Introduction: Human liver microsomal incubations are often used to predict the metabolic lability of new chemical entities. The clearance values are scaled-up from in vitro data and mathematically corrected for plasma protein binding, or in some cases the free fraction ratio of plasma to microsomes, using well-established scaling methods such as the well-stirred model. This can be time consuming for multiple compounds since it requires separate experiments to determine in vitro lability, and free fraction.
Methods: We attempted to streamline clearance predictions by combining experiments into one. Firstly, we combined the free fraction experiments into one free fraction ratio by measuring the partitioning of compound between plasma and microsomes, and by applying this experimental ratio to clearance predictions found that it performed at least as well as free fractions determined separately. We also incubated compounds with plasma added to the incubation mixture and compared the predicted clearances to values determined using traditional mathematical protein binding corrections.
Results: Consistently, incubations with added plasma resulted in CL predictions closer to literature values than incubations only mathematically corrected for protein binding. For example, incorporating plasma into a ketamine incubation resulted in a CL value of 15.1 mL/min/kg, compared with a value of 10.2 using mathematical binding corrections. The literature value is 16.4 mL/min/kg.
Discussion: This work characterizes this new method and compares it to the traditional microsomal incubation method using several literature compounds, and suggests that streamlining the methods may generate quality data faster and with less resource investment.