis-relevant abnormalities in adulthood. This points to the promising potential of cortical non-invasive neuromodulation as a preventive treatment when applied early in the course of the disease.
Conclusion
Preclinical translational-oriented findings indicate, that neuromodulation to cortical areas offers the possibility of targeting early neuropathology and through this diminish the progression of a later schizophrenic profile. Further studies are needed to investigate whether such early cortical stimulation may serve as a future preventive treatment in schizophrenia.
Schizophrenia, is a complex, highly disabling and multi-symptomatic neuro-psychiatric disorder, with a significant socio-economic impact affecting both the patient, their families, and the society as a whole [
]. According to recent views, schizophrenia is considered a neurodevelopmental disorder, in which progressive accumulation of neuropathological disruptions facilitates an outbreak of symptoms later in life [
]. The symptom profile includes both cognitive deficits, negative symptoms, and positive symptoms, with the later typically emerging during late adolescence/early adulthood [
]. Whereas for today, antipsychotic medications constitute the first line of treatment, yet 10–30% of patients respond poorly, or do not respond at all, and additional 30% of patients report only a partial relief [
World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of schizophrenia, Part 1: acute treatment of schizophrenia.
]. Further, while being to some degree effective against the positive symptoms, antipsychotics medications fail to substantially improve cognitive deficits or negative symptoms [
]. Altogether, this points to the necessity of investigating novel treatments. When considering the neurodevelopmental nature of schizophrenia alongside its devastating outcomes, the development and testing of preventive avenues is appealing. The etiology of schizophrenia is likely multifactorial and may include both heritable genetic factors as well as early environmental insults. Amongst others, prenatal insults have shown to disrupt normal brain maturation and subsequent lead to dysfunctional neuronal circuits found in schizophrenia. Consistent with the neurodevelopmental model, prenatal exposure to infections are associated with an increased risk of schizophrenia in the human offspring [
]. This association has been back-translated into the lab and resulted in the establishment of several animal models of schizophrenia, based on the introduction of amongst others prenatal maternal immune activation [
The role of maternal immune activation in altering the neurodevelopmental trajectories of offspring: a translational review of neuroimaging studies with implications for autism spectrum disorder and schizophrenia.
]. Given the ethical, clinical and methodological limitations of human trials, the concurrent use of valid animal models capturing the neurodevelopmental course of this disorder allows for an in-depth investigation into future prevention strategies.
This review focuses on the use of cortical neuromodulation in schizophrenia and its potential as a preventive treatment. We briefly present schizophrenia as a neuronal circuit disorder, by outlining the interconnected neuronal circuits involved in the pathology. This is followed by a presentation of animal models displaying a neurodevelopmental nature of schizophrenia, and in which neuromodulation techniques so far have been applied. We then provide a separate discussion of clinical and preclinical trials, investigating the use of neuromodulation in schizophrenia, including the current research on cortical non-invasive neuromodulation, and how this may offer the possibility as a future preventive treatment.
Methods
We performed a search in Medline (PubMed) in September 2020 using a combination of relevant medical subject headings (MeSH) and text words including; schizophrenia, animal models, preventive treatment, neuromodulation, deep brain stimulation (DBS), transcranial direct current stimulation (tDCS) and transcranial magnetic stimulation (TMS). The search included human and preclinical trials as well as existing systematic reviews and meta-analysis. There were no restrictions on language or date of publication. The reference list of eligible studies was also assessed for identification of further relevant studies.
Schizophrenia – A neuronal circuit disorder
The pathology of schizophrenia has traditionally been centered around a dysfunctional dopamine (DA) system, and it is widely accepted that abnormalities in this system could explain positive symptoms [
]. However, a sole focus on DA has been challenged as oversimplified, and DA dysfunction is now rather considered a pathological finding integrated alongside abnormal neuronal circuits [
]. Indeed, structural and functional imaging as well as pharmacological studies have pointed towards schizophrenia as a disorder of distinct neuronal circuits. This potentiates the application of neuromodulation techniques, in which specific brain areas are targeted, with the goal to obtain symptomatic relief [
Schizophrenia is characterized by the presence of both hypo – and hyperconnectivity in distinct circuits. Clinical investigations have shown, that both mesolimbic and mesocortical circuits collectively contribute to the spectrum of symptoms [
]. In line with the hypo/hyperconnectivity profile, the neuropathology involves opposing DA activity, as hyperactivity is found in the mesolimbic DA system, together with a hypoactive mesocortical DA signaling [
]. In patients, negative symptoms and cognitive deficits have been linked to the hypodopaminergic state of the mesocortical circuit and decreased integrity of especially the frontal cortex, which in particular involves abnormalities of the prefrontal cortex (PFC) [
]. Accordingly, mild impairment in executive functions, related to early abnormalities within the frontal cortex, is observed in patients considered to be in the prodromal phase of schizophrenia [
]. As disease progresses, this is followed by additional decline in cognitive functions and further morphological changes, which now also includes temporal areas. This shows that ongoing neuroplastic changes plays a role in the course of the disease [
]. As a part of the mesocortical circuit, the PFC has multiple connections to other subcortical areas involved in schizophrenia, including the striatum, hippocampus and dorsomedial thalamus (DM) [
]. On the contrary, positive symptoms, such as delusions and hallucinations, are linked to the hyperdopaminergic state within the mesolimbic system - with brain areas including the nucleus accumbens (Nacc) and the hippocampus [
] Similar mechanism of action has been hypothesized for the use of neuromodulation. Here, direct stimulation of the Nacc and hippocampus is thought to alleviate psychotic symptoms by dampening the hyperdopaminergic state [
]. A study by Wang et al. showed, that the disease process of schizophrenia, is related to progressive time-dependent changes in specific subcortical areas that are directly connected to the cortex, thus including the Nacc, hippocampus and thalamus. On the contrary, no progressive change was observed in subcortical areas indirectly connected to the cortex [
]. Deficit in sensorimotor gating is observed in neurodevelopmental disorders and consistently found in both patients and animal models of schizophrenia [
]. Sensorimotor gating can be operationally measured using the pre-pulse inhibition (PPI) paradigm. Findings show that PPI deficits may be linked to a combined disruptions between the mesocortical system, primarily extending between the PFC and DM, and the mesolimbic system including the Nacc [
]. On a neuroanatomical level, several volumetric changes have been found in patients with schizophrenia, including a decrease in the hippocampus and PFC as well as an increase in lateral ventricles [
As such, the neuropathology involves interconnected circuits and associated brain areas, which collectively leads to the symptomatic profile observed in schizophrenia. As described later, many of these areas have already been targeted by the use of both invasive and non-invasive neuromodulation techniques. However, the most appropriate brain target still needs to be determined, and here the timing on when stimulation is applied may be crucial. From a pathological point of view, the PFC plays a prominent and potential early role as the disease evolves. From a neuroanatomical point of view, the PFC is interconnected with several subcortical regions relevant to schizophrenia. This makes the PFC an ideal candidate for the application of neuromodulation techniques, as this may facilitate both direct modulation of the abnormal activity of the cortex as well as potentially mediate an indirect modulation of its associated networks (see Fig. 1). When seeking to construct a preventive treatment approach, it is essential to investigate how early and later deviations within the neuronal circuits emerge, and thus to determine which brain area should be targeted at a given time point. Apart from solely relying on clinical observations, which bears a certain number of limitations, further in-depth investigations are possible by the concurrent use of appropriate animal models.
Fig. 1Current brain areas relevant to schizophrenia in which neuromodulation has been applied. Several cortical, as well as subcortical brain regions, have shown to play a prominent role in the symptomatic profile of schizophrenia. Many of these areas have already been targeted in the context of schizophrenia, both in clinical and preclinical settings by use of deep brain stimulation (areas marked with red lining/red text and with references provided in brackets). A few cortical regions have been targeted by the use of non-invasive neuromodulation, either with transcranial direct current stimulation or transcranial magnetic stimulation (areas marked with yellow lining and with references provided in brackets). Here, especially the prefrontal cortex (PFC) is interconnected with several of these subcortical regions (marked with a thicker black line), which makes the PFC an ideal target for the application of non-invasive neuromodulation techniques to provide both direct stimulation of the aberrant cortical activity as well as indirect stimulation of its associated networks. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Animal models displaying a neurodevelopmental course
To understand how the neuropathology of schizophrenia progresses, animal models displaying the neurodevelopmental nature of the disease are needed. Such animal models generally rely on maternal exposure to various environmental stimuli, thereby mimicking the epidemiological findings showing an association between schizophrenia and prenatal/perinatal exposure to environmental insults in patients. Such exposures in animals tend to facilitate the progression of both pathological findings and symptoms relevant to schizophrenia, which further have proven sensitive towards antipsychotic drugs. Alongside the etiological significance, this shows the high level of both face -, predictive - and construct validity of these neurodevelopmental models. A further comprehensive review on neurodevelopmental animal models in schizophrenia can be found in the Review by Meyer et al., 2010 [
]. Currently, only two neurodevelopmental animal models of schizophrenia have been used to investigate the effect of neuromodulations techniques. These include the so-called MAM-E17 model and MIA model.
The MAM-E17 model is based on maternal injection of methylazoxymethanol acetate (MAM) on embryonic day 17, which disrupts cellular proliferation in the embryo during time of administration. MAM injections targets proliferating cells migrating to the hippocampus and cortical areas [
]. The consequences of these induced disruptions, start to occur in late puberty, eventually facilitating the manifestation of symptoms as well as neuroanatomical deficits in this model. On a behavioral level, the MAM-model displays hyperlocomotion, social withdrawal, working memory deficits, and deficits in sensorimotor gating [
]. Structural and functional abnormalities have been found in limbic areas and the neocortex, including an abnormal neuronal density in the PFC and decrease in total brain volume [
Disruption of neurogenesis on gestational day 17 in the rat causes behavioral changes relevant to positive and negative schizophrenia symptoms and alters amphetamine-induced dopamine release in nucleus accumbens.
]. The MAM-E17 model furthermore displays a loss of parvalbumin interneurons in the hippocampus and PFC, which is linked to deficits in GABAergic signaling and subsequent hypofunction within these areas [
Neurobiology of disease A loss of parvalbumin-containing interneurons is associated with diminished oscillatory activity in an animal model of schizophrenia.
Decrease in parvalbumin-expressing neurons in the hippocampus and increased phencyclidine-induced locomotor activity in the rat methylazoxymethanol (MAM) model of schizophrenia.
]. In accordance with the hyperfunctioning of the mesolimbic system, a study by Flagstad et al., 2004 showed, that amphetamine application to the MAM-E17 model facilitated a larger increase in DA release within the Nacc as compared to controls animals. When amphetamine was applied to the medial PFC (mPFC), this subsequently led to a decrease in DA release within the Nacc in control animals, yet this was not observed in the MAM-E17 model [
Disruption of neurogenesis on gestational day 17 in the rat causes behavioral changes relevant to positive and negative schizophrenia symptoms and alters amphetamine-induced dopamine release in nucleus accumbens.
]. This diminished interaction between the mPFC and Nacc observed in MAM-E17 model, is believed to reflect an inability of cortical areas to regulate subcortical areas, thus correlating with the decreased function of especially the frontal cortex observed in schizophrenia [
Disruption of neurogenesis on gestational day 17 in the rat causes behavioral changes relevant to positive and negative schizophrenia symptoms and alters amphetamine-induced dopamine release in nucleus accumbens.
The MIA model is based on the environmental risk factor of maternal infection during pregnancy, in which pregnant dams are exposed to the viral mimic polyinosinic-polycytidilic acid (Poly I:C). The offspring of these dams subsequently display a spectrum of behavioral abnormalities characteristic of schizophrenia, which similar to clinical observations, do not emerge until late adolescents/early adulthood. These include deficits in PPI, attentional control, working memory, and sensitivity to amphetamine [
Immune activation during pregnancy in rats leads to a PostPubertal emergence of disrupted latent inhibition, dopaminergic hyperfunction, and altered limbic Morphology in the offspring: a novel neurodevelopmental model of schizophrenia.
]. In addition, the neuropathology observed in the MIA model emerges gradually over time, in which some processes precede the occurrence of the behavioral deficits. On a neuroanatomical level, abnormalities in the adolescent MIA animals include a decrease in the volume and neurogenesis of the hippocampus, which is later followed by an emergence of an increased volume of lateral ventricles and decrease in PFC volumes in adulthood [
]. Furthermore, MIA affects both metabolic brain activity and neurotransmission in a maturation-dependent manner. Lower levels of DOPAC (dopamine metabolite) and activity in the PFC and ventral hippocampus have been seen prior to and with the manifestation of behavioral deficits in adult MIA animals. This reflects an early impairment of especially the PFC and points towards an early and lasting hypofunction of this area, which subsequently matches with cognitive deficits being one of the first manifestations in schizophrenia [
]. As the disease progresses, higher levels of DA in the Nacc and globus pallidus (GP) have been found to develop as a matter of age, in which the MIA model eventually displays higher levels of DA and higher metabolic activity within the Nacc as compared to controls in adulthood. As mentioned earlier, it is hypothesized, that the combined action of both the decrease in PFC neuronal activity and increase in Nacc neuronal activity eventually promotes sensorimotor gating deficits in adult rodents. Thus, this may explain the PPI deficits observed in the MIA animal [
]. The MIA model furthermore displays aberrant inflammatory properties by means of increased density of microglia and increased soma size in the hippocampus and Nacc. In the MIA model, such inflammatory abnormalities have been linked to subsequent schizophrenia-like behavior [
Collectively, this shows that both models of schizophrenia display a neurodevelopmental course and subsequent delay in symptom manifestation, in which a combined disruption in both mesolimbic – and mesocortical circuits eventually leads to an outbreak of symptoms. Such findings indicate common features of deficits between these animal models and patients with schizophrenia. This potentiates the use of such animal models for investigating how neuromodulation techniques may affect the underlying pathology relevant to schizophrenia.
Invasive neuromodulation to target existing symptoms
To improve treatment strategies, the use of neuromodulation techniques such as deep brain stimulation (DBS) is suggested as a targeted approach. DBS is a procedure that involves the implementation of electrodes in pathology-relevant brain targets. Electrical stimulation is applied through the electrodes, which alters the activity of the target alongside its associated network [
]. When using DBS, various parameters can be applied, including low-frequency stimulation (LFS) or high-frequency stimulation (HFS), as well as different pulse widths and current intensities.
Clinical trials
The use of DBS in schizophrenia has so far only been investigated in a few treatment-resistant adult patients with existing symptoms of schizophrenia. In a pilot study by Corripio et al. (2016), stimulation was applied to either the Nacc or the subgenual anterior cingulate cortex (sgACC). The stimulation was initially applied with the following parameters; 2.5 voltage, 60 μs pulse width, 130 Hz. During the stabilization phase, parameters were individually modified, guided by the degree of symptoms observed in each patient. Patients in each group showed positive results, as symptom severity was reduced following stimulation to the respective brain targets [
]. Further investigation into the metabolic effect of DBS in the same group of patients, showed a change in glucose metabolism in the targeted area as well as the associated network. Metabolic changes were most prominent following Nacc-DBS as opposed to sgACC-DBS [
]. Apart from the Nacc and sgACC, DBS has also been applied to the habenula (HB) in two adult patients with treatment-resistant schizophrenia. The parameters applied included: 60 μs pulse width, 60 Hz and 135 Hz, with the voltage being increased from 0 V to 10 V. Following treatment, both patients displayed clinical improvement. However, at 12 months follow-up, only one patient continued to benefit from treatment, whereas the other patient deteriorated. These results showed that HB-DBS may be effective in certain patients with schizophrenia [
]. An overview of the investigated targets by means of DBS in clinical settings, can be found in Fig. 1. Further insight into the use of DBS in schizophrenia is found in the review by Gault et al., 2018 [
]. The use of DBS in early stages of schizophrenia has not yet been investigated. However, in other disorders such as Parkinson, results show that if DBS is applied early in the course of the disease, this may slow down the progression of rest tremor [
Apart from being a therapeutic approach, DBS may also serve as an investigative tool in preclinical settings, as it allows for a direct assessment, on how distinct brain areas and neuronal circuits are involved in a given disorder [
Anti-anhedonic effect of deep brain stimulation of the prefrontal cortex and the dopaminergic reward system in a genetic rat model of depression: an intracranial self-stimulation paradigm study.
]. So far, the effect of DBS has only been investigated in two neurodevelopmental models - the MIA and MAM-E17 model. The study by Klein et al., 2013, was the first paper to systematically investigate the effect of different DBS parameters in the MIA model. The stimulation parameters applied included: 90 μs pulse width, 5 or 130 Hz, and constant current of 75 or 150 μA. In this study, the MIA model was utilized to test the efficacy of DBS on PPI deficits observed in the adult rat, when DBS was applied to various brain regions, using different parameters. The behavioral effect of stimulation was assessed both during stimulation and post stimulation. Areas included the Nacc, mPFC, DM, subthalamic nucleus (STN), entopeduncularis nucleus (EP) (rodent equivalent to GP internus), and the GP. The authors complemented these efforts by testing HF-DBS using the pubertal WIN administration rat model of schizophrenia. Results showed, that application of HF stimulation to the mPFC, DM and Nacc all improved PPI deficits in the MIA model, which is in line with the regulatory role of these areas in PPI as mentioned earlier. The ameliorating effects of DBS were proven to be transient as PPI deficits were measured again once stimulation was terminated. On the contrary, targeting the STN or EP did not affect PPI deficits. The translational power of results was further strengthened as HFS to the mPFC and DM was also shown to normalize PPI deficits in the WIN model [
]. Based on these findings, investigations were extended and the effects of HF-DBS to the mPFC and Nacc in the MIA model were tested in two behavioral paradigms; PPI and latent inhibition (LI). The stimulation parameters applied included: 90 μs pulse width, 130 Hz, and with a constant current of 150 μA [
]. This was coupled with an investigation into the effect of stimulation on neurocircuitries by the use of FDG-PET. DBS to the mPFC and Nacc both improved PPI and LI deficits in the MIA model. Of note, HFS-DBS to the Nacc induced deficits in both PPI and LI in control animals, indicating that the effect of stimulation is pathology dependent. Metabolic investigations into the effect of DBS on neurocircuitry showed, that even though Nacc-DBS and mPFC-DBS led to a similar behavioral response, stimulating these two areas induced an opposing metabolic response in adjacent brain circuits in the MIA model. This showed that mPFC-DBS and Nacc-DBS operate different circuits as well as different mechanisms [
]. Here, it may be speculated, that the opposing effects following DBS to the mPFC and Nacc potentially reflects the hypo/hyperfunctioning profile of these areas, found in the MIA model of schizophrenia as well as in patients. The hippocampus in conjunction with the Nacc and ventral tegmental area has amongst others been implicated in the onset of psychosis and therefore the hippocampus has also been suggested as a possible DBS target in patients with schizophrenia [
]. This has been pre-clinically assessed in the MAM-E17 model, in which HF-DBS has been applied to the ventral hippocampus. Stimulation parameters included: 100 μs pulse width, 130 Hz and a current of 200 μA. Results showed that stimulation to the ventral hippocampus reversed auditory evoked potentials in the mediodorsal thalamic nucleus and infralimbic cortex. This shows that simulation to the ventral hippocampus has the potential of reversing deficits in the processing of auditory information in schizophrenia [
Deep brain stimulation of the ventral hippocampus restores deficits in processing of auditory evoked potentials in a rodent developmental disruption model of schizophrenia.
]. An overview of the investigated targets by means of DBS in preclinical settings can be found in Fig. 1.
Collectively, both clinical and preclinical settings show, that a direct stimulation of relevant brain areas, leads to a decrease in symptoms in adulthood. However, so far, the most appropriate target still needs to be identified. The effect of DBS on brain functioning still needs further investigation, yet it is considered to involve a combined action of local inhibition mediated by direct stimulation and subsequent distant activation of associated networks [
]. In both clinical and preclinical settings, stimulation to cortical areas (sgACC and mPFC) as compared to a subcortical area (Nacc) led to differential effects in the metabolic activity of associated networks. Increased metabolic activity has been associated with a modulation of neuronal circuits, and through this linked to clinical improvement [
]. Since disease progression relies on neuroplastic changes within distinct networks, this indicates that the relevance of a given brain target may variate depending on the stage of the disease. As cortical areas are potentially involved in the early manifestation of schizophrenia, neuromodulation to cortical areas, before the disease reaches chronicity, holds a certain potential. However, the use of early invasive neuromodulation such as DBS has obvious clinical and ethical limitations. As an alternative, non-invasive neuromodulation techniques allow for a potential broader application.
Non-invasive neuromodulation as a preventive treatment
Given the neurodevelopmental course of schizophrenia, various preventive approaches have been investigated in young adolescent patients considered to be in high risk of developing psychosis. These approaches included both psychological, nutritional, and pharmacological measures, with generally encouraging results [
Omega-3 fatty acid supplementation changes intracellular phospholipase A 2 activity and membrane fatty acid profiles in individuals at ultra-high risk for psychosis.
Effects of omega-3 PUFA on immune markers in adolescent individuals at ultra-high risk for psychosis – results of the randomized controlled Vienna omega-3 study.
]. As such, the application of the atypical antipsychotic olanzapine during the prodromal phase of schizophrenia led to a subsequent clinical improvement in symptoms [
]. Based on clinical data and preclinical investigations, the cortical impairment of especially the PFC has been suggested an early and lasting deficit in the progression of schizophrenia [
]. Due to such cortical abnormalities, schizophrenia disorder is an ideal candidate for applying non-invasive neuromodulation techniques. Such techniques include transcranial direct current stimulation (tDCS) and transcranial magnetic stimulation (TMS). In these procedures, either a weak current or magnetic source is applied across the scalp, by which the cortical area considered involved in the neuropathology, is targeted. Both TMS and tDCS are considered well tolerated and safe to use in clinical settings [
There are currently no clinical trials investigating how cortical non-invasive neuromodulation affects disease progression, if stimulation is applied prior to symptom manifestation in patients with schizophrenia. However, the effect of TMS applied to either the dorsolateral prefrontal cortex (DLPFC) or temporo-parietal cortex has been investigated in adult patients with already manifested symptoms. A Cochrane review from 2015 included a total of 41 studies investigating the use of TMS compared to sham or standard treatment for schizophrenia. Low-frequency TMS (1 Hz at 80–110% motor threshold) was applied in the majority of studies investigating the effect of temporo-parietal cortex stimulation. On the contrary, a large variety of stimulation parameters were applied in the studies investigating the effect of DLPFC stimulation (stimulation ranging from 1 to 50 Hz, at a motor threshold ranging from 10 to 110%). Collectively, some positive results on symptoms were observed, however the evidence was rated as very low due to risk of bias and imprecision between the included studies. The authors concluded, that there is currently insufficient evidence to either support or refute the use of TMS in schizophrenia [
]. To move forward there is a need for further studies based on standardized stimulation protocols. The use of tDCS has also been investigated in adults with schizophrenia. A systematic review and meta-analysis from 2020 included 16 RCT, investigating the effect of tDCS on symptoms in schizophrenia. The same areas have been targeted as with TMS, thus including the temporo-parietal cortex and the DLPFC. In all the included studies, stimulation was applied at 2 mA for 20 min per session. The anode was mainly positioned at the DLPFC, and the cathode was positioned mostly at the temporoparietal junction. Results show that positive – and negative symptoms as well as auditory hallucination improved following stimulation [
]. Recently Hansbauer and colleagues (2020) reviewed studies concerning the use of tDCS and repetitive TMS for the treatment of catatonia – a syndrome observed in schizophrenia. Results were based on case reports and stimulation was applied to the DLPFC. The review concludes that both tDCS and rTMS may be beneficial in patients with catatonia that do not respond to benzodiazepines [
]. An overview of the investigated targets by means of TMS and tDCS in clinical settings can be found in Fig. 1.
Preclinical trials
The encouraging clinical results of antipsychotics leading to a potential delay in psychosis, has also been tested in the MIA model. Here, the administration of haloperidol, clozapine, and risperidone as a preventive measure, led to a reduction in later psychosis-related behavior and anatomical abnormalities [
Risperidone administered during asymptomatic period of adolescence prevents the emergence of brain structural pathology and behavioral abnormalities in an animal model of schizophrenia.
]. The MIA model is currently the only neurodevelopmental animal model in which also preventive neuromodulation techniques have been tested. In the study by Hadar et al. (2017), chronic HFS-DBS was applied to the mPFC in the MIA model during adolescence (stimulation parameters applied: 90 μs pulse width, 130 Hz; constant current at 150 μA), followed by neurobiological and behavioral testing in adulthood [
]. The mPFC was targeted based on the finding that abnormalities within this area constitutes an early impairment in the MIA model and due to its prominent role in schizophrenia pathology [
]. Results showed that chronic stimulation applied to the mPFC during adolescence reduced the emergence of several behavioral abnormalities including deficits in both PPI, executive functions and attentional control. On a neurobiological level, chronic stimulation prevented the development of later enlarged lateral ventricles as well as reduced excessive dopamine content in the GP. Chronic HF-DBS applied to both the mPFC and Nacc in adolescence, using the same stimulation parameters as Hadar et al. (2017), has furthermore shown to prevent the abnormal microglia properties in the adult MIA offspring [
]. These findings collectively show, that early neuromodulation to pathology relevant areas is capable of diminishing schizophrenia progression on both a behavioral and neurobiological level in the MIA model. A direct translation of these findings into the clinic is obviously not feasible, however these studies still bear clinical and scientific value as they point towards the possibility of diminishing neurobiological progression by the use of targeted neuromodulation techniques in adolescents.
The positive findings following early DBS application, led to further investigation into the application of tDCS as a preventive measure in the MIA model. Here, stimulation was applied to the PFC in adolescents, followed by neurobiological and behavioral assessment in adulthood. Either anodal or cathodal stimulation was applied (intensity at 50 μA, with a current density of 14.3 A/m2) for 20 min, twice a day. Results showed that application of tDCS prevented deficits in PPI and discrimination reversal testing. On a neuropathological level, tDCS prevented both enhanced mesolimbic dopaminergic neurotransmissions as observed by amphetamine-induced activity as well as the emergence of enlarged lateral ventricles. The application of tDCS did not improve social interaction deficits or anhedonia and tDCS did not affect the reduced number of parvalbumin-expressing cells otherwise observed in the MIA model. These results showed that the application of non-invasive stimulation to the PFC is indeed capable of preventing later development of positive and cognitive symptoms alongside certain associated neuropathological findings in the MIA model [
]. As a note of caution, results also showed that stimulating the PFC by use of tDCS, led to later PPI deficits in the adult control rats. These results point towards the necessity of correctly identifying patients at risk and thereby the patients that may benefit from treatment. An overview of the investigated targets by means of tDCS in preclinical settings, can be found in Fig. 1. Further studies, in which other neurodevelopmental animal models are used to investigate preventive measures, are warranted. The use of TMS as a preventive measure has so far not been tested in preclinical settings.
Collectively, preclinical investigations show that early direct stimulation of the cortex has the potential to decrease disease progression. Such preventive measures remain to be investigated in clinical settings. Non-invasive stimulation of the cortex in adult patients is currently inconclusive. Given the early pathological involvement of especially the PFC, it remains to be investigated whether cortical stimulation bears a larger therapeutic potential in patients, when applied early in the disease, as opposed to being applied when the disorder is fully manifested. Also, due to the branched connections between the PFC and relevant subcortical areas, it remains to be investigated whether early stimulation to the PFC may prevent disease progression, by means of an early regulation of downstream targets that play a larger role in the later pathology of schizophrenia.
Conclusion
The development of schizophrenia involves progressive changes within several interconnected brain areas, which eventually leads to symptom outbreak. Schizophrenia is a neuronal circuit disorder and therefore neuromodulation techniques can be applied. This allows for a treatment approach with larger spatial precision as compared to conventional antipsychotics. So far, both clinical and preclinical findings show, that DBS indeed is capable of reducing existing symptoms, when relevant brain areas are targeted. However, to move forward, further insight into the appropriate brain areas is needed. Here, the use of neurodevelopmental models provides a platform to study how neuropathological processes and symptoms progress over time, thereby highlighting which areas should be targeted at a given time point. The invasive nature of DBS bears obvious ethical, methodological and clinical limitations when it comes to applying such techniques as preventive measures in patients. As an alternative, the use of non-invasive neuromodulation may be more appropriate, as they are generally well tolerated and hold a considerable safety profile. Here, the PFC constitutes an optimal brain target, due to its early and lasting role in the schizophrenic profile, as well as due to its branched connection with several other brain areas involved in schizophrenia. There is currently insufficient data supporting the use of cortical non-invasive stimulation in adult patients with already manifested symptoms. The use of non-invasive cortical neuromodulation as a preventive treatment in adolescents has been assessed in a preclinical study, with results showing a prevention of both behavioral and neuropathological progression. Further insight into biomarkers identifying patients at risk and the subsequent use of non-invasive neuromodulation as a preventive measure is mandatory. However, if current preclinical findings are eventually translated into the clinic, this offers the potential of a non-invasive treatment that targets the evolving neuropathology and thereby relieves patients that otherwise may end up substantially burden by symptoms later in life.
Declaration of competing interest
The authors have no conflicts of interest to disclose.
Acknowledgments
The contributions of authors have partially been supported by grants from the German Research Foundation (DFG), Germany (HA 8469/2-1, WI 2140/4–1, WI 2140/5–1) and the German Federal Ministry of Education and Research (BMBF), Germany (01EE1406A, 01EE1403A).
References
Cloutier M.
Aigbogun M.S.
Guerin A.
Nitulescu R.
Ramanakumar A.V.
Kamat S.A.
et al.
The economic burden of schizophrenia in the United States in 2013.
World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of schizophrenia, Part 1: acute treatment of schizophrenia.
The role of maternal immune activation in altering the neurodevelopmental trajectories of offspring: a translational review of neuroimaging studies with implications for autism spectrum disorder and schizophrenia.
Disruption of neurogenesis on gestational day 17 in the rat causes behavioral changes relevant to positive and negative schizophrenia symptoms and alters amphetamine-induced dopamine release in nucleus accumbens.
Neurobiology of disease A loss of parvalbumin-containing interneurons is associated with diminished oscillatory activity in an animal model of schizophrenia.
Decrease in parvalbumin-expressing neurons in the hippocampus and increased phencyclidine-induced locomotor activity in the rat methylazoxymethanol (MAM) model of schizophrenia.
Immune activation during pregnancy in rats leads to a PostPubertal emergence of disrupted latent inhibition, dopaminergic hyperfunction, and altered limbic Morphology in the offspring: a novel neurodevelopmental model of schizophrenia.
Anti-anhedonic effect of deep brain stimulation of the prefrontal cortex and the dopaminergic reward system in a genetic rat model of depression: an intracranial self-stimulation paradigm study.
Deep brain stimulation of the ventral hippocampus restores deficits in processing of auditory evoked potentials in a rodent developmental disruption model of schizophrenia.
Omega-3 fatty acid supplementation changes intracellular phospholipase A 2 activity and membrane fatty acid profiles in individuals at ultra-high risk for psychosis.
Effects of omega-3 PUFA on immune markers in adolescent individuals at ultra-high risk for psychosis – results of the randomized controlled Vienna omega-3 study.
Risperidone administered during asymptomatic period of adolescence prevents the emergence of brain structural pathology and behavioral abnormalities in an animal model of schizophrenia.
00069-3&id=gr1.jpg){kind=link}