Skip to content
Clinical Development

Operationalizing Complex Innovative Trial Design (FDA’s New Approach)

  • July 21, 2020
One of the FDA’s recent initiatives centers on operationalizing complex innovative trial design.  The FDA defines complex clinical trial designs as designs that are intended to advance and modernize drug development.  The guidance emphasizes that there is no fixed definition of what constitutes a complex innovative trial design because what is considered innovative or novel today will change over time, and aspects of essential trial designs will vary by therapeutic area.

The FDA did specify six types of trials that they currently consider innovative:

  1. Incorporation of real-world data.
  2. Trial adaptations to the challenges of implementing trial designs in rare diseases
  3. Utilization of patient focused endpoints as primary endpoints for product approval
  4. Master protocol design
  5. Adaptive design protocols
  6. Bayesian statistical design and simulation protocols

Incorporation of real-world data

The most common application of real world data in clinical trials is to enrich or replace the control arm of the standard study with either historical data or real time data taken from existing data banks, such as those from Health Maintenance Organizations (HMOs) and other large healthcare organizations or from healthcare technology companies like Flatiron Health that provide data from large patient populations over time.

Trial adaptations to the challenges of implementing trial designs in rare diseases

The FDA continues to support the development of drugs for orphan rare diseases.  One means of facilitating clinical trials in rare diseases is the use of historical control data to enrich or replace the control arm. Decreasing the control arm, or in some cases replacing it, has the potential to increase participant recruitment and adherence to trials, and accelerate timelines.  In addition to expediting recruitment, because there is such a small population of potential participants, sometimes patients are more likely to sign up for a trial when they know they are likely going to receive the investigational agent rather than the placebo.

Utilization of patient focused endpoints as primary endpoints for product approval

Rare disease trials frequently utilize surrogate endpoints as these diseases tend to progress over a long period of time and are likely to be irreversible.  As such, the desire is to treat patients at an early stage of disease where a quality of life can still be maintained and prevent mortality.  Leveraging modeling and simulation may provide support for the use of these surrogate markers.

There’s been a huge emphasis in the past decade on patient-focused medicine especially on quality of life.  The patient experiences may be collected through surveys, patient interviews or focus groups.  The FDA is seeking some input and providing recommendations to sponsors on the best methodology for collecting this type of data.

Master protocol design

The over-arching goal of a master protocol is to simultaneously or iteratively evaluate multiple therapeutic agents in potentially multiple diseases or patient subpopulations within a specific disease.  Master trials have primarily been implemented in oncology.  One of the key advantages is that potential participants only go through one screening process for multiple investigational treatments.  Randomization may no longer be one-to-one randomization and patients may be preferentially assigned to the therapeutic agents that researchers feel is the most promising for their variant of the disease.

There are multiple interim analysis as they evaluate therapeutic agents with a novel trial, and that requires special statistical methods to eliminate the risk of bias.  These trials are characterized by dynamic addition and removal of treatment arms as they are found to be efficacious or not.

One type of master protocol is a basket trial.  In a basket trial one agent is being evaluated across multiple diseases or subpopulations of a disease.  One example of a disease that lends itself to basket trials is breast cancer where multiple biomarker statuses exist including hormone receptor and epithelial growth receptor status variants.

The other major structure design of a master protocol is the umbrella design where multiple treatment options on a single disease or subpopulation of a disease is being tested.  There can be one control arm for all of the study arms or some treatment arms may not have a time-matched control arm.  Master protocols may combine elements of both a basket and umbrella design.

One of the most well-known master protocols is Lung-MAP.  Largely funded by the National Cancer Institute (NCI) and administered by the SWOG Cancer Research Network, Lung-MAP enrolled its first patient in 2014 and is ongoing.  The single screening procedure includes evaluation of key biomarkers and participants are assigned to the treatment arm that is thought to be most promising for that participant.  Randomization is not balanced as they are looking specifically at biomarkers and disease sub-types in addition to therapeutic agents.

Adaptive design protocols

Adaptive trials can either be very simple or part of a very complex protocol such as part of a master protocol.  Quite simply an adaptive trial design means utilizing pre-specified decision trees to modify the ongoing trial based on accumulated data.  For example, in early clinical phase when testing multiple doses, if unexpected safety signals arise, the trial can either be stopped or the dose can be modified.  It has been a well-accepted methodology for safety to use adaptive trial designs, but it is now advanced for use in pharmacokinetics and efficacy.

Bayesian statistical design and simulation protocols

The final type of the complex design mentioned by the FDA is Bayesian statistical design and clinical trial simulation.  These methodologies leverage previous information on patient variability, differences in disease course among patients and the differences in treatment response among patients to maximize the trial design.  This gives the opportunity to estimate the probability that a specific patient at a specific time in their disease will have a response to the treatment.

What Led to these Actions?

The major impetus that led to these actions by the FDA is the 21st Century Cures Act that was enacted by Congress in December 2016.  It charged the FDA with very specific tasks on how to implement and educate sponsors about innovative clinical trial designs.

In response to passage of the 21st Century Cures Act, the sixth renewal of the Prescription Drug User Fees Act (PDUFA) was enacted in October 2017.  There are multiple features within this act that are intended to increase the number of products that are approved within the first cycle by enhanced communication between the Agency and sponsors.  There is a sustained commitment by the FDA to breakthrough therapies and surrogate endpoints.  It specifically focuses on complex innovative trial design including features for rare diseases, real world evidence, capturing the patient voice, and clinical trial modeling and simulation.

There have been multiple FDA programs implemented after PDUFA VI including adaptive trial design and complex innovative design programs.  The complex innovative trial design pilot program is a joint initiative of the Center for Drug Evaluation and Research (CDER) and the Center for Biologic Evaluation and Research (CBER).  The CDER program specifies that their program is led by the office of statistics.  One of the key take home messages is the major role of statistics in operationalizing these types of trials.

The pilot program offers sponsors two additional focused meetings with the FDA designed to specifically discuss the proposed study design.  The Agency emphasizes these meetings are not intended to replace existing pathways for interaction such as the end of Phase II meetings or meetings for break through therapy or orphan designation.

There are very specific eligibility requirements for the pilot programs.  The sponsor must have either initiated pre-IND or IND application, so this program is not for a hypothetical compound.  Further, the proposal needs to be focused on clinical effectiveness of the product and there needs to be sufficient clinical data available to inform the proposal.

The proposals accepted into the pilot program will serve as educational resources to facilitate the adoption of complex innovative designs by all sponsors.  One of the key requirements for eligibility into the program is the sponsor must agree that the FDA can publicly disclose certain study design and analysis elements even before the product is approved.

The FDA aims to select a variety of clinical design approaches from therapeutic areas in order to enhance the educational effort of a program.  They will be accepting proposals four times a year and they will select two proposals each quarter.  Selection considerations include therapeutic need addressed by the investigational agent, trial design appropriateness, the need for simulations, level of innovation of the trial design, and the value proposition of the design proposed.  Potential benefits from acceptance to the program include expedited therapeutic product development and increased clinical trial efficacy.  But the program is intended to educate all stakeholders and advance the use of an adoption of complex innovative trial designs.

The FDA announced the first two proposals that were accepted into the pilot program in January 2019.  The first proposal from Eli Lilly was a master protocol for chronic pain that is evaluating multiple products for treatment of chronic pain.  This is anticipated to facilitate more rapid clinical assessment.  Two take-away messages are that master protocols can be used for other indications besides oncology and that this program is not just intended for small or medium size companies.

The other proposal that was accepted in the first cycle of this pilot project was from Wave Life Sciences.  This drug candidate had already received an orphan disease therapeutic product and was intended for a subset of patients with Duchenne’s Muscular Dystrophy.  The elements of innovative designs addressed in this proposal were leveraging historic controls to augment the placebo arm and reduce the number of participants receiving placebo.

It is important to recognize that all the regulatory agencies are considering innovative trial designs.  As soon as you recognize the potential for trial innovative to accelerate your development program, even if your drug candidate is in preclinical development, it is advantageous to at least consider conversations with the regulatory authorities regarding your development program.  It is also possible to consider collaborative and innovative design aspects within the existing framework of product approval.  Medpace has the expertise to guide you through this process.