Enrollment of racial and ethnic minoritized groups in gynecologic oncology clinical trials: a review of the scope of the problem, contributing factors, and strategies to improve inclusion

Abstract

Racial inequities are well-documented across the gynecologic oncology care continuum, including representation of racial and ethnic minoritized groups (REMGs) in gynecologic oncology clinical trials. We specifically reviewed the scope of REMG disparities, contributing factors, and strategies to improve inclusion. We found systematic, and progressively worsening, under-enrollment of REMGs, particularly of Black and Latinx populations. Additionally, race/ethnicity data reporting is poor, yet a prerequisite for accountability to recruitment goals. Trial participation barriers are multifactorial, and successful remediation likely requires multi-level strategies. More rigorous, transparent data on trial participants and effectiveness studies on REMG recruitment strategies are needed to improve enrollment.

INTRODUCTION

Healthcare disparities in gynecology have been described along multiple axes, including race/ethnicity, age, gender socioeconomic status (SES), education, geography, sexual identity, disability, and language. Racial inequities in gynecologic oncology are particularly striking; Black individuals are nearly twice as likely to die from endometrial and cervical cancers compared to White individuals, representing two of the five largest disparities in US Black-White cancer mortality.¹ Racial disparities are well-documented across the continuum of gynecologic oncology care, from basic science research, epidemiology, screening, prevention, treatment, outcomes, participation in clinical trials, and policy.²

This review focuses on disparities related to racial and ethnic minoritized groups (REMGs) in gynecologic oncology clinical trials, recognizing race as a social construct that can be conceptualized using Doll’s three frameworks for studying racial inequity within gynecologic oncology.³ These frameworks capture how embodied structural racism may explain racial differences in cancer molecular biology and genetics, or how oppression based on the intersectionality between race and SES impacts receipt of quality care and disparities in cancer morbidity and mortality.

We review the scope of REMG disparities in gynecologic cancer clinical trials (CCTs), contributing factors, and strategies to improve inclusion. Inclusion of representative study populations is essential for scientifically reliable and generalizable results regarding safety and efficacy, as well as for the use of trial data to inform clinical guidelines and standards of care. Racial and ethnic equity in oncology trials merits particular attention because precision oncology therapies depend on broad genetic sampling to develop standards of “normal” and “pathologic.” The absence of diverse cohorts results in treatments based on disease paradigms without genetic data from patients with the highest oncologic morbidity and mortality.⁴ Furthermore, clinical trials often offer the only access to the most cutting-edge cancer treatment and provide access to improved survivorship.

SCOPE OF THE PROBLEM IN GYNECOLOGIC ONCOLOGY

Recognition of pervasively low enrollment of women, gender non-binary individuals, and REMGs in US CCTs has prompted large-scale efforts such as the enactment of the National Institutes of Health (NIH) Revitalization Act of 1993.^5,6 Yet despite this mandate, inclusion of REMGs in CCTs-- including within gynecologic oncology-- has actually worsened over time,^7,8 and reporting and analysis by race/ethnicity remains inadequate.⁹

A 2015 review of 445 Gynecologic Oncology Group (GOG) publications (67,568 participants,1985–2013) found a 2.8-fold decrease in the enrollment of Black participants over time (16% in 1994–2002 v 5.8% from 2009–2013, p<0.001), and observed enrollment of Black participants far below expected enrollment based on age-adjusted incidence of malignancies (Table 1).¹⁰ Subsequent reviews of gynecologic oncology phase 1¹¹ and immunotherapy CCTs⁴ demonstrated similar, worsening trends-- including up to 32-fold lower-than-expected enrollment of Black women in ovarian immunotherapy trials.

Table 1.

Expected vs. observed enrollment of Black vs. White patients in GOG clinical trials between 1985–2013. From Scalici J, Finan MA, Black J, et al. Minority participation in Gynecologic Oncology Group (GOG) Studies. Gynecol Oncol. 2015;138(2):441–444; with permission. (See Table 3 in original).

	White Incidence	Black Incidence	Expected W:B Ratio
Expected Incidence per 100,000
Ovary	12.9	9.5	1 to 0.74
Endometrial	24.8	22.4	1 to 0.90
Cervical	7.7	10.3	1 to 1.34
Sarcoma	3.6	7	1 to 1.94
	Observed White	Observed Black	Observed W:B Ratio
Observed Incidence
Ovary	23,269	1,258	1 to 0.05
Endometrial	9,289	850	1 to 0.09
Cervical	4,421	1,340	1 to 0.30
Sarcoma	638	238	1 to 0.37
	Expected W:B Ratio	Observed W:B Ratio	Fold Difference In Ratios
Ovary	1 to 0.74	1 to 0.05	14.8-fold
Endometrial	1 to 0.90	1 to 0.09	9.8-fold
Cervical	1 to 1.34	1 to 0.30	4.5-fold
Sarcoma	1 to 1.94	1 to 0.37	5.2-fold

While the above reviews only evaluated disparities between Black and White trial participants, other reviews additionally characterized representation of Latinx, Asian American/Pacific Islander (AAPI), and American Indian/Alaskan Native (AI/AN) individuals. A 2016 review of 156 National Cancer Institute (NCI) sponsored gynecologic CCTs (18,913 participants, 2003–2012) found lower-than-expected enrollment of Black, Latinx, and Asian American/Pacific Islander (AAPI) participants across all cancer types (ovarian, uterine, cervical), with the greatest inequities for Black participants in ovarian CCTs and Latinx participants in ovarian and uterine CCTs (4% enrolled v 11% expected for Black participants in ovarian CCTs; 4% v 18% and 5% v 19% for Latinx women in ovarian and uterine CCTs, respectively).¹² Conversely, White participants were overrepresented for all cancer types, while AI/AN enrollment matched or exceeded expected incidence. A 2022 review that included 23 uterine and cervical cancer brachytherapy trials found lower-than-expected representation of Latinx and AAPI participants for uterine and cervical CCTs, respectively (both p<0.001).¹³ A recent 2022 study assessed representation across all United States (US)-based obstetrics and gynecology (OB/Gyn) clinical trials registered on ClinicalTrials.gov and resultant publications (591,196 participants, 2007–2020) and found gynecologic oncology performs the worst of all OB/Gyn subspecialties with respect to Latinx and Black participant enrollment (multivariate regression odds ratios, p<0.05).¹⁴

These trends appear to continue based on participants included in the most recent practice-changing trials. Some, such as SOLO1 and PRIMA, failed to report race and ethnicity data, in part because of varied demographic definitions across countries.^15,16 Fifteen recent trials on PARP inhibitors under-enrolled non-Hispanic Black and Latinx participants, whereas White and AAPI participants were over-represented.¹⁷ Cutting-edge trials on lenvatinib plus pembrolizumab (KEYNOTE 146, KEYNOTE 775)^18,19 and tisotumab vedotin (innovaTV204/GOG3023)²⁰ featured over 90% White participants. Notably, many of these trials enrolled patients in European countries with predominantly White patient populations, which further dilutes the total percentage of non-White participants from more racially and ethnically diverse countries.

Reporting of participant race/ethnicity is also poor, with only 38% of GOG trials¹⁰ and 23% of phase I trials¹¹ reporting racial breakdown of participants. However, reporting did appear to improve over time.^4,10 A recent analysis of racial demographics in US-based gynecologic oncology clinical trials and resultant publications (306 trials and 140 publications, 2007–2020) demonstrated that only 50% of trials reported racial/ethnic demographic data in ClinicalTrials.gov and less than 75% of trial publications included participant race/ethnicity in results.¹⁴ Although gynecologic oncology performed better than other specialties with respect to reporting, the extent of missing data hinders progress towards reducing inequities in REMG enrollment.^14,21 It is not possible to extrapolate the racial makeup of trials without reported results.

In summary, enrollment of REMGs and reporting of race/ethnicity demographic data is poor. While reporting appears to be improving over time, enrollment has worsened. These findings persist despite nearly three decades of national efforts to improve diverse representation in CCTs. We will next summarize factors that contribute to insufficient reporting and representation in gynecologic CCTs, as well as efforts to ameliorate this critical issue.

CONTRIBUTING FACTORS

Overall enrollment of patients in CCTs has been cited to be as low as 2.5% of those eligible, and in addition to underrepresentation of REMGs, studies have cited underrepresentation of adolescents, older adults (≥65 years old), rural residents, and low SES individuals.²² The multifactorial etiologies of clinical trial homogeneity can be categorized into historical, socioeconomic, cultural, and structural barriers. Each of these categories intersects with race and ethnicity in a complex matrix that influences REMG enrollment in gynecologic oncology trials.

Historical Factors

Mistrust in the medical establishment, including participation in clinical trials, has been earned through centuries of systematic exploitation of REMG populations. Perhaps the most notorious US research study is the US Public Health Service Syphilis Study at Tuskegee.²³ Examples from the field of OB/Gyn include the development of the HeLa immortalized cell line, cells derived without consent from a cervical cancer tumor biopsy taken from Henrietta Lacks, a Black woman whose descendants continue to live in poverty without compensation for the lucrative, widespread use of her cells.²⁴ Other examples include unanesthetized surgical experimentation of obstetric fistula repair techniques by Marion Sims on Anarcha, Lucy, Betsey and other enslaved Black women.²⁵ Finally, coercive sterilization of Chicana women at the Los Angeles-USC Medical Center in the 1960s-1970s, culminating in the landmark civil rights lawsuit Madrigal v Quilligan,²⁶ is one of countless examples of reproductive abuse among REMG women²⁷ that continues to present day, with accusations of forced sterilization of Immigration Customs Enforcement (ICE) detainees as recently as 2020.²⁸

Several reviews cite mistrust of research and the medical system, perceived harms, and exploitation of REMG individuals to the benefit of institutions and White populations as barriers to enrollment among Latinx, AAPI, and Black individuals.^29,30 However, the fallibility of these obstacles have recently come into question; numerous studies have demonstrated that willingness to participate in CCTs does not vary by race and ethnicity.^31–35 Attributing low enrollment to REMG attitudes, may not only be inaccurate, but further perpetuate inequitable access to trials if other etiologies are not addressed. History should remain relevant as a scaffold for social justice and the need for research reparations that acknowledge the horrific role scientists played in the abuse of REMG individuals. However, such history should be shared with caution and appropriate context to avoid negative stereotypes and misapplication by the research industrial complex that further ostracizes REMG populations.

Socioeconomic Factors

SES is an important social determinant of health that is highly interconnected with race and ethnicity, due to centuries of discriminatory US policies and practices that systematically excluded non-White populations from financial institutions and means of socioeconomic mobility.^36,37 Accordingly, systematic reviews identify several socioeconomic obstacles to participating in clinical trials including: lack of transportation, inadequate or underinsurance, inordinate transport times, lack of childcare, difficulty attending research requirements while balancing multiple competing demands (multiple jobs, childcare, familial demands), and low health literacy.^29,30 Within gynecologic oncology, a retrospective review of patients who were eligible for, and offered CCTs, found that the second most common factor for declining participation was a socioeconomic barrier.³⁸ Individuals with low literacy and health knowledge are disproportionately burdened by complex medical-legal paperwork. Additionally, individuals with low household income are more likely to reside in “low access counties,” defined as US counties located over 50 miles from the closest gynecologic oncologist, which further restricts access to CCTs.³⁹ Collectively, these obstacles contribute to a poverty tax⁴⁰ and can result in reduced access to trials, lost wages, potential workplace penalization and incremental costs that comprise an untenable proportion of low-income individuals’ income and prevent their participation in CCTs.

Cultural Factors

Cultural inclusivity plays a key part of diversity in clinical trials and often manifests differently for distinct racial and ethnic groups. Lack of access to culturally- and linguistically-appropriate materials and research staff has been cited as a shared barrier across REMGs, particularly impacting non-English speaking populations’ trial awareness, recruitment, retention, and completion.^29,30 Other culturally-inclusive practices vary across groups; for example, several studies emphasize the need for culturally-congruent research designs that allow familial involvement in the consent process in order to successfully recruit AAPIs; others noted the importance of accounting for fears that study participation would impact patients’ immigration status.^30,41,42

Racial and cultural concordance between research staff and participants, including use of investigators and staff from the same targeted communities as participants, as well as partnerships with community organizations, can all facilitate greater inclusivity and increased REMG recruitment.^29,30,43 Lack of diversity among gynecologic oncology providers challenges the use of racially- and culturally-concordant investigators and practices. Unfortunately, this is likely to persist based on recent trainee demographics: underrepresented minorities (URM; defined as Black, Latinx, AI/AN, AAPI, and Other individuals), are least represented among gynecologic oncology fellows, compared to all other OB/Gyn, surgical and medical fields (Figure 1).⁴⁴ Accordingly, much of the literature on cultural barriers to diverse research recruitment notes lack of cultural competence, even cases of racial bias, among researchers who are predominantly White.⁴⁵ This contributes to increased time, labor and materials needed to implement culturally-congruent research designs with culturally-concordant staff.⁴⁶

Figure 1.

Three-year trend of underrepresented minority (URM) representation of Gyn Onc vs. Ob Gyn vs. All Other Surgical vs. All Other Medical specialties. From Ngo NT, Aniagolu N, Lang J, Mcdougale A, Ekwenna O. Underrepresented minority representation trends in gynecologic oncology fellowships in the United States. Gynecologic Oncology. 2021;160(2):485–491; with permission (See Figure 1 in original).

Relatedly, a systematic review of provider-related factors influencing recruitment of REMGs to CCTs found several studies in which REMGs cited physicians’ communication methods as a key barrier to trial enrollment, reporting that they felt like they were treated with a lack of compassion or respect, or felt rushed or patronized.⁴⁷ Studies showed that interpersonal racism informs stereotypes about the “ideal” patient to recruit for cancer studies and that providers hesitate to recruit REMG participants based on presumptions of mistrust and thus disinterest in enrollment, as well as unsubstantiated concerns for their adherence to trial protocols.^45,47–49 Correspondingly, providers are less likely to offer CCT participation or provide robust discussions of CCT options with REMG patients.^50,51 The combination of poorly-inclusive practices as well as interpersonal and structural racism have led to a research ecosystem that disproportionately excludes REMG participants.

Structural Factors

A 2000 review concluded that REMGs are disproportionately burdened by structural barriers to CCT participation, which relate to health and research system factors such as study design, access to care, funder influences, and health policies.⁵² Limited trial availability and trial design features, namely restrictive eligibility criteria, exclude over three-fourths of cancer patients from participation.⁵³ REMGs suffer disproportionately from chronic medical conditions included in CCT exclusion criteria (e.g. cardiac disease).^29,54–56 Furthermore, REMGs are more likely to be excluded due to subjective exclusion criteria that are susceptible to bias, e.g. perceived “mental status” or “history of noncompliance,” or false assumptions about disqualifying comorbidities.^38,57

REMGs are also less likely to receive cancer care at high volume centers⁵⁸ or with gynecologic oncologists who conduct CCTs.^59,60 Small hospitals and clinics where REMGs often receive medical care have limited research infrastructure and resources to support trial recruitment,^61,62 and frequently lack robust referral systems to centers where trials occur.^63,64 Additionally, lack of adequate health insurance and decreased access to cancer screening care delays cancer diagnoses in REMG populations, consequently delaying and/or precluding CCT enrollment.⁵²

Other trial features that limit diverse recruitment include: long study durations, strict treatment or intervention schedules, increased follow-up visits, and side effects.^29,30,52 Finally, unique challenges to gynecologic oncology include relatively low incidence of some cancers, such as vulvar cancer, or strict inclusion criteria for disease stage or severity that limit sample sizes.⁶⁵

STRATEGIES TO INCREASE INCLUSION

Several systematic reviews^30,66–68 and proceedings from high-level meetings^69,70 have been published on effective strategies to increase REMG participation in CCTs. These comment on the limited research base, including: few studies that quantify or compare strategy efficacy or effectiveness,⁶⁸ scant studies within each REMG subgroup,^30,66 and lack of studies that implement strategies to recruit multiple REMG subgroups⁶⁶-- thus limiting assessment of whether strategy success varies by REMG subgroup. A 2019 mixed methods assessment of US Cancer Centers for Excellence with above-average recruitment of REMGs⁷¹ highlighted four domains of “notable practices,” which we have adapted to organize findings from our review of the literature as: 1) Leadership, 2) Provider/Investigator, 3) Community and Patient, 4) Structural.

Leadership engagement

Cancer centers with above-average REMG recruitment had leadership committed to an institutional focus on diversity, expressed through: recruitment of REMG faculty, establishment of leadership roles dedicated to diversity issues and REMG faculty recruitment and promotion, creation of institutional partnerships between faculty and community physicians, and support of related system-wide educational campaigns and programs.⁷¹ The National Institute on Minority Health and Health Disparities (NIMHD) has provided financial support for these efforts since 2002, including funding centers engaged in health disparities research.

Provider/investigator engagement

Healthcare providers’ recommendation to join a trial has been cited as the most influential factor for patient participation.⁷¹ However, provider bias limits REMG recruitment. Strategies that increase the number of racially- and culturally-concordant faculty and research staff involved in CCTs may improve REMG trial recruitment and retention.⁴³ Cultural competency training is a widely-used strategy towards this goal, with tools available online from Enhancing Minority Participation in Clinical Trials (EMPaCT), a consortium of five institutions with both NCI-funded cancer centers and NIMHD-funded health disparities programs.^72,73 However, there is limited evidence that cultural competency training alone is sufficient.⁷⁴ Instead, more robust institutional infrastructure is needed to create an inclusive culture for women, gender non-binary individuals, and REMGs.⁷¹ The NIMHD supports such efforts through institutional grants, career development, and loan repayment programs targeted to REMG researchers.⁷⁵

Community and patient engagement

Community-based participatory research methods (CBPR)

Local partnerships and implementation of CBPR principles are strategies that can help ensure research aligns with unique social, cultural, and historical attributes of diverse patient communities. A 2021 systematic review found that CBPR methods underlied community engagement interventions that successfully improved Black CCT participation.⁷⁶ Others similarly highlighted use of CBPR principles, such as: training staff in culturally-competent trial design, transparent sharing of findings with participants and communities, and sharing power through integration of community advisory boards in trial leadership and operations.⁷¹ An NCI-funded initiative, “Walking Forward,” credited use of these methods with successful recruitment of AI participants to CCTs.⁷⁷

Community outreach

Community outreach strategies such as partnering with important community institutions, grassroots nonprofits and advocacy groups,^78,79 as well as leveraging trusted community members, can increase REMG recruitment. The multisite design of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO), a NCI-funded trial that enrolled 155,000 subjects over 22 years, permitted comparison of recruitment strategies and found that a “one size fits all” approach using mass media direct mailers resulted in underrepresentation of REMGs.^46,80 In contrast, sites that implemented CBPR methods, partnered with local organizations, highlighted unique local cultural values, provided bilingual staff, and elevated community stakeholders found success in the recruitment of Latinx, Black and Asian participants.^81–83 Of note, these studies all cited how community outreach is labor intensive and resulted in significantly higher per capita recruitment costs (in part because homogenous research staff did not represent the diverse communities from which they recruited), compared to conventional methods.⁸⁰

Lay recruiters and navigators

The use of lay community representatives such as patient navigators from REMG communities expands cadres of trusted racially- and culturally concordant research staff while simultaneously strengthening community capacity.⁷¹ A 2014 literature review identified 12 studies on the use of navigators to increase enrollment of Black, Latinx, AAPI, and AI/AN patients in CCTs, and highlighted navigators’ unique ability to assist with logistical and cultural barriers.⁸⁴ Two subsequent studies demonstrated that lay navigators nearly doubled CCT participation and completion among AAPI and Black participants, respectively (p<0.001).^85,86

Patient engagement

Low SES individuals with low literacy and health knowledge require increased support through research staffing and communication tools; strategies that are generally more labor intensive and involve more personal contacts. Three notable participatory methods are repeatedly cited: engaging patients in trial design; providing culturally and linguistically accessible, user-friendly materials; and building trust.⁷¹ This includes appropriate remuneration for trial participants with financial incentives and benefits.³⁰ Recent studies with mobile and wearable technology may present a novel approach.⁸⁷ As most individuals across class, race and ethnic groups own smartphones, these present an equitable strategy with the notable exception of geriatric populations, who have lower cell phone ownership and use.⁸⁸

Overcoming structural barriers

Two of the most important structural barriers to REMG CCT participation are limited trial availability and restrictive eligibility criteria. Developing referral networks between academic medical centers and primary care or smaller community practices that serve REMG communities,^89–93 and widening eligibility criteria to reflect the chronic diseases that disproportionately impact REMG communities could help overcome these barriers.⁵³

Additionally, standardized reporting metrics and recruitment requirements are critical to REMG recruitment improvement. Since passage of the 1993 NIH Revitalization Act, NIH agencies like the NIMHD and the NCI have expressed commitment to documenting and reporting information to improve REMG trial recruitment, as stipulated in requirements for investigators to secure NIH funding.^5,6 Yet, few accountability mechanisms or consequences exist for not meeting recruitment goals.⁹⁴ In order to comply with this legislation, NCI-designated cancer center support grant (CCSG) reporting requirements mandate accrual of women and minorities to interventional clinical trials in rough proportion to the cancer patient population of the CCSG’s primary catchment area.⁹⁵ However, CCSG recipients vary widely in how they define recruitment goals, instead benchmarking too narrowly to the demographics of their center’s existing cancer patient population, or too widely, to their city or state’s general population-- which can each vary significantly with regards to racial/ethnic makeup.⁹⁶ More standardized data definitions and collecting and reporting processes to quantify grantee’s progress on minority trial accrual, including reporting on how race, gender, and race/ethnicity are defined, determined, collected, and reported are needed.⁹⁶ The Food and Drug Administration’s recently-released draft guidance to industry sponsors on the creation of “Diversity Plans” to improve REMG enrollment in clinical trials is a promising step in this direction.⁹⁷

In addition to specialty-specific reformations to improve REMG recruitment and retention, national legislation can promote, support, and fund diversity in CCTs. For example, the 2000 Medicare policy change requiring coverage of routine care costs of CCT participants,⁹⁸ and the 2010 passage of the Affordable Care Act (ACA), which required private insurance plans to allow clinical trial participation and cover standard-of-care costs.⁹⁸ Of note however, the ACA requirement did not extend to state Medicaid plans,⁹⁹ of which over 60% of enrollees identify as Black, Latinx, Asian American or another non-White race/ethnicity.¹⁰⁰ Instead, only some states passed mandates requiring Medicaid to cover routine care costs in CCTs, thus limiting trial access.⁹⁸ There is hope however that more equitable access to CCTs will occur given recent passage of the federal Clinical Treatment Act, which went into effect January 1, 2022, and requires all Medicaid plans to cover routine costs of CCTs.⁹⁸

CONCLUSION

Diversity in clinical trials is essential for validity, generalizability, and equity. Recent reviews of gynecologic oncology trials have found systematic, and progressively worsening, under-enrollment of REMGs, particularly of Black and Latinx populations. The contrast between gynecologic oncology and other subspecialties in trials that often occur at similar academic institutions begs the question of why gynecologic oncology performs so poorly, especially when other subspecialties-- though far from succeeding-- are recruiting more representative trial participants. Transparency and reliable access to accurate data on clinical trial enrollment are prerequisites to equitable REMG representation. To improve clinical trial enrollment equity, all gynecologic oncology trialists must commit to and provide resources for accurate, inclusive enrollment metrics and practices. As the barriers to CCT participation are multifactorial, the greatest success may occur when multi-level strategies are implemented simultaneously. Once these data become more rigorous and broadly available, they can be critiqued and analyzed for areas where the gynecologic oncology community can improve enrollment and provide a baseline from which to measure progress.