On the lower predictive bound approach for noninferiority clinical trials with binary data

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacology

Background:

• Noninferiority trials often lack placebo controls due to ethical reasons.
• This necessitates indirect inference for treatment effect hypotheses.
• European Medicines Agency guidelines require indirect confidence bounds for new drugs.

Purpose of the Study:

• To develop a new statistical approach for noninferiority trials without placebo controls.
• To meet the European Committee for Medicinal Products for Human Use (CHMP) guideline requirements for indirect confidence bounds.
• To provide methods for sample size calculation and indirect statistical inference.

Main Methods:

• A novel approach using the lower predictive bound of a confidence level to determine the noninferiority margin.
• Development of methods for sample size calculation.
• Establishment of procedures for indirect statistical inference, including p-values, confidence intervals, and point estimates.

Main Results:

• The proposed lower predictive bound approach offers a method for determining noninferiority margins in placebo-absent trials.
• Methods for sample size calculation and indirect statistical inference are provided.
• The approach was assessed using an actual trial example and compared with other margin determination methods.

Conclusions:

• The developed lower predictive bound approach provides a viable solution for noninferiority trials lacking placebo controls.
• This method facilitates indirect statistical inference, meeting regulatory requirements.
• The approach is applicable for designing and analyzing clinical trials evaluating new drug efficacy.