Stimulation of Response Production

How Stimulating Production Changes an Indirect Response

An indirect response does not change simply because a drug effect is present: the response changes as production and loss become unbalanced. If a drug increases production while loss remains constant, production initially exceeds loss at baseline. The response therefore rises, and the increased response gradually restores balance by increasing the amount lost per unit time.

This tutorial formulates that mechanism and derives its initial direction and eventual response level. It focuses on the turnover model and on the consequences of holding the loss rate fixed, not on a particular concentration-effect function or a full time-course solution.

Definition: In the stimulated-production model, $R(t)$ is the response at time $t$, $k_{in}>0$ is the baseline production rate, $k_{out}>0$ is the first-order loss-rate constant, and $S(t)\geq 0$ is a dimensionless drug stimulation. The model is $$ \frac{dR}{dt}=k_{in}\bigl[1+S(t)\bigr]-k_{out}R(t). $$ The drug changes production through the multiplier $1+S(t)$; it does not directly change $k_{out}$.

The response has units of response amount or concentration, $k_{in}$ has response units per time, and $k_{out}$ has inverse-time units. Since $S(t)$ is dimensionless, both terms on the right side of the differential equation have response-per-time units. The model assumes a well-mixed response state with first-order loss: at a given $R$, the loss rate is $k_{out}R$.

Baseline, Initial Direction, and the Shifted Equilibrium

Before drug stimulation, take $S(t)=0$. At baseline equilibrium, the response is constant, so its derivative is zero. Setting the model's right side to zero gives

0=k_{in}-k_{out}R_0 \qquad\Longrightarrow\qquad R_0=\frac{k_{in}}{k_{out}}.

Suppose stimulation begins at time zero and has a fixed value $S_0>0$. The response cannot jump under this model: its state changes according to a finite rate of change. Immediately after stimulation begins, $R(0)=R_0$, while production has increased to $k_{in}(1+S_0)$. Therefore,

\left.\frac{dR}{dt}\right|_{0^+} =k_{in}(1+S_0)-k_{out}R_0 =k_{in}(1+S_0)-k_{in} =k_{in}S_0>0.

Thus the initial direction is upward. The initial slope is proportional to both baseline production and stimulation, and has units of response per time. This is an instantaneous slope, not an immediate change in response amount.

If the stimulation remains constant, the new equilibrium $R_{ss}$ is found by setting the derivative to zero again:

0=k_{in}(1+S_0)-k_{out}R_{ss} \qquad\Longrightarrow\qquad R_{ss}=\frac{k_{in}(1+S_0)}{k_{out}} =R_0(1+S_0).

For positive stimulation, this level exceeds baseline. In particular, $R_{ss}/R_0=1+S_0$: a stimulation of $0.5$, for example, raises the steady-state response by 50% relative to baseline in this linear turnover model. This is a model-based steady-state implication; it does not claim that every observed response will reach that level during a finite dosing period.

Key result: With constant $S_0>0$, fixed positive $k_{in}$ and $k_{out}$, and a response initially at baseline, the initial response slope is $k_{in}S_0>0$, and the drug-on equilibrium is $R_0(1+S_0)>R_0$. Production changes immediately, but $R(t)$ changes continuously.

Why the Response Approaches the New Level

The sign of the response rate can be expressed in terms of the distance from the drug-on equilibrium. Under constant stimulation,

\frac{dR}{dt} =k_{out}\left(R_{ss}-R\right).

When $R<R_{ss}$, the derivative is positive; when $R>R_{ss}$, it is negative. At $R=R_{ss}$, it is zero. This shows why an initially lower response rises toward the new equilibrium and why the equilibrium is stable. A short stability justification makes the conclusion precise: define $V=(R-R_{ss})^2/2$, the squared distance from equilibrium divided by two. Then

\frac{dV}{dt} =(R-R_{ss})\frac{dR}{dt} =-k_{out}(R-R_{ss})^2\leq 0.

Since $k_{out}>0$, this distance decreases whenever the response differs from equilibrium. The result justifies stable return toward $R_{ss}$ without requiring the full time-course solution. If stimulation is removed, $S$ returns to zero; at any response above baseline, production is then less than loss, so the response moves downward toward $R_0$.

Worked Example: Initial Rise and New Equilibrium

Consider a hypothetical response with $k_{in}=12$ response units per day and $k_{out}=0.60$ per day. It is initially at baseline, and a drug produces a constant stimulation $S_0=0.50$. Calculate baseline, the initial slope, and the drug-on equilibrium.

Baseline. Using $R_0=k_{in}/k_{out}$, $$ R_0=\frac{12\ \text{response units/day}}{0.60\ \text{day}^{-1}} =20\ \text{response units}. $$

Initial slope. The response is still 20 response units at the instant stimulation begins, so the loss rate is $0.60\times20=12$ response units per day. Production is $12(1+0.50)=18$ response units per day. Hence $$ \left.\frac{dR}{dt}\right|_{0^+}=18-12 =6\ \text{response units/day}. $$ The positive slope means the response starts to increase; it does not mean the response jumps upward by 6 units.

Drug-on equilibrium. $$ R_{ss}=\frac{12(1+0.50)}{0.60} =\frac{18}{0.60} =30\ \text{response units}. $$ The predicted equilibrium is 10 response units, or 50%, above baseline. These calculations use the same fixed loss-rate constant before and during stimulation.

Worked Example: Checking a Proposed Response Pattern

In a hypothetical setting, $k_{in}=8$ response units per hour, $k_{out}=0.40$ per hour, and constant stimulation is $S_0=0.25$. A proposed description says that the response is initially 20 response units and immediately falls after the drug is given. Check that description against the model.

First calculate the baseline: $$ R_0=\frac{8}{0.40}=20\ \text{response units}. $$ The starting value is consistent with baseline equilibrium. At that value, loss is $0.40(20)=8$ response units per hour, while stimulated production is $8(1.25)=10$ response units per hour. Therefore, $$ \left.\frac{dR}{dt}\right|_{0^+}=10-8 =2\ \text{response units/hour}. $$ The model predicts an initial rise, not an initial fall. A declining response immediately after stimulation would be inconsistent with the stated starting condition, positive stimulation, and fixed loss rate. The discrepancy could reflect a different mechanism, a non-baseline starting value, a lag or observation-timing issue, or model misspecification; the calculation alone cannot determine which explanation is correct.

Assumptions, Diagnostics, and Interpretation

The directional and equilibrium conclusions depend on specific assumptions. The baseline must be at equilibrium if $R_0=k_{in}/k_{out}$ is used. Both rate parameters must be positive. Stimulation must be positive for the predicted upward initial slope, and the loss-rate constant must stay fixed. The equation also assumes production responds through the stated multiplier and loss remains proportional to the current response. If any of these conditions fails, the conclusions may not apply as written.

When evaluating data or a fitted model, check whether the response pattern is compatible with these implications. For a baseline response followed by positive stimulation, an initial downward trend conflicts with the model's initial-slope prediction. Under constant stimulation, a sustained response far above the calculated equilibrium is also incompatible with a stable drug-on steady state under these parameter values. In practice, noisy or sparse observations may obscure the true initial slope or whether equilibrium has been reached. These patterns are reasons to investigate assumptions and timing, not proof by themselves that a particular alternative mechanism is correct.

Production and loss parameters can also be difficult to distinguish from response observations alone. Baseline equilibrium constrains their ratio, $k_{in}/k_{out}$, rather than identifying both rates separately. Time-course information and other evidence may be needed to support estimates of each parameter. A visually plausible curve is not, by itself, evidence that the production mechanism has been uniquely identified.

Feature	Stimulated response production	Inhibited response production
Drug action on production	Production rises above $k_{in}$	Production falls below $k_{in}$
Loss term	$k_{out}R$, unchanged	$k_{out}R$, unchanged
Initial direction from baseline	Upward for positive stimulation	Downward for positive inhibition
Equilibrium implication	Above baseline	Below baseline

The comparison with inhibited production follows from changing the sign of the production effect while retaining the same loss structure. The class of indirect response models also includes drug effects on response loss, which change a different term in the balance equation. These mechanisms should not be treated as interchangeable: a similar response direction does not establish which rate was altered. Distinguishing them requires considering the model assumptions and the observed time course, not just whether the response rises or falls.

Worked Example: Translating a Percentage Stimulation

A hypothetical model has baseline $R_0=15$ response units and a constant stimulation specified as 20% above baseline production. Determine $S_0$, the drug-on production rate relative to baseline, the equilibrium response, and the initial slope if $k_{in}=3$ response units per hour.

A 20% increase means $1+S_0=1.20$, so $S_0=0.20$. Production under drug is therefore 120% of baseline, not 20% of baseline. The equilibrium prediction is $$ R_{ss}=R_0(1+S_0) =15(1.20) =18\ \text{response units}. $$ The initial slope is $$ \left.\frac{dR}{dt}\right|_{0^+} =k_{in}S_0 =3(0.20) =0.60\ \text{response units/hour}. $$ Thus the response starts rising at 0.60 response units per hour and, if the constant-stimulation assumptions hold, tends toward 18 response units. The slope is not a prediction that the response will rise by exactly 0.60 units in every hour; the slope changes as the response changes and loss increases.

Common Pitfall

A frequent error is to treat a 50% increase in production as an immediate 50% increase in the response. The model changes the production rate, not the response state. Immediately after the change, $R$ remains at its baseline value; the rate of response change is $k_{in}S_0$. Only under constant stimulation and the stated linear turnover assumptions is the eventual equilibrium $R_0(1+S_0)$. Confusing the initial slope with the final response change mixes quantities with different units and meanings.

Key takeaway: In the stimulated-production model, positive stimulation raises production while leaving loss proportional to the current response at the same rate constant. From baseline, the response initially rises, then approaches a stable equilibrium above baseline if stimulation is maintained.

Check Your Understanding

Answer each question using the model assumptions stated in this tutorial. Brief solutions follow the questions.

If $k_{in}=10$ response units per day and $k_{out}=0.50$ per day, what is the baseline response?
Using those parameters, what is the initial slope for a constant stimulation $S_0=0.30$ when the response starts at baseline?
For the same stimulation, what is the drug-on equilibrium, and how does it compare with baseline?
Why does the response not jump immediately to the drug-on equilibrium when stimulation begins?
Name one assumption whose failure could make an observed initial decrease inconsistent with the model prediction.

Solutions. (1) $R_0=10/0.50=20$ response units. (2) The initial slope is $k_{in}S_0=10(0.30)=3$ response units per day, which is positive. (3) $R_{ss}=20(1.30)=26$ response units, 6 response units or 30% above baseline. (4) The model specifies a finite rate of change of the response; stimulation changes production, while the response state remains continuous. (5) For example, the response might not actually start at baseline, stimulation may not be positive, or loss may not remain fixed. Any of these would remove a condition used to predict an initial rise.

Feature	Stimulated response production	Inhibited response production
Drug action on production	Production rises above \(k_{in}\)	Production falls below \(k_{in}\)
Loss term	\(k_{out}R\), unchanged	\(k_{out}R\), unchanged
Initial direction from baseline	Upward for positive stimulation	Downward for positive inhibition
Equilibrium implication	Above baseline	Below baseline