Stakeholders want more clarity from FDA on using data from externally controlled trials

The US Food and Drug Administration (FDA) recently released draft guidance on the use of externally controlled clinical trials in determining the safety and efficacy of drug products. The topic generated a flood of public comments, with many stakeholders wanting examples of how external controls and RWD can be used in specific cases, particularly in rare diseases.

The draft guidance describes using externally controlled clinical trials in evidence for a drug product’s safety and efficacy, such as a group of treated or untreated people in a historical control or in another setting. FDA’s guidance details the study design of these externally controlled trials (ECTs), potential biases in validating trial results, use of patient-level data from other trials or real-world data (RWD), and communication considerations with the agency regarding ECTs.

The draft guidance does not elaborate on use of summary-level estimates as a replacement for patient-level data, the reliability of RWD sources or natural history studies, and use of external control data as a supplement for a traditional randomized controlled trial.

Most stakeholders acknowledged that the draft guidance was limited to patient-level data from other clinical trials and sources of RWD and appreciated the agency’s promise to provide more specific guidance in the future regarding real-world evidence (RWE). The American Society of Hematology (ASH) noted in their comment that they would like to see more specific examples about potential issues and challenges surrounding external controls as well as “how the FDA would like information to be prepared and presented for planning meetings with sponsors and data providers around external controls” and potential outcomes from meetings with the agency.

Commenters also identified several potential issues with the draft guidance, such as not completely addressing the issue of using external controls, not describing the use of hybrid control data, unclear proposals for documentation surrounding RWD and RWE, issues involving study design, and concerns about utilization for pediatric and rare diseases.

Some stakeholders said the agency did not fully answer the question of how the use of external controls would be a potential solution for researching and identifying new treatments for patients. In their comment, EMD Serono asked for additional examples of using external control arms because “the guidance as currently written provides little recognition of when, where and why external control arms could provide such a solution.”

The American Society of Gene & Cell Therapy (ASGCT) agreed, noting in their comment that the guidance “does not meaningfully advance the use of external controls, including controls based on real-world evidence (RWE), potentially limiting sponsors’ ability to overcome practical and ethical challenges in developing treatments for rare disease.”

ASGCT also expressed concern that the draft guidance does not align with Congressional direction to use RWE. “Given that the Congressional direction to issue this guidance is over six years old and that the advancement of RWE has been supported through multiple cycles of [Prescription Drug User Fee Act] funding, we expect that FDA’s views on how RWE may support drug development would be significantly more mature,” they said.

Rather than focusing on the challenges of adopting RWE, ASGCT said, the focus of the draft guidance should be on how it can be used to “to support product development, and recommendations on the collection, analysis, and submission of control data that would best support regulatory approval particularly when traditional trial design is unethical or unfeasible.”

Some stakeholders noted that FDA should include a discussion of hybrid control arms in this draft guidance or in future documents. “Many of the considerations discussed in this guidance apply to external control arms; investigators and sponsors will benefit from clearly understanding FDA’s perspective on how implementation of supplemental controls might differ from external controls, FDA’s overall approach to risk assessment of differing control types, and data analytic approaches,” the American Society for Clinical Oncology (ASCO) said in their comments.

Other commenters wanted more information on documentation and retention requirements for RWD. In their comment, the Association of Clinical Research Organizations (ACRO) said the retention periods, records and reports required for RWE is not addressed in the guidance, and requested more information about whether this data is subject to the rule requiring sponsors to retain records and reports for 2 years after the approval of a marketing application or discontinuation of investigational use.

“ACRO asks for the final guidance to clarify that the retention period for RWE data is no longer than that required for records and reports of conventional clinical trials conducted under IND authority,” they said, as well as the agency’s expectations for retention of this data.

ASCO also pointed out that RWD documentation may not be as robust compared to traditional clinical trials. “[W]e believe it is important for the FDA to provide clarity about its approach to determining impact when there are unaccountable but potentially important, imbalances between the treatment arm and the external control arm,” they said.

In their comment, IQVIA noted the challenges around RWD concerning data privacy, data access, and data transformation. “Real world data is importantly differentiated from data gathered through a randomized-controlled trial in that RWD are recorded for purposes not specific to the study question, and as such, systems and quality controls may be differently defined,” they said.

In their comment, the non-profit Public Responsibility in Medicine and Research (PRIM&R) said that based on the draft guidance, it is currently unclear what to include in Institutional Review Board (IRB) protocols concerning external controls.

“[W]e recommend the guidance include a section that both specifies factors that the IRB should take into consideration when evaluating the appropriateness of a proposed external control and advises investigator on the type of information they should include in their IRB protocols to describe the control selected and the rationale for selecting it,” they said. In addition, PRIM&R noted a directive to have sponsors and investigators collect data in current trials so that data can serve as an external control for a future study may be appropriate.

PRIM&R also raised the issue of ethical concerns surrounding external controls. “[T]he draft guidance does not address ethical issues in studies involving external controls,” they said. “Specifically, the guidance is silent on respecting individuals’ consent for the use of their data that were collected for other purposes (for example, data from other studies, or real-world data). The use of these data poses distinct challenges to institutional review boards (IRBs) charged with overseeing such studies.”

The Alliance for Regenerative Medicine (ARM) explained in their comment that use of external controls may not be feasible for some rare diseases.

“Some of the guidance on the use of retrospective natural history studies as a control arm could be challenging for treatments for diseases with small populations, as is the case for many of the diseases that cell and gene therapies would be able to treat,” they explained. “These historical control groups are critical for evaluating gene and cell therapy products in patient populations that heretofore had no effective therapies and would now have the potential to receive a product that would prevent disease progression or provide a cure rather than symptomatic treatment benefits.”

An FDA proposal to require the inclusion of patient-level data in marketing applications may prevent the use of external controls for some rare diseases, the American Brain Tumor Association said in their comment. “[I]ndividual patient-level data (IPD) is not always available for many cancers and other rare diseases,” they wrote. “This is due to many barriers that interfere with data sharing in healthcare.”

There is also no acknowledgment in the draft guidance on how to utilize external controls in pediatric populations, ACRO noted. “These may be important to include as randomized studies may be difficult to execute for ethical reasons and so external control arms may play an important role,” they said. “For life-limiting conditions, inclusion of data from other jurisdictions or historical sources may be required to act as the only viable reference group.

“It would also be helpful to include comment regarding acceptable external control group approaches for extremely rare conditions where variability of treatment decisions may mean that electronic health records are not sufficient to act as controls,” they added.

Draft guidance

Stakeholders want more clarity from FDA on using data from externally controlled trials

Related topics

Recommended content

Welcome to the new RAPS Digital Experience