Disposition pathway-dependent approach for predicting organic anion-transporting polypeptide-mediated drug-drug interactions

Hepatic transporter drug-drug interactions: an evaluation of approaches and methodologies

INTRODUCTION

Drug-drug interactions (DDIs) continue to account for 5% of hospital admissions and therefore remain a major regulatory concern. Effective, quantitative prediction of DDIs will reduce unexpected clinical findings and encourage projects to frontload DDI investigations rather than concentrating on risk management ('manage the baggage') later in drug development. A key challenge in DDI prediction is the discrepancies between reported models. Areas covered: The current synopsis focuses on four recent influential publications on hepatic drug transporter DDIs using static models that tackle interactions with individual transporters and in combination with other drug transporters and metabolising enzymes. These models vary in their assumptions (including input parameters), transparency, reproducibility and complexity. In this review, these facets are compared and contrasted with recommendations made as to their application. Expert opinion: Over the past decade, static models have evolved from simple [I]/k_i models to incorporate victim and perpetrator disposition mechanisms including the absorption rate constant, the fraction of the drug metabolised/eliminated and/or clearance concepts. Nonetheless, models that comprise additional parameters and complexity do not necessarily out-perform simpler models with fewer inputs. Further, consideration of the property space to exploit some drug target classes has also highlighted the fine balance required between frontloading and back-loading studies to design out or 'manage the baggage'.

Comparison of the static in vivo approach to a physiologically based pharmacokinetic approach for metabolic drug–drug interactions prediction

Background: The in vivo mechanistic static model (IMSM) and the physiologically based pharmacokinetic (PBPK) model are two approaches used to predict the magnitude of drug–drug interactions (DDIs). The aim of this study was to evaluate the performance of IMSM and to compare IMSM with the PBPK approach implemented in Simcyp. Methods: The predictive performances of IMSM were evaluated on a panel of 628 DDIs. Subsequently, the IMSM and PBPK approaches were compared on a set of 104 DDIs. Results: The IMSM yielded 85% of predictions within 1.5-fold of the observed value on the 628 DDIs panel. The predictive performances of IMSM were better than those of the PBPK approach (median fold error 1 vs 0.86 on 104 studies; p = 0.02). Conclusion: The IMSM approach is an alternative tool for metabolic DDIs prediction.

A Prediction Model of Drug Exposure in Cirrhotic Patients According to Child-Pugh Classification

BACKGROUND AND OBJECTIVE

Prediction of drug clearance in liver cirrhosis patients is currently based on in vitro-in vivo extrapolation and physiologically-based pharmacokinetic models. No static model for this purpose has been described. The objectives of this study were to (1) derive a static model for predicting drug exposure in cirrhotic patients, and (2) to evaluate the model on a large set of published data.

METHODS

The impact of cirrhosis was characterized by the ratio of the total and unbound drug area under the concentration-time curve (AUC) in cirrhotic patients to the AUC measured in healthy subjects These ratios were predicted for Child-Pugh classes A, B, and C. The AUC ratios observed in published data were compared with AUC ratios predicted by the model.

RESULTS

Among 171 drugs examined, 83 published AUC ratios for 45 drugs in cirrhotic patients were available for analysis. The mean ± standard deviation relative prediction error for the total and unbound AUC ratios was 0.22 ± 0.58 and 0.24 ± 0.56, respectively. There were four outliers among the 83 predicted values. Simulations showed that the prediction error was negligible provided that the hepatic extraction coefficient was less than 0.8.

CONCLUSIONS

For mild and moderate cirrhosis (classes A and B), the predicted unbound AUC ratio is typically approximately 2 and 3.5, respectively, for most drugs. In the absence of data in cirrhotic patients, the drug dose might be empirically reduced by these factors. In severe cirrhosis (class C), our model may help clinicians to adjust their prescriptions.

Physiologically based pharmacokinetic modelling and in vivo [I]/K(i) accurately predict P-glycoprotein-mediated drug-drug interactions with dabigatran etexilate

BACKGROUND AND PURPOSE

In vitro inhibitory potency (Ki )-based predictions of P-glycoprotein (P-gp)-mediated drug-drug interactions (DDIs) are hampered by the substantial variability in inhibitory potency. In this study, in vivo-based [I]/Ki values were used to predict the DDI risks of a P-gp substrate dabigatran etexilate (DABE) using physiologically based pharmacokinetic (PBPK) modelling.

EXPERIMENTAL APPROACH

A baseline PBPK model was established with digoxin, a known P-gp substrate. The Km (P-gp transport) of digoxin in the baseline PBPK model was adjusted to Km (i) to fit the change of digoxin pharmacokinetics in the presence of a P-gp inhibitor. Then 'in vivo' [I]/Ki of this P-gp inhibitor was calculated using Km (i) /Km . Baseline PBPK model was developed for DABE, and the 'in vivo' [I]/Ki was incorporated into this model to simulate the static effect of P-gp inhibitor on DABE pharmacokinetics. This approach was verified by comparing the observed and the simulated DABE pharmacokinetics in the presence of five different P-gp inhibitors.

KEY RESULTS

This approach accurately predicted the effects of five P-gp inhibitors on DABE pharmacokinetics (98-133% and 89-104% for the ratios of AUC and Cmax respectively). The effects of 16 other P-gp inhibitors on the pharmacokinetics of DABE were also confidently simulated.

CONCLUSIONS AND IMPLICATIONS

'In vivo' [I]/Ki and PBPK modelling, used in combination, can accurately predict P-gp-mediated DDIs. The described framework provides a mechanistic basis for the proper design of clinical DDI studies, as well as avoiding unnecessary clinical DDI studies.