Professor UCSF School of Pharmacy San Francisco, California
Our recent online paper [Benet, L.Z., Sodhi, J.K. Investigating the Theoretical Basis for In Vitro–In Vivo Extrapolation (IVIVE) in Predicting Drug Metabolic Clearance and Proposing Future Experimental Pathways. AAPS J 22, 120 (2020). https://doi.org/10.1208/s12248-020-00501-9] examines the boundary conditions for predicting clearance for metabolized drugs that are not clinically relevant transporter substrates. We show that prediction of in vivo clearance from in vitro measures of metabolism violates the basic assumptions employed in IVIVE and could never give correct predictions for ~2/3rds of the drugs investigated in the literature. We propose that this is due to utilizing a chemistry approach in determining the in vitro intrinsic clearance, where the rate constant of elimination is essentially multiplied by the fixed volume of the in vitro incubation, where this latter term is drug independent and measureable. But what we are trying to predict is an in vivo pharmacokinetic metabolic intrinsic clearance that is a function of a drug specific volume of distribution, which is drug dependent and not measureable. For most drugs, which are lipophilic substances, the volume of distribution in the liver will be greater and not the same as the volume of distribution of the metabolic enzymes in the liver. We also point out that, using boundary conditions, the increased underprediction as clearance increases for high clearance drugs is consistent with the relevant hepatic blood flow in humans being about 2.5 greater than the measured hepatic blood flow. Thus, one might expect that if we do such a poor job in predicting metabolic clearance for drugs that are not clinically relevant transporter substrates, the clearance prediction outcome would be even poorer when transporter effects are relevant. But, in fact, this is not true if transporters are the rate limiting clearance parameter. Under conditions where efflux of drug from the liver back to the systemic circulation is negligible compared to hepatic drug elimination (including both metabolism and transporter mediated biliary elimination) or where both basolateral efflux and and hepatic drug elimination are both negligible, and clearance becomes rate limited by hepatic transporter uptake, transporter mediated clearance can be reasonably predicted. This has been nicely demonstrated for HMG-CoA inhibitors (statins) that are rate limited by OATP hepatic uptake transporters. However, from a clinical perspective, we are not attempting to predict clearance for a new molecular entity (NME). Rather we are trying to predict exposure (AUC) for different doses, how exposure changes due to pharmacogenomic variance and as a result of drug-drug interactions. For an IV dose, where AUC is inversely related to clearance, we will be able to predict exposure for these OATP substrates. However, this will not be true for oral drug dosing of the BDDCS Class 1 and 2 statins (where the major route of elimination, but not the rate limiting step, is metabolism) where gut metabolism (Fg) can be a major determinant of exposure. We still require further advances in the basic science IVIVE approaches in predicting exposure for both metabolic and transporter substrate NMEs.
Upon completion, participants will be able to understand why IVIVE predictions exhibit more than 2-fold errors for metabolized drugs that are not clinically relevant substrates for transporters.
Upon completion, participants will be able to understand why for certain select transporter substrates this inability to predict clearance for metabolized drugs will not be true for some OATP substrates.
Upon completion, participants will be able to recognize that prediction of clearance may not be the relevant parameter one is hoping to predict for an NME, but rather exposure is the clinically relevant outcome to predict.
Upon completion, participants will be able to explain why for OATP rate limited substrates clearance prediction will successfully predict exposure for IV dosed drugs, but not for many orally dosed drugs.
Upon completion, participants will recognize that the values for gut availability (Fg) are not reasonably predictable for drug substrates that are rate limited by hepatic uptake, but where the major route of elimination is metabolism.