Neuroprogression in schizophrenia: Pathways and underpinning clinical staging and therapeutic corollaries

Schizophrenia continues to be one of the most diverse yet devastating clinical conditions that psychiatrists have to manage. Despite decades of extensive research, our current armoury of treatments remains limited. So what are we up to and where do we go next? Davis et al. (2014) propose a neuroprogressive pathway in schizophrenia, an idea that has gained increasing popularity since experts first noted similarities between the life course of schizophrenia and that of other, more classic neuroprogressive disorders, such as multiple sclerosis. For example, in both conditions, earlier disease onset and increased length and frequency of episodes have been associated with an inferior response to medication and a poorer prognosis (Berk et al., 2011). Currently, one popular viewpoint is that schizophrenia is a disorder of disconnection. In other words, it is a problem with the way the nerve cells and neural circuits talk to one another, but the question remains: How does this neural machinery become dismantled and why does it fail to develop properly? The purpose of the Davis et al. (2014) paper was to synthesise the current knowledge of neuroprogressive processes that may occur in schizophrenia. In particular, it amalgamates new knowledge from genetics, imaging, immunology and neurochemistry to help us understand how schizophrenia becomes established in the brain.

Davis et al. (2014), together with others in the field, believe that the clinical syndrome of schizophrenia arises in susceptible people via an accumulation of ‘multiple hits’. These initial ‘hits’ are thought to occur in utero and in the perinatal period, and involve complex epigenetic changes, which together lead to the establishment of abnormal inflammatory responses. Specifically, they propose that these downstream epigenetic effects result in the increased expression of inflammatory markers such as tryptophan catabolites (TRYCATS) and reactive oxygen species (ROS), both of which are damaging to the developing brain. When these markers are combined with further ‘hits’, such as an under-performing N-methyl-d-aspartate (NMDA) receptor and/or methylation abnormalities (i.e. when the wrong bits of a gene are silenced), they propose that this toxic combination can cause the growth and development of the abnormal neuronal circuits seen in schizophrenia.

The dopamine hypothesis has been pivotal to schizophrenia research for many decades. However, in vivo changes in dopamine activity has only been definitely established in recent years. Modern research has been heavily influenced by the Howes and Kapur (2009) paper, ‘The dopamine hypothesis of schizophrenia: version III’, which proposed a final common pathway on the premise that dopamine remains the ‘wind behind the psychotic fire’. Howes and Kapur attempted to integrate our understanding of how genetic abnormalities, recreational drugs and social stressors all come together to cause abnormalities in the dopamine pathways. Despite this resurgence of interest in dopamine, Davis et al. (2014) re-emphasise that dopamine antagonists (i.e. the great majority of our antipsychotic repertoire) fail to control the negative and cognitive symptoms of the disorder. Thus, a great strength of this paper is that it also reviews some of the novel evidence-based treatments for schizophrenia, as well as modes of action of the existing neuroleptics that may work on the neuroprogressive mechanisms outlined.

One such treatment is the use of omega-3 polyunsaturated fatty acids (PUFAs), which Davis et al. reviewed as an adjunct therapeutic intervention. The idea of feeding the brain with essential fatty acids is becoming increasingly popular and there are promising clinical trials that PUFAs play an important role in the conversion of the prodromal state to full-blown psychosis. Amminger et al. (2010) conducted a double-blind randomised controlled trial of 81 patients and studied the effects of taking 1.2 g/day of omega-3 PUFA over 1 year. The difference between the groups in the cumulative risk of progression to overt psychosis was an impressive 22.6%. PUFAs also significantly reduced positive symptoms (p=0.01), negative symptoms (p=0.02) and, overall, they improved functioning significantly (p=0.002) compared with placebo. These results are extremely encouraging and, given our current dopamine antagonists’ failure to improve negative symptoms, PUFAs should be considered in the prodromal phase.

However, Davis et al. (2014) do not tackle the ‘multiple hits’ that occur in later life, after onset of illness. ‘Hits’ prior and proximal to the onset of illness may serve as windows for intervention both to prevent the onset of illness and to improve the prognosis in those who do become unwell. One worry of focusing on ‘hits’ early in life, is how one can intervene in a timely way and avoid therapeutic pessimism. We know that a variety of later ‘hits’ can contribute to the aetiology and progression of schizophrenia. These include use of cannabis and social adversity, such as bullying, migration, family stress, social isolation, as well as affective change, which can also play an important role. Therefore, a plausible model of the onset of psychosis, we would argue, needs to draw not only on neuroscience, but also on the insights of social psychiatry and cognitive psychology (Broome et al., 2005) and how such processes interact.

The Davis et al. (2014) paper is important: it is a timely state-of-the-art review that synthesises knowledge from a diverse range of research methodologies and proposes direct mechanisms that may underpin the disruption of neuronal architecture. The review highlights the challenges and possibilities in gaining further understanding of this disorder and finding new treatments to help our patients. But they do not attempt to answer one important question: What happens if an individual only receives a few of these ‘hits’, and not enough to develop the full clinical syndrome? Future research might want to consider how varying ‘hit’ combinations could lead to other clinical or sub-clinical phenotypes.

See Review by Davis et al., 2014, 48(6): 512–529.