Application of unbound liver-to-plasma concentration ratio to quantitative projection of cytochrome P450-mediated drug-drug interactions using physiologically based pharmacokinetic modelling approach

Antifungal Drugs TDM: Trends and Update

PURPOSE

The increasing burden of invasive fungal infections results in growing challenges to antifungal (AF) therapeutic drug monitoring (TDM). This review aims to provide an overview of recent advances in AF TDM.

METHODS

We conducted a PubMed search for articles during 2016-2020 using "TDM" or "pharmacokinetics" or "drug-drug-interaction" with "antifungal," consolidated for each AF. Selection was limited to English language articles with human data on drug exposure.

RESULTS

More than 1000 articles matched the search terms. We selected 566 publications. The latest findings tend to confirm previous observations in real-life clinical settings. The pharmacokinetic variability related to special populations is not specific but must be considered. AF benefit-to-risk ratio, drug-drug interaction (DDI) profiles, and minimal inhibitory concentrations for pathogens must be known to manage at-risk situations and patients. Itraconazole has replaced ketoconazole in healthy volunteers DDI studies. Physiologically based pharmacokinetic modeling is widely used to assess metabolic azole DDI. AF prophylactic use was studied more for Aspergillus spp. and Mucorales in oncohematology and solid organ transplantation than for Candida (already studied). Emergence of central nervous system infection and severe infections in immunocompetent individuals both merit special attention. TDM is more challenging for azoles than amphotericin B and echinocandins. Fewer TDM requirements exist for fluconazole and isavuconazole (ISZ); however, ISZ is frequently used in clinical situations in which TDM is recommended. Voriconazole remains the most challenging of the AF, with toxicity limiting high-dose treatments. Moreover, alternative treatments (posaconazole tablets, ISZ) are now available.

CONCLUSIONS

TDM seems to be crucial for curative and/or long-term maintenance treatment in highly variable patients. TDM poses fewer cost issues than the drugs themselves or subsequent treatment issues. The integration of clinical pharmacology into multidisciplinary management is now increasingly seen as a part of patient care.

Consideration of the Unbound Drug Concentration in Enzyme Kinetics

The study of enzyme kinetics in drug metabolism involves assessment of rates of metabolism and inhibitory potencies over a suitable concentration range. In all but the very simplest in vitro system, these drug concentrations can be influenced by a variety of nonspecific binding reservoirs that can reduce the available concentration to the enzyme system(s) under investigation. As a consequence, the apparent kinetic parameters, such as K_m or K_i, that are derived can deviate from the true values. There are a number of sources of these nonspecific binding depots or barriers, including membrane permeation and partitioning, plasma or serum protein binding, and incubational binding. In the latter case, this includes binding to the assay apparatus as well as biological depots, depending on the characteristics of the in vitro matrix being used. Given the wide array of subcellular, cellular, and recombinant enzyme systems utilized in drug metabolism, each of these has different components which can influence the free drug concentration. The physicochemical properties of the test compound are also paramount in determining the influential factors in any deviation between true and apparent kinetic behavior. This chapter describes the underlying mechanisms determining the free drug concentration in vitro and how these factors can be accounted for in drug metabolism studies, illustrated with case studies from the literature.

Direct Comparison of Total Clearance Prediction: Computational Machine Learning Model versus Bottom-Up Approach Using In Vitro Assay

The in vitro-in vivo extrapolation (IVIVE) approach for predicting total plasma clearance (CL_tot) has been widely used to rank order compounds early in discovery. More recently, a computational machine learning approach utilizing physicochemical descriptors and fingerprints calculated from chemical structure information has emerged, enabling virtual predictions even earlier in discovery. Previously, this approach focused more on in vitro intrinsic clearance (CL_int) prediction. Herein, we directly compare these two approaches for predicting CL_tot in rats. A structurally diverse set of 1114 compounds with known in vivo CL_tot, in vitro CL_int, and plasma protein binding was used as the basis for this evaluation. The machine learning models were assessed by validation approaches using the time- and cluster-split training and test sets, and five-fold cross validation. Assessed by five-fold validation, the random forest regression (RF) and radial basis function (RBF) models demonstrated better prediction performance in eight attempted machine learning models. The CL_tot values predicted by the RF and RBF models were within two-fold of the observed values for 67.7 and 71.9% of cluster-split test set compounds, respectively, while the predictivity was worse in the time-split dataset. The predictivity of both models tended to be improved by incorporating in vitro parameters, unbound fraction in plasma (f_u,p), and CL_int. CL_tot prediction utilizing in vitro CL_int and the well-stirred model, correcting for the fraction unbound in blood, was substantially worse compared to machine learning approaches for the same cluster-split test set. The reason that CL_tot is underestimated by IVIVE is not fully explained by considering the calculated microsomal unbound fraction (cf_u,mic), extended clearance classification system (ECCS), and omitting high clearance compounds in excess of hepatic blood flow. The analysis suggests that in silico machine learning models may have the power to reduce reliance on or replace in vitro and in vivo studies for chemical structure optimization in early drug discovery.

Challenging the Relevance of Unbound Tissue-to-Blood Partition Coefficient (Kpuu) on Prediction of Drug-Drug Interactions

PURPOSE

To examine the theoretical/practical utility of the liver-to-blood partition coefficient (Kp_uu) for predicting drug-drug interactions (DDIs), and compare the Kp_uu-approach to the extended clearance concept AUC_R-approach.

METHODS

The Kp_uu relationship was derived from first principles. Theoretical simulations investigated the impact of changes in a single hepatic-disposition process on unbound systemic (AUC_B,u) and hepatic exposure (AUC_H,u) versus Kp_uu. Practical aspects regarding Kp_uu utilization were examined by predicting the magnitude of DDI between ketoconazole and midazolam employing published ketoconazole Kp_uu values.

RESULTS

The Kp_uu hepatic-disposition relationship is based on the well-stirred model. Simulations emphasize that changes in influx/efflux intrinsic clearances result in Kp_uu changes, however AUC_H,u remains unchanged. Although incorporation of Kp_uu is believed to improve DDI-predictions, utilizing published ketoconazole Kp_uu values resulted in prediction errors for a midazolam DDI.

CONCLUSIONS

There is limited benefit in using Kp_uu for DDI-predictions as the AUC_R-based approach can reasonably predict DDIs without measurement of intracellular drug concentrations, a difficult task hindered by experimental variability. Further, Kp_uu changes can mislead as they may not correlate with changes in AUC_B,u or AUC_H,u. The well-stirred model basis of Kp_uu when applied to hepatic-disposition implies that nuances of intracellular drug distribution are not considered by the Kp_uu model.

Estimation of the Effect of OAT2-Mediated Active Uptake on Meloxicam Exposure in the Human Liver

Active uptake mediated by organic anion transporter 2 (OAT2) has been previously hypothesized as a key player in hepatic disposition of its substrates. Previous studies have shown that another hepatic uptake transporter, organic anion transporting polypeptides (OATP) 1B1, significantly elevates liver concentrations of drugs transported by it. As tissue concentration typically governs pharmacodynamics, drug-drug interactions, and toxicity in the liver, it is important to understand if OAT2 functions similarly to OATP1B1 in raising liver exposure. Since this is a research problem that cannot be easily assessed in clinical studies at this time, here we estimated human liver exposure of an OAT2 substrate meloxicam using a deduction method based on physiologically based pharmacokinetic (PBPK) modeling of clinical systemic exposure data. Although in vitro data suggest that OAT2-mediated active uptake is involved in meloxicam disposition, the modeling result concludes that its unbound liver exposure is unlikely significantly different from its unbound systemic exposure. This conclusion is further supported by data and modeling from a terminal monkey study and in vitro hepatocyte studies with bovine serum albumin. Overall, based on currently available data, we do not expect that OAT2 has a strong impact on the liver exposure of meloxicam.