Dose-Interval Effects Under Saturable Target Binding

Why Equal Average Dose Does Not Mean Equal Target Engagement

Two regimens can deliver the same total antibody dose per unit time while producing very different concentration profiles. A longer interval generally gives a higher peak and lower trough; a shorter interval tends to smooth those fluctuations. When binding is saturable, occupancy is a nonlinear function of concentration, so equal average concentration does not imply equal average occupancy. If bound antibody is also internalized, the changing concentration and target availability can affect clearance as well.

A useful comparison therefore separates three questions: how the regimen changes concentration fluctuations, how those fluctuations translate into occupancy, and how target renewal and complex removal feed back on drug disposition. The calculations below first isolate the occupancy effect under rapid equilibrium, then show why target turnover and saturable elimination require additional care.

Definition: The average dose rate is dose divided by dosing interval. At periodic steady state, the interval-average concentration is the concentration integrated over one dosing interval and divided by that interval. Target occupancy is the fraction of target bound by drug. Under rapid reversible equilibrium with negligible depletion effects, occupancy at concentration \(C\) is \(O(C)=C/(K_D+C)\), where \(K_D\) is the dissociation constant.

Concentration Fluctuations at Periodic Steady State

Begin with a deliberately simple model: instantaneous intravenous bolus doses, one-compartment concentration, and first-order elimination with rate constant \(k\). Binding does not affect drug disposition in this first calculation. Let \(\tau\) be the dosing interval and \(\Delta C\) the concentration increment from each dose. Between doses, concentration decays exponentially. At periodic steady state, the peak just after a dose and the trough just before the next dose are

C_{\max}=\frac{\Delta C}{1-e^{-k\tau}}, \qquad C_{\min}=C_{\max}e^{-k\tau}.

Integrating the exponential decline over an interval gives the interval-average concentration:

\bar C_{\tau} =\frac{1}{\tau}\int_0^\tau C_{\max}e^{-kt}\,dt =\frac{C_{\max}(1-e^{-k\tau})}{k\tau} =\frac{\Delta C}{k\tau}.

Thus equal average dose rate gives equal \(\bar C_{\tau}\) in this linear model, because \(\Delta C\) is proportional to dose and the average dose rate is proportional to \(\Delta C/\tau\). It does not give equal peaks or troughs. These statements rely on linear first-order elimination, constant volume, and periodic steady state; they do not automatically hold when target-mediated disposition materially changes clearance.

Worked Example: Equal Average Concentration With Two Intervals

Consider hypothetical regimens with \(k=0.1\) per day and an interval-average concentration of \(100\) mg/L. Compare intervals of 28 days and 7 days. The model assumes bolus dosing and linear elimination, with both regimens at periodic steady state.

For either interval, \(\Delta C=\bar C_{\tau}k\tau\). At 28 days, \(\Delta C=100(0.1)(28)=280\) mg/L. Therefore,

C_{\max,28}=\frac{280}{1-e^{-2.8}}=298.13\ \text{mg/L}, \qquad C_{\min,28}=298.13e^{-2.8}=18.13\ \text{mg/L}.

At 7 days, \(\Delta C=100(0.1)(7)=70\) mg/L, giving

C_{\max,7}=\frac{70}{1-e^{-0.7}}=139.05\ \text{mg/L}, \qquad C_{\min,7}=139.05e^{-0.7}=69.05\ \text{mg/L}.

Both regimens have an interval average of \(100\) mg/L by construction. The 28-day regimen has the higher peak and lower trough. The dimensionless quantity \(k\tau\) summarizes the amount of first-order elimination occurring during an interval: here it is 2.8 versus 0.7.

From Concentration to Average Occupancy

For rapidly equilibrating binding with negligible target depletion and internalization effects on binding, use \(O(C)=C/(K_D+C)\). Because this function is nonlinear, average occupancy must be calculated from the concentration profile, not by substituting the average concentration into the occupancy equation. For the exponential profile above, integration gives

\overline O_{\tau} =\frac{1}{\tau}\int_0^\tau \frac{C_{\max}e^{-kt}}{K_D+C_{\max}e^{-kt}}\,dt =\frac{1}{k\tau} \ln\left(\frac{K_D+C_{\max}}{K_D+C_{\min}}\right).

Worked Example: Comparing Peak, Trough, and Average Occupancy

Use the concentrations from the previous example and suppose, hypothetically, that \(K_D=100\) mg/L. The rapid-equilibrium occupancy formula is assumed to apply throughout each interval.

For the 28-day regimen, peak occupancy is \(298.13/(100+298.13)=0.749\), and trough occupancy is \(18.13/(100+18.13)=0.1535\). Its interval-average occupancy is

\overline O_{28} =\frac{1}{2.8}\ln\left(\frac{100+298.13}{100+18.13}\right) =\frac{1}{2.8}\ln\left(\frac{398.13}{118.13}\right) \approx 0.4339.

For the 7-day regimen, peak occupancy is \(139.05/(100+139.05)=0.582\), and trough occupancy is \(69.05/(100+69.05)=0.408\). Its interval-average occupancy is

\overline O_{7} =\frac{1}{0.7}\ln\left(\frac{100+139.05}{100+69.05}\right) =\frac{1}{0.7}\ln\left(\frac{239.05}{169.05}\right) \approx 0.4950.

In this example, the shorter interval has lower peak occupancy but higher trough and average occupancy, despite the same average concentration. Occupancy is dimensionless; the concentrations and \(K_D\) are in mg/L. These values illustrate a specified rapid-equilibrium model, not a general prediction for every target-mediated system.

Target Turnover Adds a Timescale

Rapid binding is only part of the system when target is synthesized, degraded, or removed with bound drug. A minimal one-to-one binding model can represent free drug \(C\), free target \(R\), and drug–target complex \(B\). For a shared concentration scale, one form is

\begin{aligned} \frac{dC}{dt} &= \text{Input}(t)-k_{\mathrm{el}}C-k_{\mathrm{on}}CR+k_{\mathrm{off}}B,\\ \frac{dR}{dt} &= k_{\mathrm{in}}-k_{\mathrm{out}}R-k_{\mathrm{on}}CR+k_{\mathrm{off}}B,\\ \frac{dB}{dt} &= k_{\mathrm{on}}CR-(k_{\mathrm{off}}+k_{\mathrm{int}})B. \end{aligned}

Here \(k_{\mathrm{in}}\) is target production, \(k_{\mathrm{out}}\) is free-target loss, and \(k_{\mathrm{int}}\) is complex internalization. Without drug, the baseline target concentration is \(R_0=k_{\mathrm{in}}/k_{\mathrm{out}}\), and the free-target turnover timescale is \(1/k_{\mathrm{out}}\). Binding and internalization can change both free target and drug loss. Consequently, occupancy may depend on the history of exposure, not just on the instantaneous concentration.

Worked Example: Comparing the Interval With Target Recovery Time

Suppose a hypothetical target has a free-target half-life of 7 days. After complete depletion, assume for this calculation that drug no longer suppresses target production or affects its loss. The fraction of baseline target restored after time \(t\) then follows first-order recovery:

\frac{R(t)}{R_0}=1-e^{-k_{\mathrm{out}}t}, \qquad k_{\mathrm{out}}=\frac{\ln 2}{7}=0.09902\ \text{per day}.

After 7 days, the recovered fraction is \(1-e^{-0.09902(7)}=1-0.5=0.500\), or 50.0%. After 28 days, it is \(1-e^{-0.09902(28)}=1-0.0625=0.9375\), or 93.75%. This comparison shows that the intervals span one versus four target half-lives. It does not predict actual target levels during dosing: that requires accounting for ongoing binding, drug exposure, and complex internalization. It does show why a target that is depleted during exposure may have very different availability before the next dose under the two regimens.

Consequences for Saturable Clearance

A common reduced description of target-mediated elimination is a saturable rate \(v(C)=V_{\max}C/(K_m+C)\), where \(V_{\max}\) is the maximum elimination rate and \(K_m\) is the concentration at half that rate. This representation is useful when its reduction assumptions are appropriate, but a dynamic target model may be needed if target abundance changes substantially over the interval.

The saturable rate is concave in concentration for positive \(C\), since

\frac{d^2v}{dC^2} =-\frac{2V_{\max}K_m}{(K_m+C)^3}<0.

For a concentration profile with a fixed interval mean, concavity implies that the average saturable elimination rate is no greater than the rate evaluated at that mean: \(\overline{v(C)}\leq v(\bar C_{\tau})\). This follows from Jensen's inequality. In words, high concentrations spend part of the interval in the saturated region, where additional concentration contributes less elimination than it would in the unsaturated region. This comparison concerns a fixed profile mean. Under the same dose rate, the eventual mean concentration can itself change as the nonlinear elimination changes.

Worked Example: Interval-Average Saturable Elimination

For a numerical illustration, take the hypothetical profiles above and let \(K_m=100\) mg/L and \(V_{\max}=10\) mg/L per day. Assume the reduced saturable-rate model applies and the target availability represented by its parameters is constant over each interval. Then the average rate is \(V_{\max}\) times the average occupancy-like factor \(C/(K_m+C)\).

For the 28-day profile, the calculated average factor is 0.4339, so \(\overline v_{28}=10(0.4339)=4.339\) mg/L per day. For the 7-day profile, it is 0.4950, so \(\overline v_{7}=10(0.4950)=4.950\) mg/L per day. At the common mean concentration of \(100\) mg/L, the rate evaluated at the mean is \(v(100)=10(100)/(100+100)=5.000\) mg/L per day, above both profile averages. This calculation isolates the curvature effect at an imposed common mean concentration; it does not assert that both regimens with the same dose rate will have that same steady-state mean when this clearance is active.

How to Compare Regimens and Diagnose a Poor Approximation

A practical comparison should keep the question and assumptions explicit. First compare dose rate, interval, and the dimensionless elimination quantity \(k\tau\). Then calculate peak, trough, and interval-average occupancy only if binding is fast enough for the equilibrium approximation and the target state is not changing in a way that invalidates it. If target abundance or complex internalization changes materially, compare simulated or observed trajectories of free drug, total drug, free target, and complex over repeated intervals.

The principal alternatives answer different questions:

Approach	Useful when	Main limitation
Equilibrium occupancy from concentration	Binding is rapid and target availability is approximately fixed	Does not represent target recovery or binding history
Reduced saturable-clearance rate	Target dynamics can be summarized by stable effective parameters	Can miss changing target abundance and complex fate
Dynamic drug–target model	Binding, target turnover, or internalization evolves on relevant timescales	Requires more parameters and informative measurements

Warning signs include a predicted trough near zero when observed drug remains measurable, systematic errors in peak-to-trough shape, different patterns in free and total drug, or target recovery that does not match the assumed turnover rate. Check units and mass balance, and verify that dose and interval actually represent the same average dose rate. A reduced model may fit concentrations while giving misleading occupancy if the assay measures total rather than free drug or target.

Common pitfall: “The average dose rate is the same, so average occupancy and target-mediated clearance must also be the same.” This is false because occupancy and saturable elimination are nonlinear functions of concentration, and target availability can change over time. Compare interval profiles or integrate the relevant rate over the interval; do not substitute the mean concentration into a nonlinear function and call it the average response.

The simple calculations have important limits. They assume bolus input and a single exponential concentration decline; subcutaneous absorption, distribution phases, and FcRn-mediated recycling can alter the profile. The occupancy integral assumes rapid equilibrium and fixed effective binding properties. The saturable-clearance comparison assumes a time-invariant \(V_{\max}\) and \(K_m\); target turnover and complex internalization can make effective clearance vary within and between intervals. Accordingly, interval choice should be evaluated with a model that matches the timescales and measurements of the application, rather than inferred from average dose alone.

Check Your Understanding

Use the assumptions stated in each question. Solutions follow each problem.

If \(k\) is fixed, what happens to peak-to-trough variation as \(k\tau\) increases? Solution: The trough-to-peak ratio is \(e^{-k\tau}\), so it decreases as \(k\tau\) increases. The profile has a larger proportional decline within the interval.
Why is \(O(\bar C_{\tau})\) not generally equal to \(\overline O_{\tau}\)? Solution: Because \(O(C)=C/(K_D+C)\) is nonlinear. The average of a nonlinear function of a varying concentration need not equal the function evaluated at the average concentration.
A target half-life is 14 days. What fraction of baseline target is restored 14 days after complete depletion under the drug-free first-order recovery assumption? Solution: \(1-2^{-14/14}=1-0.5=0.500\), or 50.0%.
For a fixed mean concentration profile, why can increasing fluctuation lower average Michaelis–Menten elimination? Solution: The rate \(v(C)=V_{\max}C/(K_m+C)\) is concave because \(v''(C)<0\). Jensen's inequality gives \(\overline{v(C)}\leq v(\bar C_{\tau})\); more variability can lower the average rate at the same mean.