Researchers identified for the first time a polygenic risk gradient spanning schizophrenia (SCZ) and bipolar disorder (BD), indexed by the level of mood-incongruent psychotic symptoms, according to a new study in JAMA Psychiatry.
“We hypothesized that BD with psychosis would be associated with higher polygenic risk for SCZ and that this association would be stronger when mood-incongruent psychotic features were present, given their phenotypic similarity to the psychotic symptoms of prototypic SCZ,” wrote primary author Judith Allardyce, MRCPsych, PhD, Medical Research Council Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, Cardiff, Wales, and coauthors.
About one-third of all psychotic features in BD—such as delusions or hallucinations—are characterized as mood incongruent and are similar to typical symptoms of SCZ. These mood-incongruent features predict poor lithium response and worsened prognosis.
The authors posit that BD with psychosis and mood-incongruent psychotic features could be designated a subgroup with stronger causal ties to SCZ. Although previous stratified linkage and candidate-gene studies examining BD associations with SCZ genetics showed more powerful effects in psychosis and mood-incongruent samples, these findings are inconsistent thus lending weak support to a “causal heterogeneity hypothesis.”
Genome-wide association studies (GWAS) support a strong polygenic component in the risk of BD and SCZ, with substantial genetic variance shared by the two illnesses. This shared genetic basis can be calculated using polygenic risk score (PRS).
Dr. Allardyce and colleagues analyzed the correlation between SCZ polygenic liability and psychotic BD presentations using PRSs gathered from a robust SCZ-GWAS discovery set. “The PRSs were constructed using alleles derived from a larger SCZ-GWAS discovery set, which reduces the measurement error and improves power from both this sample and the larger BD sample,” they wrote.
In this case-control study, researchers compared 4,436 BD cases (67% women; mean age 46 years) with the genotypic data for 4,976 schizophrenia cases and 9,012 controls. The team employed multinomial logistic regression to approximate differential PRS associations spanning case and control categories.
The researchers found that an exposure-response gradient existed along all clinical phenotypes. They found the greatest PRS association for schizophrenia (RR=1.94, 95% CI 1.86-2.01), followed by schizoaffective BD (RR=1.37, 95% CI 1.22-1.54), bipolar type I (RR=1.30, 95% CI 1.24-1.36), and bipolar type II (RR=1.04, 95% CI 0.97-1.11) in decreasing order. Furthermore, in BD cases, the effect gradient followed the nature of psychosis—with mood-incongruent psychotic features displaying the strongest correlation (RR=1.46, 95% CI 1.36-1.57) followed by mood-congruent features (RR=1.24, 95% CI 1.17-1.33), and no psychotic symptoms (RR=1.09, 95% CI 1.04-1.15).
The researchers view their results as an advance toward precision medicine in psychiatry. “These findings represent a step toward the goal of reconceptualizing phenotypic definitions using richer clinical signatures, measured across quantitative or qualitative domains, including symptom loading and biomarker expression.”
Of interest, they acknowledged that phenotypic misclassification could have been a limitation of this study, despite detailed case review, interviews, and strong inter-rater reliability.
“To our knowledge, this study is the first to show a gradient of polygenic liability across SCZ and BD, indexed by the occurrence and level of mood incongruence of positive and disorganized psychotic symptoms,” the researchers concluded.