Schizophrenia wrecks the lives of millions worldwide – and has defeated researchers looking for a single cause. Time for complex new thinking.

PAUL is 21. He thinks the voices started a couple of years ago, but it’s hard to remember exactly because they just seemed to fade in. They whisper insistently, commenting on his actions, trying to control his thoughts and feelings. Living with them is a constant battle, causing him to drop out of college and stop seeing friends. He has been treated in hospital and is being prescribed antipsychotic drugs, but he sees all this as part of a conspiracy.

Paul’s world view is informed by psychosis. This mental state disrupts perception and the interpretation of reality, and is characterised by hallucinations and delusions. Doctors recognise psychosis as a marker for many medical conditions ranging from those caused by electrolyte disturbance to epilepsy, dementia and rare autoimmune disorders.

In Paul’s case these conditions are rapidly excluded. After other short-lived, mood or drug-related causes are also excluded, Paul is diagnosed with schizophrenia - one of a group of disorders characterised by psychosis. But schizophrenia also affects Paul’s emotional and verbal responsiveness, motivation and insight. And it is these functional symptoms that are its most disabling features because they erode the ability to interact with others, maintain social contacts and work.

So what is schizophrenia? In the late 19th century German psychiatrist Emil Kraepelin identified the symptoms and presentation of a disease later called schizophrenia by Eugen Bleuler, a Swiss psychiatrist. Bleuler saw it as an umbrella term for a collection of diseases. Despite attempts to define subtypes or identify specific forms, schizophrenia is still treated broadly as a single disease, and it affects around 1 per cent of adults.

So a shorter, more honest answer to the question of what schizophrenia is would be that we won’t really know until we can define its neurobiological basis. For now, psychosis represents a major frontier in neuroscience because it shakes our certainties about the way we see the world - and understand the brain.

We have learned some things, though. The disease is substantially heritable, and is more common in the offspring of older men. Its occurrence increases after complications in birth, and other risk factors include childhood abuse, early teenage cannabis use and social marginalisation. We also know that people who develop schizophrenia tend to have had subtle delays in language, motor and social development in early childhood, predating the onset of symptoms by many years. This suggests the cause is neuro-developmental.

Being able to distinguish psychoses is important because it allows researchers to develop treatments for schizophrenia, including antipsychotic drugs, and makes psychological intervention possible. But despite treatment, people with schizophrenia end up with serious health problems - their average life expectancy is 20 years below the norm for people in the developed world.

One reason progress has stalled is that people with schizophrenia are a diverse group. This is not surprising since the symptoms include hallucinations, delusions, impaired thinking, disorganised behaviour, or problems with motivation - and someone may only be affected by two of these problems. Also, some 20 per cent of patients suffer one episode, while nearly a third are chronically ill, responding only partially to treatment.

Despite this variability, most researchers and doctors still see schizophrenia as a single disease. Now researchers, including my team, are challenging this assumption, using what we know about the human genome. As a “book of life”, the genome is poorly edited, littered with typographical errors, and has many deleted or duplicated pages. But this is a boon for researchers because it is from those pages that the first breakthroughs have emerged.

Over the past five years, genomic tools have shown that structural mutations (where a part of the genome is duplicated or deleted, possibly disrupting or altering the production of proteins) are more common in people with schizophrenia. Although each mutation is rare in individuals with schizophrenia, researchers have found in about 5 per cent of patients a dozen genomic regions where such variation occurs. Depending on the region, this variation increases the risk of schizophrenia between three and 20-fold. Some regions are large and include many genes, but some are small, bringing a single gene into focus.

Or so we thought. In fact, each recurrent structural mutation, called a copy number variation (CNV), increases the risk of a range of developmental disorders that include autism, learning disabilities and epilepsy. But while CNVs may contribute significantly to human disease, they are only one type of variation in the genome, so other, even more common forms of variation may also be important. As it is now possible to sequence genomes on a large scale, we can look for changes at the level of single nucleotide bases, which account for most of the genetic difference between people.

This variation ranges from nucleotide differences present in 5 to 50 per cent of the population, to uniquely individual changes. Whether the changes have an impact depends largely on whether they occur within genes or affect the gene’s function. As you might expect, the most damaging changes are the most rare. We all carry quite a few, but studies suggest people with schizophrenia carry more than average. Again, many of the critical genes are implicated in disorders such as autism.

Last month, researchers at the World Congress on Psychiatric Genetics in Hamburg, Germany, heard that the genes implicated in schizophrenia do not appear to fit into a single pathway but point to critical processes, such as the workings of the immune system, the plasticity of synapses, and the regulation of sets of genes during early brain development.

This prompted debate as to whether Bleuler was right, and whether we are starting to identify the diseases in his syndrome. If so, understanding the genetics will be key not only diagnostically but in tailoring treatments that target molecular mechanisms rather than clinical symptoms. For Paul, this could mean no longer being seen as someone with schizophrenia likely to respond to a generic treatment. Instead, he might have, say, a rare mutation requiring a specific therapy. And if Paul has a brother with depression or a niece with epilepsy sharing the mutation, they too might respond to the same treatment.

So what next? We have a growing list of rare mutations, with more likely to be found. But the healthcare impact will depend on two things: first, the proportion of people with at least one rare mutation (and estimates for schizophrenia are about 5 per cent, and more than 10 per cent for autism). Second, and more important, is the cumulative impact of each mutation on the risk of disease.

In some cases, a mutation will profoundly affect a key molecular mechanism, causing or vastly increasing the risk of schizophrenia. For most, the disease is likely to be a consequence of many inherited or new genetic mutations coupled with environmental factors such as stress, rather than the result of a single mutation. In fact, the genetic background risk is likely to involve hundreds or even thousands of more subtle genetic effects called single nucleotide polymorphisms, each making a tiny contribution to the risk.

The findings also pose big questions. Are those most severely affected the people with the most severe mutations? How many genetic mechanisms are involved? What are the best targets for therapies? Are there people with low genetic risk where the causes or triggers depend on exposure to cannabis, say, or life’s stresses? What will we learn about normal brain development and function? The answers will be important for all of us.

by Aiden Corvin, NewScientist