Futility in clinical research

Editor-In-Chief: C. Michael Gibson, M.S., M.D. [1]; Associate Editor-In-Chief: Gonzalo A. Romero, M.D. [2]

Overview

During the conduction of a clinical trial several interim analysis are performed in order to obtain data that could be used for early termination of the trial or adjusting the study's protocol due to futility, adverse effects profiles or insignificant statistical efficacy of the intervention being studied. This data is reported to the Data Monitoring Committee (DMC) which can help in decision making of the trial.

Futility Definition

Futile is used to describe any intervention to achieve a result that is possible, but experience suggests that cannot be systematically obtained. A futile action is one that cannot achieve the intended goals. It is important to distinguish it from hopelessness; futility refers to the objective quality of an action, hopelessness is a subjective term.

Medical Futility

In medicine a treatment or intervention is considered futile if it does not improve prognosis, comfort, well-being or general state of health or any other clinical end-point. Example of futile treatments could be the ones used in vegetative state patients and children with leukemia or Hodgkin lymphoma which are often described as "pushing ahead with futile therapies". Some futile treatments are used in terminally ill patients to facilitate their well being. It is important to highlight that carefully analyzing some futile treatments could eventually lead to clinical studies that might contribute to medicine in knowledge.

Data and Safety Monitoring Board

The Data and Safety Monitoring Boards (DSMBs), also known as Data Monitoring Committees (DMCs), or Data and Safety Monitoring Committees (DSMCs), is a group of subjects with important experience which reviews on a periodically basis accumulating data from ongoing clinical trials. They have several responsibilities such as monitoring that the study continues until the planned completion of follow-up, regardless of the duration of treatment. Their members assess for safety of trial participants through Interim Monitoring. Also they monitor effectiveness, monitor study conduct, make recommendations, maintain meeting records.The DSMB establish a Charter describing the Operational procedures which includes meetings scheduling, data presentation, interim data access, assessment of conflict of interest, and methods of interim reports to the DSMB.

The FDA recommends that a DSMB should have access to the treatment assignment for each group, but other authors recommend that the information should be coded to avoid the introduction of bias by unblinding the DMC, assuring a greater objectivity of the interim review. They may recommend termination of a study if a group shows a significant adverse event when compared to control or if the investigated treatment shows to be futile (futility analysis) through the interim assessment and statistical methods used. The Sponsor and DSMB should establish a process to unblind the treatment codes to the DSMB members when needed in case of an emergency.

Trial Monitoring

There are two types of monitoring Clinical Trials:

1. One involves the oversight of the quality of the trial; which includes that the protocol is followed correctly, the data is properly accrued, evaluating of dropout rates and adherence, therefore study conduct mainly. It does not access the information to compare different treatment arms (unblinding) and does not change the Type I error chance. The sponsor is responsible for this, and can be performed by the sponsor or by a group selected by the sponsor.
2. The interim analysis involves the accruing of comparative treatment results requiring unblinding or access to treatment group assignment and comparative treatment group summary information. Therefore, the protocol (or appropriate amendments prior to a first analysis) should contain statistical plans for the interim analysis to prevent certain types of bias.

Interim Monitoring and Analysis

It is an analysis intended to compare treatment groups in regards to efficacy, safety or futility at any time before the formal trial completion, which gives important information about the study progress. The frequency of this analysis is variable. Their intention is to monitor Safety, Efficacy and Futility. If the treatment difference is smaller than some pre-established statistical value, the clinical trial is stopped; meaning a trial that would not have shown statistical significance if they had gone onto completion. A DSMB guided by a monitoring plan acceptable to both the DMC and the study leadership, will generally be in charge with recommending early termination on the basis of a positive result only when the data is promising and the false positive risk is low. 1 In the past some clinical trials have been stopped for futility. Example of this were large trials in critical care medicine.

- Safety is assessed to determine a specific number of serious adverse events towards one group. Some adverse effects appear only in larger studies.

- Efficacy to determine the difference between the intervention being studied and the control. There are different approaches to assess efficacy such as likelihood methods, Bayesian methods, Frequentist Methods, The O'Brien-Fleming approach, and group sequential methods.

- Futility to determine if the novel drug is not more efficacious than the control. If this is the scenario the trial should be stopped.

Futility Assessment

Various approaches exist to assess futility including stochastic curtailment, asymmetric stopping boundaries and predictive power.

1. The Stochastic Curtailment used in early trials when a Committee reviews the results of a trial (assuming about actual success rates of the treatments) and calculates the probability that the trial will give a significant result if finished. If this probability was small the trial should be terminated. A safe way to do this is to use the original difference that was used initially to calculate the sample size.

2. Using Asymmetric Stopping Boundaries. The futility boundary can be based on how quickly you want to stop the trial if the treatment is ineffective. The faster you stop for an ineffective treatment, however, the larger the sample size needs to be to detect a difference if it is there.

3. Using Conditional Power is a method that quantifies the statistical power to yield an answer different from the one obtained through interim analysis. If the quantitive result is significantly small, it could be concluded that continuing the study would be futile. A Conditional Power could be set from 10 to 30%, any Z-value obtained by an Interim Analysis below the set up threshold shows futility.

After several Interim Analysis are performed, the data collected would serve to determine if there is a positive or a negative trend in the ongoing Trial.

Data-dependent Stopping

Describes any statistical or administrative cause to stop a trial. It could lead to early termination or protocol changing.

The Statistical Criteria give guidelines for stopping the trial because the decision to stop a trial is not based just on statistical data collected on one endpoint, judgement should also be considered. The statistical methods for interim analyses have some similarities, such as:

• Require the researchers to establish the objectives specifically and in advance of the trial,
• Usually have similar performance qualities,
• Establish a penalty for terminating early,
• Assume that the basic study design is sound,
• Require structure to the problem beyond the information currently collected,

Consequences of Stopping a Trial due to Futility

Advantages of Futility Stopping

Nevertheless, stopping a trial due to Futility might have positive consequences or advantages, for example :

1. Time saving
2. Appropriate use of research funding
3. Reducing the participants exposure to the harmful treatment
4. Allows information dissemination about the treatments as soon as possible

These resources can be used towards more promising research studies that might generate important discoveries.

Disadvantages of Futility Stopping

Stopping a trial after obtaining data that suggests Futility originates several disadvantages, such as:

1. Leaves the primary research question without a definitive answer.
2. Difficult interpretation of a negative result.
3. Wider confidence bounds than if the study was finished which leads to the estimated treatment difference to be biased downward.
4. Increases the risk of imbalance in prognostic factors.
5. Alters the analyses of secondary outcomes.
6. Adverse events data is limited.
7. Researchers will unlikely conduct the trial stopped again.
8. Clinical investigators have a responsibility to consider the effects of their research upon the totality of literature in their field.

- There may be pressures to continue the study in order to increase precision, reduce errors with the final goal of achieving enough statistical power. Also to assess subgroups, and obtain information on secondary clinical endpoints.