NEW YORK – Rare variants contributing to schizophrenia appear to be overrepresented in severe schizophrenia cases and in other psychotic conditions with cognitive symptoms, according to new research from a team at Massachusetts General Hospital and Harvard Medical School.
As part of the Finland-based SUPER psychosis study, the researchers set out to more fully characterize the phenotypes that coincide with common or rare variants associated with risk of schizophrenia and other psychiatric conditions. For this, they analyzed data for 11,367 Finnish control individuals and 10,417 Finnish participants with schizophrenia, bipolar disorder, psychotic depression, or other schizoaffective or psychotic conditions. Nearly 5,800 of the cases were individuals with schizophrenia, and more than 1,600 had a bipolar disorder diagnosis.
"The SUPER psychosis study is a major nationwide effort to gather both genetic and phenotypic information," said Tarjinder Singh, an instructor at MGH, who noted that five Finnish university hospital districts are involved in the effort.
In an analysis that Singh presented at the virtual American Society of Human Genetics annual meeting on Friday, for example, the team focused on rare variants, looking at the risk factors that were present or absent in psychosis patients with shared clinical features.
"Given patients that carry rare variants, do they present themselves in particular ways? Are there particular phenotypic dimensions that are particularly associated? And within individual gene clusters and individuals who carry mutations in a single gene, do we see also particular patterns?" Singh said.
Many common variants and rarer copy number variants or predicted protein-truncating mutations have been linked to schizophrenia and related conditions in studies over the past decade, he and his co-authors noted in an abstract accompanying the ASHG presentation.
Still, they argued, "the degree to which clinical presentation is influenced by different types of genetic risk has not been fully explored, with such efforts hampered by limited data in current studies and exacerbated by the imprecision of diagnostic criteria across disorders."
Along with array-based genotypes for most of the cases and controls, the team used 18,237 exomes and 6,072 whole-genome sequences — including control sequences generated for a handful of past population studies — to look for rare variants that lined up with specific disease features.
That was possible because patients enrolled in the SUPER psychosis study complete a set of standardized psychiatric interviews and questionnaires, Singh explained, adding to the data that could be identified based on individuals' prescription history, primary care, hospital visits, and the like.
The investigators also had access to information on participants' other diagnoses, ranging from hypertension, heart disease, and diabetes to learning difficulties or delays.
As in past studies, the team found that cases carried an overrepresentation of rare protein-truncating loss-of-function mutations in "constrained" genes that tend to be relatively mutation-resistant in control individuals, particularly in genes previously linked to schizophrenia risk.
Such associations were particularly pronounced in individuals with schizophrenia, though they also turned up in individuals diagnosed with bipolar disorder, psychotic depression, and other forms of psychosis, Singh reported.
Likewise, genes identified in past studies of schizophrenia appeared prone to rare protein-truncating or missense alterations across the conditions considered in the SUPER psychosis study, though the effect was especially pronounced in the schizophrenia cases, in individuals with severe schizophrenia (based on the use of a drug prescribed for treatment-resistant schizophrenia), those with cognitive delays or disability pension reliance, and patients with younger disease diagnoses.
The investigators also explored the psychotic phenotypes that accompanied protein-truncating mutations in specific risk genes, distinguishing between genes contributing across shared psychiatric conditions and those specific to cases marked by cognitive or intellectual delay features.
"Not only, through these studies, can we learn global patterns of enrichment and associations with phenotypes, but we're beginning to see differences between gene carriers in a way that we couldn't do before," Singh said. "This gives us much more information to annotate and understand each individual [risk] gene."
More broadly, these and other findings so far "show the immense value of standardized and comprehensive phenotyping when building new collections for advancing gene discovery," the authors suggested, "especially when studying highly heterogeneous syndromes and disorders."