Prediction of human efficacious antidepressant doses using the mouse forced swim test

Abstract

The forced swim test (FST) is a commonly used preclinical animal behavioural model for prediction of antidepressant activity in humans. While the FST may qualitatively predict efficacy, less is known about the quantitative translation of FST data to human efficacious doses. Assessing quantitative translation allows better predictions of human efficacious doses and a higher chance of success in the drug development process.Dose-response and time-course FST experiments were carried out on mice using four marketed antidepressants (citalopram, desipramine, bupropion, desvenlafaxine) in addition to ketamine, all with varying mechanisms of action. Population pharmacokinetic (PK)/pharmacodynamic (PD) analysis methods were applied to analyse the PK and immobility data, and the accuracy of the translation of FST data to human doses was evaluated using both area under the curve (AUC) and concentration-based approaches.The results showed that for the five antidepressants, average human AUC at clinically relevant doses were up to 38-fold higher than mouse AUC at doses associated with 50% of maximal efficacy in the FST (ED50). Using a concentration approach, human peak and trough drug concentrations at clinically relevant doses were generally associated with concentrations of at least 65% (EC65) and 20% (EC20) of maximal effect in mice, respectively.The FST is a useful tool to predict antidepressant efficacy across a variety of drugs with different mechanisms of actions. However, human doses can be over-or under-predicted many fold when using the traditional approach of estimating based upon ED50 AUC in mice. It is recommended that a concentration approach be used, where concentrations associated with 80% (EC80) and 30% (EC30) of maximal effect in the mouse are used as general targets for human maximum and trough concentrations, respectively, in the prediction of clinically efficacious doses of new, potential antidepressant agents.