Related Experiment Videos

Optimal sample size for a series of pilot trials of new agents

T J Yao1, C B Begg, P O Livingston

  • 1Department of Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, New York 10021, USA.

Biometrics
|September 1, 1996
PubMed

Insights

This study introduces a new method for optimizing sample sizes in therapeutic agent screening trials. It suggests smaller, continuous trials are more efficient for identifying promising agents over time.

Area of Science:

  • Biostatistics
  • Clinical Trial Design
  • Pharmacological Screening

Background:

  • Therapeutic agent screening is an ongoing process, necessitating adaptive strategies.
  • Traditional fixed total sample size approaches may not be optimal for continuous screening.
  • Minimizing the time to identify promising agents is a key objective in drug discovery.

Purpose of the Study:

  • To develop a novel approach for determining optimal sample sizes in sequential screening trials.
  • To address the limitations of fixed sample size designs in the context of continuous agent screening.
  • To minimize the time required for identifying effective therapeutic agents.

Main Methods:

  • An empirical Bayes formulation was used to optimize individual sample sizes.
  • Error rates were fixed, while sample sizes were optimized to reduce identification time.
  • Bootstrapping techniques were employed to assess the reliability of the proposed method.

Main Results:

  • The new approach demonstrated that relatively small individual screening trials are optimal.
  • Application to historical vaccination trial data validated the method's efficiency.
  • The empirical Bayes formulation effectively balances error rates and sample size optimization.

Conclusions:

  • The proposed method offers a more efficient strategy for sample size determination in continuous screening.
  • Optimizing individual trial sizes in sequential screening can accelerate the identification of therapeutic agents.
  • This approach provides a statistically sound framework for adaptive clinical trial design in drug discovery.

Related Concept Videos