We thank Sudhakar Selvaraj and colleagues1 for their interest in our work. They offer several critiques of the relevance of PET occupancy data to withdrawal effects. However, we think their claim that changes in plasma levels of drug would not be associated with withdrawal symptoms has little evidential support. The more abrupt reduction in plasma levels of antidepressants with shorter half-lives is understood to cause their more severe and quick-onset withdrawal symptoms.2 Selvaraj and colleagues state that a study by Michelson and colleagues3 shows no correlation between change in blood concentration and withdrawal symptoms for any individual SSRI; however, they also found “percentage reduction in plasma concentrations across drug groups was statistically significantly correlated with new adverse events.”3 This suggests that the study was underpowered to find an effect for individual agents—an effect which became clear when the three SSRIs were grouped together.3 Additionally, reduced brain receptor exposure to antidepressants is likely to be the key factor in withdrawal symptoms, and plasma concentration might not accurately reflect brain exposure because of a number of individual pharmacokinetic effects, another limitation to this analysis.3