Role of Polygenic Risk Score in the Familial Transmission of Bipolar Disorder in Youth

Importance: Establishing genetic contributions to the transmission of bipolar disorder (BD) from parents to offspring may inform the risk of developing this disorder and further serve to validate BD in youth. Objective: To evaluate the specific association of BD polygenic risk scores (PRSs) on the familial transmission and validity of pediatric BD. Design, Setting, and Participants: This community-based case-control longitudinal study (Pittsburgh Biological Offspring Study) included parents with BD I/II and their offspring and parents without BD (healthy or non-BD psychopathology) and their offspring. Participants were recruited between March 2001 and May 2007, and analysis took place from December 2020 to September 2021. Exposures: PRSs for BD, major depressive disorder, schizophrenia, and attention-deficit/hyperactivity disorder. Main Outcomes and Measures: Participants were prospectively evaluated using standardized interviews blind to parental diagnosis. DNA was extracted from saliva and genotyped. PRSs were constructed based on independent large-scale genome-wide association studies. Results: A total of 156 parents with BD I/II and 180 parents without BD (mean SD age, 39.6 7.9 years; 241 female 72%) as well as 251 offspring of parents with BD and 158 offspring of parents without BD (mean SD age, 10.4 4.7 years; 213 female 52%) of European ancestry were analyzed. Participants were assessed a mean of 6.7 times during a mean (SD) of 13 (3.4) years of follow-up (84% retention). More offspring of parents with BD developed BD (58 23.1% vs 8 5.1%; P < .001) and depression (126 50.2% vs 52 32.9%; P < .001) compared with offspring of parents without BD. BD PRS was higher in both parents and offspring with BD than parents and offspring without BD (parents: odds ratio, 1.50; 95% CI, 1.19-1.89; P < .001; explained 4.8% of the phenotypic variance vs offspring: hazard ratio, 1.34; 95% CI, 1.03-1.7; P = .02; explained 5.0% of the phenotypic variance). BD PRS did not differ across BD subtypes. In a model combining parental and offspring BD PRS, the parental BD PRS association with offspring BD was fully mediated by offspring BD PRS (hazard ratio, 1.40; 95% CI, 1.05-1.86; P = .02). Parental BD had a stronger direct association than parental or offspring BD PRS with offspring BD risk (hazard ratio, 5.21; 95% CI, 1.86-14.62; P = .002), explaining 30% of the variance. Parental and offspring BD PRS explained 6% of the BD onset variance beyond parental diagnosis. There were no significant between-group differences in PRSs for major depressive disorder, schizophrenia, and attention-deficit/hyperactivity disorder in parents or offspring and they were not significantly associated with BD onset. Conclusions and Relevance: The findings of this study add to the extant clinical validation of BD in youth. Parental BD and offspring BD PRS independently associated with the risk of BD in offspring. Although this is promising, the association of BD PRS was relatively small and cannot be used alone to determine BD risk until further developments occur.