Response to lithium in patients with bipolar disorder is associated with clinical and transdiagnostic genetic factors. The predictive combination of these variables might help clinicians better predict which patients will respond to lithium treatment.

To use a combination of transdiagnostic genetic and clinical factors to predict lithium response in patients with bipolar disorder.

This study utilised genetic and clinical data (n = 1034) collected as part of the International Consortium on Lithium Genetics (ConLi+Gen) project. Polygenic risk scores (PRS) were computed for schizophrenia and major depressive disorder, and then combined with clinical variables using a cross-validated machine-learning regression approach. Unimodal, multimodal and genetically stratified models were trained and validated using ridge, elastic net and random forest regression on 692 patients with bipolar disorder from ten study sites using leave-site-out cross-validation. All models were then tested on an independent test set of 342 patients. The best performing models were then tested in a classification framework.

The best performing linear model explained 5.1% (P = 0.0001) of variance in lithium response and was composed of clinical variables, PRS variables and interaction terms between them. The best performing non-linear model used only clinical variables and explained 8.1% (P = 0.0001) of variance in lithium response. A priori genomic stratification improved non-linear model performance to 13.7% (P = 0.0001) and improved the binary classification of lithium response. This model stratified patients based on their meta-polygenic loadings for major depressive disorder and schizophrenia and was then trained using clinical data.

Using PRS to first stratify patients genetically and then train machine-learning models with clinical predictors led to large improvements in lithium response prediction. When used with other PRS and biological markers in the future this approach may help inform which patients are most likely to respond to lithium treatment.

Our aim was to investigate bipolar patients in order to test the validity of various outcome measures and to identify prognostic predictors for pharmacological treatment.

One hundred patients were interviewed using a computerized life-charting program in a descriptive, retrospective analysis. The concept “Burden of illness” was defined as a combination of severity and duration of episodes. Response to treatment was defined as the difference in burden before and after treatment, a low burden during treatment, and freedom of episodes for at least 3 years after insertion of treatment.

The absence of mixed episodes and a high initial burden predicted a good response measured as the difference in burden. If remission for 3 years or a low burden during lithium treatment was used, the absence of rapid cycling and of mixed episodes were the most important predictors. The severity of illness before treatment had no impact.

We suggest the use of absolute measures of severity during treatment as the most appropriate measure of the outcome. Furthermore, our data provide corroboration that treatment with lithium ameliorates the prognosis of the illness, but that mixed episodes and rapid cycling predict a poorer response to lithium.