In a conversation hosted by the Friends of Cancer Research and the American Society of Clinical Oncology, cancer researchers spoke with regulators and industry to lay out ways to modernize cancer clinical trial eligibility criteria. Further, advances in use of real-world evidence and remote data collection may eventually reduce reliance on the classical clinical trial, according to some attendees.
During the 9 April webinar, attended by US Food and Drug Administration acting chief Janet Woodcock, MD, among others, Ned Sharpless, MD, director of the National Cancer Institute (NCI), said that the pandemic spurred necessary innovation in cancer clinical trials.
“In consultation with FDA, we tried a lot of these things like consent by phone and shipping oral [investigational new drug] medicines to the home, and they were wildly successful,” said Sharpless. “I think we’ve learned a lot of things that you do – perhaps we should have been doing all along – that will persist after the pandemic.” (RELATED: Woodcock: Post-COVID, new flexibility could boost cancer trial diversity, Regulatory Focus 09 April 2021)
Sharpless then turned to another topic of interest to FDA, real world evidence: “There are things that you can learn from different datasets, if you’re clever about how you collect those data, that can inform clinical practice,” he said. During the pandemic, NCI engaged with several contractors to conduct data aggregation on COVID-19 serology “using these novel, tokenized, privacy-preserving relational linkage software approaches – and that’s the future, man!” said Sharpless.
“I’m telling you that you can answer questions in the real world by aggregating many large disparate datasets in ways that I never thought possible,” continued Sharpless, adding that there may be instances where the classic clinical trial involving individual consent may not be required. “I know the FDA’s been pushing that envelope for a long time and it hasn’t worked perhaps as well as we all envisioned, but it will someday,” reaping considerable economic and time savings.
Sharpless also addressed another major push in cancer research, lowering the eligibility bar. The American Society of Clinical Oncology (ASCO) and the Friends of Cancer Research have had a joint program to address eligibility criteria for participation in cancer clinical trials. Efforts included reaching out to FDA, NCI, investigators, sponsors and patient advocates. Through these and other efforts, said Sharpless, “I think we’ve changed the mindset of medical oncologists about brain metastasis and HIV status and borderline renal insufficiency…they date from a different era when clinical trials, I think, worked in a different way.”
Though academic medicine and the national research institutes are buying in to the importance of expanding trial participation to ensure that patients in cancer trial research resemble the patients who will receive the drugs clinically, “pharma was sort of the last to come on board,” said Sharpless. But now, he sees those attitudes changing as well, with the pharmaceutical industry coming to accept that restrictive inclusion criteria are “bad for everybody.”
Work presented at ASCO’s 2019 annual meeting was shared during the webinar, showing that broadening trial may greatly increase trial participation and make it more likely that participants look more like the patients who will eventually be receiving the therapies under investigation.
The goals of broadening trial eligibility, said R. Donald Harvey, MD, of Emory University’s Winship Cancer Institute, are threefold: to enable more patients to participate in trials; to make the population studied in trials more representative, thus making results more generalizable; and to accelerate trial accrual, “so we can get answers to the questions we’re asking more quickly.”
The retrospective study went back to look at deidentified electronic health record data maintained by ASCO over the period from 2011 to 2018. Investigators looked at real-world data for trials involving patients with non-small cell lung cancer (NSCLC), choosing this disease because comorbidities and more advanced disease are common in patients with NSCLC and many trials are available; also, the ASCO data for these patients were well curated. Patients were included if they received therapy after being diagnosed with NSCLC.
In the study, Harvey and his colleagues compared actual trial participation with what would have been seen with broadened criteria that included patients who had prior or concurrent cancers, patients with brain metastases, and those with renal impairment.
Applying the broadened criteria to the NSCLC cohort of patients nearly doubled the cohort size, Harvey said, adding another 4,851 patients to the 5,495 who were actually included in the trials. Looking at the data another way, 47.7% of the total NSCLC cohort were excluded by application of the traditional criteria. In contrast, just 1.5% of the cohort were excluded when the broadened criteria were applied; these patients all had severe renal compromise.
Further, application of the broadened criteria would have increased participation of older adults in the clinical trials, said Harvey, upping the percentage of those aged 75 years and up from about 16% to over 22% of participants. “The effect of broadening criteria needs to be looked at in the context and prism of age representation,” he said. “The majority of patients with cancer are advanced in age and have tended to be underrepresented in our clinical datasets currently.”
“These criteria really are likely to result in participants that are more reflective of the broader population,” added Harvey. “Narrower criteria to exclude patients really should only be used based on compelling scientific rationale.”
ASCO-Friends virtual meeting