Model-Informed Drug Development (MIDD)

Understand how pharmacometric models are used across drug development to support dosing, trial design, and regulatory decisions.
Tip

What you’ll build today: a clear understanding of how pharmacometric models are used in real drug development to inform decisions across the lifecycle.

Learning Objectives

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

  • Define Model-Informed Drug Development (MIDD)
  • Explain how models are used across development stages
  • Connect exposure–response and simulation to real decisions
  • Understand the role of PMx in regulatory and clinical strategy

Key Ideas

Model-Informed Drug Development (MIDD) is the use of models to:

  • integrate data
  • generate predictions
  • support decisions

Across drug development.

MIDD = using models to make better decisions, earlier and more efficiently

The MIDD Workflow

Model-informed development turns data into decisions.

Conceptually:

Data

↓

Model

↓

Evaluation

↓

Simulation

↓

Decision

Each stage answers a different question:

  • Data → What happened?
  • Model → Why did it happen?
  • Evaluation → Can we trust it?
  • Simulation → What could happen?
  • Decision → What should we do?

This workflow is one of the foundations of modern pharmacometrics.

Why This Lesson Matters

Everything you’ve learned so far builds to this:

  • PK → understanding exposure
  • Models → explaining data
  • Estimation → fitting models
  • Diagnostics → validating models
  • Simulation → predicting outcomes

Now:

MIDD connects all of these into real-world impact

Where MIDD is Used

Early Development (Phase 1)

  • First-in-human dose selection
  • Safety and exposure assessment

Mid Development (Phase 2)

  • Exposure–response analysis
  • Dose optimization

Late Development (Phase 3)

  • Confirm dosing strategy
  • Support labeling decisions

Worked Example: Dose Selection

Imagine:

A new drug shows:

  • efficacy above AUC = 50
  • increasing toxicity above AUC = 120

Several doses are simulated.

Results:

Dose Target Attainment
Low Many patients underexposed
Medium Most patients in target
High More patients exceed safety limits

Decision:

👉 Select the medium dose.

This decision was made without running additional clinical studies.

Expanding the Example

MIDD changes the questions we ask.

Instead of:

Which dose worked?

we ask:

Which dose is most likely to work across future patients?

This shift allows development teams to:

  • compare scenarios
  • quantify uncertainty
  • make decisions earlier

Models do not replace studies.

They improve how studies are designed and interpreted.

Insight

MIDD shifts drug development from reactive to predictive.

Note

The goal is not just to analyze data, but to anticipate outcomes.

Key Components of MIDD

  • PK/PD models
  • Exposure–response relationships
  • Simulation tools
  • Clinical data integration

Why This Matters for Decisions

MIDD supports:

  • dose selection
  • trial design
  • risk–benefit assessment
  • regulatory communication

Strategies

  • Focus on decision-relevant questions
  • Use validated models
  • Communicate uncertainty clearly
  • Integrate multiple data sources

Common Mistakes

  • Treating models as definitive answers
  • Ignoring uncertainty
  • Misusing simulations
  • Disconnecting models from decisions

Practice Problems

  1. What is MIDD?
  2. Where is it used in development?
  3. Why is it valuable?
  1. The use of models to inform drug development decisions
  2. Across all phases (early, mid, late)
  3. It improves efficiency and decision quality

What MIDD Looks Like in Practice

In practice, MIDD may support:

  • selecting first-in-human doses
  • reducing unnecessary trial arms
  • evaluating special populations
  • supporting regulatory discussions
  • updating labeling recommendations

The specific tools may change.

The underlying idea stays the same:

use evidence and models to improve decisions.

Summary

MIDD:

  • integrates modeling and decision-making
  • supports drug development across stages
  • improves efficiency and outcomes

It represents one of the clearest examples of how pharmacometrics creates real-world impact.

  • MIDD = models + decisions
  • Use across development stages
  • Focus on prediction and uncertainty
  • Always connect models to decisions