Back to Explorer

Multiple Endpoints in Clinical Trials: Guidance for Industry

FinalCenter for Biologics Evaluation and Research Center for Drug Evaluation and Research10/21/2022
MultiplicityType I error rate

Description

This guidance provides sponsors and review staff with the Agency’s thinking about the problems posed by multiple endpoints in the analysis and interpretation of study results and how these problems can be managed in clinical trials for human drugs, including drugs subject to licensing as biological products. Most clinical trials performed in drug development contain multiple endpoints to assess the effects of the drug and to document the ability of the drug to favorably affect one or more disease characteristics. As the number of endpoints analyzed in a single trial increases, the likelihood of making false conclusions about a drug’s effects with respect to one or more of those endpoints becomes a concern if there is not appropriate adjustment for multiplicity. The purpose of this guidance is to describe various strategies for grouping and ordering endpoints for analysis and applying some well-recognized statistical methods for managing multiplicity within a study in order to control the chance of making erroneous conclusions about a drug’s effects. Basing a conclusion on an analysis where the risk of false conclusions has not been appropriately controlled can lead to false or misleading representations regarding a drug’s effects.

Key Topics

Terms and concepts identified from this document

Scope & Applicability

Product Classes

2
Biological products

development program for drug and biological products

Combination vaccines

Trials of combination vaccines are a situation in which co-primary endpoints are applicable.

Stakeholders

1
sponsors

Target audience for recommendations on assessment of overall survival

Regulatory Context

Regulatory Activities

1
Marketing application

Submission for drug approval

Document Types

6
Labeling

Cybersecurity information should be included in device labeling

Prospective analysis plan

Detailed plan for dissolution profile comparison

FDA-approved labeling

Describes findings of safety, purity, and potency

Master protocol

A protocol designed with multiple substudies to evaluate medical products

statistical analysis plan

Basic statistical principle for clinical trials

Guidance for Industry

The type of document providing recommendations for CIPN drug development.

Attributes

8
Type I Error

Trial design should provide strong control of Type I error

Alpha

The significance level, typically 0.05.; Each vertex is allocated an initial amount of alpha.

Alpha allocation

Testing could proceed with alpha allocation across multiple hypotheses

Type II Error Rate

Considerations for sample size; FDA is also concerned with the risk of making a Type II error.

Type I error rate

Statistical parameter requiring strong control when testing multiple endpoints

significance level

Threshold (alpha) below which the Type I error rate should be controlled; Endpoints are tested at the endpoint-specific significance levels.

p-value

Probability of observing a result at least as extreme as the study result assuming the null hypothesis is true; The probability value used to determine statistical significance.; The Holm procedure directs testing the smaller p-value first.

Sample size

Determined during the planning of a research study

Technical Details

Substances

1
Losartan

Drug used in clinical trial references for renal and cardiovascular outcomes

Testing Methods

10
Bonferroni Method

Statistical method for managing multiplicity; A single-step procedure commonly used for multiplicity adjustment.

Graphical approach

The graphical approach is a means for developing and evaluating multiple analysis strategies.

Permutation

A resampling approach for multiple endpoints.

Bootstrap

An Introduction to the Bootstrap

Resampling-Based Multiple-Testing Procedures

Statistical approach using bootstrap and permutation for multiple endpoints.

Prospective Alpha Allocation Scheme

A single-step method that has a slight advantage in power over the Bonferroni method.

Superiority test

Tested after non-inferiority without multiplicity adjustment if it is the only endpoint; To test for superiority, one should first establish non-inferiority.

Non-inferiority test

A primary endpoint can be tested for non-inferiority followed by superiority; Figure A1 shows a flow diagram for non-inferiority and superiority tests.

Mixture gatekeeping procedures

Developed to account for complex logical/hierarchical relationships among endpoints

Hierarchical testing approach

Used when endpoints are ordered based on clinical importance

Clinical Concepts

10
Myocardial infarction

For studies in which the outcome or outcomes of interest (e.g., myocardial infarction or stroke) include fatal outcomes.

Type 2 Diabetes

evaluation of a drug for the management of Type 2 Diabetes

Hypoglycemia

patients with type 2 diabetes may face unique safety issues such as a risk for hypoglycemia

HbA1c

Normal range of HbA1c used as a response criterion in diabetes trials.

Death

Section D.9 regarding patient mortality

Stroke

correlated with female sex in VAD study

Primary Endpoint

The main outcome measure in a clinical trial

Secondary Endpoint

The hierarchy of families of endpoints; The endpoint with inadequate power for detection is classified as a secondary endpoint.

Composite Endpoints

incorporates multiple aspects of the disease into a single endpoint

Surrogate endpoints

reasonably likely or expected to predict a clinical benefit

Standards & References

External Standards

1
American College of Rheumatology scoring system

Used for rheumatoid arthritis multi-component endpoints.

Specifications

3
primary endpoints

Endpoints critical to establish effectiveness for approval

secondary endpoints

Provide useful description to support primary endpoints or demonstrate additional clinically important effects

exploratory endpoints

Used for research purposes or new hypotheses generation; do not need multiplicity adjustment

ICH References (2)

ICH E9

Statistical Principles for Clinical Trials; Discourages deterministic procedures due to high risk of bias; Notes that the use of Bayesian methods in clinical trials may be considered.

ICH E9(R1)

Statistical Principles for Clinical Trials: Addendum: Estimands and Sensitivity Analysis in Clinical Trials

Related CFR Sections (2)

Related MFDS Guidelines

Korean regulatory guidelines covering similar topics

See Also (8)

Multiple Endpoints in Clinical Trials: Guidance for Industry | Guideline Explorer | BioRegHub