Worked Example: Receptor-Occupancy MABEL

From Receptor Occupancy to a Dose

A receptor-occupancy MABEL calculation links mechanistic evidence to a predicted human dose: select an occupancy level intended to represent minimal pharmacological effect, calculate the concentration associated with that occupancy, and use predicted human pharmacokinetics (PK) to find the dose expected to produce it. The key practical detail is that receptor occupancy depends on free drug concentration, while a PK prediction may report total drug concentration. When the amount of target is not negligible relative to drug, some drug is bound to receptors, so the total concentration needed to reach the target occupancy is higher than the free concentration alone.

The earlier material on receptor occupancy established the basic binding relation. Here, the calculation makes receptor abundance explicit and carries the result through a hypothetical human PK prediction. The chosen occupancy threshold is an input to the calculation, not a universal biological constant: its suitability depends on the mechanism, pharmacology, and evidence for what level of engagement could produce an effect.

Definition: For a reversible, one-to-one drug–receptor interaction at equilibrium, let \(K_D\) be the equilibrium dissociation constant, \(C_f\) the free drug concentration, and \(R_T\) the total concentration of accessible receptor. The fractional receptor occupancy is \(\theta\), the fraction of receptor sites occupied. Under the assumptions stated here, \(\theta=C_f/(K_D+C_f)\), and the concentration of drug bound to receptor is \(R_T\theta\).

The binding relation assumes a single class of accessible sites, a one-to-one interaction, and equilibrium between free and bound drug. It also treats the concentrations as referring to the same relevant compartment. \(K_D\) and \(C_f\) must have the same concentration units; \(R_T\) must use those units as well. If the target is in tissue, a plasma concentration is not automatically the concentration at the target site.

Include Receptor Abundance Through Mass Balance

The occupancy equation gives the free concentration required for a chosen occupancy. If the PK model predicts total concentration in the same compartment, mass balance is needed to account for bound drug. In this simplified model, total drug concentration is free drug plus receptor-bound drug:

C_T=C_f+R_T\theta

For a specified threshold \(\theta^*\), the occupancy equation can be rearranged to find the required free concentration. Substituting that result into mass balance gives the total concentration threshold:

C_f^*=K_D\frac{\theta^*}{1-\theta^*}, \qquad C_T^*=K_D\frac{\theta^*}{1-\theta^*}+R_T\theta^*

This derivation assumes \(0\leq\theta^*<1\). It shows why receptor abundance matters when converting an occupancy target to total drug concentration: at the chosen occupancy, the bound amount is \(R_T\theta^*\). If receptor abundance is negligible compared with the free concentration, then \(C_T^*\) is approximately \(C_f^*\). If PK predicts free concentration directly, occupancy is calculated from that free concentration and receptor abundance does not enter the occupancy equation itself.

A useful check in the reverse direction is to calculate occupancy from a predicted total concentration. Under this model, the physical solution is the root between zero and one of:

R_T\theta^2-(C_T+K_D+R_T)\theta+C_T=0

This follows by substituting \(C_f=C_T-R_T\theta\) into the binding relation and rearranging. In a dose calculation, however, it is often clearer to calculate \(C_T^*\) from the chosen occupancy first, then compare that threshold with the PK prediction.

Worked Dose Calculation

Worked Example: A 1% Occupancy Threshold

All data are hypothetical. Suppose a reversible one-to-one ligand has \(K_D=0.0200\) nM and the accessible receptor concentration in the modeled compartment is \(R_T=0.00200\) nM. A mechanistic assessment selects \(\theta^*=0.0100\), or 1.00%, as the low-occupancy threshold for this calculation. A linear human PK prediction gives a total concentration of \(0.00200\) nM per mg of administered dose at the exposure time relevant to the anticipated effect. Find the dose predicted to reach the threshold.

First calculate the free concentration needed for 1.00% occupancy:

C_f^*=0.0200\ \mathrm{nM}\times\frac{0.0100}{1-0.0100} =\frac{0.000200}{0.9900}\ \mathrm{nM} =0.000202020\ldots\ \mathrm{nM}

At 1.00% occupancy, the bound drug concentration is:

R_T\theta^*=0.00200\ \mathrm{nM}\times0.0100 =0.0000200\ \mathrm{nM}

Therefore, the total concentration required is:

C_T^*=0.000202020\ldots+0.0000200 =0.000222020\ldots\ \mathrm{nM}

Using the stated PK coefficient, the threshold dose is:

D^*=\frac{0.000222020\ldots\ \mathrm{nM}} {0.00200\ \mathrm{nM/mg}} =0.111010\ldots\ \mathrm{mg} \approx0.111\ \mathrm{mg}

The result is a hypothetical occupancy-based MABEL dose under the specified model and threshold. The units reduce to mg because concentration is divided by concentration per mg. At this predicted dose, the model gives the total concentration required for 1.00% occupancy; it does not establish that 1.00% occupancy will produce a particular clinical effect.

Why Receptor Abundance and PK Must Match

Worked Example: Changing the Target Abundance

All data are hypothetical. Keep \(K_D=0.0200\) nM, \(\theta^*=0.0100\), and the PK coefficient \(0.00200\) nM per mg from the previous calculation, but suppose the accessible receptor concentration is \(R_T=0.0200\) nM. Calculate the dose threshold and explain the change.

The free concentration required by the occupancy threshold is unchanged, because \(K_D\) and \(\theta^*\) are unchanged:

C_f^*=0.0200\ \mathrm{nM}\times\frac{0.0100}{0.9900} =0.000202020\ldots\ \mathrm{nM}

The bound concentration is now:

R_T\theta^*=0.0200\ \mathrm{nM}\times0.0100 =0.000200\ \mathrm{nM}

Thus the total threshold concentration and dose are:

C_T^*=0.000202020\ldots+0.000200 =0.000402020\ldots\ \mathrm{nM}

D^*=\frac{0.000402020\ldots\ \mathrm{nM}} {0.00200\ \mathrm{nM/mg}} =0.201010\ldots\ \mathrm{mg} \approx0.201\ \mathrm{mg}

The dose is higher because, at the same fractional occupancy, the larger receptor pool binds more drug. This comparison is valid only if \(R_T\), the binding concentrations, and the PK concentration refer to a mutually consistent compartment. It would be misleading to combine a tissue receptor abundance with plasma PK as though they described one well-mixed compartment without evidence supporting that link.

Translate a Concentration Threshold Through Predicted PK

Let \(C_T(t;D)\) denote the predicted total drug concentration at time \(t\) after dose \(D\). Under linear, dose-proportional PK, suppose \(C_T(t;D)=a(t)D\), where \(a(t)\) is the predicted concentration per unit dose at that time. If occupancy is expected to matter at a specified time, the corresponding dose is \(D^*=C_T^*/a(t)\). If the minimal-effect criterion concerns avoiding a threshold throughout a time course, the relevant exposure metric and time window must be stated; using a peak concentration is appropriate only when the threshold is tied to peak-driven effect.

Worked Example: Using a Predicted Concentration-Time Profile

All data are hypothetical. Use the first example’s \(C_T^*=0.000222020\ldots\) nM. Suppose a linear human PK prediction gives \(0.00400\) nM per mg at the predicted peak and \(0.000800\) nM per mg at 24 hours. If the concern is that occupancy might reach the selected threshold at the peak, calculate the dose threshold using the peak prediction.

The dose corresponding to the predicted peak is:

D^*_{\mathrm{peak}}= \frac{0.000222020\ldots\ \mathrm{nM}} {0.00400\ \mathrm{nM/mg}} =0.0555051\ldots\ \mathrm{mg} \approx0.0555\ \mathrm{mg}

At 24 hours, the dose that would produce the same total concentration threshold is:

D^*_{24\ \mathrm{h}}= \frac{0.000222020\ldots\ \mathrm{nM}} {0.000800\ \mathrm{nM/mg}} =0.277525\ldots\ \mathrm{mg} \approx0.278\ \mathrm{mg}

The peak-based dose is lower because the predicted concentration per mg is higher at the peak. Under the stated objective, \(0.0555\) mg is the dose predicted to reach the threshold at peak. It is not a dose predicted to maintain the threshold at 24 hours. The two calculations answer different exposure questions, so the selected PK time or summary must match the biological rationale.

Assumptions, Checks, and Alternatives

Before treating the arithmetic as a MABEL estimate, check that its assumptions are defensible for the intended use:

Binding: the one-to-one equilibrium model is an adequate representation at the concentrations of interest. If binding changes materially over the PK time course, equilibrium may be inadequate; binding kinetics may need to be represented.
Target abundance: the value is a relevant estimate of accessible receptor concentration in the compartment being modeled, not simply a total receptor amount from a different tissue or assay.
Concentration basis: \(K_D\), \(R_T\), and \(C_T\) use compatible molar units. The calculation treats \(C_T\) as free plus receptor-bound drug in the modeled compartment; additional binding or distribution pools are not included.
PK mapping: the dose-to-concentration prediction applies to the contemplated dose and population. The examples assume linear dose proportionality; nonlinear PK or target-mediated disposition can invalidate a constant concentration-per-dose coefficient.
Effect threshold: the selected occupancy is justified as a conservative marker of minimal effect for the mechanism. A numerical occupancy threshold alone does not establish biological irrelevance below it or a clinical effect at it.

A diagnostic calculation is to reconstruct the target occupancy from the dose and predicted total concentration. Convert the dose to \(C_T\), solve the mass-balance equation for the root \(0\leq\theta\leq1\), and check that it agrees with the intended threshold. Also check units at every step and confirm that the predicted concentration is evaluated at the time or exposure metric used to define the criterion.

A common mistake is to insert total concentration directly into \(\theta=C/(K_D+C)\) when receptor-bound drug is not negligible. That formula uses free concentration: total concentration includes drug already bound to receptors. The correction is to calculate the free threshold from \(K_D\) and \(\theta^*\), add \(R_T\theta^*\) to obtain total concentration, and then use a total-concentration PK prediction. If receptor binding is negligible, the simpler calculation may be a good approximation, but that should be justified rather than assumed.

An occupancy-based MABEL is one mechanistic approach to choosing a low starting exposure. A functional in vitro assay may instead identify a low concentration associated with a measured response; that response concentration can be informative, but it is not automatically equivalent to a receptor-occupancy threshold. A toxicology-based approach uses a different evidence base and does not replace the mechanistic occupancy calculation. These approaches can answer different questions, and this occupancy calculation alone is not a complete starting-dose decision.

Important limitations include uncertainty in \(K_D\), receptor abundance, the selected occupancy threshold, and the human PK prediction. The calculation also simplifies the biology to one accessible receptor pool and an equilibrium interaction. If these simplifications are not credible, the numerical dose should not be presented as precise merely because the arithmetic is precise; a more suitable mechanistic model or evidence-based concentration-response criterion is needed.

Key takeaway: For an equilibrium one-to-one model, a chosen occupancy threshold implies a free concentration \(K_D\theta^*/(1-\theta^*)\). Add \(R_T\theta^*\) when translating that threshold to total concentration, then use a matching human PK prediction to calculate the dose.

Check Your Understanding

Try each question, then compare your reasoning with the solutions.

If \(K_D=0.0500\) nM and the chosen occupancy threshold is 2.00%, what free concentration is required?
For that threshold, suppose \(R_T=0.0100\) nM. What is the total concentration threshold?
If the total concentration threshold is \(0.00110\) nM and predicted concentration is \(0.00100\) nM per mg, what dose reaches the threshold?
Why can a total-concentration PK prediction not always be inserted directly into the free-concentration occupancy equation?

Solutions. (1) \(C_f^*=0.0500\times(0.0200/0.9800)=0.0010204\ldots\) nM, or 0.00102 nM. (2) The bound concentration is \(0.0100\times0.0200=0.000200\) nM, so \(C_T^*=0.0010204\ldots+0.000200=0.0012204\ldots\) nM, or 0.00122 nM. (3) \(D=0.00110\ \mathrm{nM}/(0.00100\ \mathrm{nM/mg})=1.10\) mg. (4) Occupancy depends on free drug. Total drug can include receptor-bound drug, so first account for that bound amount when translating between total concentration and occupancy.