Summarizing historical information on controls in clinical trials.

Area of Science:

• Clinical Trial Design
• Biostatistics
• Pharmacoeconomics

Background:

• Historical data is crucial for designing clinical trials and can be integrated into analyses.
• Exploiting past information from comparable control groups is a relevant strategy.

Purpose of the Study:

• To discuss summarizing historical control data as prior information in Phase IV and proof-of-concept trials.
• To emphasize the importance of assessing the similarity of control parameters for fair evaluation.

Main Methods:

• Utilized a meta-analytic-predictive methodology.
• Quantified heterogeneity of control parameters using between-trial variation.
• Determined prior effective sample size and its upper bound (prior maximum sample size).

Main Results:

• A Phase IV trial showed small heterogeneity, yielding a prior effective sample size of ~90 patients.
• Proof-of-concept trials exhibited moderate to substantial heterogeneity, resulting in prior effective sample sizes of ~20 and ~7 patients.
• Prior effective sample size depends on the quantity of historical data and low between-trial heterogeneity, capped by the prior maximum sample size.

Conclusions:

• The meta-analytic-predictive approach assumes exchangeability of control parameters across trials.
• Sensitivity analysis is recommended due to challenges in quantifying between-trial variability.
• Historical control data offers a valuable, efficient option, particularly for nonconfirmatory trials.