2008 - Marseille - France

PAGE 2008: Applications- CNS
Sophie Gisbert

Analysing Raw Count data from an In Vitro Target Occupancy Assay to Select Doses for Human Safety/Tolerability Trial

S.Gisbert, B. Laurijssens, C. Chen

GlaxoSmithKline, UK

Introduction: Compound A is being developed to treat chronic pain in conditions such as osteoarthritis. As concentration effect relationship obtained in the rat PK/PD model was not clear, it was decided to use human in vitro target occupancy as a biomarker to predict the human pharmacology of this compound.

Objective: The objective of this work was to help selecting doses for the initial human safety/tolerability trial using human in vitro target occupancy data.

- An in vitro competition radioligand binding assay was used to investigate the target occupancy of compound A on its target and repeated in 5 experiments.
- The raw count data were analysed using NONMEM. A sigmoidal model was applied to link concentrations to radioactivity taking into account the non specific binding. Inter-experiment variability was associated to the non specific binding.
- Subsequently simulations of the relationship between dose and target occupancy were performed in Berkeley Madonna software.
- The PK parameters were provided by estimates from a human microdose study and absorption characteristics predictions using Gastroplus software.

- Parameters of the relationship between concentrations and raw radioactivity, Ki and gamma, were estimated with high precision and the model fitted the data well.
- IC50 was estimated from Ki using Cheng-Prusoff equation.
- For the simulations an inter-subject variability in the IC50 of 30% was used to reflect potential variability in the transport of the compound to the target site.
- Since compound A is an antagonist minimum effect was predicted at 0.3 mg (20% target occupancy) and the pharmacological dose was predicted to be above 50 mg ( >98% target occupancy).

Conclusion: Analysis of raw count data from a human in vitro Target Occupancy Assay allowed predicting the range of pharmacological doses for initial Human Safety/Tolerability Trials.

Reference: PAGE 17 (2008) Abstr 1393 [www.page-meeting.org/?abstract=1393]
Poster: Applications- CNS