Delayed Response and ODE Intuition

Introduce delayed pharmacodynamic responses and develop intuition for turnover and ordinary differential equations.
Tip

Big picture: Not all responses occur immediately. Delayed responses motivate dynamic models and introduce ODE thinking.

Learning Objectives

By the end of this lesson, you will be able to:

  • explain delayed pharmacodynamic response
  • distinguish direct and delayed effects
  • interpret turnover concepts
  • understand the meaning of an ODE
  • explain the roles of production and loss

Key Ideas

  • concentration does not always equal response
  • response may accumulate gradually
  • delayed response introduces state change
  • ODEs describe rates of change

Setup

library(tidyverse)

Why Direct Models Are Not Always Enough

Previously we assumed:

Concentration → Response

But many biological systems respond gradually.

Examples:

  • clotting factors
  • biomarkers
  • cell populations

Instead:

Concentration

↓

Biological Change

↓

Response

This introduces delay.

Worked Example 1: Immediate versus Delayed Response

Compare two systems.

Direct effect:

Concentration → Effect

Delayed effect:

Concentration

↓

Biological Process

↓

Effect

Question:

Why might effect continue after concentration changes?

Answer:

The biological system may have memory.

Worked Example 2: Turnover Intuition

Many responses behave like:

Production

↓

Response

↓

Loss

Examples:

  • biomarker synthesis
  • protein degradation
  • clotting factor turnover

Interpretation:

Response changes because:

Input ≠ Output

When:

Production = Loss

the system stays stable.

Worked Example 3: Introducing ODEs

To describe change through time, we introduce an ordinary differential equation (ODE).

The simplest turnover model:

\[ \frac{dR}{dt} = k_{in} - k_{out}R \]

Interpretation:

Component Meaning
\(R\) response
\(k_{in}\) production rate
\(k_{out}\) loss rate

Read this as:

Rate of Change = Production − Loss

Question:

If production exceeds loss,
what happens?

Response increases.

Worked Example 4: Simulate Turnover Behavior

Simulate a simple turnover process.

time <- seq(0, 30, by = 0.1)

kin <- 20
kout <- 0.2

resp <- numeric(length(time))
resp[1] <- 0

dt <- diff(time)[1]

for(i in 2:length(time)) {

  dRdt <-
    kin -
    kout * resp[i - 1]

  resp[i] <-
    resp[i - 1] +
    dRdt * dt
}

turnover_tbl <-
  tibble(
    TIME = time,
    RESPONSE = resp
  )

ggplot(
  turnover_tbl,
  aes(
    TIME,
    RESPONSE
  )
) +
geom_line() +
labs(
  title = "Turnover Response",
  x = "Time",
  y = "Response"
)

Interpretation:

Response changes gradually.

This pattern is consistent with a turnover process in which response accumulates over time and approaches a steady state.

For now, we focus on the behavior of the system.

Later in the curriculum, we will revisit these ideas when introducing ordinary differential equation (ODE) models and simulation-based approaches.

Worked Example 5: Drug Effect on Turnover

Drugs may influence:

Production → Stimulation or Inhibition

or:

Loss → Stimulation or Inhibition

Examples:

Mechanism Example
inhibit production lower response
stimulate production higher response
inhibit loss sustained response
stimulate loss faster decline

This becomes the foundation of indirect response models.

Connecting to PK/PD

Direct models assumed:

Exposure → Immediate Response

Delayed models assume:

Exposure

↓

Dynamic System

↓

Observed Response

Next lesson combines:

  • PK
  • PD
  • multiple endpoints

using a full PK/PD example.

Strategies

  • think about biology
  • think about accumulation
  • focus on interpretation

Common Mistakes

  • treating ODEs as abstract math
  • expecting instantaneous response
  • ignoring turnover

Practice Problems

  1. What creates delayed response?

  2. What does:

\[ \frac{dR}{dt} = 0 \]

mean?

  1. What happens if production exceeds loss?

  2. Why can direct models fail?

  3. Give one biological example of turnover.

Problem 1

Response depends on intermediate biology.

Problem 2

System equilibrium.

Problem 3

Response increases.

Problem 4

They cannot represent delay.

Problem 5

Examples:

  • clotting factors
  • biomarkers

Summary

  • delayed response introduces dynamics
  • ODEs describe change
  • turnover explains delay
  • production and loss determine behavior
  • Delay ≠ variability
  • ODE = rate of change
  • Turnover matters