Principal stratum strategy: Potential role in drug development.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmaceutical Research

Background:

• Randomized trials estimate causal treatment effects in overall and sub-populations.
• Post-randomization events, termed intercurrent events, can complicate causal effect estimation.
• Standard analyses may fail to provide causal estimates when intercurrent events are treatment-related.

Purpose of the Study:

• To address challenges in estimating causal treatment effects when intercurrent events occur post-randomization.
• To introduce the principal stratum strategy as an alternative analytical approach.
• To advocate for the ICH E9(R1) estimand framework for handling intercurrent events.

Main Methods:

• Utilizing the principal stratum strategy to classify subjects based on potential intercurrent event occurrence.
• Applying the ICH E9(R1) estimand framework to formulate and address clinical questions.
• Illustrating the approach with examples from drug development.

Main Results:

• Principal strata questions are common in drug development.
• The ICH E9(R1) estimand framework enhances transparency and adequacy of analyses.
• Key assumptions for principal strata estimation are identified, many of which are unverifiable.

Conclusions:

• The principal stratum strategy and ICH E9(R1) framework provide a more robust approach to intercurrent events.
• Transparent assumptions and adequate analyses are crucial for reliable conclusions.
• Sensitivity analyses are essential to evaluate the robustness of findings based on unverifiable assumptions.