An Overview and Analysis Regarding the Use of Adjudication Methods in EU and US Drug Approvals

Abstract

Background:

Several regulatory guidelines recommend that assessments of endpoints supporting drug approval should be verifiable by applicants and the regulatory agencies to minimize the potential for bias. This becomes especially critical when assessments are not based on measurable data but are derived from the interpretation of measurements, when they require the application of complex endpoint assessments, or when a study cannot be blinded. To make such interpretation more robust, a verification of (subjective) assessments by an independent panel of experts is frequently utilized. The objective of this paper was to analyze how often adjudicated methods across efficacy and safety assessments were used in drug approvals in the European Union and United States in 2013 and early 2014.

Methods:

A total of 35 new molecular entities (NMEs) approved by the US Food and Drug Administration (FDA) and 88 European Medicines Agency (EMA) approvals in Europe were included in this analysis.

Results:

An adjudication method was used in phase III development programs in 69% of the NMEs approved in the United States and 41% of EMA approvals. Drugs developed for oncology and endocrinology typically used an independent review committee (IRC) in line with recommendations made in relevant regulatory guidance, whereas nervous systems, antivirals, and vaccines drugs typically did not. Central reading was most frequently used for efficacy endpoints or in a combination of efficacy endpoints and safety measures. Overall, approximately 20% to 30% of the primary endpoints analyzed in the US/EMA documentation were classified as subjective endpoints that were based on clinician-dependent (and subject-dependent) assessments. The remaining 70% to 80% were more robust endpoints that were reviewed by a central committee and/or were based on objective (measurable) endpoints, including laboratory tests.

Conclusion:

While no one size fits all, the need to include an IRC depends on the subjectivity of the primary endpoint, the therapeutic area concerned, the clinical trial design, the need to assess reliability of marginal positive events, or if a critical assessment is required for adverse event accuracy.