Residence Time and Occupancy Persistence

Why Occupancy Can Persist After Exposure Falls

A drug concentration can fall while target occupancy remains substantial. One possible explanation is slow dissociation: ligand already bound to the target leaves gradually. Another is sustained free exposure: ligand continues to be available to bind, so dissociated target can become occupied again. These mechanisms can produce similar occupancy observations, but they imply different relationships between concentration, binding kinetics, and the duration of target engagement.

This tutorial relates the dissociation rate constant \(k_{\mathrm{off}}\) to target residence time and derives occupancy decline after free ligand is removed. The distinction is mechanistic: a long residence time describes the lifetime of an individual bound complex under a first-order dissociation model; it does not by itself establish how long occupancy will remain high in a biological system.

Occupancy Decline Under Complete Washout

The binding model for a single reversible ligand and a homogeneous target population is

Here \(k_{\mathrm{on}}\) is the association rate constant and \(C_{\mathrm{free}}(t)\) is the free ligand concentration. The first term describes binding to unoccupied target, while the second describes dissociation from occupied target. During ideal washout, free ligand is removed rapidly and remains zero, so the association term is zero. The resulting equation is \(dR/dt=-k_{\mathrm{off}}R\).

For an initial occupancy \(R(0)=R_0>0\), separation of variables and integration give the exponential decline below. The positive initial value is required for the logarithm in this separation. If \(R_0=0\), the equation directly gives the constant solution \(R(t)=0\), which is also consistent with the exponential expression.

Because the exponential factor is dimensionless, \(k_{\mathrm{off}}t\) must be dimensionless: if time is measured in hours, \(k_{\mathrm{off}}\) is measured in \(\mathrm{h}^{-1}\). The occupancy half-life follows by setting \(R(t)/R_0=1/2\). The time to decline to any positive threshold \(q\leq R_0\) follows by solving \(q=R_0e^{-k_{\mathrm{off}}t}\).

The occupancy half-life is independent of \(R_0\) under this model, but the time to a fixed absolute threshold depends on the starting occupancy. In the ideal first-order model, the bound-state lifetime has an exponential distribution with survival probability \(P(T>t)=e^{-k_{\mathrm{off}}t}\). Its mean is \(1/k_{\mathrm{off}}\), whereas its median is \(\ln(2)/k_{\mathrm{off}}\). Thus “residence time” should be specified as the mean lifetime when used in this standard sense; it is not the time for all bound ligand to dissociate.

Worked Washout Calculation

Applying the exponential equation requires more than knowing a dissociation rate. The washout must make free ligand negligible at the target, association must therefore be negligible, and dissociation must be adequately described by a single first-order rate constant over the interval. The target population is treated as homogeneous, with no synthesis, degradation, or interconversion changing the amount of available target. These are model assumptions, not automatic consequences of measuring a declining concentration.

Residual Free Ligand Changes the Prediction

If free ligand remains, occupancy does not generally decline according to the zero-concentration washout equation. For constant \(C_{\mathrm{free}}>0\), the binding equation can be written as a linear differential equation. Define the equilibrium occupancy at that constant concentration as \(R_{\mathrm{eq}}=k_{\mathrm{on}}C_{\mathrm{free}}/(k_{\mathrm{on}}C_{\mathrm{free}}+k_{\mathrm{off}})\). Solving the equation gives a trajectory toward \(R_{\mathrm{eq}}\), rather than a decline toward zero:

The exponential rate is the sum of association and dissociation contributions. When free concentration is zero, \(R_{\mathrm{eq}}=0\), and this solution reduces to the complete-washout equation. When free concentration is positive, ongoing association can offset dissociation. Therefore, persistent occupancy alone is not evidence of a long residence time.

Interpreting Data and Recognizing Model Failure

Under complete washout, the model implies that \(\ln(R(t)/R_0)\) is a straight line against time with slope \(-k_{\mathrm{off}}\), provided \(R(t)>0\) and occupancy is measured on a suitable scale. A roughly linear pattern is compatible with the model but does not prove its assumptions. Curvature, a changing apparent slope, or a nonzero late plateau can flag model mismatch. Possible explanations include residual free ligand, target turnover, more than one dissociation process, changing assay performance, or random measurement error. A pattern alone does not identify which explanation is correct.

Useful checks include verifying that free concentration has fallen sufficiently at the relevant target site, examining whether the occupancy assay can reliably quantify values near the predicted lower range, and checking whether target amount changes over the observation period. The washout curve should be interpreted alongside concentration and target measurements where available. In particular, a plasma concentration below an assay’s quantification limit does not establish that free concentration is zero at the target site.

Several approaches answer related but different questions. An equilibrium occupancy calculation describes the relationship between concentration and occupancy under equilibrium assumptions; it does not, by itself, describe how quickly occupancy changes after exposure falls. A concentration-time analysis describes exposure but does not establish target dissociation kinetics. A binding model that includes both association and dissociation can describe occupancy when free concentration changes or persists, but it requires information about the concentration driving binding as well as the rate constants. The simple washout exponential is useful when complete washout is a defensible approximation; it should not be substituted for a concentration-driven model when ongoing association is plausible.

Limitations and a Common Mistake

The formula \(R(t)=R_0e^{-k_{\mathrm{off}}t}\) assumes reversible binding to a homogeneous target, a constant first-order dissociation rate, negligible rebinding, and no relevant change in target abundance. It describes fractional occupancy of the target population represented by the model. It does not incorporate target synthesis or loss, ligand redistribution, or multiple target states. If these processes matter, the measured occupancy may fall faster or slower than the simple prediction. The rate constant estimated in one experimental setting may also fail to represent dissociation in another setting if the binding environment differs.

Common mistake: “Occupancy remains high for several hours, so the ligand must have a long residence time.” This conclusion is not justified if free ligand is still present, because ongoing association can replace ligand that dissociates. First establish whether the zero-free-concentration assumption is credible; then interpret the washout trajectory using the dissociation model. Conversely, a long mean residence time does not imply that occupancy remains high indefinitely: under ideal washout, the exponential decline continues, and the occupancy eventually becomes small.