From pharmacokinetic genes to polygenic scores and multi-omic integration: advances toward precision psychiatry

Pharmacological treatment is the mainstay in the acute and long-term management of severe mental disorders such as major depressive disorder, schizophrenia, and bipolar disorder. However, there is large interindividual variability in clinical response, with around one-third of patients presenting treatment-resistance.

This review provides a comprehensive overview of genes that modulate the efficacy or safety of antidepressants, antipsychotics, or mood stabilizers based on a high or moderate level of evidence and for which clinical recommendations are available. Next, we highlight novel methodological and analytical approaches such as polygenic scores, pleiotropic analysis and the analysis of multiomic data with machine learning methods that might allow to explain a larger proportion of genetically driven interindividual variability in clinical response to psychotropic medications.

To date, a high level of evidence is only available for metabolizer phenotypes of a limited number of pharmacokinetic genes for antidepressants and antipsychotics (CYP2D6, CYP2C19, and CYP2B6), and selected HLA alleles for the mood stabilizer carbamazepine. However, transdiagnostic polygenic scores as well as machine learning models based on the integration of clinical determinants with multiomic data represent a promising strategy to move us closer to precision psychiatry.