Review articleThe use of polygenic risk scores to identify phenotypes associated with genetic risk of bipolar disorder and depression: A systematic review
Introduction
Mood disorders (major depressive disorder (MDD) and bipolar disorder (BD)) are common, highly heritable psychiatric conditions as evidenced by twin, adoption and family studies (Bienvenu et al., 2011, Oswald et al., 2003, Shih et al., 2004, Sullivan et al., 2000). Family history of a mood disorder is a strong risk factor for future development of a mood disorder. High-risk studies of offspring of parents with a mood disorder have shown that offspring have an increased lifetime risk for BD (Duffy et al., 2014) or MDD when compared to controls (Rice et al., 2002). Additionally, they are also at increased risk for developing other psychiatric disorders (Rasic et al., 2014). Previously, information about how genetic risk for these disorders was manifest in the population was obtained from high-risk studies that followed up offspring of parents with these disorders. However, a major limitation of these high-risk studies was the relatively small sample sizes, which meant studies were inadequately powered to detect small effect sizes.
The advent of genome-wide association studies (GWAS) has revolutionised identification of genetic variants contributing to psychiatric disorders. GWAS technology can examine many genetic variants in the genome simultaneously, without an a priori hypotheses. Approaches have been developed to harness this information, allowing us to study how individuals with different burdens of genetic risk differ from one another. Through GWAS, several risk variants associated with MDD and BD have been identified (Major Depressive Disorder Working Group of the Psychiatric Genetics Consortium et al., 2017, Sklar et al., 2011, Stahl et al., 2017, Sullivan, 2013). Though the number of risk variants is fewer than for schizophrenia (SZ) (Purcell et al., 2009, Ripke et al., 2014), this is likely due to the smaller discovery samples used in the GWAS analyses, and the lower heritability of MDD compared to schizophrenia (SZ). Whilst individual single nucleotide polymorphisms (SNPs) have very small effect on disease risk, summing the weighted allelic dosage across all SNPs, creating a single polygenic risk score (PRS) has enabled the exploration of how genetic risk is manifest directly in individuals across different populations (Fig. 1) (Wray et al., 2014).
Since the first GWAS were published, many studies have examined whether the MDD-PRS or BD-PRS are associated with a range of different phenotypes and outcomes. Though other methods such as linkage disequilibrium (LD) score regression or genomic residual maximum likelihood (GREML) analysis exist to examine genetic correlation, the focus of this systematic review was to identify and summarise studies that used the PRS approach to examine how genetic risk for MDD and BD is manifest in clinical and population-based samples.
Section snippets
Methods
We undertook a systematic review following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al., 2009) (Supplementary table 1).
Results
This review includes 25 articles that examined associations between the MDD-PRS/BD-PRS and a measurable phenotype (see Table 1 for a summary of associations with broad phenotypic outcomes). Most studies derived their PRS from the same discovery sample: the first GWAS from the Psychiatric Genetics Consortium (PGC) for MDD and BD (Sklar et al., 2011, Sullivan, 2013). Individual studies used different p-value thresholds (PTs) to assess the relationship between genetic risk for the disorder and
Discussion
To the best of our knowledge, this is the first paper to systematically review how genetic risk for MDD and BD is associated with a broad range of phenotypic outcomes. Other reviews have focussed on the application of PRS methodology to psychiatric disorders, particularly SZ (Wray et al., 2014). More recently we have reported phenotypes associated with genetic risk for SZ (Mistry et al., 2017).
Higher MDD and BD polygenic risk scores were associated with increased risk of different
Acknowledgements
We would like to thank the Cardiff University librarians who assisted with our search strategy.
Funding
SM is funded by Mental Health Research UK. DJS is a Lister Institute Prize Fellow (2016–2021). JRH is funded by the Wellcome Trust GW4 Clinical Academic Training Fellowship. This study was supported by the NIHR Biomedical Research Centre at University Hospitals Bristol NHS Foundation Trust and the University of Bristol. The views expressed in this publication are those of the authors and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health.
Role of funding source
SM is funded by Mental Health Research UK. DJS is a Lister Institute Prize Fellow (2016–2021). JRH is funded by the Wellcome Trust GW4 Clinical Academic Training Fellowship. SZ is supported by the NIHR Bristol Biomedical Research Centre. The views expressed in this paper are those of the authors and not necessarily the NIHR or any other funders.
References (67)
- et al.
Polygenic risk scores in bipolar disorder subgroups
J. Affect. Disord.
(2015) A continuity between bipolar II depression and major depressive disorder?
Progress. Neuro-Psychopharmacol. Biol. Psychiatry
(2006)- et al.
Investigating the genetic variation underlying episodicity in major depressive disorder: suggestive evidence for a bipolar contribution
J. Affect. Disord.
(2014) - et al.
Pharmacogenetics of antidepressant response: a polygenic approach
Progress. Neuro-Psychopharmacol. Biol. Psychiatry
(2017) - et al.
Reduced educational attainment in bipolar disorder
J. Affect. Disord.
(2006) - et al.
Neurocognitive functioning in the premorbid stage and in the first episode of bipolar disorder: a systematic review
Psychiatry Res.
(2015) - et al.
A critical overview of the clinical evidence supporting the concept of neuroprogression in bipolar disorder
Psychiatry Res.
(2016) - et al.
Heterogeneity in cognitive functioning among patients with bipolar disorder
J. Affect. Disord.
(2008) - et al.
Genomic predictors of combat stress vulnerability and resilience in U.S. Marines: a genome-wide association study across multiple ancestries implicates PRTFDC1 as a potential PTSD gene
Psychoneuroendocrinology
(2015) - et al.
Differentiation of bipolar I and II disorders by examining for differences in severity of manic/hypomanic symptoms and the presence or absence of psychosis during that phase
J. Affect. Disord.
(2013)