Statistical Methodologies Advancement and Policy Change for Enhancing Diversity and Inclusion in Clinical Trials

Meizi Liu (Takeda Pharmaceuticals), Ruben G.W. Quek (Regeneron Pharmaceuticals Inc), Rachael Liu (Takeda Pharmaceuticals)

1. Introduction

Diversity and inclusion are becoming crucial priorities in clinical trials, where underrepresentation remains a persistent challenge, limiting the generalizability of trial results and potentially compromising patient safety. This lack of diversity in clinical trials limits collection of data on how well treatments work across different populations, impacting health equity and trust in healthcare systems, particularly among underrepresented communities. Addressing this gap is essential not only for scientific validity but also to ensure equal access to safe, effective treatments for all.

Speakers and organizers for parallel session PS21 at RISW 2024: Rachae Liu, Meizi Liu, Atasi Poddar, and Ruben Quek (from left to right).

This article summarizes the parallel session PS21 at ASA Biopharmaceutical Section Regulatory-Industry Statistics Workshop 2024 (RISW 2024), which focuses on enhance diversity and inclusion in clinical trials from multiple perspectives. It explores the new guidelines and action plans established by the FDA, the innovative statistical methods that can be leveraged to improve representation, and the integration of patient input to make trials more relevant and accessible. Together, these perspectives underscore a growing commitment across the pharmaceutical and healthcare sectors to improve diversity and inclusion in clinical trials, driving toward a future where trial results are more applicable, ethical, and beneficial for all.¹

2. Regulatory Strategies for Promoting Diversity in Clinical Trials

From a regulatory standpoint, the FDA is committed to increasing the participation of underrepresented populations in clinical trials. Some notable examples include issuing guidance with recommendations to include representative populations [1], encouraging the use of innovative trial designs, promoting fit-for-purpose technologies that make clinical trials more accessible [2], and conducting solution driven public workshops [3]. To further promote transparency, the FDA also publishes demographic data from pivotal clinical trials that supported the original FDA approval of new drugs on its website, thereby providing insight into current representation gaps [4].

Despite these efforts, multiple challenges continue to hinder diverse participants in clinical trials. Some examples were discussed at the meeting. The eligibility criteria of the clinical trials determine who should be able to participate in clinical trials. The use of template of commonly used eligibility criteria, which may not have strong clinical and scientific justification in the context of a trial, may limit participation of representative populations. Logistical challenges of participating in clinical trials including distant geographic locations of the trial sites, frequent trial site visits, financial burden of participation, and language barriers may further restrict the inclusion of participants. For rare disease trials, to enroll and to retain a broad spectrum of the patient population could be uniquely challenging as rare diseases often affect small, geographically dispersed patient populations who may have disease-related travel limitations.

Examples of strategies to promote diversity in clinical trials were discussed at the workshop. The FDA recommends sponsors to: 1) broaden eligibility criteria to increase diversity in enrollment through inclusive trial practices, trial design and methodological approaches, and enrichment strategies; 2) improve enrollment, recruitment, and retention practices to promote inclusivity; and 3) adopt recruitment and retention practices specific to trials for drugs intended to treat rare diseases.

Additionally, in a guidance issued in 2022, the FDA recommends sponsors of the medical products to submit Diversity Action Plans for the enrollment and retention of a clinically relevant study population, to help ensure adequate representativeness of study participants that reflect different age groups, sexes, and racial and ethnic demographic characteristics [5]. The guidance satisfies section 3602 of FDORA which requires that FDA updates or issues guidance relating to the format and content of Diversity Action Plans. The plan must include: 1) goals for enrollment in the clinical study, disaggregated by race, ethnicity, sex, and age group of clinically relevant study populations; 2) rationale for such goals; and 3) explanation of how the sponsor intends to meet such goals. For drugs and biological products, FDA recommends that sponsors submit the plan to the relevant Investigational New Drug applications as soon as practicable but no later than the date on which the sponsor submits the protocol to FDA for the phase 3 study or, as appropriate, other pivotal study. The plan is applicable for certain clinical studies of devices, including those intended to serve as the primary basis for the FDA’s evaluation of the safety and effectiveness and benefit-risk determination of the device. The plans are intended to increase clinical study enrollment of participants of historically underrepresented populations and improve the data the agency evaluates to determine if a new medical product can be marketed in the US. In summary, these strategies contribute to a regulatory framework that promotes equitable clinical research participation.

3. The Role of Statisticians in Promoting Diversity in Clinical Trials

Statisticians play a crucial role in promoting diversity, equity, and inclusion in clinical trials. By carefully considering population diversity during the trial design and analysis phases, statisticians ensure that data collected can better represent the full spectrum of populations affected by diseases. Without careful planning and considerations in study design, trials risk underrepresenting key demographic groups. The findings from such trials may lack generalizability, limiting their application to real-world settings. By collaborating with various stakeholders, statisticians contribute to developing study designs, protocols, diversity action plans, and statistical analysis plans. Their work lays the foundation for scientifically rigorous and inclusive research, helping to address disparities and promote equitable health outcomes across diverse communities.

From a statistical methods perspective, overcoming challenges in enrolling diverse subgroups, involves leveraging advanced methodologies such as adaptive enrichment designs and Bayesian dynamic borrowing. Adaptive enrichment designs allow trials with interim decisions to enrich subgroup representation, while Bayesian borrowing incorporates real-world evidence, effectively increasing sample sizes for underrepresented groups. These approaches, supported by the FDA’s Complex Innovative Trial Design (CID) Pilot Program [6], facilitate trial designs that prioritize diversity without compromising reliability, ensuring that clinical findings are applicable to all affected populations.

Case Study

During the session, we shared a hypothetical case study on how the innovative statistical methodologies can be applied for more efficient drug development for a post-marketing commitment (PMC) study. In this case study, targeted therapy A is a novel drug in oncology which was granted approval in a phase III confirmatory study. However, due to multiregional study, only 8.7% of the safety population in the phase III study is African American. And the safety data raised concerns as the rate of a particular adverse events (AEs), was notably high among the African American population. Table 1 provides an overview of the hypothetical safety population data from the confirmatory study.

To further investigate these AEs, the sponsor plans a PMC study, specifically focusing on African American patients by conducting a single-arm trial. The goal is to assess the incidence of AEs within this subgroup. In the traditional approach, without borrowing from prior data, the sample size could be determined using a standard non-inferiority trial framework. Assuming a non-inferiority margin of 0.12, 62 African American patients would be required to achieve the desired power of 80%.

By applying the multisource exchangeability model (MEM)[7], a Bayesian borrowing method, we can reduce the sample size to 55 patients while maintaining the same level of statistical power as the standard non-inferiority design (See Figure 1A). The MEM method is similar to other Bayesian hierarchical models, as it leverages not only data from African American patients but also from other racial groups with similar AE rates. Additionally, we ensure that the type I error is reasonably controlled below 0.2, making this approach both statistically sound and efficient (See Figure 1B).

Further details regarding MEM can be found in [7, 8]. This case study demonstrates how innovative statistical methods, such as Bayesian borrowing, provide opportunities to improve the overall effectiveness of study design. These methods support more inclusive and efficient drug development, particularly for underrepresented populations.

Statisticians are at the forefront of efforts to improve diversity and inclusion in clinical trials. By leveraging adaptive designs, Bayesian methods, and real-world evidence, they can ensure that trials are more inclusive and generate results that are generalizable to broader populations. In doing so, they help bridge the gap in healthcare disparities, ensuring that all patients—regardless of race, ethnicity, or gender—benefit from advances in medical research. The involvement of statisticians in diversity and inclusionefforts is essential, not only to enhance the scientific validity of clinical trials but also to foster equity in access to treatment and care.

4. Enhancing Diversity and Inclusion in Clinical Trials – Considerations For A Checklist For Master Protocols and Beyond

Another important topic covered in the session was the critical role of patient engagement in drug development. Patients and their caregivers can inform drug development about the impact of disease and the ability to participate in clinical trials. Patients are the experts in multiple aspects of their disease, including symptom burden, quality of life, and potential treatment challenges. Their perspectives can inform key trial design elements, such as benefit-risk preferences (i.e., patient preference), tolerance for side effects, and desired treatment outcomes. Early engagement with patients is especially impactful, as it shapes clinical trial designs that are aligned with patients' values and needs.

The FDA has issued four guidance documents on patient-focused drug development (PFDD) [9]. Examples of potential benefits to patients awaiting treatment through more efficient clinical trial designs, such as master protocols, can include: 1) treatments and trials designs that better fit patient’s needs; 2) availability of more potential treatments; and 3) faster access to treatments. These make it possible to develop innovative therapies to minimize untoward side effects through careful dose escalation practices and lowering the chance of enrolling in a placebo arm for clinical trial participants. This may lead to an increase in patient compliance and data quality from trials. This is especially important for rare diseases and/or unique subpopulations, e.g., pediatrics, oncology genomic sub-types.

In 2020, FDA published guidance on “Enhancing the Diversity of Clinical Trial Populations — Eligibility Criteria, Enrollment Practices, and Trial Designs Guidance for Industry” [1]. This brings front and center the importance of patient engagement and Diversity/ Equity/ Inclusion/ Accessibility (DEIA) early in the drug development, especially in underrepresented communities. The 2022 Food and Drug Omnibus Reform Act (FDORA), further mandates industry sponsors to submit Diversity Action Plans (DAP) for Phase 3 or other pivotal trials to improve enrollment from diverse population [5]. However, in spite of all the guidance from the FDA, a recent interview by PinkSheet with former US Surgeon General Jerome Adams, revealed skepticism that clinical trial diversity will improve in the US without stronger government requirements [10].

In 2022, the FDA advisory committee panel feedback provided a prime example for what not to do when relying predominantly on foreign data to pursue US FDA approval [11]. Innovent and Eli Lilly sought approval for treatment of advanced non-small cell lung cancer (aNSCLC) based on a single Phase III trial, ORIENT-11, conducted entirely in China. In addition, the primary endpoint and comparator agent in that trial were inappropriate given the current therapeutic landscape for first-line aNSCLC in the US, the agency said. The FDA advisory committee panel voted 14-1 in favor of additional trials needed to demonstrate applicability of the ORIENT-11 results to US patients and medical practice. Later, the FDA issued a complete response letter denying approval of the treatment [12]. In Sept 2024, the FDA issued draft guidance regarding considerations for generating clinical evidence from oncology multiregional clinical development programs [13]. The guidance stated that Sponsor should at minimum, consider the following factors, when assessing need for multiregional clinical trials:

• Patient-related factors (e.g., exposure to disease risk factors, genetic ancestral background)

• Disease-related factors (e.g., prevalence of disease subtypes, frequency and distribution of certain molecular drivers of oncogenesis in the population)

• Health care system factors that can impact prior treatments received and available treatments following a clinical trial (e.g., access to specialized oncology care, cancer screening practices, availability and affordability of cancer treatments)

• Socio-cultural factors (e.g., diets, cultural beliefs regarding use of “alternative” therapies).

In recognition of the importance of patient engagement and representation, members of the American Statistical Association (ASA)-DahShu IDSWG Master Protocol Working Group (MPWG) was charted in 2021. One of the sub-teams, People and Patient Engagement (PE) Subteam, has been tasked to incorporate guidelines, standards and recommendations on how to include patient input into clinical trials, especially complex and innovative clinical trials such as master protocols (MPs), and to establish education and communication channels to patient advocacy groups (PAGs) [14].

During the RISW 2024 session, we shared and discussed a draft checklist to Guide DEIA considerations for MPs and explore effective strategies for enhancing diversity and inclusion in clinical trials. The checklist covers aspects pertaining to 5 Phases (See Figure 2). Details of the checklist will be part of a forthcoming book “Patient Representation and Accessibility In Master Protocol”.

Looking ahead, it is crucial to evaluate the trade-offs of striving for diverse trial participation. While the FDA may prioritize US representation and enrollment demographics, assessing trials/ sponsors efforts on inclusive outreach and recruitment practice may be important to account for the challenges that sponsors encounter when executing DAP. Importantly, investigators and sponsors have to recognize that more inclusive trials will inevitably lead to smaller sample sizes in certain subgroups, potentially complicating the detection of statistically and clinically meaningful results. As discussed by Schwartz et al, building trust of the design and conduct of clinical trial can be as important for patients as the actual results [15]. Further resources published by the Clinical Trials Transformation Initiative [16], Friends of Cancer Research [17] and US FDA Office of Minority Health and Health Equity [18] discusses the challenges and strategies for enhancing DEIA in clinical trials to ultimately improve equitable access, diverse patient participation in clinical trials and improve the generalizability of trial findings.

Figure 2: Enhancing Diversity and Inclusion in Clinical Trials: Phases of A Checklist for Master Protocols

5. Conclusion

In conclusion, advancing diversity and inclusion in clinical trials is not just an ethical imperative but a scientific necessity to ensure that trial results are broadly applicable to diverse patient populations. Regulatory initiatives, innovative industry practices, and the incorporation of patient perspectives are laying the groundwork for more inclusive trials. These efforts address the long-standing challenges of underrepresentation while driving the development of therapies that reflect the needs of all patients.

As this session highlighted, collaboration across regulators, industry leaders, and patient advocates is essential for sustaining momentum in enhancing diversity in clinical trials. Continued progress will require commitment, creativity, and a willingness to embrace new strategies. By fostering inclusivity, the clinical trial ecosystem can better fulfill its mission of delivering effective healthcare solutions to diverse communities worldwide.

References

1. U.S. Food and Drug Administration. (2020). Enhancing the diversity of clinical trial populations—Eligibility criteria, enrollment practices, and trial designs guidance for industry. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/enhancing-diversity-clinical-trial-populations-eligibility-criteria-enrollment-practices-and-trial

2. U.S. Food and Drug Administration. (2023). Digital health technologies for remote data acquisition in clinical investigations. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/digital-health-technologies-remote-data-acquisition-clinical-investigations

3. Clinical Trials Transformation Initiative. (2023, October 18-19). Virtual public workshop to enhance clinical study diversity. U.S. Food and Drug Administration. https://ctti-clinicaltrials.org/virtual-public-workshop-to-enhance-clinical-study-diversity/

4. U.S. Food and Drug Administration. (n.d.). Drug trials snapshots. https://www.fda.gov/drugs/drug-approvals-and-databases/drug-trials-snapshots

5. U.S. Food and Drug Administration. (2024). Diversity action plans to improve enrollment of participants from underrepresented populations in clinical studies. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/diversity-action-plans-improve-enrollment-participants-underrepresented-populations-clinical-studies

6. U.S. Food and Drug Administration. (n.d.). Complex innovative trial design meeting program. https://www.fda.gov/drugs/development-resources/complex-innovative-trial-design-meeting-program

7. Kaizer, A. M., Koopmeiners, J. S., & Hobbs, B. P. (2018). Bayesian hierarchical modeling based on multisource exchangeability. Biostatistics, 19(2), 169–184.

8. Wei, W., Blaha, O., Esserman, D., Zelterman, D., Kane, M., Liu, R., & Lin, J. (2024). A Bayesian platform trial design with hybrid control based on multisource exchangeability modelling. Statistics in Medicine, 43(12), 2439–2451.

9. U.S. Food and Drug Administration. (n.d.). FDA patient-focused drug development guidance series: Enhancing the incorporation of the patient's voice in medical product development. https://www.fda.gov/drugs/development-approval-process-drugs/fda-patient-focused-drug-development-guidance-series-enhancing-incorporation-patients-voice-medical

10. Karlin-Smith, S. (2024, October 25). You shall not pass: Former surgeon general on why U.S. government should push trial diversity. Citeline. Retrieved November 8, 2024, from https://insights.citeline.com//PS155063/You-Shall-Not-Pass-Former-Surgeon-General-On-Why-US-Government-Should-Push-Trial-Diversity/

11. LaFraniere, S. (2022, February 10). FDA rejects cancer drug that relied on clinical trial conducted in China. The New York Times. https://www.nytimes.com/2022/02/10/health/fda-cancer-drug-china.html

12. Eli Lilly and Company. (2022, March 24). Lilly announces complete response letter for sintilimab in combination with pemetrexed and platinum chemotherapy for the first-line treatment of people with nonsquamous non-small cell lung cancer [News release]. https://bit.ly/3umij6U

13. U.S. Food and Drug Administration. (2024). Considerations for generating clinical evidence from oncology multiregional clinical development programs. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-generating-clinical-evidence-oncology-multiregional-clinical-development-programs

14. Huml, R. A., Collyar, D., Antonijevic, Z., Beckman, R. A., Quek, R. G., & Ye, J. (2023). Aiding the adoption of master protocols by optimizing patient engagement. Therapeutic Innovation & Regulatory Science, 57(6), 1136-1147.

15. Schwartz, A. L., Alsan, M., Morris, A. A., & Halpern, S. D. (2023). Why diverse clinical trial participation matters. New England Journal of Medicine, 388(14), 1252-1254.

16. Corneli, A., Hanlen‐Rosado, E., McKenna, K., Araojo, R., Corbett, D., Vasisht, K., ... & Calvert, S. B. (2023). Enhancing diversity and inclusion in clinical trials. Clinical Pharmacology & Therapeutics, 113(3), 489-499.

17. Friends of Cancer Research. (2024) Improving equity in oncology clinical trials: challenges and strategies for setting diversity enrollment goals. https://friendsofcancerresearch.org/wp-content/uploads/Improving_Equity_in_Oncology_Clinical_Trials.pdf

18. Araojo, R., Lee, C., & Lolic, M. (2024). Bridging the Health Equity Gap: An Update from the US Food and Drug Administration Office of Minority Health and Health Equity. Health Equity, 8(1), 235-237.