Image courtesy of Pixabay
Lack of diversity in clinical trials has long been an issue, driven by challenges with recruitment and participation. In recent years, pharmaceutical companies have prioritized recruiting more diverse patient groups for their trials. And in some areas, it is working. In the past ten years, the representation of Black and African American patients in U.S.-based clinical trials has improved. Currently, data from Phesi show that 14.9% of clinical trial participants self-identify as being in this group. That proportion is slightly greater than the 13.4% of the U.S. population that identifies as Black and African American, according to the 2019 U.S. Census.
To develop truly effective therapies for all, clinical trials must be carried out in populations representative of those who will receive the new treatments. However, certain patient subpopulations are significantly and consistently underrepresented in clinical trials. A recent analysis by Phesi of 1.3 million clinical trial participants in the U.S. found that Asian, Hispanic and Latino, Native Americans and Alaska natives, Native Hawaiians and other Pacific Islanders were significantly and consistently underrepresented in 1,580 Phase 2 and 3 clinical trials. Patients within these groups need effective and suitable medical interventions. So why are they being overlooked when it comes to clinical trial recruitment?
Now, thanks to new ways of using data to support clinical trial design, we can turn the tide on underrepresentation. Insights from industry-collected data – both historical and live – can be applied to drive change and support both pharmaceutical companies and patient groups.
Diversity and clinical trials
The safety and efficacy of treatment vary depending on several factors, including sex, race, ethnicity and age. Suppose a treatment is not adequately assessed across all of these factors during clinical trials. In that case, the developers may be held accountable for any adverse reactions identified after the drug has gone to market. Not only does this have cost and safety implications, but the best treatment for certain subgroups may not be identified, compromising the quality of care for patients.
One such example is the anti-platelet drug Clopidogrel. Used to reduce a person’s risk of heart disease, heart attacks and stroke, Clopidogrel is broken down into its active form by a group of enzymes known as CYP enzymes. Although widely prescribed, different ethnic groups respond differently to Clopidogrel due to their genetics. For example, almost a third of South Asians have variation in the gene responsible for producing CYP enzymes, causing them to metabolize Clopidogrel more rapidly – which can increase the risk of bleeding. To mediate this, it is recommended that lower doses of Clopidogrel are given to this group. Conversely, almost a quarter of Japanese people are ‘poor metabolizers’ of Clopidogrel, and this treatment is rendered entirely useless for those with the genetic variation that causes this effect.
Identifying variations during the clinical trial stage of drug development allows the best treatment strategies to be designed for all groups with a particular condition, saving money and improving patient experiences.
Reporting and the challenges
According to demographic data collected by the FDA, in the clinical trials that led to marketable drugs in 2020, 8% of trial participants were Black or African American. However, the representation of Black and African American patients within different areas of clinical research varies widely. For example, in 2016, less than 3% of trial participants for cancer and cardiovascular therapies were Black or African American, despite these diseases often being more severe in this demographic. Trial groups for studies focusing on psychiatric diseases, on the other hand, were made up of 24.2% Black and African American participants.
As well as inconsistency in representation in different areas of clinical research, reporting of trial participant demographics are not standardized. Government bodies, regulators and pharmaceutical companies do not use the same metrics when reporting patient and participant demographics. This makes accurate representation in study groups a further challenge. Aligning with the categories used by government bodies and census data will allow clinical research teams to design trial groups that are more closely representative of the existing population. As this approach becomes standardized across pharmaceutical research organizations, clinically-meaningful demographics can be identified, and better trials designed.
Synthetic patient profiles
The more diverse trial groups are, the more data are available on the traits associated with certain ethnic groups. These data can then be used to develop synthetic patient models. With the extensive clinical data sets available, synthetic patient arms can be used in place of real-world patients – acting as a control group for those demographics where enough data are available.
Not only do synthetic patient models eliminate the need for control patients in clinical trials – overcoming a range of ethical issues surrounding placebo use – but they also lead to a more efficient drug development process. Having synthetic patient models for everyone, not just certain demographics, is key to plugging the diversity gap in clinical trials. But this can only be achieved with sufficient, detailed and consistent information about different ethnic groups. Again, consistency in reporting patient demographics is key in all groups benefiting from the advances in control group automation.
Gen Li, Ph.D.
By properly collecting, analyzing and applying data on patient demographics, truly representative clinical trials can be designed. This will have benefits across the scope of clinical research, with trials becoming more efficient, cost-effective and equitable. In addition, diversity in trial groups is key to restoring trust in the clinical trial process and, most importantly, improving health outcomes for everyone.
Gen Li, Ph.D., is the President of Phesi.