Poster Abstract
This poster features a collaboration with EMD Serono.
Objectives
The serotonin reuptake transporter (SERT) is responsible for the removal and recycling of the neurotransmitter serotonin from neuronal synapses and is an important pharmacological target for treating a variety of CNS disorders. However, increased levels of extrasynaptic serotonin resulting from SERT inhibition has been associated with a spectrum of adverse events (AEs) termed “serotonin syndrome”, that range in severity from mild to life-threatening. Thus, we sought to develop a model to predict the incidence proportion of tremors, a principal component of serotonin syndrome, using existing data from marketed drugs.
Methods
To quantify the incidence of serotonin syndrome across drugs that exhibit either on- or off-target SERT inhibition, we focused primarily on the tremor incidence proportion. Tremors are a characteristic manifestation of the serotonin syndrome, and the tremor incidence proportion is frequently reported for drugs inhibiting SERT. We investigated whether the incidence proportion of tremors is correlated with SERT target coverage. Since target coverage is not typically directly measured, we estimated it for various drugs and doses by computing the unbound brain exposure/SERT IC50, considering:
- in vivo data on plasma and/or brain exposure (Cavg, Cmax, and brain:plasma partitioning);
- in vitro data on potency (SERT IC50) and on nonspecific binding (fraction unbound) in plasma and brain; and
- predicted pH-dependent lysosomal partitioning in the brain
Results
Using this approach, a translational model-based meta-analysis was performed on tremor incidence data compiled for 20 serotonergic drugs, considering the brain target coverage estimated at the respective dosages. A relationship between tremor incidence and brain unbound exposure / SERT IC50 was observed and described by an Emax model. This relationship was observed considering both Cavg and Cmax. Subset analyses of serotonergic drugs based on their range of targets and selectivity to SERT (e.g., exclusion of drug classes with relatively low selectivity for SERT, such as TCAs and opioids) also revealed relationships between tremor incidences and estimated target coverage. Together, our analyses of tremor incidence data across drug classes identify the estimated brain unbound exposure / SERT IC50 to be a valuable predictor of tremor incidence proportion.
Conclusions
A model-based meta-analysis from diverse literature data of various marketed drugs was performed in order to reconcile disparate incidence proportions of serotonin syndrome-associated tremors, based on estimated brain SERT coverages. The relationship identified by the meta-analysis allows for the tremor incidence to be predicted for investigational drugs which exhibit brain exposure and either on- or off-target SERT inhibition, based on their predicted SERT coverage. Potential extensions of this work include: (1) inclusion of additional symptoms/characteristics of serotonin syndrome, and (2) correction for binding competition for combinations of two inhibitors (or parent + active metabolite) that bind to the same site on SERT.