Exploring the Use of Serum-Derived Small Extracellular Vesicles as Liquid Biopsy to Study the Induction of Hepatic Cytochromes P450 and Organic Anion Transporting Polypeptides.

Liver-derived small extracellular vesicles (sEVs), prepared from small sets of banked serum samples using a novel two-step protocol, were deployed as liquid biopsy to study the induction of cytochromes P450 (CYP3A4, CYP3A5, and CYP2D6) and organic anion transporting polypeptides (OATP1B1 and OATP1B3) during pregnancy (nonpregnant (T0), first, second, and third (T3) trimester women; N = 3 each) and after administration of rifampicin (RIF) to healthy male subjects. Proteomic analysis revealed induction (mean fold-increase, 90% confidence interval) of sEV CYP3A4 after RIF 300 mg × 7 days (3.5, 95% CI = 2.5-4.5, N = 4, P = 0.029) and 600 mg × 14 days (3.7, 95% CI = 2.1-6.0, N = 5, P = 0.018) consistent with the mean oral midazolam area under the plasma concentration time curve (AUC) ratio in the same subjects (0.28, 95% CI = 0.22-0.34, P < 0.0001; and 0.17, 95% CI = 0.13-0.22, P < 0.0001). Compared with CYP3A4, liver sEV CYP3A5 protein (subjects genotyped CYP3A5*1/*3) was weakly induced (≤ 1.5-fold). It was also possible to measure liver sEV-catalyzed dextromethorphan (DEX) O-demethylation to dextrorphan (DXO), correlated with sEV CYP2D6 expression (r = 0.917, P = 0.0001; N = 10) and 3-hour plasma DXO-to-DEX concentration ratio (r = 0.843, P = 0.002, N = 10), and show that CYP2D6 was not induced by RIF. Nonparametric analysis of liver sEV revealed significantly higher CYP3A4 (3.2-fold, P = 0.003) and CYP2D6 (3.7-fold, P = 0.03) protein expression in T3 vs. T0 women. In contrast, expression of both OATPs in liver sEV was unaltered by RIF administration and pregnancy.

Physiologically Based Pharmacokinetic Modeling to Identify Physiological and Molecular Characteristics Driving Variability in Drug Exposure.

Prospectively defining the physiological and molecular characteristics most likely driving between-subject variability (BSV) in drug exposure provides the opportunity to inform the assessment of biomarkers to account for this variability. A physiologically based pharmacokinetic (PBPK) model was constructed and verified for dabrafenib. This model was then used to evaluate the physiological and molecular characteristics driving BSV in dabrafenib exposure. The capacity to discriminate a steady-state dabrafenib trough concentration >48 ng/mL was also evaluated. The mean simulated/observed ratios for the parameters describing dabrafenib exposure in single-dose, multiple-dose, and drug-drug interaction studies were between 0.78 and 1.23. Multivariable analysis indicated that consideration of baseline weight, body mass index, and CYP2C8, CYP3A4, and P-gp abundance strongly predicts steady-state dabrafenib trough concentration above 48 ng/mL (ROC AUC = 0.94; accuracy = 86%). This is the first study to use a verified PBPK model to identify baseline physiological and molecular characteristics driving BSV in drug exposure.