Association Between Polygenic Risk Scores and Treatment Response to Antidepressants, Benzodiazepines, and Antihistamines in Anxiety and Depression

Abstract

Background

Anxiety and depression are the most prevalent mental health disorders. The first-line treatment is antidepressants, such as serotonin reuptake inhibitors, but benzodiazepines and antihistamines are also used to treat anxiety. Only one-third of patients achieve remission with first-line treatment. Identifying responders and nonresponders to monotherapy prior to treatment could increase remission rates and reduce dropout. The aim of the current study was to predict response to antidepressants, benzodiazepines, and antihistamines from polygenic risk scores (PRSs) in individuals with anxiety and/or depression symptoms.

Methods

We identified 2515 individuals in a genotyped cohort in the Swedish Twin Registry who had been prescribed drugs for anxiety and/or depression. Of these individuals, 1037 received monotherapy (555 with antidepressants, 169 with benzodiazepines, and 313 with antihistamines). The remaining 1478 individuals switched or added more drugs during the assessment period (2005–2018). The accuracy of 42 PRSs for psychiatric diagnoses as well as for nonclinical phenotypes in predicting mono- versus multitherapy was assessed using logistic regression.

Results

Monotherapy with benzodiazepines was predicted by a PRS for depressive symptoms indexed by the Patient Health Questionnaire (odds ratio [OR] = 1.29), while monotherapy with antihistamines was predicted by a PRS for lifetime anxiety disorder (OR = 1.25) and a PRS for schizophrenia (OR = 1.24). None of the investigated PRSs significantly predicted monotherapy with antidepressants.

Conclusions

Real-world data suggest that monotherapy with benzodiazepines or antihistamines can be predicted from PRSs related to anxiety, depression, and schizophrenia.