Examining the curious overlap between motor neurone disease and schizophrenia

Neurologists and psychiatrists formally parted company in the 1800s, and neurological and psychiatric conditions are treated very differently. The work of Sigmund Freud (originally a neurologist, but also the father of psychoanalysis) has greatly influenced how we train young doctors about brain disorders. Put simply, we neurologists are trained to deal with the ‘brain’, while our colleagues in psychiatry are trained to deal with the ‘mind’, as though the two are entirely separate entities. However, as our scientific understanding of neuroscience advances, it is time to re-evaluate this dichotomy and to move towards an integration of knowledge across these closely-related, though separately regarded, disciplines. Our recently observed relationship between the neurodegenerative condition that is amyotrophic lateral sclerosis (ALS), also known as motor neurone disease (MND), and neuropsychiatric conditions such as schizophrenia, is a good example of how neurological and psychiatric conditions can overlap at a biological level and how combined analyses can provide greater insights into both conditions.

<h3><strong>MND</strong></h3>

At present, we teach that MND is a ‘pure’ neurological condition, characterised by combined upper and lower motor neurone degeneration with progression to death within three years. The peak age of onset is in the early 60s. The incidence is around 2.6/100,000 in Ireland, with approximately 130 new diagnoses every year. The prevalence in Ireland is eight/100,000, or approximately 350 living cases. Based on data from the Irish MND Register, we estimate that the overall lifetime risk in Ireland for developing MND is approximately 1:300. Diagnosis of MND is clinical and requires the presence of both upper and lower motor neurone signs, with exclusion of other conditions that could mimic the clinical presentation. Neuroimaging is performed to exclude structural pathology, and tests for other conditions such as multiple sclerosis and neurophysiological studies are performed to confirm the presence of acute and chronic denervation, the absence of sensory findings, and to ensure that mimic syndromes such as multifocal motor neuropathy are excluded.

<h3><strong>Schizophrenia </strong></h3>

By contrast, schizophrenia is a typical psychiatric condition that generally affects young people and is characterised by a combination of both positive and negative symptoms that include auditory hallucinations, thought disorder and blunted affect, respectively. Management is by psychopharmacology with additional psychological support, such that many people with schizophrenia can live high-quality, independent lives.

And yet there is much that remains to be understood about this condition. Schizophrenia is known to be polygenic in inheritance. Some patients with schizophrenia exhibit executive impairment and reduced social cognition and there is a higher rate of metabolic syndrome in later years. While neurologists are not involved in the care of patients with ‘typical schizophrenia’, patients with later-onset symptoms fulfilling the criteria for psychosis and associated with neurodegeneration (eg, those with dementia associated with Parkinson’s disease and those with Huntington’s disease (HD)) often come under the care of neurologists because they have ‘organic’ brain disease with psychiatric symptomatology.

<h3><strong>Interface</strong></h3>

Our group has recently begun to explore the interface between neurodegenerative disease (MND) and neuropsychiatric disease by undertaking family aggregation studies of kindreds in which the proband has MND, by comparing genome-wide association studies of DNA from MND patients and controls and also by performing whole-genome sequencing of patients with MND. We have also sought to measure changes in structural and dynamic brain function by harnessing modern imaging and signal analysis technologies, such as MRI and spectral EEG.

First, the family aggregation studies. We know that at least 10-to-20 per cent of patients with MND have a family history of MND, and that up to 30 genes of major effect have been associated with the condition. The most common gene variant associated with MND is a repeat expansion in C9orf72, which accounts for approximately 10 per cent of all patients with MND in Ireland. C9orf72 is also associated with frontotemporal dementia, and has been described in patients with features of HD who are negative for the HD expansion, and some kindreds with bipolar affective disorder and others with obsessive compulsive disorder.

In 2013, we published a study in <em>Annals of Neurology</em>, in which we showed that kindreds of patients with ALS are also more likely than controls to have extended family members with schizophrenia/psychosis and to have a close family member who had committed suicide. In our study of over 12,000 relatives from 400 families, we found that people with MND are nine times more likely expected to have other family members with schizophrenia and 16 times more likely to have had another family member who had taken their own life. Because families share the same DNA, the higher rates of psychiatric disease in family members of people with MND suggested that MND and schizophrenia might share genetic risk. With our colleagues from psychiatry, we tested DNA samples from Irish patients with schizophrenia and did not find evidence of the C9orf72 gene expansion within this group, suggesting that the overlap we had identified in Irish MND kindreds could not be explained by the presence of the C9orf72 expansion.

In 2017, we published a second study of MND families in <em>JAMA Neurology</em> and showed that the kindreds of patients with MND were also more likely to exhibit other neuropsychiatric traits such as impulse dyscontrol, obsessive compulsive disorder, autism spectrum disorder and bipolar affective disorder, than were control kindreds. In this study, we found that up to 30 per cent of people with MND had three or more family members with a neuropsychiatric condition, which we consider to be an ‘endophenotype’ of MND. The remaining families did not have higher rates of neuropsychiatric illness compared to controls. Again, this overlap was not caused by the presence of the C9orf72 repeat expansion.

<h3><strong>Heterogeneous</strong><strong></strong></h3>

Our observation of clustering of MND and neuropsychiatric conditions within kindreds was important for two reasons. Firstly, this provides evidence that MND is heterogeneous. Our failure to recognise this heterogeneity may be a significant factor in the limited success of the development of new therapeutics, as our clinical trials to date have not accounted for different patterns of disease pathogenesis. Secondly, the presence of a family history of neuropsychiatric disorders is likely to reflect a subtype of MND with a specific set of genetic susceptibilities that are shared between MND and psychiatric conditions. Our study suggested the possibility of genetic pleiotropy (a gene variant associated with a wide range of different phenotypes), or of oligogenic inheritance (a number of different gene variants that are inherited together with a higher frequency than that expected by chance). Either way, the work demonstrates that the presence of a strong family history of neuropsychiatric disease should trigger a detailed study of family genomics within these kindreds — a project that is now underway in our group in association with the international Project MinE consortium (www.projectminE.com).

To further test the overall hypothesis that MND and neuropsychiatric conditions might be biologically linked at a genetic level, we also combined genome-wide association (GWAS) data from our international MND Consortium with data from the Psychiatric Genome Consortium and we published our findings in <em>Nature Communications</em>. In total, we analysed GWAS data from 13,000 MND patients and over 30,000 schizophrenia patients. Our combined analysis has shown a 14 per cent polygenic overlap in the genetic basis of schizophrenia and MND, suggesting that these two very different conditions — one of which is a disorder of the ‘mind’ and exclusively ‘psychiatric’, and the other a ‘brain disorder’ and exclusively neurological — are also biologically linked at a polygenic level, suggesting the presence of shared pathways of disruption within the brain. As our data fully confirmed, our hypothesis that there is a genetic link between MND and neuropsychiatric conditions was correct. We must now seek to understand the biological processes that underpin this overlap of seemingly very different conditions.

<h3><strong>Neural networks</strong><strong></strong></h3>

We have proposed that these conditions all represent disorders of neural networking. There is already an extensive literature demonstrating that brain networking is disrupted in schizophrenia. However, the concept of MND as a ‘network disorder’ is less well established, although the new technologies of imaging and signal analysis have changed our perspective. We already know that MND is not just a disorder of motor neurones. Our neuropsychological studies in MND have demonstrated that up to 50 per cent of patients develop significant changes in executive function, social cognition and behaviour and that a small proportion can become psychotic. Our recent experimental studies using MRI have shown extensive extra motor involvement in MND and our recently-published studies in the journal <em>Cerebral Cortex</em> of brain connectivity using spectral EEG have shown areas of both increased and decreased brain connectivity in MND. These changes are in regions of the brain that correlates with structural changes on MRI and corresponds with neuropsychological deficits.

A clear implication of this work is that the development of drugs that focus exclusively on motor neurones may not be the right approach for human MND, and that there is a biological, plausible rationale for the development of drugs that target changes in brain networks. It could also be the case that some drugs used within the neuropsychiatric domain could hold promise for conditions like MND.

<h3><strong>Conclusion</strong></h3>

This link between MND and neuropsychiatric diseases, such as schizophrenia, shows that the divide between psychiatry and neurology is artificial. We need to recognise that brain disease has many different manifestations, and the best way to develop new treatments is to understand the biological factors that are driving this overlap. This will influence how we think about diseases going forward, and in turn how we train our future doctors in both psychiatry and neurology.

<strong>References on request</strong>