Why Unbound Fractions Matter in Kp Prediction
A tissue-to-plasma partition coefficient, \(K_p\), compares total drug concentrations in tissue and plasma. Mechanistic prediction methods estimate how drug distributes among tissue water, lipids, and other binding environments. Their outputs can depend on unbound fractions, but only if the concentration basis is clear: a partition term referenced to unbound drug is not automatically a total tissue-to-total plasma \(K_p\).
The essential bookkeeping is simple. Plasma binding determines how much total plasma drug corresponds to a given unbound plasma concentration. Tissue binding determines how much total tissue drug corresponds to a given unbound tissue concentration. At equilibrium, if the unbound concentrations in the two compartments are equal, their total-concentration ratio is governed by the ratio of those binding capacities.
Unbound fractions are dimensionless and lie between zero and one when defined as fractions of the same total concentration. In practice, the tissue value must be tied to a clear basis, such as intact tissue or a corrected homogenate estimate. A value from one basis should not be silently substituted for a value from another.
The Equilibrium Link Between Binding and Kp
Assume passive distribution has reached equilibrium, the measured total concentrations refer to compatible plasma and tissue samples, and unbound drug has the same concentration in the two compartments. Also assume that the unbound fractions are applicable at the concentrations and conditions represented by the prediction. Write \(C_{u,p}\) and \(C_{u,t}\) for unbound plasma and tissue concentrations, and \(C_p\) and \(C_t\) for their corresponding total concentrations. By definition, \(C_{u,p}=f_{u,p}C_p\) and \(C_{u,t}=f_{u,t}C_t\). Under the equal-unbound-concentration assumption, \(C_{u,p}=C_{u,t}\), so:
This relation is a useful consistency check, not a universal shortcut for every prediction method. It applies when equal unbound concentrations are a reasonable equilibrium description and when both unbound fractions use compatible definitions. Ionization and pH differences can be represented through a more detailed mechanistic model; transport processes or failure to reach equilibrium can make equal unbound concentrations an unsuitable assumption.
Composition-based methods often estimate tissue binding capacity rather than taking \(f_{u,t}\) as a direct input. Let \(B_t\) denote the predicted ratio of total tissue concentration to free concentration in tissue water. A method might construct it from tissue components:
Here \(\phi_j\) is the fraction of tissue volume assigned to component \(j\), and \(K_j\) is that component's concentration relative to free aqueous drug concentration. The sum is an illustrative composition-based bookkeeping form: the actual terms and their definitions depend on the selected method. The expression for \(K_p\) follows because the method estimates \(C_t/C_{u,t}=B_t\), while plasma contributes \(C_{u,p}/C_p=f_{u,p}\); with equal unbound concentrations, \(C_t/C_p=B_t f_{u,p}\). If the method instead estimates \(f_{u,t}\), then \(B_t=1/f_{u,t}\), giving the same equilibrium relation.
Worked Examples
Worked Example: Converting a Tissue Binding Capacity to Kp
Suppose a hypothetical composition-based calculation estimates \(B_t=18.0\), meaning total tissue concentration is predicted to be 18.0 times the tissue free concentration. Let \(f_{u,p}=0.060\), and assume equal unbound tissue and plasma concentrations at equilibrium. Calculate the total tissue-to-plasma \(K_p\).
The method's tissue estimate is referenced to unbound concentration, so apply the plasma unbound fraction once:
The predicted total tissue concentration is 1.08 times the total plasma concentration under these assumptions. The corresponding tissue unbound fraction implied by this binding capacity is \(f_{u,t}=1/18.0=0.0556\). Checking the equilibrium ratio gives \(f_{u,p}/f_{u,t}=0.060/0.0556\approx1.08\), consistent with the composition calculation.
Worked Example: Effect of Changing Plasma Binding
A hypothetical tissue model gives the same \(B_t=25.0\) under two plasma-binding scenarios. In scenario A, \(f_{u,p}=0.080\); in scenario B, \(f_{u,p}=0.032\). Assume the tissue estimate is referenced to free aqueous drug and that the equilibrium assumptions hold. Calculate and compare the predicted \(K_p\) values.
For scenario A:
For scenario B:
The plasma unbound fraction in scenario B is \(0.032/0.080=0.400\) times that in scenario A, and the predicted \(K_p\) is also 0.400 times as large. This follows because the tissue binding capacity was held fixed: lower \(f_{u,p}\) means less unbound drug per unit total plasma concentration, and therefore a lower tissue-to-total-plasma ratio at equal unbound concentrations. It does not imply that tissue binding itself changed.
Worked Example: Detecting a Double Binding Correction
A hypothetical method reports \(K_p=1.60\) as a total tissue-to-total plasma ratio. Its documentation states that the plasma unbound fraction \(f_{u,p}=0.040\) was already included. An analyst multiplies the reported result by \(0.040\) again and reports \(0.064\). Identify the error and recover the method's implied tissue binding capacity.
Because the reported value already includes plasma binding, it is the total-concentration \(K_p\); it must not be multiplied by \(f_{u,p}\) again. Under the stated equilibrium formulation, the implied tissue capacity is:
Applying the plasma factor once to that capacity returns the original result: \((0.040)(40.0)=1.60\). The analyst's extra multiplication changed the quantity to \(0.040(1.60)=0.064\), which has no justification as the same total-to-total \(K_p\). The mistake arose from treating a completed \(K_p\) as though it were still a tissue-to-unbound concentration term.
How to Keep Method Conventions Straight
Rodgers–Rowland, Schmitt, and Berezhkovskiy approaches organize composition and binding contributions differently. Earlier material on these methods establishes their tissue-component terms and corrections. When using a particular implementation, inspect what its reported quantity represents rather than relying on the method name alone. An equation may calculate a tissue capacity relative to free aqueous drug, include plasma binding in a final conversion, or express partitioning relative to a plasma reference that already accounts for binding.
| Quantity supplied or reported | Interpretation | Binding step to check |
|---|---|---|
| \(B_t=C_t/C_{u,t}\) | Tissue total-to-free capacity | Multiply by \(f_{u,p}\) to obtain total-to-total \(K_p\), if equal unbound concentrations apply. |
| \(f_{u,t}\) | Unbound fraction on a stated tissue basis | Use \(K_p=f_{u,p}/f_{u,t}\) only under the equilibrium and compatible-basis assumptions. |
| Total-concentration \(K_p\) | Predicted \(C_t/C_p\) | Do not apply a further plasma binding correction if already included. |
| Component partition term | Drug concentration in a component relative to a stated reference | Confirm whether the reference is free aqueous drug, unbound plasma, or total plasma. |
A reliable calculation records three items beside every input: its physical meaning, the concentration basis, and whether it has already been used in a correction. In particular, “unbound fraction” alone is not a complete specification. The relevant fraction may be for plasma or tissue; a tissue value may also depend on the assay and its correction conventions.
Assumptions, Diagnostics, and Limitations
The simple fraction relationship is most defensible when equilibrium has been reached, unbound concentrations are equal across the two compartments, and the unbound fractions remain appropriate for the conditions being modeled. A composition-based calculation adds assumptions about tissue composition and how each component contributes to drug concentration. If a method accounts explicitly for ionization, pH, or specific component interactions, its own terms should be used consistently rather than replacing them with a single \(f_{u,t}\) without justification.
Several checks can expose a convention or implementation problem:
- Check the direction of the ratio. For \(K_p=C_t/C_p\), a smaller \(f_{u,p}\) lowers \(K_p\) when tissue capacity is fixed. Reversing the ratio reverses that interpretation.
- Check whether \(f_{u,p}\) is already included. Multiplying a total-to-total prediction by \(f_{u,p}\) again is double correction.
- Check reference concentrations. A tissue-to-free-plasma term and a tissue-to-total-plasma \(K_p\) are different quantities even if both are described informally as partitioning.
- Check tissue basis and units. Fractions are dimensionless, but tissue capacity and \(K_p\) are meaningful only when the concentration and volume conventions in their definitions agree.
- Check plausibility without treating it as proof. A value inconsistent with the stated \(f_{u,p}\), \(f_{u,t}\), or \(B_t\) may indicate a calculation error or violated assumptions. Plausibility alone cannot establish equilibrium or validate the model.
The alternatives answer related but not identical questions. A composition-based method estimates tissue distribution from tissue components and physicochemical properties; an experimentally determined tissue unbound fraction can provide a direct binding input, subject to its assay basis. The simple ratio \(f_{u,p}/f_{u,t}\) is a compact equilibrium relationship, not a substitute for component modeling when the components, ionization, or other mechanisms are needed to explain tissue distribution. Conversely, a detailed composition model does not make inconsistent binding conventions harmless: its output still needs a clearly defined reference basis.
The ratio can also fail as a practical prediction if equilibrium is not attained, unbound concentrations are not equal, or the chosen fractions do not represent the conditions of interest. A tissue homogenate fraction, for example, may not equal the effective unbound fraction in intact tissue unless the experimental method and any corrections support that interpretation. These are reasons to interpret a calculated \(K_p\) as conditional on its model and inputs, rather than as an automatic consequence of measuring two fractions.
Check Your Understanding
Answer each question using the stated definitions and assumptions. Short solutions follow the questions.
- A method estimates \(B_t=12.0\), and \(f_{u,p}=0.050\). Under equal unbound concentrations, what is \(K_p\)?
- If \(K_p=0.75\) and \(f_{u,p}=0.030\), what tissue capacity \(B_t\) is implied by the composition-based relationship?
- A measured \(f_{u,p}=0.10\) and \(f_{u,t}=0.025\) are used under the equilibrium assumptions. Calculate \(K_p\).
- Why is multiplying a reported total-concentration \(K_p\) by \(f_{u,p}\) potentially wrong?
- Name one assumption that could make \(K_p=f_{u,p}/f_{u,t}\) an unsuitable description of an observed tissue-to-plasma ratio.
Solutions. 1. \(K_p=f_{u,p}B_t=(0.050)(12.0)=0.600\). 2. \(B_t=K_p/f_{u,p}=0.75/0.030=25.0\). 3. \(K_p=0.10/0.025=4.00\). 4. The reported total-concentration ratio may already include plasma binding, so another multiplication would apply the correction twice. 5. For example, equilibrium may not have been reached, unbound concentrations may not be equal, or the fractions may use incompatible concentration bases.