Drug Development Tool Qualification: US FDA Offers Three-Step Review Process

FDA's proposed review timeline to qualify biomarkers, clinical outcome assessments, and animal models can actually be pretty quick – assuming submissions clear the initial "reviewability" hurdle. 

The US FDA has set up a three-step review process to qualify tools for use in drug development. But before beginning this procedure, the agency first determines the "reviewability" of a sponsor's submission.

FDA officials described how the agency is implementing the new qualification process for drug development tools, established by the 21^st Century Cures Act, at a Dec. 11 public meeting. Enacted in December 2016, the statute required the agency to convene the meeting to solicit input on the qualification process.

The Cures Act added Section 507 to the Food, Drug, and Cosmetic Act, which lays out a three-stage process for qualification of drug development tools. A drug development tool can be a method, material or measurement that can aid in drug development and regulatory review. The Cures Act specifies that these tools include biomarkers and clinical outcome assessments; FDA added animal models to the program.

The new process replaced the agency's existing qualification process, referred to as the legacy qualification program, which FDA initiated in 2004.

Chris Leptak, director of the Center for Drug Evaluation and Research's Biomarker Qualification Program, noted the benefits of the new process, which remains voluntary.

"Rather than under the old process or other aspects of FDA activities where basically you put something over the fence, we look at it behind our closed doors, and then we give you some type of a response back," the Cures process has given the agency a much more collaborative role, he said.

Leptak said the agency would like to hear from sponsors as to whether they think the process is of value. He noted that FDA has found that it has opened the door to involvement from a broader community.

"Ultimately, if you are engaging with FDA in the drug specific space, if you're not a drug developer there is not a natural pathway for you to have a conversation with us. If you are an academic, if you are a clinical provider, if you're a patient group, there is not a natural mechanism whereby you could bring forward useful information that we could use as part of drug development" and aid in public health.

Three Submissions

Leptak described the three sequential stages of submissions for qualification of a drug development tool (DDT) and the information FDA expects to receive.

Sponsors must first submit a letter of intent that explains the need for the tool and why it could potentially be helpful. Leptak said that if there are other tools in this space that do something similar or the same, a sponsor should say why the new tool is better. He said FDA also wants to know the information to support the tool, the context of use, and its feasibility, especially around any analytics that may need to be developed.

If FDA accepts the letter, the sponsor then submits a qualification plan. Leptak said this allows FDA to take a deeper look at the current scientific understanding and the sponsor's interpretation of that scientific information.

"When you are trying to bring forward published literature to support your idea, it's important for us to have a very candid conversation around the value and potential limitations of the available data. Because if we can't agree on that it's hard to define the knowledge gaps" to decide what must be done to get to your goal, Leptak explained.

Once the plan is accepted and the sponsor collects and analyzes the data, the sponsor submits a full qualification package to FDA. If it is successful, the tool can be used in the IND, NDA and BLA space without the need to resubmit the information and FDA will not need to re-review the information for context of use.

Leptak noted that some sponsors have asked if they could submit just a full qualification package. The answer is no; all submission stages must be completed.

Leptak said FDA's review of submissions consists of a three-step process. The first step is an assessment of the submission, in which the agency determines if the sponsor has covered all the content elements and the proposed tool fits within the scope of its programs.

"There's an opportunity for us, based on our past history of developing tools in this space, to help you tell the best story you can to bring it forward for further discussion," Leptak said. "The point of the first tier of review is to understand what your goal is and give you the language and tools to articulate that."

The submissions are then reviewed by a committee of subject matter experts, which include officials from the Office of New Drugs. They evaluate the submission based on regulatory precedent, current disease-specific challenges, and the level of impact on drug development programs.

The third tier of review is conducted by a new center-wide DDT committee, which includes senior leadership throughout CDER. It will make a final decision on whether to accept the submission, drawing on the historical regulatory knowledge of its members.

The Cures Act required FDA to establish a reasonable timeline for review of submissions but did not define what that would be. Leptak said FDA is proposing that the letter of intent be reviewed within three to four months; the qualification plan be reviewed within six to eight months; and the full qualification package be reviewed within eight to 10 months.

FDA's Elektra Papadopoulos, of the Clinical Outcome Assessment Qualification Program, said the timelines are goal posts and if the agency can review submissions faster it will do so. "We thought these were reasonable timelines at least to begin with," she said. "I could envision if we find something that requires more data to be submitted some adjustments in the timeline would be needed."

To date, FDA has qualified a total of seven biomarkers or biomarker panels. There are another 27 publicly disclosed biomarker submissions for which qualification decisions have not been made, according to data on FDA's website. The list of 27 includes two submissions for which a letter of intent was not accepted due to reasons related to scientific merit.

The agency has qualified four clinical outcome assessments, with another 34 publicly disclosed submissions to process. (Also see "US FDA Drug Development Tool Qualification Takes Center Stage At December Meeting" - Pink Sheet, 12 Nov, 2018.)

'Reviewability' Determination

Before FDA begins the review process, it first determines the "reviewability" of a submission, assessing whether it addresses the needed elements of a tool for the particular program so FDA can decide on its acceptability in a predictable and timely manner.

Leptak said FDA drew on the concept of "file-ability" in the IND/NDA space in creating the concept. He noted that since the new qualification process has been in effect, FDA has found that some submissions need no work and can go to tier 2, while others have had six back and forth exchanges before they could be submitted to a committee.

"We think the reviewability concept is a really valuable one and we'd like to see if you agree with us," he said.

If FDA determines a submission is not reviewable, it sends a non-reviewable letter to the sponsor with detailed comments on how it can be improved for resubmission. When the proposal is reviewable, the agency sends a letter to the sponsor and the review clock starts. Leptak said sponsors must respond to recommendations FDA makes in subsequent submissions. He noted that FDA had done a spot check of legacy projects and found that it would make a series of recommendations, repeatedly raising the same issues which were ignored by the sponsor.

Value Of Transparency And Sharing Data

The Cures Act also mandated that the qualification submissions and FDA responses be posted on the agency's website. Leptak noted that this has been beneficial for stakeholders. He said that under the legacy process, groups would not know who else was working on a drug development tool. He said only a quarter of tool developers were willing to make the information public.

"The outcome of this is that we have a shared learning culture that's developed," he said. He noted that one of the challenges of tool development has been access to data, and now academic or patient groups can identify sponsors and offer to share data or collaborate with them.

Before "we tried to do a dating game" and ask people if they would be willing to talk to each other, Leptak said. "But now you guys can do your own dating and we don't have to be the wingman in the process."

Speaking from the audience, Alex Bush, associate director of regulatory strategy at Sarepta Therapeutics Inc., noted that a lot of DDT work is done in the context of the IND, NDA and BLA drug development process and asked how work done outside the DDT process will impact the ability of other organizations to use it.

Responding from the clinical outcomes assessment perspective, Papadopoulos said when work is done under an individual drug developer's IND and the drug fails, the instrument will die along with it and others won't be able to use it.

"I would encourage sponsors to submit to qualification so that others can leverage the work that's already been done," she said.

Papadopoulos added that another benefit of having a common instrument is that it can allow FDA to make comparisons among products that it cannot do if each sponsor is using their own proprietary instrument.

FDA Needs To Hire More People Instead Of Playing 'Musical Chairs'

In a panel on identifying key themes and next steps, Sonya Eremenco, associate director of the Patient-Reported Outcome Consortium at the Critical Path Institute, said she would like clarity as to who is on the subject expert committee.

John-Michael Sauer, program officer of the Biomarkers Program and Executive Director of the Predictive Safety Testing Consortium at the Clinical Path Institute, noted that it has been 10 years since the first qualification in 2008 and only a handful of biomarkers have been qualified. "We need to think creatively how we can move this forward faster," he stated.

Leptak responded that FDA could commit to faster timelines if it could hire more people.

"If it takes us two years to fill a position, how can we commit to something faster?" he said. "Ultimately, what ends up happening internally is that when new positions or new roles are created, because it's so difficult to bring in somebody from the outside we just take and do musical chairs of existing employees and keep moving them around."

"How can we help because we want to speed up qualification. Is that a letter writing campaign?" Sauer asked.

Leptak replied, "You as stakeholder that FDA engages with have a voice that is important. So, think of how you could use that voice to address this problem."

FDA's docket for public comment on the drug development tool qualification process is open until Jan. 31. Leptak cited topics the agency would like input on: the pros and cons of the concept of reviewability; if the proposed timelines are reasonable; what kind of engagement should happen between submissions; and input on the public posting of information on FDA's website.