Objective First, Method Second: Why the Estimand Definition Comes First in Pharmacometric Covariate Modeling

Area of Science:

• Pharmacometrics
• Quantitative Pharmacology
• Clinical Pharmacology

Background:

• Pharmacometric covariate analyses are crucial for understanding drug behavior and informing clinical decisions.
• Two primary objectives exist: mechanistic understanding and clinical decision-making, each requiring different analytical approaches.
• The distinction between these objectives and their analytical targets has historically been unclear.

Purpose of the Study:

• To clarify the distinct objectives of pharmacometric covariate analyses.
• To reconcile mechanistic modeling with regulatory and clinical needs using a structured framework.
• To emphasize the importance of estimand definition in covariate modeling.

Main Methods:

• Adoption of the International Council for Harmonisation E9(R1) estimand framework.
• Distinguishing between conditional covariate effects (mechanistic) and unconditional covariate effects (clinical/regulatory).
• Prioritizing estimand definition (the objective) over the estimator in covariate modeling.

Main Results:

• The ICH E9(R1) estimand framework provides a clear distinction between mechanistic and clinical objectives in covariate analyses.
• This framework reconciles the differing needs of mechanistic modeling and regulatory/clinical decision-making.
• Estimand definition is identified as the primary choice in covariate modeling, not the specific estimation method.

Conclusions:

• A structured approach using the ICH E9(R1) estimand framework optimizes pharmacometric covariate analyses.
• This approach enhances the quantitative evidence used for communicating clinical impact.
• Clear estimand definition leads to improved patient care through better-informed clinical decisions.