Impact of Delayed Effects in the Exposue-response Analysis of Clinical QT Studies
View Presentation View Presentation
*Arne Ring, Boehringer Ingelheim Pharma GmbH & Co. KG 

Keywords: QT interval, pharmacodynamics, hysteresis, mixed modelling, lag time

In addition to the "E14 analysis", the analysis of PK concentration and QT change from baseline has gained some scientific attention. This PK-QT analysis depends on some underlying modelling assumptions, one of it the direct relationship between concentration and effect. Indirect relationships, such as delays, might occur in case of "deep effect compartments" or unexpected metabolites. // The current investigation examines the detection of potential effect delays and their impact on the estimates of the QT change from baseline. For the detection, different metrics are explored which shall quantify the hysteresis curve between both parameters. // As a result, delays of up to 30 to 60 minutes (depending on the kinetics of the drug) do not effect the estimation of the QT prolongation to a relevant extent, while delays of more than 2 hours lead to biased estimates of more than 20%.