A year before the COVID-19 pandemic began, a team of clinical statisticians at the University of Texas MD Anderson Cancer Center sat together in small office for a year, painstakingly hand coding data from the US clinical trials database, www.clinicaltrials.gov.
They were trying to answer a simple question: Why are cancer-drug trials enrolling too few patients over the age of 65?
More than 300 trials and 262,354 patients later, the research team confirmed that participants in clinical trials were, on average, 6.5 years younger than the population for whom the drug was intended.
“We found marked disparities across different disease sites…. The patients that are enrolling on studies are markedly younger than the average patient seen in the population with those same conditions,” said team leader Ethan Ludmir, MD, assistant professor, Division of Radiation Oncology at the University of Texas.
And this age disparity was significantly greater in industry-funded trials.
Researchers have known for 20 years that cancer trial participants are not representative of the wider cancer population, and numerous government guidance documents have been issued on the matter. However, this Texas team’s findings were the first unambiguous evidence that pharmaceutical companies seem to be selecting younger patients to test their drugs.
“If we’re being generous then perhaps the answer is: They’re looking for some element of homogeneity, which is to say they don’t want competing risks to make the signal-to-noise ratio uninterpretable,” said Ludmir.
Laura Bothwell, PhD, assistant professor, Yale School of Public Health, recently coauthored a 259-page consensus report for the National Academies of Sciences, Engineering and Medicine on how to increase the research involvement of under-represented groups.
Bothwell said, “The problem with industry funded research is that…it’s an inevitable conflict of interest that exists. They want the research to show that their products work. And older populations…have a lot more complications, which leads to potentially less favorable results.”
The MD Anderson findings were published in JAMA Oncology. “That was the starting point in our journey,” said Ludmir. For the next 3 years, the researchers mined their painstakingly constructed database to understand what was preventing greater numbers of older patients from enrollment in cancer trials.
Meanwhile, answers were coming from elsewhere. In parallel with the work at MD Anderson, a team in California led by Mina Sedrak, MD, a medical oncologist at the City of Hope National Medical Center, had also started investigating age disparities in clinical trials.
Sedrak, who also serves as deputy director of Clinical Trials at the Center for Cancer and Aging, said he had become increasingly concerned that he did not have adequate information on new cancer therapies for his older patients.
“I was caring for a large number of people who were…older adults,” said Sedrak, “But the data that was being used to get the standard-of-care treatment for cancer did not include older adults. And so there was this lack of applicability.”
He summed up the challenges in a 2021 review paper: “Most of what we know about cancer therapeutics is based on clinical trials conducted in younger, healthier patients.”
By 2030, it is estimated that 70% of all new cancer diagnoses will be in patients 65 years old and older. By contrast, patients over age 65 still account for only 40% of patients in cancer trials registered with the FDA (2015 figures) and older adults make up only 44% of participants in practice-changing cancer trials, according to a 2022 study.
So what is going on? Are studies specifically designed to squeeze out older patients?
Surprisingly, patients are not being kept out of trials by formal age limits, according to Ludmir. His team found that only 10% of phase 3 trials over the past 30 years had an upper limit for age, and age restrictions have been dropping by 1% a year. (For example, 16% of trials that enrolled in 2002-2005 had an upper age limit, compared with just 8% of trials that started in 2010-2014.)
Sedrak’s team found that “clinician bias” may be a factor, a situation in which trial investigators — particularly academic oncologists — are subconsciously picking younger, healthier patients for trials and excluding older, sicker patients to protect them from drug toxicities.
Ludmir said this was understandable, especially in the case of industry-driven trials, which tend to have demanding endpoints and “an overall posture of more treatment aggressiveness.”
“These are typically not trials where they’re saying, ‘Hey, if we add acupuncture…are we going to see improved patient reported outcomes?'” Ludmir explained. “You’re asking…, I’ve got this cocktail of two pretty rough chemos: I want to see what happens if I add an immunotherapy to that. If I’m the clinician in clinic, I might reasonably, subconsciously, say, is the 75-year-old really who I want on this?”
What about patient bias? Perhaps fewer older patients wish to join clinical trials?
Not so, at least not at community cancer centers, said Sedrak. His team’s analysis of the National Cancer Institute Community Oncology Research Program database for 2016-2019 revealed that older patients were just as keen as the younger patients to participate in trials (68% of patients aged 50-69 years and 65% of patients 70+; P = .28).
However, drug companies may be excluding older patients by more subtle means. One-fifth of patients over 65 have had a prior cancer. Ludmir and coauthor Roshal Patel, MD, used their hand-coded www.clinicaltrials.gov database to look at prior malignancy exclusion criteria (PMEC). The analysis found “pervasive utilization” of PMEC in phase 3 trials, cropping up in 41% of studies over the past 30 years.
PMEC was significantly associated with age disparities and was significantly more common in industry-funded trials.
When asked whether PMEC are “age restriction by stealth” on the part of drug companies, Ludmir was reluctant to assign blame, but stood by his data: “The wider you restrict people in terms of having a prior cancer, the wider the age disparities in the subsequent studies, which to me is about as strong, in terms of causal understanding of these phenomena, as you can reasonably get at this level.”
In March the FDA released a guidance document titled Inclusion of Older Adults in Cancer Clinical Trials. However, its recommendations are “nonbinding” and “do not have the force and effect of law.”
To fix the issues, said Sedrak, the FDA must be given teeth.
“Okay, you write guidelines,” he said. “But if you don’t actually hold people accountable to following the guidelines, how are we going to implement and make sure that we’re transforming policy into action?”
Bothwell of Yale’s School of Public Health agreed. “Accountability has been the weakest link for decades now.”
She concluded, “In medicine there’s a tendency to believe that a therapy, because it exists and it has been tested and it’s shown some efficacy, it’s useful. But we don’t know the answer to that question unless we have statistically valid research in the population that we’re using it in.”
Bothwell and Ludmir report no conflicts of interest. In his publications, Sedrak reports industry grants from Seattle Genetics, Eli Lilly, Novartis, and Pfizer Foundation.
This article originally appeared on MDedge.com, part of the Medscape Professional Network.
