Will the Pharmacologically Meaningful Half-Life Please Stand Up?

In the 21st century, much emphasis in the field of clinical pharmacology has been placed on pharmacometrics and systems biology—the mathematical modeling of complex biologic processes. These principles have been applied to clinical trial simulations, prediction of drug– drug interactions using physiological pharmacokinetic models, optimization of dosing regimens, modeling of disease state progression, and Bayesian forecasting of clinical trial outcomes. Although these approaches have undoubtedly contributed to advances in drug development and individualized dosing, it seems appropriate to revisit certain basic principles of clinical pharmacology that are relevant to pharmacotherapy. Drug elimination half-life is directly related to distribution volume, an equilibrium concept, and is inversely related to total body clearance, which in turn depends on variables such as blood flow to eliminating organs and intrinsic clearance, as described elsewhere.1,2 The clinical pharmacology literature cites different descriptors of half-life: distribution and elimination half-lives, effective half-life, and the so-called context-sensitive half-life. Distribution and elimination half-lives usually pertain to initial and terminal loglinear portions of concentration–time profiles, respectively. For certain drugs with multicompartment pharmacokinetic characteristics, there can also be a (gamma) terminal phase half-life that is reflective of binding and dissociation from a receptor complex or very slow release from a deep peripheral compartment. Effective half-life pertains to drug accumulation following repeated dosing, and context-sensitive half-life to duration of administration. As a result of the kinetics of drug distribution to effect sites, it is important to recognize that for various drugs, the clinically or pharmacologicallymeaningful half-life does not always equate to the elimination half-life. This is contrary to the belief of many clinicians, where the assumption is that the longer the elimination half-life, the longer the duration of pharmacologic effect. The aim of this commentary is to illustrate, by way of relevant examples, the necessity of use of an appropriate descriptor of half-life (vide supra), and not solely the elimination half-life, to account for the duration of pharmacologic effect and dosing interval for individual drugs or within drug classes. In fact, the elimination half-life is irrelevant to the duration of pharmacologic effect in numerous instances and is affected by the absorption rate constant (for oral dosing), sampling interval and duration, and assay sensitivity (among other aspects). For the purpose of this discussion, it is considered that the underlying rate processes are first order and that the duration of drug action correlates with the pharmacologically relevant half-life and accumulation at appropriate receptors in the biophase. Although elimination and terminal halflife can also be relevant to drug toxicity, an in-depth description of this aspect is beyond the scope of this commentary. Regarding the relevance of drug distribution, the benzodiazepine diazepam is highly lipophilic, with a short distribution half-life (1 to 2 hours) and a long elimination half-life (21 to 37 hours), and equilibrates rapidly with brain tissue upon intravenous administration.3 Diazepam has a short duration of pharmacologic effect because of the rapid redistribution from the brain to peripheral tissues. However, the more polar drug lorazepam has a shorter elimination half-life (10 to 20 hours), smaller volume of distribution, and is more slowly distributed into brain tissue,