Most of the time, evidence of a cancer therapy’s success in clinical trials is reassuring. This time, it was not.
Most of the time, evidence of a cancer therapy’s success in clinical trials is reassuring. This time, it was not.
Narjust Duma, MD, was treating a patient at the Mayo Clinic, in Rochester, Minnesota, where she was a hematology–medical oncology fellow. The patient asked the usual questions, and she responded with the usual answers: data describing the therapy’s overall odds of success and the odds of the cancer returning following therapy.
“What about me?” the patient asked.
Unfazed, Duma explained that the data she was quoting were based on cancer stage and thus represented a reasonable expectation of a patient’s odds with that stage of cancer.
The patient, however, was not satisfied. He kept asking, over and over again, “What about me?”
“And when I finally stopped and I asked, ‘What do you mean?’” she said, “he [answered], “What about me? What about Black men like me?’
She did not know how to answer his question, Duma realized. “And I remember, my stomach was hollow for, like, 2 days,” she said.
The patient knew that as a Black man, his life expectancy was already shorter than that of his White peers. He wanted to know what the trial data showed about the success of the therapy in Black men, not just White men.
Duma tried to find out, but when she examined the study data, she found that of the thousands of patients recruited, just 5 were Black men—not nearlyenough to draw reasonable conclusions about the safety and efficacy of the therapy in Black men such as her patient.
She relayed the rest of the story to Targeted Therapies in Oncology. “I went back,” Duma said, “and I told my patient, ‘I don’t know.’ ”
The clinical trial process has birthed a revolution in cancer care, allowing a host of cutting-edge therapies to make it from the laboratory to the clinic. However, the success of the modern drug development pipeline obscures one important fact: The trials that are meant to prove the safety and efficacy of potential therapies consistently overrepresent White patients and often underrepresent racial and ethnic minorities, women, and individuals of lower socioeconomic status.
The problem connects to issues of health care access and equity, but it also underscores an uncomfortable truth: In many cases, the trial data available for minority populations are much less robust than the data available for members of majorities.
The problem is evident from the data. The FDA’s 2018 Drug Trials Snapshots, for instance, showed that of the more than 5000 patients that year who participated in oncology trials that led to approvals, just 4% were Black and 4% were Hispanic, despite the fact that 13.4% of the US population was Black and 18.1% was Hispanic.1 Women were also under-represented: They made up 38% of the trial population1 but 51% of the US population.2 At the same time, Asian patients were overrepresented, making up 15% of the trial population but just 5.8% of the US population.1
The problem is not new. In 1993, Congress passed the National Institutes of Health Revitalization Act, which directed the agency to develop guidelines to boost inclusion of women and minorities in clinical trials. Those guidelines were published the following year. However, in a landmark 2004 study, future US Surgeon General Vivek H. Murthy, MD, MBA, and colleagues reported that patients who were female, older, or Black or Hispanic were still considerably less likely than male, younger, and White patients to enroll in cooperative group cancer trials. They also found that Black participation in such trials had been decreasing, not increasing, in recent years.3
The problem exists not just in a macro sense but also within specific cancer types. For instance, Black patients have 2 to 3 times the risk of being diagnosed with multiple myeloma (MM) compared with White patients, and Black patients also tend to be diagnosed with MM at a younger age than White patients. However, a survey of MM pivotal trials supporting FDA applications between 2003 and 2017 found that just 4.5% of patients in clinical trials were Black.4,5
In a 2020 examination of racial disparities in MM risk and outcomes, corresponding author Timothy R. Rebbeck, PhD, of the Dana-Farber/Harvard Cancer Center in Boston, and colleagues said the lack of Black enrollment led to significant limitations in the conclusions that could be drawn from MM studies.5
“The small numbers of Black [patients] enrolled in clinical trials not only limit the confidence in the reported results, but they raise questions as to whether the trends observed in the Black patients who did enroll are generalizable to broader patient populations,” Rebbeck and colleagues wrote.
or Duma, now an assistant professor at the University of Wisconsin School of Medicine and Public Health in Madison, the interaction with the patient was a call to action. She had seen the results of Murthy’s 2004 study, but she also knew that the study was based on National Cancer Institute (NCI)–funded trials, and since then, industry-funded trials had become more common.
“So I was wondering, are we doing better now that the whole drug development platform has changed?” she said.
Duma and colleagues decided to search the ClinicalTrials.gov database for US-based oncology clinical trials completed between 2003 and 2016. Of the 1012 trials involving the 7 most common cancers, 31% reported ethnicity, for a total of 55,689 patients. They found that Black and Hispanic patients were still significantly less likely than White patients to be enrolled in clinical trials.
When they compared their results with the historical data from 1996 to 2002 reported by Murthy,3 they found that enrollment had actually dropped for both Black (from 9.2% to 6.0%) and Hispanic patients (from 3.1% to 2.6%). Duma and colleagues also found that women were disproportionately under-enrolled in studies focusing on melanoma, lung cancer, and pancreatic cancer.6
“Your race and your background will affect you if you are included in a trial and ultimately will affect the long-term outcome [of your disease],” Duma said.
That is one reason real-world evidence (RWE)—data collected outside of clinical trials—has grown in importance, according to a 2019 paper coauthored by Atul Batra, MBBS, MD, DM, and Winson Cheung, MD, MPH, both of the University of Calgary, in Alberta, Canada. Looking specifically at colorectal cancer, they wrote that RWE is an important complement to data from randomized controlled trials (RCTs) because it gives insights into the actual patient population, instead of just the potentially unrepresentative population included in the trials.7
“Although RCTs remain the ‘gold standard’ for evaluating the efficacy of novel interventions in cancer, there is general acceptance that RCTs alone might not offer all the answers to relevant clinical or research questions,” Batra and Cheung wrote.
For community oncologists, the primary clinical questions revolve around safety, efficacy, and dosage. Duma said it is impossible to confidently answer such questions when treating patients whose demographic groups were not adequately represented in the associated trials.
“Safety is always the No. 1 rule in anything with medicine,” Duma said. “You need to be safe for your patients, but it’s unsafe to give a drug after it’s approved, but was [not tested] in the group [to which your patient belongs.]”
According to Batra and Cheung’s research, one issue that leads to disparities between trial populations and the general population is the use of stringent eligibility criteria in clinical trials. “Real-world” patients may have comorbidities and other traits that can lead to their exclusion from cancer trials that are designed for maximum clarity of data.7
Lori J. Pierce, MD, professor and vice provost for Academic and Faculty Affairs at University Hospital, University of Michigan, Ann Arbor, and president of the American Society of Clinical Oncology (ASCO), has made increasing diversity in trials a focal point of her tenure at the organization. She and ASCO have worked with the NCI and FDA on efforts to expand trial eligibility criteria, hoping that broadening the potential pool of enrollees would help diversify trials without sacrificing the integrity of the results.
“They found out that by changing just 3 common eligibility criteria—renal function measures, the presence of brain metastases, and having a history of a prior cancer—the eligibility for clinical trials increased by almost 2-fold,” Pierce said during an interview with Targeted Therapies in Oncology.
Those results, noted Pierce, show that using more relaxed exclusion criteria can make a major difference in trial populations.
Another limiting factor that has seen progress is insurance coverage. Private insurers in the United States are required to cover costs associated with clinical trials, such as costs for blood tests and x-rays. However, the requirement does not extend to Medicaid. Although 15 states have long directed their state Medicaid programs to cover the costs, there was no federal coverage mandate.8 That was a problem, Pierce noted, because underrepresented minorities tend to rely on Medicaid at disproportionately high rates.
ASCO has been pushing for Congress to extend the coverage mandate to Medicaid for a number of years. The effort finally paid off in December 2020, when Congress included the Clinical Treatment Act in its year-end coronavirus disease 2019 relief bill.9
“We were popping the cork—I personally was popping the champagne cork—when this [legislation] was passed, because I really think that it will make a big difference,” Pierce said.
Duma agreed that the lack of Medicaid coverage has been a significant part of the problem, but she noted that individuals on the economic margins will still face significant cost burdens. For instance, an analysis conducted in 2019 found parking fees at NCI-designated cancer treatment centers can be very expensive. Such treatment centers tend to be located in the downtown areas of major cities, where parking can cost as much as $40 per day, although some cancer centers provide free parking.10 Duma said that trial sponsors should help cover parking and other transportation costs to make participation more feasible for all patients.
Although some hurdles can be overcome with straightforward policy adjustments, others require broader changes. One problem, Pierce noted, is that many clinicians do not have the time or the infrastructure in place to participate in trials, or to get their patients involved in trials. She suggested that better training of community oncologists to help them understand how to become involved in trials would help. So, too, would cultural awareness training and translation services to facilitate better communication with potential enrollees who do not speak English. However, she declared that success will require involvement by more than just physicians.
“Not everything needs to rest on the shoulders of the physician,” Pierce said. “It takes a village, right?”
In this case, the “village” means patient navigators, physician’s assistants, nurse practitioners, and others who can spend time getting to know patients, figuring out which patients would be a good fit for trials, and helping to walk those patients through fears and past barriers.
At the same time, such interventions will work only if patients trust the health care system.
“Sometimes that lack of trust in the health care system is seen as though it doesn’t have a basis [in fact],” pointed out Duma. “It does have a basis.”
Duma noted, for instance, a 2020 whistleblower complaint in which a nurse working at an immigration detention center alleged that mostly Spanish-speaking women detained by US Immigration and Customs Enforcement were undergoing hysterectomies at alarming rates, raising questions about whether medical professionals there had consistently obtained informed consent.11
One way to build up trust is by explaining and offering options to patients. This is good practice in general; it also helps those with distrust of the system to feel that they are in control of their own health care decisions.
Pierce said there is ample evidence that one way to boost diversity in clinical trials is simply to do a better job of informing patients about the option to participate. She pointed to a study by Joseph M. Unger, PhD, of the Fred Hutchinson Cancer Research Center in Seattle, and colleagues. They analyzed 35 cancer studies published between 2000 and 2020, comprising almost 10,000 patients.12They found that 55% of patients who were asked to participate in a trial agreed to do so, a figure that held firm for both treatment and cancer control trials. Notably, Black patients were just as likely as White patients (58% vs 55%) to say “yes” when asked (TABLE 11).
Pierce said the Unger study should serve to contradict concerns that minority patients are reluctant to enroll in trials; the data show that they are not unwilling, when given ample opportunity and information. Although it is important to acknowledge the roots of mistrust, physicians can succeed in enrolling a diverse patient population if they take the time to explain studies and the safeguards.
“From a scientific point of view, to be able to prove the safety and prove the efficacy, and from a trust point of view...to be able to say, yes, members of various ethnic and racial groups were represented on this trial, I think are all very important,” Pierce said.
Last year, ASCO and the Association of Community Cancer Centers (ACCC) announced a partnership to develop solutions to the problem of underrepresentation of minorities in clinical trials.13 The partnership’s steering committee, which is cochaired by Pierce, is currently at work on a 3-pronged process to try to address the issue (FIGURE13).
Their plan will join reports by the Enhancing Minority Participation in Clinical Trials (EMPaCT) consortium, which has published a number of papers and training tools for health care workers. Also finding success are other local efforts, such as a program by Andrea Silber, MD, to recruit more women in the New Haven, Connecticut, area into trials.
Pierce said the ASCO-ACCC initiative expects to unveil its plan, currently under development, in early 2022, adding that she hopes such efforts can make meaningful strides in addressing what has proven to be a durable problem. However, Pierce said that in the end, success may come not from overhauling systems, but from taking the current strategies that work and making them the norm.
“I [am not convinced] that we’ve got to come up with some brand-new idea that never has been thought of before,” she said. “Practical solutions that are consistently applied may be the solution here. So, I’m very optimistic.”