Protein binding-dependent decreases in hERG channel blocker potency assessed by whole-cell voltage clamp in serum

In vitro hERG blocking potency is measured in drug discovery as part of an integrated cardiovascular risk assessment. Typically, the concentrations producing 50% inhibition are measured in protein-free saline solutions and compared with calculated free therapeutic in vivo Cmax values to estimate a hERG safety multiple. The free/unbound fraction is believed responsible for activity. We tested the validity of this approach with 12 compounds by determining potencies in voltage clamp studies conducted in the absence and presence of 100% dialyzed fetal bovine serum (FBS). Bath drug concentrations in saline solutions were measured to account for loss of compounds due to solubility, stability, and/or adsorption. Protein binding in dialyzed FBS was measured to enable predictions of serum IC50s based on the unbound fraction and the saline IC50. For 11 of 12 compounds, the measured potency in the presence of dialyzed FBS was within 2-fold of the predicted potency. The predicted IC50 in dialyzed FBS for one highly bound compound, amiodarone, was 9-fold higher than the measured serum IC50. These data suggest that for highly bound compounds, direct measurement of IC50s in the presence of 100% serum may provide a more accurate estimate of in vivo potencies than the approach based on calculated serum shifts.