Dissecting Bipolar Disorder Heterogeneity Through Familial Aggregation, Symptom Dimensions, and Genetic Risk

Abstract

Bipolar disorder (BD) is a highly heritable but clinically diverse condition, complicating efforts to identify its biological basis. This dissertation investigates familial patterns, symptom dimensions, and genetic risk. A systematic review and meta-analysis identified 14 clinically relevant familial traits, notably age of onset, lithium response, and psychotic features. Principal component analysis in two BD family cohorts revealed consistent latent dimensions, episode frequency and age of onset, that showed greater similarity among relatives, suggesting heritability. Finally, polygenic scores (PGS) were examined in relation to symptom dimensions. While some associations emerged (e.g., PGS for schizophrenia was inversely related to episode frequency), they did not remain significant after correction. Together, these findings support a dimensional, data-driven approach to BD, moving beyond categorical subtypes to capture clinical and genetic complexity. This work contributes to defining biologically meaningful subgroups and informing precision medicine in BD.