PBPK for Transporter-Mediated Drug Interactions

1. Why Do Transporters Matter in Drug Interactions?

Transporter-mediated drug interactions occur when one drug changes the activity or expression of a membrane transporter and thereby changes the absorption, distribution, or elimination of another drug.

Transporters can move drugs into cells or out of cells. Their effects can therefore occur at several important sites, including the intestine, liver, kidney, and other tissues. A transporter may influence systemic exposure directly, or it may change the availability of a drug to a metabolic enzyme.

A physiologically based pharmacokinetic (PBPK) model provides a framework for representing these processes mechanistically. Instead of treating an interaction simply as an empirical fold-change in clearance, a PBPK model can represent where the transporter acts, how strongly it acts, and how its activity interacts with passive permeability, metabolism, blood flow, and other disposition pathways.

2. What Are Drug Transporters?

Drug transporters are membrane proteins that facilitate the movement of endogenous substances and drugs across biological membranes. In pharmacokinetics, they are commonly grouped into uptake transporters and efflux transporters.

The exact transporter set relevant to a drug should be determined from its physicochemical properties, experimental data, known disposition pathways, and the scientific question being addressed.

3. How Does a Transporter-Mediated DDI Occur?

A transporter-mediated DDI generally begins with a perpetrator drug altering the activity or expression of a transporter involved in the disposition of a victim drug.

For example, inhibition of a hepatic uptake transporter can reduce the entry of a victim drug into hepatocytes. The consequence could be increased plasma concentrations if hepatic uptake normally contributes substantially to hepatic elimination.

However, the result is not universally an increase in exposure. If hepatic uptake is necessary for access to a metabolic pathway, inhibiting uptake can reduce metabolism and increase plasma exposure. But if transporter-mediated uptake primarily represents a route into a protected or storage compartment, the consequences can be different.

Similarly, inhibition of an intestinal efflux transporter such as P-gp can increase absorption for some substrates, whereas inhibition of a renal secretory transporter can reduce renal clearance.

4. Why Use PBPK for Transporter-Mediated DDIs?

Traditional DDI analysis can summarize an interaction as an observed change in exposure, such as an AUC ratio. PBPK modeling attempts to explain that change using physiological and drug-specific mechanisms.

A PBPK model divides the body into physiologically meaningful compartments or organs and represents drug movement between them. Transport processes can then be incorporated at the tissue interfaces where they occur.

PBPK therefore provides a way to ask not only how large an interaction might be, but also why it occurs and under what conditions it may change.

5. Where Can Transporter-Mediated Interactions Occur?

Intestine

Transporters expressed in the intestinal epithelium can affect the fraction of an orally administered dose that crosses into the systemic circulation. Efflux transporters can move drug from enterocytes back toward the intestinal lumen, while uptake processes can facilitate movement into enterocytes.

Liver

Hepatic uptake transporters can determine how rapidly a circulating drug enters hepatocytes. Once inside the hepatocyte, the drug may undergo metabolism, biliary excretion, intracellular binding, or other processes.

Kidney

Renal transporters can contribute to active secretion or reabsorption. Inhibition can therefore alter renal clearance even when glomerular filtration remains unchanged.

Other tissues

Transporters can also affect drug distribution into tissues and access to pharmacological targets. The importance of a transporter therefore depends on both its expression and the role it plays in the overall disposition of the drug.

6. Uptake and Efflux Transporters

The simplest conceptual distinction is whether a transporter moves drug into a cell or out of a cell. The pharmacokinetic consequence depends on what happens to the drug after that transport step.

Consider hepatic uptake. If transporter-mediated uptake is a major route into hepatocytes and the intracellular drug is subsequently metabolized, transporter inhibition can decrease hepatic clearance.

By contrast, inhibition of a hepatic efflux transporter may increase intracellular exposure and reduce biliary excretion. The same inhibition event can therefore have different consequences depending on the victim drug's complete disposition network.

7. Representing Transport in a PBPK Model

At a conceptual level, transporter-mediated movement can be represented as a clearance-like process or as a mechanistic transport process using transporter-specific parameters.

A simple linear uptake representation can be written as:

where \(CL_{\text{uptake}}\) represents an effective uptake clearance under the conditions of the model.

For saturable transporter activity, a Michaelis-Menten form is often more appropriate:

Here, \(V_{\max}\) describes the maximum transport capacity and \(K_m\) is the concentration associated with half-maximal transport rate.

When transporter activity is inhibited, the model can modify transporter parameters or the effective transport capacity according to the mechanism of inhibition.

8. Transporter Inhibition in PBPK

A perpetrator can inhibit a transporter through several mechanisms. The simplest case is reversible inhibition, where the inhibitory effect depends on perpetrator concentration.

A simplified competitive inhibition relationship can be represented as:

where \(I\) is the inhibitor concentration and \(K_i\) characterizes inhibitor potency under the assumed model.

For a transporter process represented through a clearance term, a simplified inhibition factor can instead be expressed as:

These equations are conceptual representations. Actual PBPK implementations may use more detailed transporter kinetics and may account for free rather than total concentrations, multiple inhibitors, time-varying concentrations, or other mechanisms.

Why concentration matters

A transporter inhibitor may produce a weak effect at one concentration and a substantial effect at another. PBPK models naturally account for this because perpetrator concentrations change with dose, absorption, distribution, metabolism, and elimination.

9. Transporter Induction and Changes in Expression

Not all transporter DDIs result from immediate inhibition. A perpetrator can also alter transporter expression or activity over time.

In a mechanistic PBPK model, transporter abundance or activity can therefore be represented as a dynamic quantity rather than as a fixed constant.

The time course of induction may depend on perpetrator concentration and the turnover of the transporter or regulatory system.

This creates an important difference between inhibition and induction:

For transporter-mediated DDI prediction, the distinction between direct inhibition and changes in expression is therefore important when interpreting both the timing and magnitude of the interaction.

10. Intestinal Transporters and Oral Bioavailability

Transporters in the intestinal wall can influence the fraction of an orally administered dose that reaches systemic circulation.

For an orally administered drug, a simplified relationship is:

where \(F\) is systemic bioavailability, \(D\) is dose, and \(CL\) is systemic clearance under the assumptions of a linear system.

Transporters can affect \(F\) by altering the balance between movement into enterocytes, movement back into the intestinal lumen, metabolism within the gut wall, and passage into portal blood.

For example, inhibition of intestinal P-gp can increase absorption for a drug whose net intestinal efflux is substantial. The resulting change in systemic exposure depends on whether the victim drug is absorption-limited, permeability-limited, metabolized in the gut wall, or governed by other competing processes.

11. Hepatic Uptake and Efflux

Hepatic transporter processes are especially important because transport into and out of hepatocytes can be closely connected to metabolism and biliary excretion.

Consider a drug whose hepatic elimination requires uptake through OATP1B1 or OATP1B3 followed by intracellular metabolism. A simplified conceptual pathway is:

If uptake is strongly inhibited, less drug enters the hepatocyte. Depending on the relative contributions of hepatic uptake, metabolism, renal elimination, and other pathways, systemic exposure may increase.

For transporter substrates with parallel uptake and passive diffusion, the effect may be smaller because passive entry can partially compensate for reduced transporter activity.

Transporter-enzyme interplay

One of the most important reasons to use PBPK is that transporter and enzyme processes can be coupled. A transporter may control access to an intracellular metabolic enzyme, making transporter inhibition an indirect mechanism for changing metabolic clearance.

12. Renal Transporters and Active Secretion

Renal elimination is not limited to glomerular filtration. Drugs can also undergo active secretion and reabsorption.

A simplified renal clearance relationship can be expressed as:

Transporters such as OAT1, OAT3, OCT2, MATE1, and MATE2-K can contribute to active renal secretion for particular substrates.

For a drug undergoing active secretion, inhibition of a renal uptake transporter can decrease entry into renal tubular cells. Inhibition of an efflux transporter responsible for movement from the tubular cell into urine can have a similar net effect on secretion, depending on the transporter arrangement.

Because renal transporter effects can change renal clearance without necessarily changing glomerular filtration, PBPK models can be useful for separating these mechanisms.

13. Transporters Rarely Act Alone

A common modeling mistake is to treat a transporter as if it were an isolated pathway. In reality, drug disposition is usually the result of multiple parallel and sequential processes.

This network explains why a transporter DDI can sometimes have an apparently unexpected result. Inhibiting one pathway can redistribute drug through competing pathways.

For example, if hepatic uptake decreases, renal elimination may become relatively more important. If metabolism decreases, unchanged renal or biliary elimination may partially compensate. PBPK models can represent these competing pathways simultaneously.

14. From In Vitro Transporter Data to PBPK DDI Predictions

A mechanistic transporter DDI assessment commonly integrates information from several sources.

1. Identify relevant transporters. Determine which transporters contribute meaningfully to the victim drug's disposition and which transporters may be inhibited or induced by the perpetrator.
2. Characterize transporter activity. Use appropriate in vitro systems to determine whether the victim is a substrate and whether the perpetrator inhibits or modulates the transporter.
3. Characterize potency. Estimate parameters such as \(K_i\), \(IC_{50}\), \(K_m\), or \(V_{\max}\), as appropriate to the experimental system and mechanistic model.
4. Scale transporter activity. Translate in vitro information into the human PBPK framework while accounting for relevant transporter expression and tissue localization.
5. Build the victim model. Represent absorption, distribution, metabolism, transport, and elimination before adding the perpetrator.
6. Build the perpetrator model. Predict perpetrator concentrations at the relevant transporter sites.
7. Simulate the DDI. Run the victim alone and with the perpetrator under the same dosing conditions.
8. Compare exposure. Calculate predicted ratios such as \(AUC_{\mathrm{DDI}}/AUC_{\mathrm{control}}\) and \(C_{\max,\mathrm{DDI}}/C_{\max,\mathrm{control}}\).
9. Evaluate uncertainty. Examine which assumptions and parameters have the greatest influence on the predicted interaction.

The resulting AUCR is an output of the mechanistic model. It is not itself a transporter parameter.

15. Worked Example: A Simplified Hepatic Uptake DDI

Consider a hypothetical oral drug whose elimination depends substantially on hepatic uptake. For illustration, suppose a simplified model attributes the drug's systemic clearance to two parallel pathways:

• Hepatic uptake-dependent clearance: 6 L/h
• Other clearance pathways: 4 L/h

The baseline total clearance is therefore:

Now suppose a perpetrator reduces the effective hepatic uptake pathway by 50%.

Step 1: Adjust transporter-dependent clearance

Step 2: Calculate total clearance during the DDI

Step 3: Estimate the exposure ratio

For a simplified linear system in which bioavailability is unchanged, exposure is inversely proportional to clearance:

Thus, this simplified model predicts an approximately 1.43-fold increase in AUC.

Step 4: Why this is only a teaching example

A real PBPK model would generally be more detailed. It might represent intestinal absorption, plasma and tissue concentrations, transporter kinetics, hepatic blood flow, protein binding, metabolism, biliary excretion, renal elimination, and the perpetrator's time-varying concentration.

16. Modeling the Victim and Perpetrator Drugs

A transporter-mediated DDI model contains two interacting systems.

This distinction is particularly important for transporter DDIs because the perpetrator concentration at the site of interaction may differ substantially from its plasma concentration.

A PBPK model can account for this difference by predicting concentrations in the relevant tissue rather than relying solely on a single systemic concentration.

17. A Practical Workflow for Transporter-Mediated PBPK

1. Define the DDI question. Identify the perpetrator, victim, route, dose, and clinical scenario.
2. Map the victim's disposition. Identify absorption, transporter, metabolic, renal, and biliary pathways.
3. Identify relevant transporters. Focus on transporters with plausible contributions to the victim's disposition or perpetrator's interaction potential.
4. Review in vitro evidence. Evaluate substrate and inhibitor data, assay systems, concentration ranges, and experimental limitations.
5. Build and verify the victim model. Establish that the model reproduces relevant clinical PK observations before adding the DDI.
6. Build the perpetrator model. Verify the perpetrator's concentration-time behavior and, where possible, relevant tissue exposure.
7. Implement the transporter mechanism. Represent inhibition, induction, or another interaction mechanism at the appropriate anatomical site.
8. Perform sensitivity analysis. Determine whether transporter parameters, binding, permeability, metabolism, or other assumptions dominate the prediction.
9. Simulate the DDI. Compare victim-only and victim-plus-perpetrator scenarios.
10. Evaluate against available clinical data. Where clinical DDI data exist, compare observed and predicted exposure changes.
11. Explore the intended clinical scenario. Once adequately qualified, use the model to examine doses, populations, schedules, or conditions that may not have been directly studied.

18. Common Challenges and Limitations

Transporter PBPK can be powerful, but it is also sensitive to uncertainties that are less prominent in simpler compartmental models.

• In vitro-to-in vivo translation. Transporter activity measured in an experimental system may not directly represent activity in human tissue.
• Transporter abundance. Expression levels can differ among tissues, individuals, and experimental systems.
• Free versus total concentration. Inhibitory potency may depend on the concentration relevant to the transporter rather than simply the measured total plasma concentration.
• Multiple transporters. A victim drug may be a substrate of several uptake and efflux transporters simultaneously.
• Transporter-enzyme interplay. Transport can determine intracellular access to metabolic enzymes, making pathways interdependent.
• Compensating pathways. Inhibition of one route may be partly offset by passive diffusion, metabolism, renal elimination, or another transporter.
• Parameter uncertainty. Several combinations of transporter parameters can sometimes produce similar clinical PK profiles.
• Clinical extrapolation. A model validated under one condition may require additional evaluation before being applied to a substantially different population or dosing scenario.

19. Transporter PBPK in Drug-Interaction Assessment

Transporter-mediated DDIs are part of contemporary drug-interaction evaluation. International guidance addresses both experimental characterization and the use of model-based approaches to support DDI assessment.

The ICH M12 framework provides recommendations for evaluating enzyme- and transporter-mediated pharmacokinetic interactions, including in vitro studies, clinical studies, predictive modeling, interpretation, risk assessment, and risk management.

For PBPK analyses, the model should be sufficiently documented that reviewers can understand the scientific basis of the simulation. Important elements include:

• Structural model and physiological assumptions.
• Drug-specific physicochemical and PK inputs.
• Transporter expression and activity assumptions.
• In vitro transporter data and scaling methods.
• Perpetrator inhibition or induction mechanism.
• Model verification and qualification.
• Clinical DDI predictions and observed comparisons where available.
• Sensitivity and uncertainty analyses.
• Intended application and limitations of extrapolation.

The FDA's PBPK reporting guidance similarly emphasizes clear documentation of the model, methods, results, discussion, and supporting information when PBPK analyses are submitted for regulatory purposes.

20. Why Sensitivity Analysis Is Essential

Transporter PBPK models can contain many parameters. Sensitivity analysis helps determine which assumptions actually control the DDI prediction.

Suppose the predicted AUCR is strongly sensitive to hepatic uptake clearance but relatively insensitive to passive permeability. That result suggests that uncertainty in transporter activity deserves particular attention.

Conversely, if the predicted AUCR changes very little across a wide range of transporter inhibition parameters, the clinical prediction may be robust to that particular uncertainty.

Sensitivity analysis is therefore not merely a statistical exercise. It helps identify which biological assumptions need the strongest experimental support.

Where to Go Next

A natural progression is to study specific transporter systems in greater detail. Useful next topics include P-gp and BCRP in PBPK, OATP-mediated hepatic uptake, OAT and OCT renal transport, MATE-mediated renal secretion, transporter inhibition models, and transporter-enzyme interplay.

From there, the next level is to build complete perpetrator-victim PBPK models and examine how transporter DDIs interact with CYP-mediated metabolism, UGT metabolism, renal clearance, protein binding, and nonlinear disposition.

References

1. U.S. Food and Drug Administration. M12 Drug Interaction Studies. Final Guidance for Industry. 2024.
2. European Medicines Agency / ICH. ICH M12 Guideline on Drug Interaction Studies. Step 5. 2024.
3. U.S. Food and Drug Administration. Physiologically Based Pharmacokinetic Analyses — Format and Content Guidance for Industry. 2018.
4. U.S. Food and Drug Administration. In Vitro Metabolism- and Transporter-Mediated Drug-Drug Interaction Studies Guidance for Industry.
5. European Medicines Agency. Guideline on the Investigation of Drug Interactions. Revision 1.
6. International Transporter Consortium. Giacomini KM, Huang S-M, Tweedie DJ, et al. Membrane transporters in drug development. Nature Reviews Drug Discovery.