P133 DPX2 Cells as a Model System for Detecting PXR-Mediated Induction of CYP3A Enzymes by Therapeutic Agents; Utility for DDI Predictions

Odette A. Fahmi , Pharmacokinetics and Drug Metabolism, Pfizer, Groton, CT
Judy Raucy , Puracyp, Inc, Carlsbad, CA
Elsa Ponce , Puracyp, Inc, Carlsbad, CA
Jerome M. Lasker , Puracyp, Inc, Carlsbad, CA
The increase in P450 enzyme activity noted after exposure to certain therapeutics can elicit marked drug-drug interactions (DDI) that, in turn, may ultimately result in poor clinical outcome or adverse drug effects.  The enhanced oxidative metabolism observed in response to drug treatment is mediated primarily by CYP3A4, the most prevalent hepatic P450 enzyme whose expression is regulated by the orphan nuclear receptor PXR.  In vitro systems capable of rapidly and accurately determining whether a given agent activates PXR, and is thus capable of enhancing CYP3A-mediated metabolism, are highly sought-after tools for drug discovery screening.  To that end, we assessed the utility of DPX2 cells, a HepG2-derived cell line stably integrated with a PXR expression vector plus a luciferase reporter, to detect therapeutics that not only cause PXR activation/CYP3A induction but also elicit clinical DDI.  DPX2 cells were found to express mRNAs encoding CYP1A2, CYP2B6, CYP3A4, CYP3A5 and CYP3A7.  Expression of CYP3A mRNA was inducible by treatment with rifampicin (RIF) and carbamazepine, although only CYP3A7 protein was detected after RIF treatment.  CYP3A7 was catalytically active in DPX2 cells, as indicated by the RIF-inducible (29-fold) metabolism of luciferin isopropyl acetal, a bioluminescent CYP3A substrate.  DPX2 cells expressed markedly higher levels of PXR than non-transfected HepG2 cells, and also contained appreciable levels of the key influx and efflux ABC transporters MDR1, MRP2, MRP3, MRP4, NTCP, OCT1, OAT2 and OATP2B1.  The capacity of DPX2 cells to accurately identify clinical agents capable of eliciting CYP3A enzyme induction in vivo was then investigated with 22 clinical inducers and 12 clinical non-inducers.  All 22 therapeutic agents with known in vivo inducing properties proved capable in DPX2 cells of activating PXR, with Emax values ranging from 3.5- to 13-fold.  Of the 12 therapeutics without in vivo inducing effects, 6 or 50% acted as PXR activators while the other 6 failed to activate this receptor. The relative induction score (RIS), which was calculated by combining DPX2 cell activation parameters (EC50 and Emax) for eleven drugs (7 clinical inducers and 4 clinical non-inducers) with their efficacious total plasma concentrations demonstrated a strong correlation (r2 = 0.90) between RIS and the magnitude of induction of midazolam clearance.  However, application of the RIS derived from the midazolam curve with other object drugs; such as nifedipine, simvastatin, triazolam and ethinyl estradiol resulted in predictions (mostly over-predictions) with varying degrees of accuracy.  Additionally, PXR activation data obtained with the 12 clinical non-inducing agents did not translate into DDI predictions that were significant.  In conclusion, the DPX2 cell-based assay exhibits the sensitivity and specificity needed for detecting potential clinical CYP3A inducers, and offers a more rapid and economical tool compared to human hepatocytes for use in drug discovery screening and early development.