Methods to control the empirical type I error rate in average bioequivalence tests for highly variable drugs

Area of Science:

• Pharmacokinetics and Drug Metabolism
• Biostatistics
• Regulatory Science

Background:

• Average bioequivalence (ABE) tests assess generic drug efficacy against original drugs.
• Highly variable drugs require additional criteria in bioequivalence studies due to high intra-subject variance.
• Current ABE tests for highly variable drugs may not adequately control the type I error rate, increasing consumer risk.

Purpose of the Study:

• To propose two novel, statistically sound methods for average bioequivalence testing in highly variable drugs.
• To ensure control of the empirical type I error rate without relying on sample estimators in null hypotheses.
• To provide reliable bioequivalence assessment for drugs with high intra-subject variability.

Main Methods:

• The first method sequentially tests intra-subject variance and then average drug effect.
• The second method simultaneously evaluates both average drug effect and intra-subject variance.
• Both methods incorporate statistical controls to manage error rates.

Main Results:

• Simulations demonstrate that the proposed methods effectively control the empirical type I error rate.
• The new methods offer improved reliability for bioequivalence assessments of highly variable drugs.
• Performance analysis confirms the statistical soundness of the developed approaches.

Conclusions:

• The proposed methods provide statistically robust alternatives for average bioequivalence testing of highly variable drugs.
• These methods enhance the accuracy of bioequivalence assessments, ensuring consumer safety.
• Pharmaceutical manufacturers and regulatory agencies can benefit from these improved testing strategies.