Close this search box.
Close this search box.

The importance of diversity in clinical trials

In this article, Dr Harriet Gray-Stephens, Pharmaceutical Physician in the Clinical and Medical Affairs team at Boyds, discusses the importance of diversity in clinical trials and the steps that are being taken to improve the quality of data derived from clinical research.

Diversity in clinical trials

Different people experience the same disease differently: Clinical trials need to involve a variety of people so that the data collected is representative of the population intended to benefit from a treatment. Historically, clinical trials were conducted on almost exclusively white, male study participants. This has created bias within clinical trial data including gaps in our understanding of disease and the drug at both a disease population and individual level.

Product development programmes for both drugs and devices should consider the medical and demographic factors that impact the generalisability of study results to the patient population that will use the product. However, this remains a problem for drug developers and patients alike. The US Food and Drug Administration (FDA) states that we must “ensure drugs work and are safe for all those who may use them,” and to achieve that, we need diverse trials. Indeed, on 13 April 2022, the FDA issued a draft guidance, “Diversity Plans to Improve Enrolment of Participants from Underrepresented Racial and Ethnic Populations in Clinical Trials.”

Typically, diversity is thought to include racial and ethnic inclusion, as these are an important factor in determining a patient’s likely genotype. Clinical researchers are used to selecting “Hispanic ethnicity” as a check box for most clinical trials, but enquiries rarely go further than this.

Increasingly, diversity is being seen as more than just genetics. Clinical trial sites also need to think about other types of diversity including age and LGBTQ+.

Genetics and ethnicity

It is thought that up to 20% of drugs behave differently based on racial or ethnic differences. One example of this is alcohol. Several variants exist of the gene that produces aldehyde dehydrogenase, an enzyme found in the liver that metabolises alcohol. One of these gene variants, which generates a nonfunctional enzyme, is often present in Asian people but rarely in Caucasian and African American people. Individuals with two defective copies of this gene are less able to metabolise alcohol and experience more intense reactions to alcohol. Key differences in drug metabolism exist in a lot of enzymes and may help us to understand how we, as individuals, will react to a drug.

CYP P450 enzymes which are responsible for the metabolism of many drugs, show different frequencies of polymorphisms depending on ethnicity. Different polymorphisms metabolise drugs at different rates: some increase the rate of drug elimination (ultrametabolisers), leading to faster metabolic clearance and potentially reduced effectiveness and a need for high doses. Other polymorphisms decrease drug metabolism, which may increase the potential for drug reactions, higher concentrations and more adverse events. One of the most common cytochrome P450 polymorphisms occurs in the CYP2D6 isozyme: CYP2D6 poor metaboliser is less common in Asian populations than in Western ones, and is found in around 1% of Thai, Chinese and Japanese populations compared to 5-10% in Caucasians. Differences in the ethnic composition of clinical trial participants will result in different proportions of slow metabolisers, meaning subjects have a higher risk of having more severe adverse reactions related to higher drug concentrations.

COVID-19 has disproportionately affected rational and ethnic minorities and research is being undertaken to understand these differences. However, despite intense efforts, including by the NIHR to recruit ethnic minorities to better understand vaccine effectiveness in these populations, recruitment has been slower in these populations. This demonstrates that there is still a need to promote clinical research involvement in certain populations, to provide the best data.

Issues around ethnicity

Many ethnic minorities are often under-represented in clinical trials: African American and Black patients are estimated to make up 13.4% of the population but only 5% of trial participants. Why is there such a difference in clinical trial participation in different ethnic groups? Historically, there is reduced trust in some ethnic groups due to unethical research conducted before the implementation of good clinical practice (GCP), such as the involvement of black-African men in Syphilis studies without adequate treatment for their disease.

The location of clinical trial conduct also needs to be conducted. For example, in the UK, early-phase healthy volunteer studies are often conducted in areas of relative social deprivation where the population diversity is not representative of the overall population. Individuals of different ethnic minorities may be less able to commit to the significant time burden of contributing to a clinical trial, owing to work and personal commitments. Thus, the reasons for the clinical trial population being different from, and less diverse than the real patient population is complex and difficult to address.

The FDA has recognised that output from clinical trials is less relevant or applicable to under-represented populations because less information exists relating to individuals of that ethnic origin. This problem is further compounded as individuals from these populations may have a disproportional disease burden for certain illnesses, such as diabetes and cardiovascular disease within the African American and Black populations. The FDA and other regulatory authorities acknowledge this is an issue and have declined marketing authorisation applications (new drug applications, NDA) based on insufficient information. It is becoming increasingly recognised that a lack of information on ethnic minorities is a frequent reason for refusal, and thus lack of clinical trial ethnic diversity is becoming a problem for drug developers.

Solutions to increase ethnic diversity

The FDA has issued several sets of recommendations to improve clinical trial diversity including the collection and analysis of racial and ethnic data. As of April 2023, the FDA requires that a Race and Ethnicity Plan is submitted as part of any Investigational New Drug (IND), New Drug Application (NDA), or device marketing submission which details incentives “to enrol representative numbers of participants from historically underrepresented racial and ethnic populations.” It is important to consider ethnic diversity early in clinical research, to enable review by the end of phase 2 (FDA). The plan should consider the target population, to look at distribution by ethnic group and pharmacokinetic factors, to understand routes of metabolism and the potential for these to be affected by genetic polymorphisms.

Small changes in the proposed clinical development plan can assist with the recruitment and retention of patients from a variety of ethnic minorities. A review of the location of clinical trial sites, and the inclusion of more sites near areas of higher representation of ethnic minorities can facilitate the recruitment of patients who live nearby and would not otherwise be able to afford to travel to access studies due to financial or time constraints. Simplification of the eligibility criteria to enable more patients of extremes of age, or concomitant diseases may help recruitment. Patient engagement is also key to understanding how to recruit particular ethnic groups to a trial. Reducing the burden of clinical trials through decreasing the number or duration of visits or assessments, and increasing the flexibility of visit windows can be helpful. The use of technology is going to be critical to this strategy: for example, enabling remote visits through decentralised clinical trials, electronic consent (eConsent), electronic diaries (eDiaries), and the use of local service providers, such as blood sampling at their family physician, can also help to reduce the burden of participating in clinical trials.

Age in clinical trials

Older adults (patients over 65 years of age) are a growing portion of the population and represent an area of specific health needs. Despite this, older adults are often under-represented in clinical trials.

Researchers are used to seeing age restrictions as part of eligibility criteria for clinical trials. In the early phase, healthy volunteer research, adults over 50 and sometimes even over 30 years old are actively excluded. Later phase studies have more relaxed eligibility criteria for patient age, with 30% of trials having an upper age limit for eligibility. Many studies indirectly inhibit the recruitment of older adults as they have an increased burden of co-morbidities which often makes these patients not eligible based on other eligibility criteria.

These criteria have been set by the investigators to reduce the variability of the population and reduce the risk of drug-drug interactions with their existing concomitant medications, however, result in a non-diverse patient population which does not represent the target population. Additionally, a lack of digital literacy may present a barrier to elderly patients who are less used to electronic forms of data collection. For these reasons, the mean age of patients in clinical trials is often lower than that of the population with the disease of interest. This challenge is only increasing as the average population age increases.

Issues around age

Age is an important factor to consider with regard to drug action and particularly pharmacokinetics. As individuals get older, the rate of enzyme metabolism slows down, resulting in higher concentrations and thus increased risk of adverse events. Additionally, body fat increases with increasing age, and total body water, as well as lean body mass, decreases, which reduces the distribution of hydrophilic drugs and increases the half-life of lipophilic drugs – factors which both influence the pharmacokinetics of drugs. Moreover, as older adults tend to have more co-morbidities, concomitant medications may be metabolised through the same pathways as the investigational product, or interact with enzymes thus changing the rate of metabolism of the investigational product. It is important to investigate the effect of these variables on drug activity, to best understand it in the population of interest.

Solutions to increase diversity in patient age

Regulators recognise the need for age diversity within clinical trials. This has led to an increase in the inclusion of geriatric populations (subjects over 65) in what is often called healthy older adult volunteers – subjects with limited comorbidities. These studies enable a comparison of drug pharmacokinetics and pharmacodynamics with younger patients. Some may argue that this approach is not sufficient as healthy older adult studies do not investigate the effect of comorbidities on drug responses, and comorbidities are both more common in the older population and can have a significant impact on drug metabolism including through drug interactions.

The FDA issued guidelines in 1997 requiring a minimum of 100 geriatric patients, defined as people over 65, for drug trials in diseases present in the older population, allowing researchers to detect significant differences between age groups. However, this guidance is outdated and does not represent enough older-adult representation, particularly in diseases where there is a high burden of disease in the older adult population such as oncology. The International Conference of Harmonisation (ICH) guidelines on geriatrics (E7) should be reviewed, and region-specific guidelines for increasing age-diversity in clinical trials are desperately needed.

Drug developers should carefully design clinical trial protocols to include older individuals, including less prescriptive exclusion criteria for concurrent illnesses, and a removal of an upper age limit. Regulators should look for a more product-specific benefit-risk evaluation and flexible use of additional technologies such as remote/home visits as part of decentralised clinical trials to give patients maximum flexibility to choose visit types which are most suited to their requirements. Recruitment strategies could be specified towards recruiting age-diverse cohorts.

Other diversity factors

Historically, data on sexual orientation or gender has not been collected as part of clinical trials which has made it difficult to quantify the inclusion of sexual and gender minority persons (transgender and gender non-binary individuals). This has made it difficult to know if individuals in the LGBTQ+ or non-binary gender populations are well represented in clinical research studies. Some participants may feel uncomfortable reporting their preferences, further contributing to incomplete or biased data, or avoiding participation in clinical trials entirely. Transgender patients are routinely excluded from clinical trials, with research suggesting that only 9.1% of clinical trials in the US initiated in 2022 specify transgender patients. In some cases, this may be due to the perceived risk of gender-reassignment drugs on the investigational medicinal product’s pharmacokinetic parameters.

Drug developers need to carefully evaluate the data collected from trials to ensure that the information reflects all populations likely to use the product if approved. A more inclusive and supportive clinical trial ecosystem needs to be developed to encourage clinical trial participant diversity.

Looking to the future

Increasingly, clinical research or data collection is being undertaken post-authorisation as part of phase IV studies looking at real-world safety and efficacy data. Phase IV trials are conducted to determine long-term safety and effectiveness and to identify adverse effects that may not have been apparent in prior trials.

Research and patients alike are working to prioritise diversity within clinical trials. By expanding the range of people involved, the data better reflects the realities of society and will lead to more valid and generalisable trial data, and thus drug development that applies to patients in need of treatment. Whilst diversity is primarily focused on racial and ethnic diversity, the scope needs to be increased to include other types of diversity including age and sexuality.

The challenge of age homogeneity in clinical research will only increase as the global population ages, and if not tested on all age groups, there is a risk that treatments negatively interact with concomitant medications or negatively impact patients with underlying conditions less common in younger generations. Sponsors need to consider the impact of DCTs and technologies which may increase ethnic minority inclusion in clinical trials on the recruitment of older adults and take a stratified approach to their target audience to optimise the recruitment of multiple different minority groups.

Clinical trial diversity has been acknowledged as important by regulators and is directly impacting drug developers. Early in clinical development, a highly homogeneous and non-diverse population is often sought to minimise inter-patient variability and maximise the power of a study to detect an effect. However, as development goes on, increasing diversity is required to develop a broader understanding of the product. However, this comes at the risk of exposing patients to new confounding drugs or conditions as well as reducing the power of the study to detect and effect. A careful benefit-risk balance must be struck during development, working towards a common aim of generating the best data which is accurate and representative of the target population. Sponsors must remember this during drug development and take steps wherever possible to maximise subject diversity.