This story is part of a series on Diversity in Clinical Trials, which will explore the impacts of drug development using data that does not reflect the diverse U.S. population.
To read more in this series, please visit the "Diversity in Clinical Trials" Issue in Focus page here.
The healthcare industry thrives on innovation, but has hardly budged in terms of equity in clinical trials, where first-in-class medicines are typically brought to market based on data from majority white, male patients.
Particularly as the healthcare industry zeroes in on precision medicine, therapies tailored to patients' genetic makeup, a better understanding of different populations' biological differences is vital in drug testing.
"Across the board we see either underrepresentation relative to the representation in the U.S. population, or underrepresentation relative to burden of disease," said Jennifer Alvidrez, program official at the U.S. National Institute on Minority Health and Health Disparities.
Women and minorities have demanded and achieved greater representation in government, pop culture and industry, but year after year, researchers have flagged a lack of diverse inclusion in drug trials. The result is medications that simply do not work the same way in women as they do in men, or prove more harmful to certain races than to the white participants in whom the drugs were tested.
The U.S. Food and Drug Administration approved a record 59 new therapies in 2018, but among the 53 drugs with U.S. clinical trials, only 12 had enough data to provide some knowledge of the drug's effects across races, the FDA's Drug Trials Snapshots reported. Among these limited records, the majority of participants were white.
Merck & Co. Inc.'s trials for HIV-1 drug Pifeltro evaluated 907 white patients, or 70% of the total study participation. African American or black patients made up 21%, Asian patients comprised 9% and "other" patients totaled 9%. According to its trial snapshot, Pifeltro was deemed to perform similarly across all races studied.
But in the real world, African American or black individuals in the U.S. are more often diagnosed with HIV than white patients, tallying about 16,690 diagnoses in 2017, compared to 10,048 diagnoses in non-Hispanic whites, according to the Centers for Disease Control and Prevention.
Certain medicines, like Sanofi and Bristol-Myers Squibb Co.'s blood thinner Plavix, depend on a particular gene mutation or enzyme to be effective, which may not be found uniformly across races.
Plavix, which is activated by a certain liver enzyme, was found to have varied effects across races after its approval in 1997, spurring the FDA to add a boxed warning to the drug's label in 2010. Clinical trials for the medicine had not picked up on racial differences. Approximately 5,000 individuals who had experienced little to no positive effects from the anticoagulant drug — or adverse effects like internal bleeding injuries — sued the drugmakers. While no significant settlements have been made, two state cases in Hawaii and New Mexico are still ongoing, a Bristol-Myers spokesperson confirmed.
Even when instances of disparate drug performance are identified, researchers do not always know the underlying reasons for differing drug effects — an issue that will only be exacerbated as medicines become more genetically specific.
Health data on white patients is abundant and has continued to accumulate because of their high participation in trials. But data on women for conditions such as cardiovascular disease has only just begun to ramp up. For African Americans, Hispanics or Latinos and Native Americans, the data is largely lacking or completely unreported, whether in basic genetic data, disease statistics or drug reactions.
The U.S. will cease to be majority white between 2040 and 2050, according to U.S. Census data. The face of the U.S. is changing, and with it, so is the addressable patient population.
Barriers to entry
The onus often falls on patients to seek innovative treatments. Still, the government's clinical trials database, Clinicaltrials.gov, is difficult to navigate, chock full of medical jargon, and lacks important information on language fluency requirements.
Limited English-speaking ability, especially prevalent in the growing population of older people, can bar otherwise eligible patients from clinical trials. Alvidrez said no policies address clinical trial inclusion of non-fluent English speakers.
Meanwhile, physicians can carry their own implicit biases. They do not always pass clinical trial information on to minority patients due to assumptions of their economic ability and attitudes toward research studies, or other reasons, physician-patient communication studies have showed.
Among African Americans, distrust of the healthcare system stems from the Tuskegee syphilis experiments, which took place from 1932 to 1972. For African American men older than 65, the memory is still fresh, a 2018 study by the Center for Healthy African American Men through Partnerships reported.
But men interviewed by the researchers emphasized a desire for health education and prevention, underscoring the lack of inclusion on the side of healthcare providers.
The FDA, which regulates clinical trials, has provided suggestions on its website and guidance for industry for more diverse participation in studies but has not mandated policies on diversity and inclusion. According to an FDA spokesperson, the agency lacks authority to enforce the inclusion of minorities in clinical trials, even when a disease area disproportionately affects certain racial and ethnic groups.
Recruitment and enrollment of women in trials has improved; participation reached 56% in pivotal trials from 1998 to 2000 and has remained close to 50% since 2001, the FDA spokesperson said.
No such improvement has been recorded for racial minorities. An analysis of cancer trials conducted by Mayo Clinic fellow Narjust Duma found that, from 2003 to 2016, publicly and privately supported trials enrolled 83% white patients, 6% African American, 5.3% Asian, 2.6% Hispanic and 2.4% "other." The study revealed that African American and Hispanic enrollment declined compared to data from 1996 to 2002.
While the FDA began requiring trial participation demographics from sponsors in 1998, experts said the rule is neither enforced nor followed. Hispanic participants are sometimes not recorded at all in clinical trials demographic data, as shown by the FDA's snapshots data.
The U.S. National Institutes of Health issued the Inclusion of Women and Minorities as Subjects in Clinical Research guidelines in 1993. The policy requires including women and minorities in NIH-funded research "unless a clear and compelling rationale and justification" exists.
But the NIH's policy does not set forth quantifiable goals and is "under-enforced," Alvidrez said.
From talk to action
A handful of pharmaceutical companies are working on this issue. Efforts tend to be concentrated among the largest pharma companies but are not widely known or publicized, nor are they necessarily reflected in publicly reported clinical trial data.
Johnson & Johnson, the world's largest healthcare company, created an internal working group on diversity in clinical trials, which has since expanded into a strategy focusing on access, awareness and trust, according to Staci Hargraves, vice president and global head of research and development operations at J&J unit Janssen Pharmaceuticals Inc. Because the strategy is in early development, no specific trial recruitment goals have been set, Hargraves said.
Eli Lilly and Co. has its own diversity criteria, mandating that every trial with more than 25 sites must have at least two sites that enroll more than 25% nonwhite patients.
Merck worked with the National Medical Association to increase clinical research training among African Americans, according to a 2011 workshop on clinical trial engagement. Merck was unable to provide details on the program as of October.
Certain health startups are looking to address the issue in their own way. Driver, Inc., an app for finding cancer trials, is attempting to make trial information easier to find for patients based on indication and location.
DxTerity Diagnostics Inc. is also looking to relieve the burden of underrepresented populations through its home monitoring and self-collected blood samples that provide for early biomarker detection.
Consumer genomics company 23andMe Inc., which holds a massive trove of genetic data from consumers who have volunteered for research, has joined forces with TrialSpark, a healthcare technology company that leverages data and algorithms to expand study sites into community clinics and recruit more participants. The two will apply their combined resources in select company-sponsored clinical trials.
"It's not new territory, but we're moving past the talk into action," Hargraves said.