Pathway-Specific Polygenic Scores for Predicting Clinical Lithium Treatment Response in Patients with Bipolar Disorder

Nigussie Tadesse Sharew; Scott Richard Clark; Sergi Papiol; Urs Heilbronner; Franziska Degenhardt; Janice M. Fullerton; Liping Hou; Tatyana Shekhtman; Mazda Adli; Nirmala Akula; Kazufumi Akiyama; Raffaella Ardau; Bárbara Arias; Roland Hasler; Hélène Richard-Lepouriel; Nader Ali Perroud; Lena Backlund; Abesh Kumar Bhattacharjee; Frank Bellivier; Antonio Benabarre; Susanne Bengesser; Joanna M. Biernacka; Armin Birner; Cynthia Marie-claire A; Pablo Cervantes; Hsi‐Chung Chen; Caterina Chillotti; Sven Cichon; Cristiana Cruceanu; Piotr M. Czerski

doi:10.1016/j.bpsgos.2025.100558

Genetic & Genomic MedicineBiological Psychiatry Global Open ScienceJun 25, 2025

Peer-reviewed journal article

Pathway-Specific Polygenic Scores for Predicting Clinical Lithium Treatment Response in Patients with Bipolar Disorder

Nigussie Tadesse Sharew iD, Scott Richard Clark iD, Sergi Papiol iD, Urs Heilbronner iD, Franziska Degenhardt iD, Janice M. Fullerton iD, Liping Hou iD, Tatyana Shekhtman iD, Mazda Adli iD, Nirmala Akula iD, Kazufumi Akiyama iD, Raffaella Ardau iD, Bárbara Arias iD, Roland Hasler iD, Hélène Richard-Lepouriel iD, Nader Ali Perroud iD, Lena Backlund iD, Abesh Kumar Bhattacharjee, Frank Bellivier iD, Antonio Benabarre iD, Susanne Bengesser iD, Joanna M. Biernacka iD, Armin Birner iD, Cynthia Marie-claire A iD, Pablo Cervantes iD, , Layla Kassem

Abstract

BackgroundPolygenic scores (PGSs) hold the potential to identify patients who respond favourably to specific psychiatric treatments. However, their biological interpretations remain unclear. In this study, we developed pathway-specific PGSs (PSPGS) for lithium response and assessed their association with clinical lithium response in patients with bipolar disorder (BD). MethodsUsing sets of genes involved in pathways affected by lithium, we developed nine PSPGSs and evaluated their associations with lithium response in the International Consortium on Lithium Genetics cohort (ConLi+Gen: N = 2367), validated in the combined PsyCourse (N = 105) and BipoLife (N = 102) cohorts. Lithium responsiveness was assessed using the Retrospective Assessment of the Lithium Response Phenotype Scale (ALDA scale), for categorical outcome (good vs poor response) and continuous ALDA total score. Logistic and linear regressions, adjusting for age, sex, chip type, and the first four genetic principal components, were used to test associations, after multiple testing corrections (p<0.05). ResultsResponse to lithium was associated with PSPGS for acetylcholine, GABA, calcium channel signalling, mitochondria, circadian rhythm, and GSK pathways, R{superscript 2} ranging from 0.29% to 1.91%, with R{superscript 2} of 3.71% for the combined PSPGS. Associations for GABAPGS and CIRPGS were replicated. In decile-based stratified analysis, patients with the highest genetic loading (10th decile) for acetylcholine pathway genetic variants were 3.03 times (95%CI: 1.95 - 4.69) more likely to have a good lithium response than the lowest decile (1st decile). ConclusionPSPGSs achieved predictive performance comparable with conventional genome-wide PGSs, with more biological interpretability and using a smaller list of genetic variants, facilitating further investigation into the interaction of variants and biological pathways underlying lithium response.

PDF DOI