In vitro-in vivo scaling of CYP kinetic data not consistent with the classical Michaelis-Menten model

Strategies for the prediction of in vivo drug clearance from in vitro drug metabolite kinetic data are well established for the rat. In this animal species, metabolism rate-substrate concentration relationships can commonly be described by the classic hyperbola consistent with the Michaelis-Menten model and simple scaling of the parameter intrinsic clearance (CL(int) - the ratio of V(max) to K(m)) is particularly valuable. The in vitro scaling of kinetic data from human tissue is more complex, particularly as many substrates for cytochrome P450 (CYP) 3A4, the dominant human CYP, show nonhyperbolic metabolism rate-substrate concentration curves. This review critically examines these types of data, which require the adoption of an enzyme model with multiple sites showing cooperative binding for the drug substrate, and considers the constraints this kinetic behavior places on the prediction of in vivo pharmacokinetic characteristics, such as metabolic stability and inhibitory drug interaction potential. The cases of autoactivation and autoinhibition are discussed; the former results in an initial lag in the rate-substrate concentration profile to generate a sigmoidal curve whereas the latter is characterized by a convex curve as V(max) is not maintained at high substrate concentrations. When positive cooperativity occurs, we suggest the use of CL(max), the maximal clearance resulting from autoactivation, as a substitute for CL(int). The impact of heteroactivation on this approach is also of importance. In the case of negative cooperativity, care in using the V(max)/K(m) approach to CL(int) determination must be taken. Examples of substrates displaying each type of kinetic behavior are discussed for various recombinant CYP enzymes, and possible artifactual sources of atypical rate-concentration curves are outlined. Finally, the consequences of ignoring atypical Michaelis-Menten kinetic relationships are examined, and the inconsistencies reported for both different substrates and sources of recombinant CYP3A noted.