Generally, a drug’s effects on enzyme and transporter activity are examined independently in a drug development program, but what if results from one study can predict the other? Recent research discussed by Dr. Justin Lutz, of Gilead Sciences, at the 22nd International Conference on Drug-Drug Interactions in Seattle, Washington earlier this year indicates we may be able to use CYP3A4 induction data to categorize the likely magnitude of induction of a common transporter, P‑glycoprotein (P‑gp).
Understanding the enzymes and transporters involved in the disposition of a drug candidate is an important component of ADME/DMPK studies, which informs drug developers of safety and pharmacokinetic properties to predict risk and dosing considerations. Induction studies seek to understand how a drug can induce, or upregulate, the activity of important enzymes. One of the key players in drug metabolism is the cytochrome P450 enzyme, CYP3A4. It is integral to the biotransformation of many small-molecule drugs, and as such, is routinely evaluated in preclinical in vitro studies for potential induction to assess potential drug-drug interactions (DDI) by a drug candidate. It has been established that induction of this enzyme often portends induction of another family of CYPs, CYP2Cs, which are co-regulated by at least one nuclear receptor (namely PXR). As discussed at the presentation, “Drug Transporter Induction – Can We Leverage P450 Data to Streamline Our Clinical Pharmacology Programs?” there may be evidence that, similarly, CYP3A4 induction could indicate induction of P-gp. If so, what are the potential clinical implications?
These questions were explored by Dr. Lutz and his colleagues with Gilead Sciences, Inc. by using a modeling approach based on comparison of the ratio of area under the plasma concentration-time curve of various probe substrates (AUCR) in clinical studies.1 First, they plotted the AUCR of a known CYP3A probe substrate, midazolam (MDZ), as a function of varying doses of treatment with a known CYP3A/P-gp inducer, rifampin (RIF), to look at CYP3A induction. Next, they looked at induction of P-gp based on AUCR of P‑gp substrate, total dabigatran (TDAB, the sum of conjugated and unconjugated active species present from dabigatran etexilate, DE) as a function of the same RIF dose treatment conditions used to investigate CYP3A induction. Comparison of these two plots showed that P‑gp was less inducible than CYP3A, providing evidence that induction of P-gp is one DDI category weaker than CYP3A (i.e., strong CYP3A inducers are likely to be moderate P-gp inducers). Combination of these plots created a RIF CYP3A–P‑gp relationship model curve, by plotting TDAB AUCR as a function of MDZ AUCR, that can predict qualitative categorization of CYP3A or P‑gp induction for a tested substrate, as well as potential DDI between the two probe substrates plotted based on areas of similar induction. To test the model curve for accuracy, it was applied to known moderate CYP3A inducers carbamazepine (CBZ) and rifabutin (RBT). Applying the model to these substrates accurately categorized them as moderate P-gp inducers. To follow, the AUCR of sofosbuvir (SOF), a P‑gp substrate, was predicted from the TDAB AUCR model. Using the new SOF AUCR as a function of the MDZ AUCR model, RIF (a prototypical CYP3A4 inducer) was accurately categorized as a strong inducer of CYP3A4, and CBZ and RBT were categorized as moderate inducers of CYP3A4 (based on MDZ AUCR), but weak inducers of P‑gp (based on SOF AUCR).
But what about the potential to predict induction of other transporters?
Further modeling by the Gilead Sciences group determined that strong CYP3A induction does not only correlate to moderate CYP2C and P-gp induction, but also moderate induction of OATP. It is noted that this correlation is true when PXR [pregnane X receptor, a.k.a. steroid and xenobiotic sensing nuclear receptor (SXR)] is the primary nuclear receptor regulating the enzymes and transporters involved in the compound’s disposition. It was also noted, however, that in cases where there may be additional non-PXR agonism or activation, CYP3A may under-predict the induction magnitude of OATP. This appears to be the case because, while CYP3A transcription is upregulated by PXR activation, OATP transcription is regulated not only by PXR but also by several other nuclear receptors that may have an influence on its induction magnitude. Overall, this modeling approach has enormous potential clinical application, and with more evidence in the future we may be able to use CYP3A4 induction data to predict if a drug induces P-gp, OATP or CYP2Cs.
Dr. Lutz posed an interesting viewpoint regarding decision-making and risk evaluation: if assumed parity between CYP3A induction and transporter activity is not true and a drug-drug interaction study is performed following positive CYP3A induction results, the study could be unnecessary. On the other hand, if it is not performed and the relationship does exist, developers may need to be prepared for restrictive, conservative labeling. As developers plan their preclinical programs, efficiencies and observed relationships between data cannot be undervalued; as research continues to strengthen understanding of DDI relationships, developers are better equipped to get effective therapies to market faster.
Learn more about:
 Lutz JD, Kirby BJ, Wang L, Song Q, Ling J, Massetto B, Worth A, Kearney BP, Mathias A. Cytochrome P450 3A Induction Predicts P-glycoprotein Induction; Part 1: Establishing Induction Relationships Using Ascending Dose Rifampin. Clin Pharmacol Ther. 2018 Dec;104(6):1182-1190; and Lutz JD, Kirby BJ, Wang L, Song Q, Ling J, Massetto B, Worth A, Kearney BP, Mathias A. Cytochrome P450 3A Induction Predicts P-glycoprotein Induction; Part 2: Prediction of Decreased Substrate Exposure After Rifabutin or Carbamazepine. Clin Pharmacol Ther. 2018 Dec;104(6):1191-1198
About the Authors
Subscribe to our Newsletter
Stay up to date with our news, events and research
Do you have a question or a request for upcoming blog content?
We love to get your feedback