Restrictive eligibility criteria may limit or even jeopardize data generalizability of the trial results and opportunities for patients to participate in a clinical trial. 
And while the primary purposes of eligibility criteria are designed to protect the safety of patients participating in a clinical trial and to define the trial population, excessive or overly restrictive eligibility criteria can slow trial accrual, and limit understanding of the intervention’s benefit-risk profile.
A study, funded in part by the American Society of Clinical Oncology (ASCO) scanning more than 10,500 health records of patients with advanced non-small cell lung cancer (aNSCLC) included in the CancerLinQ database found that the use of expanded clinical trial inclusion criteria, as proposed by ASCO and Friends of Cancer Research in 2017, would nearly double the percentage of patients eligible to enroll in clinical trials.
Enrollment would, according to the study, featured during the annual meeting of the American Society of Clinical Oncology, held May 31, – June 4, 2019 in Chicago, Ill, jump from 52.3% to 98.5%.
The expanded clinical trial eligibility criteria would allow aNSCLC patients with brain metastases, previous or concurrent cancers, and limited kidney function to enroll in clinical trials.
Narrow eligibility criteria
“Ongoing use of historic, narrow eligibility criteria based on antiquated safety concerns place potentially unnecessary restrictions on trial participants, thus making it increasingly difficult to conduct the clinical trials necessary to demonstrate safety and efficacy of new therapies,” explained lead study author R. Donald Harvey, PharmD, BCOP, FCCP, FHOPA, Director of the Winship Cancer Institute of Emory University’s Phase I Clinical Trials Section, Atlanta, GA.
“Changes to eligibility criteria are particularly important as we step further into the era of targeted therapies, including immunotherapies, that have different safety profiles than highly toxic systemic chemotherapies,” Harvey noted.
The researchers looked at electronic health records in ASCO’s CancerLinQ database from 2011 to 2018 and focused on adults with aNSCLC who had two or more oncologist visits and at least one dose of a systemic treatment after diagnosis. CancerLinQ® collects and analyzes real-world data from patients at practices nationwide, drawing from electronic health records, to improve the quality of cancer care.
CancerLinQ Discovery® provides access to high-quality, de-identified datasets derived from patient data to academic researchers, non-profit organizations, government agencies, and others in the oncology community.
Using this data, researchers evaluated the number of patients eligible for clinical trials using traditional criteria and using the proposed criteria from ASCO and Friends of Cancer Research, which would allow for brain metastases, previous or current cancer diagnoses, and creatinine clearance levels as low as 30 milliliters per minute. Low levels of creatinine clearance can indicate kidney damage or impaired function. Traditional criteria, on the other hand, do not allow for these conditions and exclude patients with creatine clearance levels less than 60 milliliters per minute.
In November 2018, the National Cancer Institute revised its clinical trial protocol template to broaden eligibility criteria for cancer clinical trials based on the recommendations of ASCO and Friends of Cancer Research. More time is needed, however, to assess if these broader criteria are being widely adopted.
Sixty percent of the people had an advanced, stage IV diagnosis, and 80% were former or current smokers. The median age of patients was 67.6 years; 56% were male and 44% female. When the researchers applied traditional clinical trial enrollment criteria, 5,005 (47.7% of patients) would not meet trial eligibility criteria. If, however, expanded criteria were adopted, only 154 (1.5% of patients) would not meet eligibility criteria. Use of the expanded criteria would allow 4,851 more people to meet these eligibilty criteria, resulting in nearly twice as many people with a NSCLC who would.
- Adoption of expanded criteria compared with traditional criteria would also enhance the characteristics of aNSCLC patients in the following ways:
- Older patients, raising the median age from 66.1 to 67.5 years.
- Females, raising from 40% to 44%.
- Stage IV diagnoses, raising from 55% to 60%.
- Non-squamous types of lung cancer, raising from 45% to 47%.
- Never smokers, raising the rate from 13% to 16%.
The researchers are currently performing other analyses to look at differences between people who have been treated for their disease and remained stable, and people with ongoing brain metastases.
The study researchers are also doing sub-analyses to further refine which patients would be eligible for modern-day clinical trials. They note that none of these efforts would be possible without the use of ‘big data’ via CancerLinQ.
Harvey remarks that what is most needed is greater knowledge about drug-specific and population-specific outcomes in this disease in order to quickly and safely expand eligibility criteria.
“Expanded eligibility should also help to reduce disparities in enrollment, both socially and economically,” he concluded.
David L. Graham, MD, FACP, FASCO, a physician administrator at Levine Cancer Institute in Charlotte, North Carolina and associate editor for the ASCO University Editorial Board, as well as a member of the Cancer Communications Committee and the Practice Guidelines Implementation Network (PGIN), noted: “Only about 3% of patients with cancer in the United States currently enroll in a clinical trial, and restrictive eligibility criteria established in an era before advances we have made in supportive care is one reason for this low number.
“This study makes the case for universal adoption of broader clinical trial eligibility criteria, such as that proposed by ASCO and Friends of Cancer Research, so that more lung cancer patients and others have the opportunity to safely participate in potentially lifesaving research,” Graham concluded.
The results of the study emphasize the same conclusion reached in a 2017 study by Suanna S. Bruinooge, Caroline Schenkel. In this study the authors concluded that to maximize the generalizability of clinical trial results, eligibility criteria should strive for inclusiveness to enroll participants who are representative of the intended users of the intervention under study in a timely manner.
In this study, published in the Journal of Clinical Oncology, Bruinooge and Schenkel argued that the rationale for excluding patients should be clearly articulated and reflect expected toxicities associated with the therapy under investigation based on existing data.