Randomized Withdrawal Design and Randomized Discontinuation Trial

FDA recently issued its new guidance to industry “EnrichmentStrategies for Clinical Trials to Support Approval of Human Drugs andBiological Products”. The section D of the guidance detailed the Randomized Withdrawal Design as one of the enrichment strategies.

Randomized Withdrawal study design is not new and has been practically used in many drug trials and in NDA or BLA applications. It may be called 'randomized discontinuation trial' or 'randomized discontinuation design' in some literature. Randomized Withdrawal study design is especially popular in CNS and neurology areas.

According to ICH ICH Topic E 10 “Choice of Control Group in Clinical Trials”, the Randomized Withdrawal design is defined as:

 “In a randomized withdrawal trial, subjects receiving a test  treatment for a specified time are randomly assigned to continued treatment with the test treatment or to placebo (i.e., withdrawal of active therapy).  Subjects for such a trial could be derived from an organized open single-arm study, from an existing clinical cohort (but usually with a protocol-specified "wash-in" phase to establish the initial on-therapy baseline), from the active arm of a controlled trial, or from one or both arms of  an active control trial.  Any difference that emerges between the group receiving continued treatment and the group randomized to placebo would demonstrate the effect of the active treatment.  The pre-randomization observation period on treatment can be of any length; this approach can therefore be used to study long-term persistence of effectiveness when long-term placebo treatment would not be acceptable.  The post-withdrawal observation period could be of fixed duration or could use early escape or time to event (e.g., relapse  of depression) approaches.  As with the early escape design, careful attention should be paid to procedures for monitoring patients and assessing study endpoints to ensure that patients failing on an assigned treatment are identified rapidly. 

Randomized Withdrawal design is recommended in ICH E12A “PRINCIPLES FOR CLINICAL EVALUATION OF NEW ANTIHYPERTENSIVE DRUGS “

In 2001, Nation Academies published a report “Small Clinical Trials: Issues and Challenges”. It stated the advantages and disadvantages of the Randomized Withdrawal Design:

 “The advantages of this study design are that individuals receiving the experimental intervention continue to do so only if they respond, whereas individuals receiving the placebo do so only until their symptoms return. Disadvantages include carryover effects, difficulties assessing whether the underlying disease process is still active, and long lag times to adverse events if the disease is in remission. This design is more appropriate in phase I and II trials involving healthy volunteers because it is less likely that effective treatments are being withdrawn from those who need it. In some studies, however, measurement of the placebo effect is essential (e.g., studies of drugs for the treatment of depression), and such studies might require the use of a randomized withdrawal design. In those cases, voluntary, informed consent is essential, as is the provision of care during the withdrawal period.”

Additional advantage of this design is to study the long-term efficacy or safety (withdrawal effect) and additional disadvantage is the longer overall study period (additional period is needed to identify responders).

Some of the practical cases of using the Randomized Withdrawal Design in clinical trial are:

• Randomized withdrawal design used in the clinical trials for ADHD
• Cystic Fibrosis-Dependent Exocrine Pancreatic Insufficiency trial used an open label period to establish the baseline for the randomized withdrawal period. The endpoint is the change from the baseline after the open label period to the end of the randomized withdrawal period
• Flibanserin (a so-called Female Viagra drug) for the treatment of hypoactive sexual desire disorder (PSDD) in premenopausal women. One of the pivotal trials began with a 24-week open label phase, which then enrolled only “responders” into the randomized withdrawal phase
• IGIV-C for the treatment of chronic inflammatory demyelinating polyradiculoneuropathy (ICE study): the trial had two randomization. The second randomization was only for responders (to IGIV-C or Placebo) from initial (the first) randomized period. In extension period, the randomized withdrawal design was employed. The responders were re-randomized to IGIV-C or Placebo to study the long-term efficacy by comparing the relapse rates and the time to relapse. Notice that the responders from Placebo group in initial (first) randomized period were also included in the randomized withdrawal period. This is purely for maintaining the blinding.

Additional Reference:     

• Enrichment Strategies Robert Temple, MD
• Kopec (1993) Randomized Discontinuation Trial: Utility and Efficiency
• Strdler (2007) The Randomized Discontinuation trial: a phase II design to assess growth-inhibitory agents
•