May 24, 2021
Creating a New Cancer Research Paradigm by Broadening Eligibility Criteria
ASCO and Friends of Cancer Research (Friends, a cancer advocacy organization focused on advancing science, policy, and regulation) have been collaborating to promote broader, more inclusive cancer clinical trial eligibility criteria since 2016. On April 9, 2021, ASCO and Friends co-hosted a virtual event on the topic. Stakeholders from research sites, government, patient advocacy, and industry convened to discuss the real-world benefits and challenges to implementing a new cancer research paradigm where patients are eligible for clinical trials as the default, and only excluded when there is justifiable concern for patient safety.
This is the first in a series of articles in which event and project participants will address audience questions submitted during the April 9 event (some questions have been edited for clarity or length). For more information about the ASCO-Friends collaboration and recommendations for broadened eligibility criteria, please visit the ASCO and Friends websites.
In part 1, questions are answered by Julia A. Beaver, MD. Dr. Beaver is the chief of medical oncology in the Oncology Center of Excellence at the U.S. Food and Drug Administration (FDA) and deputy director (acting) in the Office of Oncologic Diseases in the Center for Drug Evaluation and Research at FDA. She served as the FDA representative on the ASCO-Friends project taskforce. (Read part 2, with responses from Dr. Michael A. Thompson, and part 3, with responses from Dr. Lori J. Pierce.)
Should there be robust statistical input when eligibility criteria are being developed to broaden the study population? A more heterogeneous population that may result from broadening eligibility will also impact the clinical meaning of your study endpoints. Is there any discussion or guidance regarding this?
JB: Yes, statistical input is critical for successful trial development in general and would be very important to provide advice regarding efficacy analysis of a separate subgroup as part of a broader population. If outcome is anticipated to differ based on enrollment of a more heterogeneous population, calculations for sample size and outcome can be prospectively adjusted. We encourage discussion of these potential issues in meetings with the FDA and industry early, as clinical trials are being designed, in order to identify and address any concerns in the development stage prior to trial conduct. Ultimately the clinical meaningfulness of results on all endpoints studied in a more representative population would be more relevant to the real world and to those who will use the drug after approval.
How do broader eligibility criteria impact the use of historical controls or use of older clinical trials for benchmarking?
JB: If the prospective trial employs broader eligibility criteria, it is likely that more sources of data may be used for establishing the benchmark rates or control treatment efficacy assumptions for sample size calculations. This may even reduce the probability that a trial will fail due to inappropriate assumptions for the underlying treatment effects. Data in a wider setting that can provide additional information for determining a comparative benchmark for standard-of-care therapy include observational registry data, other real-world data, and data from clinical trials in the academic setting (i.e., those not intended to support a marketing application).
What is the FDA doing to simplify clinical research efforts?
JB: FDA has multiple initiatives on improving efficiencies of clinical trials and clinical research. As an example, the Oncology Center of Excellence is working on a pilot project to learn from the current system of remote assessments being conducted due to the COVID-19 pandemic. We are collecting information regarding decentralized clinical trial type modifications and evaluating the impact, if any, on trial data. We are hoping through this effort we will be able to continue many of the remote assessments to further advance decentralized trials in oncology.
In addition, we have a program that is fostering early-stage oncology product clinical research—assisting and simplifying development through scientific discussion, education, guidance, and regulatory engagement with academic life science incubators and accelerators as well as small pharmaceutical companies.
And a last example is that we continue to encourage and discuss seamless trial design, master protocols, and other innovative statistical methods to more efficiently enroll and achieve results of clinical research.