A simple numerical approach towards improving the two one-sided test for average bioequivalence.

Area of Science:

• Biostatistics
• Pharmacokinetics
• Drug Development

Background:

• Average bioequivalence testing is crucial for pharmaceutical development to ensure drug product similarity.
• The traditional two-one sided test (TOST) is commonly used but can be conservative, leading to reduced statistical power.
• Existing improvements to TOST aim to enhance power without compromising type I error control.

Purpose of the Study:

• To derive and evaluate a novel, simple statistical test for average bioequivalence.
• To compare the performance of the new test against the traditional TOST and other improved methods.
• To demonstrate the practical application of the new test with a real-world example.

Main Methods:

• A new critical value function, dependent on sample standard deviation, was developed.
• The critical value function is non-increasing and continuous, determined numerically.
• Extensive simulations were conducted to assess type I error probability and statistical power.

Main Results:

• The new test effectively controls type I error rates and avoids the conservatism of TOST.
• It demonstrates superior statistical power compared to TOST, particularly in high variance scenarios.
• The power of the new test is comparable to other advanced methods, with added simplicity.

Conclusions:

• The proposed simple test offers an improved alternative for average bioequivalence assessment.
• It provides enhanced power and maintains acceptable type I error rates, making it practical for regulatory submissions.
• The test's ease of use, similar to TOST, facilitates its adoption in pharmaceutical research and development.