Planning early for diversity helps contain costs and avoid delays

By Xoli Belgrave, Senior Director, Head of Clinical Trial Diversity and Inclusion

7 min

Planning early for diversity helps contain costs and avoid delays

New therapies must be tested in the patient populations they are intended to treat. It’s imperative scientifically, medically, and ethically. And we’re not there yet.

Just 20 percent of FDA approvals between 2014 and 2021 were based on clinical trial data that included treatment benefits and side effects for Black patients.¹ Even though Black and Latino Americans comprise 30 percent of multiple sclerosis (MS) patients in the United States, they comprise less than 5 percent of participants in MS clinical trials.² This is a problem because research suggests African and Hispanic American MS patients experience higher overall disability and symptom severity than Whites.³ To combat persistent gaps in clinical data, regulators are requiring diverse trials. But it’s also essential for societal health. When studies are not inclusive,  they deepen health disparities for under-represented communities—contributing to early deaths, poor health, and a lack of trust.⁴ For companies, if the findings are not generalizable, they can negatively impact reimbursement and market uptake and hinder innovation. We asked Parexel’s Senior Director and Head of Patient Inclusion, Xoli Belgrave, how creating a diversity plan early in development can make drug development more efficient.

Haven’t we already come a long way in achieving diversity in clinical trials?

Belgrave: We have indeed made progress, particularly in enrolling an equal number of men and women into trials versus primarily testing new drugs in white males, as in the past. However, data show persistent racial and ethnic inequities in clinical trial participation.⁵

Diversity, equity, and inclusion (DEI) now encompass many special populations and groups routinely excluded from early-stage research. People over 65, children, women who are pregnant, lactating, or of reproductive age, LGTBQIA+ communities, patients with low socioeconomic status, obese patients, and those with disabilities are among the populations who face barriers to participating in clinical research. Other factors we consider include rural versus urban geography, technology literacy, or patients’ facility with clickable or tappable devices.

Does a diversity plan require an upfront investment?

Belgrave: A successful diversity program requires time, effort, and investment. However, an intentional strategy, adopted early in development, saves money in the long run because it can keep timelines on track, preventing costly rework and delays in market access.

At Parexel, we are increasingly helping sponsors who have been asked by regulators to conduct post-marketing trials to cover racial and ethnic populations underrepresented in the clinical evidence package. In the case of partial approvals, we can fulfill the requirements through a combination of additional trial data and real-world evidence (RWE). Parexel data show that redoing an entire study can cost $10 to $20 million, and reworking recruitment strategies can cost up to $50,000 per site. It can also cost sponsors $120 per patient to acquire real-world evidence (RWE) and construct an external control arm to analyze patient subgroups. On top of that, a recent study estimated that it costs $600,000 to $800,000 each day that the FDA delays its review of a new drug or biologics license application.⁶

What costs are associated with a diversity plan?

Belgrave: Once companies have a thorough understanding of the epidemiology of the target disease, they may find that their “go-to” investigative sites can’t enroll a representative patient population. In this case, sponsors need to allocate funds for identifying and engaging with new key opinion leaders (KOLs) and community-based sites that have access to more diverse patients. Engaging ethnic minority researchers to serve as principal investigators (PIs) and site nurses can help enroll more ethnically diverse patients.

Some community-based or academic PIs and sites may be less experienced with clinical trials and require more training. This needs to be built into the research budget from day one. Remember: sponsors don’t need every site in a trial to be diverse; they need the overall program to be diverse. Bundling their existing “comfort zone” KOLs and sites with more diverse, less experienced ones—facilitating an exchange of ideas and expertise—can be an effective approach. We advise sponsors to let each site do what it’s good at.

How can an early device strategy save money?

Belgrave: Say a sponsor is developing a therapy for breast cancer, and they know they need to enroll a certain percentage of Hispanic women. If they conduct homogenous Phase 1 and 2 trials without building diversity into recruitment and planning, they will scramble at Phase 3 to find investigators and sites. In oncology, it is feasible to build diversity into Phase 1 trials. It is more economical than trying to do it later because Phase 1 cancer trials can enroll patients rather than healthy volunteers, enabling investigators to aggregate the data from every development phase. Considering inclusion at the end of Phase 2 trials is too late and expensive.

By planning early at the program level, you can create efficiencies for later in clinical development. For example, using the same sites from Phase 1 to 3 will build stronger partnerships with the communities involved.

If you start early in Phase I and do a comparator or second-line treatment study later, you already have a patient pool to work from instead of starting from scratch. This is also true if you study metastatic versus earlier-stage cancers.

How soon will regulators require diverse trials?

Belgrave: In 2022, we saw new diversity guidance documents from the FDA and Health Canada urging sponsors to recruit diverse patients, balancing race, ethnicity, gender, and age. Then, in December 2022, U.S. lawmakers passed an appropriations bill that included the Food and Drug Omnibus Reform Act of 2022 (FDORA).⁷ FDORA mandates that sponsors submit diversity action plans for pivotal clinical trials. As part of this process, the FDA will host a public workshop on November 29 and 30, 2023, to solicit input from stakeholders on enhancing diversity in clinical research.⁸ Then, in December 2023, the FDA will issue its final guidance, and sponsors will have 180 days to implement it.

At Parexel, we advise clients to immediately prepare a diversity action plan for all their assets in development because the May 2024 deadline is fast approaching.

The FDA's draft guidance on diversity (April 2022)⁹ recommends that sponsors create a diversity plan before submitting a pivotal trial design by the end of Phase 2 studies. That’s because an NDA or BLA must present effectiveness and safety data by gender, age, and race and identify dose modifications or intervals needed for a specific subgroup. However, establishing physiologic differences or genetic variation requires evidence. A solid scientific and medical rationale exists for population selection as early as possible. Collecting pharmacokinetic, pharmacodynamic, and pharmacogenomic data from a relevant and representative group of patients can inform drug exposure and response analyses. This can and should begin in Phase 1 or 1/2 testing if feasible.

In Europe, the UK’s Medicines and Healthcare products Regulatory Agency (MHRA) is leading the charge on diversity. The agency is preparing a draft guidance that closely aligns with the FDA’s requirement for diversity plans. Parexel has participated in the MHRA’s consultations with the industry and urged them to produce guidelines consistent with the FDA’s to make it easier for sponsors to comply.

What strategies can ensure that diverse trials succeed?

Belgrave: Inclusive clinical trials present some challenges. Consider how patient communities differ in different regions or socioeconomic areas of the United States alone, then multiply that to account for the rest of the world. What works in downtown Cincinnati may not work in rural Wyoming. What works in the United States may not work in France, and what works in France may not work in Korea. Companies must research this because trials that don’t have sites near where most patients are will struggle to hit enrollment targets. 

We recently helped a client run a Phase III study for a severe autoimmune disease. To maximize recruitment from African-American and Latino communities with a high disease incidence, we pursued a two-pronged strategy to engage patients and sites. We deployed patient education tools with diverse representation in imagery and appropriate health literacy terminology. We also offered transportation and childcare support to patients. We engaged with sites through clinical enrollment managers, who map referral pathways and share best practices for their local communities. We recruited 38% of study participants from ethnically and racially diverse backgrounds. This exceeded the FDA’s 20% diversity target for the condition and the industry average of 12%.

Accommodate the needs and limitations of diverse patients with practical study designs that minimize dropouts. Engage early with patients and healthcare professionals and listen to them. Site nurses are experts in scheduling study visits and how to accommodate various populations, such as the elderly and their caregivers versus toddlers and their parents. At Parexel, we use their input to optimize protocol designs that work for patients and communities. This improves recruitment and retention.

Translate trial materials into multiple languages based on the locality. Recently, we missed an opportunity to enroll a patient with a rare cancer because we did not translate the study information and informed consent forms into Mandarin Chinese, a prevalent language in their part of California. In one New York City borough, we translated documents into nine languages, including Creole and Russian. Feasible and culturally sensitive study designs are more efficient. In the current environment, they are a prerequisite—not an aspirational goal—for effective drug development.

Contributing Expert