Power Analysis and Sample Size Determination for Crossover Trials with Application to Bioequivalence Assessment of...

Area of Science:

• Pharmacokinetics and Drug Development
• Biostatistics
• Ophthalmology

Background:

• Bioequivalence assessment is crucial for generic topical ophthalmic drugs.
• Crossover designs with serial sampling are often employed.
• Accurate sample size determination is essential for trial efficiency and validity.

Purpose of the Study:

• To develop methods for sample size calculation in bioequivalence studies of topical ophthalmic drugs.
• To utilize crossover designs with serial sampling schemes.
• To provide reliable tools for sample size determination.

Main Methods:

• Derivation of power functions for Fieller-type and asymptotic confidence intervals.
• Application in crossover designs with serial-sampling data.
• Evaluation of derived power functions through simulation studies.

Main Results:

• Power functions provide precise estimates under normality and conservative estimates under log-normality.
• Intra-correlation positively impacts bioequivalence test power.
• The choice between Fieller-type and asymptotic intervals depends on the expected ratio of AUCs.

Conclusions:

• The derived Fieller-type and asymptotic power functions are suitable for determining sample sizes in crossover bioequivalence trials.
• These methods support the bioequivalence assessment of generic topical ophthalmic drugs.
• Sample size can be calculated using numerical iteration with the developed power functions.