Drug Transporters in PBPK Models

1. What Are Drug Transporters?

Drug transporters are membrane proteins that facilitate the movement of endogenous compounds, drugs, metabolites, and other substances across biological membranes. Unlike passive diffusion, transporter-mediated movement depends on the properties and activity of specific proteins.

Transporters can influence drug absorption, distribution, metabolism, and excretion. In a PBPK model, they can therefore provide a mechanistic connection between molecular properties and drug concentrations in specific organs or tissues.

2. Why Are Transporters Important in PBPK Models?

A conventional compartmental PK model may summarize disposition with parameters such as clearance and volume of distribution. A PBPK model instead attempts to represent physiological organs and tissues explicitly.

Because transporters are located at particular biological barriers, their effects can be represented at the corresponding physiological site. For example:

• Intestinal transporters can influence oral absorption and intestinal availability.
• Hepatic uptake transporters can facilitate movement of drug from blood into hepatocytes.
• Hepatic efflux transporters can move drug or metabolites from hepatocytes toward bile or back toward blood.
• Renal transporters can contribute to active tubular secretion or reabsorption.
• Blood-brain barrier transporters can influence drug entry into or removal from the central nervous system.

This spatial information is one of the principal reasons transporters can be important in mechanistic PBPK modeling.

3. Major Drug Transporter Systems

Transporters are commonly grouped into two broad superfamilies: ATP-binding cassette (ABC) transporters and solute carrier (SLC) transporters.

These systems are among the transporters specifically considered in modern regulatory drug-interaction guidance. The exact set of transporters included in a PBPK model should depend on the drug, organ, route, available evidence, and scientific question.

4. Uptake and Efflux Transporters

A crucial distinction is whether a transporter moves drug into a cell or out of a cell.

Uptake transport

Uptake transporters facilitate movement from the extracellular environment into cells. Hepatic uptake transporters such as OATP1B1 and OATP1B3 can increase the delivery of some drugs from blood into hepatocytes.

Efflux transport

Efflux transporters move substances from cells toward an extracellular compartment. P-gp and BCRP are important examples of efflux transporters.

The balance between uptake and efflux can be more informative than either process considered alone. A drug may have substantial uptake capacity but relatively little net intracellular accumulation if efflux is also strong.

5. Transporters at Physiological Barriers

PBPK models become especially useful for transporters because the location of a transporter is part of the model structure.

A transporter should therefore not be inserted into a PBPK model simply because it is known to interact with the drug. Its anatomical location and membrane orientation determine how the transporter can affect drug disposition.

6. How Are Transporters Represented Mathematically?

Transporter-mediated movement is often represented using a saturable kinetic relationship. A common starting point is a Michaelis-Menten-type expression:

Here, \(V_{\max}\) represents the maximum transport capacity, \(K_m\) is the concentration associated with half-maximal transport rate under the model, and \(C\) is the relevant drug concentration.

At concentrations much lower than \(K_m\):

Transport therefore behaves approximately linearly when the transporter is far from saturation.

At concentrations much greater than \(K_m\):

The transporter approaches its capacity and additional increases in concentration produce progressively smaller increases in transport rate.

7. Passive Diffusion Versus Transporter-Mediated Transport

Many PBPK models represent membrane movement as a combination of passive diffusion and active transporter processes.

A simplified passive component can be written as:

where \(PS\) is a permeability-surface-area term.

A transporter can then add a saturable contribution. For an uptake process:

An efflux transporter can be represented in the opposite direction using the relevant intracellular concentration.

This separation is useful because passive permeability and transporter activity have different biological meanings and can respond differently to changes in drug properties or interacting compounds.

8. Transporters in the Intestine

For an orally administered drug, intestinal transporters can influence how much drug crosses the intestinal wall and reaches the systemic circulation.

Efflux transporters such as P-gp and BCRP can reduce net transcellular movement into the portal circulation for susceptible substrates. Uptake transporters can act in the opposite direction for particular compounds.

In a mechanistic absorption model, transporter activity can therefore contribute to the relationship between luminal drug concentration, enterocyte concentration, and drug entering the portal circulation.

For example, increased intestinal efflux could reduce systemic exposure for a transporter substrate if efflux becomes an important barrier to absorption. The magnitude of the effect depends on passive permeability, transporter capacity, intestinal expression, dose, dissolution, and other processes represented by the PBPK model.

9. Hepatic Uptake and Efflux

The liver is one of the most important sites for transporter-mediated disposition.

For some drugs, passive diffusion is sufficient to provide hepatocyte access. For others, uptake transporters can make a substantial contribution to the movement of drug from blood into hepatocytes.

Once inside the hepatocyte, the drug may undergo metabolism, biliary secretion, intracellular binding, or return to the systemic circulation.

OATP1B1 and OATP1B3 are examples of hepatic uptake transporters. Several ABC transporters, including P-gp and BCRP, can contribute to efflux processes. Other transporter systems can also be relevant depending on the compound.

10. Transporters in the Kidney

Renal elimination is not simply filtration. Drug may be filtered at the glomerulus and may also undergo active tubular secretion or reabsorption.

Active secretion commonly involves coordinated transport at the basolateral and apical membranes of renal tubular cells.

For organic anions, OAT1 and OAT3 can contribute to uptake into proximal tubular cells. For organic cations, OCT2 can contribute to basolateral uptake, while MATE1 and MATE2-K can contribute to apical efflux into urine.

A mechanistic renal model can therefore represent active secretion as a transporter-mediated clearance pathway rather than assuming that renal clearance is determined entirely by glomerular filtration.

11. Transporters at Tissue Barriers

Transporters can also influence drug distribution into tissues. The blood-brain barrier is a particularly important example because endothelial cells form a highly selective interface between systemic blood and the central nervous system.

Efflux transporters such as P-gp and BCRP can limit brain exposure for susceptible substrates. In a PBPK model, these processes can be represented as part of the permeability and transporter properties of the blood-brain barrier.

The resulting brain concentration may therefore depend on:

• Passive membrane permeability.
• Transporter-mediated uptake.
• Transporter-mediated efflux.
• Unbound drug concentrations.
• Blood flow and tissue physiology.
• Binding within plasma and tissue.

This illustrates a broader PBPK principle: the same plasma concentration can produce different tissue concentrations depending on the physiological barriers and transport processes present at each tissue.

12. Transporter-Mediated Drug-Drug Interactions

A transporter-mediated drug-drug interaction (DDI) occurs when one drug changes the disposition of another through transporter inhibition, induction, competition, or other changes in transporter activity.

There are two important perspectives:

Victim drug

The victim drug is a substrate whose exposure changes because a transporter is inhibited or otherwise altered.

Perpetrator drug

The perpetrator drug is the compound that causes the change in transporter activity or expression.

For example, inhibiting a hepatic uptake transporter may reduce hepatocyte exposure for a transporter substrate. The resulting systemic exposure could increase or decrease depending on the relative importance of hepatic uptake, intracellular metabolism, biliary secretion, renal elimination, and other pathways.

This is why transporter DDIs should not be interpreted from a single pathway in isolation.

13. Transporters and Metabolic Enzymes Can Interact

Transporters and drug-metabolizing enzymes frequently operate sequentially within the same organ.

Consider a drug that must enter a hepatocyte before it can be metabolized:

Changing transporter activity can therefore change the intracellular concentration available to the enzyme. Conversely, changing metabolic activity can alter the amount of parent drug available for transport.

This creates potential transporter-enzyme interplay. A PBPK model can explicitly represent these linked processes when sufficient information is available.

14. The Main Transporter Parameters in PBPK

A transporter model may require several parameters. The exact parameterization depends on the software platform and mechanistic model.

One of the major practical challenges in transporter PBPK modeling is translating measurements from experimental systems into parameters that are representative of the relevant human physiological system.

15. From In Vitro Transport to In Vivo PBPK Parameters

Transporter activity is frequently measured in experimental systems such as transfected cells, membrane vesicles, primary hepatocytes, or other in vitro preparations.

The PBPK model, however, requires parameters representing transport in a human physiological context.

A conceptual scaling process is:

This process is not simply a unit conversion. Differences between experimental and physiological systems can include transporter abundance, membrane composition, competing pathways, protein binding, experimental substrate concentration, and cellular architecture.

Therefore, transporter scaling can introduce substantial uncertainty into PBPK predictions.

16. Transporter Saturation and Nonlinear PK

Because transporter-mediated movement can be saturable, transporter activity can produce nonlinear pharmacokinetics.

Suppose transporter-mediated elimination follows:

At low concentration, the apparent transporter clearance is approximately constant. As concentration increases and the transporter approaches saturation, the transporter contributes progressively less clearance per unit concentration.

This can produce an exposure relationship in which increasing the dose results in more-than-proportional increases in concentration or AUC.

17. How Is Transporter Inhibition Represented?

A simplified competitive inhibition relationship can be represented as:

where \(I\) is the inhibitor concentration and \(K_i\) represents an inhibition constant under the assumptions of the model.

As inhibitor concentration increases, the apparent concentration required to produce a given transport rate increases under this simple competitive model.

More complex inhibition mechanisms may require different equations. In addition, the concentration used for \(I\) in a PBPK model should correspond to the biologically relevant concentration at the transporter site rather than automatically being assumed to equal a single systemic plasma concentration.

Modern DDI frameworks combine in vitro transporter studies, clinical information, and mechanistic modeling to determine whether transporter-mediated interactions require further evaluation.

18. Transporter Induction

Transporter-mediated interactions are not limited to acute inhibition. Changes in transporter expression can alter transport capacity over time.

A simplified conceptual representation is:

In a PBPK model, this may be represented through a change in transporter abundance or effective capacity over time.

Induction can be particularly important when the transporter contributes materially to absorption or elimination. However, the magnitude and time course of induction should be supported by appropriate experimental and clinical evidence rather than assumed solely from an in vitro observation.

19. Worked Example: A Saturable Hepatic Uptake Transporter

Consider a hypothetical drug that undergoes hepatic uptake through a transporter before intracellular metabolism.

Suppose the transporter has:

• V_max = 100 mg/h
• K_m = 10 mg/L
• Hepatic extracellular drug concentration C = 5 mg/L

Step 1: Calculate transporter-mediated uptake

The transporter therefore carries approximately 33.3 mg/h at this concentration under the assumed model.

Step 2: Examine a higher concentration

If concentration increases to 50 mg/L:

The concentration increased tenfold, but transport increased only from 33.3 to 83.3 mg/h because the transporter is approaching its maximum capacity.

Step 3: Examine the limiting behavior

This illustrates the central consequence of transporter saturation: once transport capacity is approached, additional increases in concentration produce relatively small increases in transporter-mediated flux.

20. Worked Example: Transporter Inhibition in a PBPK Model

Now suppose the same hypothetical transporter is inhibited by a perpetrator drug. Assume a simplified competitive inhibition model with:

• K_m = 10 mg/L
• V_max = 100 mg/h
• C = 5 mg/L
• I = 10 mg/L
• K_i = 10 mg/L

Step 1: Calculate the inhibition factor

Step 2: Calculate inhibited transport

Under this simplified model, transporter-mediated uptake decreases from approximately 33.3 mg/h to 20 mg/h.

Step 3: Interpret the result

The immediate model result is a reduction in hepatic uptake flux. The resulting plasma exposure cannot be determined from this transporter equation alone because the effect depends on the rest of the disposition system.

If hepatic uptake is rate-limiting for intracellular metabolism, systemic exposure may increase. If other elimination pathways compensate, the systemic effect may be smaller.

21. How Transporters Fit Into a PBPK Model

A mechanistic PBPK model can be viewed as a network of organs connected by blood flow. Transporters modify the drug flux across selected membranes within that network.

This allows the model to distinguish, for example, an intestinal efflux effect from hepatic uptake inhibition even if both ultimately change plasma exposure.

22. Sensitivity Analysis for Transporter Parameters

Transporter parameters can be uncertain. Sensitivity analysis can determine whether that uncertainty materially affects the model prediction.

Parameters commonly examined include:

• Transporter abundance.
• V_max.
• K_m.
• Passive permeability.
• Inhibition constants.
• Fraction of clearance mediated by the transporter.

A simple local sensitivity measure can be expressed conceptually as:

where \(Y\) is an output such as AUC and \(\theta\) is a model parameter.

A parameter can be uncertain without being important if changes in that parameter have little effect on the prediction. Conversely, a parameter with relatively modest uncertainty can become important when the model output is highly sensitive to it.

23. How Should Transporter PBPK Models Be Evaluated?

A transporter PBPK model should be evaluated against evidence relevant to the intended application.

1. Verify the structural implementation. Confirm that transporter direction, location, and equations are implemented correctly.
2. Check in vitro behavior. Determine whether the model reproduces the experimental transport characteristics used to parameterize it.
3. Evaluate clinical PK. Compare predictions with observed human concentration-time data when available.
4. Evaluate transporter DDIs. Where applicable, assess whether the model reproduces known clinical interaction studies.
5. Perform sensitivity analysis. Identify parameters that materially affect predictions.
6. Test extrapolation carefully. Predictions for untested conditions should remain within a scientifically justified domain of applicability.

Regulatory PBPK guidance emphasizes documenting model assumptions, inputs, methods, results, and limitations. A transporter model should therefore be transparent enough that reviewers can understand how transporter evidence contributes to the final prediction.

24. Sources of Uncertainty in Transporter PBPK Models

Transporter models can be especially sensitive to biological uncertainty because transporter systems are complex and tissue-specific.

25. Transporters in Regulatory PBPK Applications

Transporter-mediated drug interactions are now explicitly incorporated into international regulatory DDI frameworks. The ICH M12 guideline provides recommendations for evaluating enzyme- and transporter-mediated pharmacokinetic drug interactions, including in vitro studies, clinical studies, and model-based approaches.

ICH M12 identifies circumstances in which clinical transporter DDI evaluation should be considered. For example, P-gp and BCRP may be particularly relevant when intestinal absorption is limited or when biliary excretion or active renal secretion is an important elimination pathway. OATP1B1/OATP1B3 are relevant when hepatic elimination or hepatic transporter-mediated access is important, while OAT1/OAT3, OCT2, MATE1, and MATE2-K are important considerations when active renal secretion contributes materially to clearance.

FDA PBPK guidance also emphasizes documenting the scientific rationale, model structure, input parameters, verification, results, and limitations when PBPK analyses are used for regulatory purposes.

Regulatory acceptance remains application-specific: the scientific question, quality of the data, model qualification, and intended use all matter.

26. A Practical Workflow for Building a Transporter PBPK Model

1. Define the scientific question. Are you predicting absorption, organ exposure, clearance, or a transporter-mediated DDI?
2. Identify relevant transporters. Consider substrate evidence, tissue expression, elimination pathways, and known DDI information.
3. Determine transporter location and direction. Identify whether the transporter is apical, basolateral, sinusoidal, canalicular, or located at another physiological barrier.
4. Characterize transporter kinetics. Determine whether available data support linear or saturable transport and identify relevant kinetic parameters.
5. Characterize passive permeability. Separate transporter-mediated movement from passive diffusion where possible.
6. Scale transporter activity. Translate experimental activity to the relevant human tissue and physiological context.
7. Integrate with metabolism. Determine whether transporter activity controls access to metabolic enzymes or other intracellular processes.
8. Integrate renal and biliary pathways. Represent transporter-mediated elimination alongside filtration, metabolism, and other routes.
9. Build the DDI mechanism if required. Represent inhibition or induction using evidence-supported parameters.
10. Verify the model. Compare predictions with independent experimental or clinical observations.
11. Perform sensitivity and uncertainty analysis. Determine which assumptions materially affect the prediction.
12. Document the model. Clearly describe assumptions, data sources, scaling methods, equations, verification, and limitations.

27. Common Mistakes When Modeling Transporters

Mistake 1: Treating every transporter as a clearance term

A transporter can influence absorption, distribution, intracellular access, or elimination. Its effect should be represented at the appropriate physiological interface.

Mistake 2: Ignoring transporter direction

Uptake and efflux have opposite effects on membrane flux. Direction is part of the biological mechanism.

Mistake 3: Assuming an in vitro V_max is directly usable in humans

Experimental transporter capacity generally requires appropriate scaling before being interpreted as a physiological human parameter.

Mistake 4: Ignoring passive diffusion

Transporter activity should generally be considered alongside passive permeability rather than as an isolated mechanism.

Mistake 5: Assuming one transporter explains the entire phenotype

Drugs can interact with multiple transporters, metabolic enzymes, binding processes, and elimination pathways simultaneously.

Mistake 6: Equating transporter inhibition with a specific AUC change

The final systemic effect depends on the complete disposition network. A reduction in transporter flux does not by itself determine the direction or magnitude of the plasma exposure change.

Mistake 7: Overparameterizing the model

Adding poorly identified transporter parameters can increase complexity without increasing predictive reliability.

28. Putting the Pieces Together

Consider a hypothetical orally administered drug with the following properties:

• Moderate passive intestinal permeability.
• Intestinal P-gp efflux.
• Hepatic OATP-mediated uptake.
• Intracellular CYP metabolism.
• Renal secretion involving an uptake/efflux transporter pathway.

A mechanistic PBPK model could represent the drug's journey as:

Transporters enter at several specific points in this chain. P-gp can influence intestinal efflux; OATP transporters can influence hepatic uptake; intracellular metabolism determines one component of hepatic elimination; and renal transporters can influence active secretion.

Now imagine that a perpetrator inhibits both intestinal P-gp and hepatic OATP uptake. The net plasma exposure cannot be inferred by simply adding the two transporter effects. Increased intestinal availability could increase systemic input, while reduced hepatic uptake could reduce intracellular metabolism. The PBPK model allows these mechanisms to interact within the same physiological system.

29. Transporter PBPK Versus an Empirical PK Model

An empirical PK model might describe the observed consequence of transporter activity through an estimated clearance or absorption parameter. A PBPK model attempts to represent where and why that effect occurs.

The tradeoff is important. Mechanistic detail can improve the ability to connect data across conditions, but it also introduces additional assumptions and parameters that must be justified and evaluated.

Where to Go Next

A natural progression after transporter fundamentals is to study hepatic uptake and biliary transport in PBPK, followed by renal active secretion, intestinal efflux, transporter-mediated drug-drug interactions, and transporter-enzyme interplay.

These topics build toward full mechanistic DDI modeling, where transporter activity, CYP metabolism, organ physiology, and perpetrator concentrations are integrated into a single PBPK framework.

References

1. FDA / ICH. M12 Drug Interaction Studies. Final guidance, August 2024. FDA guidance page →
2. FDA. In Vitro Metabolism- and Transporter-Mediated Drug-Drug Interaction Studies Guidance for Industry. FDA guidance PDF →
3. FDA. Physiologically Based Pharmacokinetic Analyses — Format and Content Guidance for Industry. 2018. FDA guidance page →
4. FDA. The Use of Physiologically Based Pharmacokinetic Analyses — Biopharmaceutics Applications for Oral Drug Product Development, Manufacturing Changes, and Controls. 2020 draft guidance. FDA guidance page →
5. EMA / ICH. ICH M12 Drug Interaction Studies. Guideline addressing enzyme- and transporter-mediated pharmacokinetic drug-drug interactions, including in vitro studies, clinical studies, and model-based evaluation. EMA guideline page →
6. FDA. Drug Development and Drug Interactions: Table of Substrates, Inhibitors and Inducers. FDA transporter and DDI tables →