Participation in U.S.‐Based ALS Clinical Trials by Sex and Race

ABSTRACT

Introduction/Aims

In global amyotrophic lateral sclerosis (ALS) trials, women and men appear to be proportionately enrolled, but quantification of enrollment by sex and race in U.S.‐based ALS trials is limited. The objective of this study was to evaluate the sex and race of participants enrolled in U.S.‐based ALS clinical trials.

Methods

Participant demographics were extracted from recent U.S.‐based Phase 2 and 3 ALS trials identified in literature and on ClinicalTrials.gov. The Centers for Disease Control and Prevention (CDC) National ALS Registry and a 2014 State Surveillance Project were used as proxies for prevalence. Participation‐to‐prevalence (PPR) ratios were calculated for sex and race. A modified time‐trend analysis was performed for race for participants enrolled before and after 2020.

Results

A total of 11 trials met criteria for inclusion with a total of 1153 patients enrolled. Compared to the CDC Registry, the PPR was 0.76 (95% CI: 0.63–0.90) for women, 1.26 (95% CI: 1.23–1.29) for White participants, 0.34 (95% CI: 0.17–0.51) for Black participants, and 0.35 (95% CI: 0.14–0.56) for all other races/multiracial. The modified time trend analysis showed no significant difference in the PPRs before and after 2020 (White: t = 1.44, p = 0.22; Black: t = −0.99, p = 0.37; Other races/multiracial: t = −1.50, p = 0.21). The comparison to the 2014 State Surveillance Project yielded similar findings.

Discussion

U.S.‐based ALS trials significantly under‐enroll non‐White participants, and there are trends toward slight underrepresentation of women. Efforts to broaden trial enrollment amongst people with ALS will help with the generalizability of trial results and hasten trial completion.

1. Introduction

The pace of amyotrophic lateral sclerosis (ALS) drug trials has increased substantially, and innovative trial designs are increasing trial efficiency and throughput [1]. As a result, more people living with ALS have the opportunity to participate in trials, and more participants are necessary to continue to enroll rapidly. Given that ALS is an uncommon disease, enrolling from the population of all people affected by ALS could hasten trial completion and improve the generalizability of future trial results.

In global ALS trials, enrollment disparities appear to exist by race and ethnicity, but not by sex [2]. Based on the Pooled Resource Open Access Clinical Trial Database (PRO‐ACT) and the optional, self‐enrolled portion of the Centers for Disease Control and Prevention (CDC) National ALS Registry, global ALS trials seem to enroll women proportionally, but there is too little data to analyze trial participation by race in these databases [3, 4]. In Alzheimer's and other neurodegenerative disease trials, Black and Hispanic people, and those with less education, are underrepresented in U.S.‐based trials [5], raising an opportunity to reach these patients and enroll trials more rapidly.

We tabulated sex and race data from participants in U.S.‐based ALS clinical trials and compared the results to two studies estimating the U.S. prevalence of ALS.

2. Methods

2.1. Trial Selection

ALS trials were identified by querying ClinicalTrials.gov using the disease term “ALS” and/or “amyotrophic lateral sclerosis.” Additionally, ALS clinical trials completed more recently were identified in a hand review of references from Tornese et al. [6]. Observational studies and device trials were excluded, as were Phase 1 trials. Since the purpose of this study was to evaluate disparities in U.S. trials specifically, those that included international sites were excluded as well. The clinical trials in Tornese et al. were cross‐referenced using the ClinicalTrials.gov registry to obtain further details on study dates, site locations, and patient demographics. Prior to 2010, race data was rarely published for ALS trials, so only trials completed between 2010 and 2023 were included in our analysis.

To estimate trial participation demographics, we recorded participant sex (female, male), race (American Indian, Asian, Native Hawaiian, African American, White/Caucasian, more than one race, unknown/not reported), and ethnicity (Hispanic or Non‐Hispanic) based on the trial publications and/or ClinicalTrials.gov.

To estimate the prevalence of ALS in the U.S., we recorded prevalence estimates from two different population surveys. First, the full CDC National ALS Registry (“Registry”), which incorporates both self‐reported data and data drawn automatically from Medicare/Medicaid and the Department of Veterans Affairs to estimate ALS prevalence [7]. The Registry includes sex and race (White, Black, and Other/multiracial) data, but ethnicity (Hispanic/Non‐Hispanic) was not available.

Second, a surveillance project (“Surveillance”) that gathered data from three states and eight metropolitan areas between 2009 and 2011 to estimate the prevalence of the disease [8]. This surveillance study included more robust racial and ethnic category reporting than the CDC Registry.

2.2. Statistical Methods

The participation‐prevalence ratio (PPR) is the ratio of a sub‐population of interest in a trial to the prevalence of that sub‐population in the general population [9]. For example, the proportion of female trial participants/the proportion of female ALS patients in the country. Because population prevalence is not known with certainty, we used both the Registry data and the Surveillance data as two estimates of the true prevalence in the U.S. population.

PPRs were calculated for the Registry using female sex, White race, Black race, and other races/multiracial, and for the Surveillance project using female sex, White race, Black race, Asian race, other races/multiracial, and Hispanic ethnicity. To calculate PPRs, we averaged enrollment numbers across trials (Table S1). We also calculated 95% confidence intervals for these PPRs. A PPR of less than 0.8 indicates underrepresentation in a trial; over 1.2 indicates over‐representation [9, 10]. The further a PPR is from the 0.8 to 1.2 range, the more dramatic the effect. Participants with unreported race or sex were included in the denominator of trial populations [2, 11].

The two registry studies estimate the prevalence of ALS across the U.S., but the racial makeup of regions, municipalities, and communities varies across the U.S. Multicenter trials, particularly ones with only a few enrolling sites, might estimate local racial makeup accurately but differ from the registry estimates. To determine whether the racial makeup of the region surrounding specific trial sites was the major driver of racial enrollment patterns, we conducted a substudy calculating the PPR for the four Healey Platform trials, which were the trials in our dataset with the largest number of enrolling sites (> 50). With so many sites, these trials are most likely to mimic the racial estimates of the ALS registries if site geography is the major driver of racial diversity in ALS trials.

We also performed a modified version of a time‐trend analysis to assess for any significant differences in the PPR of racial groups enrolled in studies prior to 2020 and those that began enrollment in 2020 or after. We compared the PPR of sex and racial groups in trials prior to 2020 to those after 2020 using a two‐tailed t‐test. All statistical analyses were performed using Microsoft Excel.

2.3. Standard Protocol Approvals, Registrations, and Patient Consents

No institutional review board approval was required because the data is publicly available.

3. Results

3.1. Trial Participation Demographics

Eleven trials met criteria for inclusion. All were phase 2 or 2/3 trials. Five of the trials started enrollment between 2015 and 2018. The remaining 4 started enrolling in 2020 as part of the HEALEY ALS Platform Trial. Among the trials, there was a total of 1153 participants. Of those, 424 (36.8%) were female. One trial did not report race; among the trials that did report race, there were 1122 participants. A total of 20 (1.8%) were Asian, 26 (2.3%) were Black, and 1057 (94.2%) were White. There were no Native American or Hawaiian/Pacific Islander participants. Four participants (0.4%) were of other races/multiracial, and 15 participants (1.3%) did not have race reported (Figure 1).

FIGURE 1.

Racial breakdown in ALS prevalence estimates and ALS trial participation. Proportion of people, by race, in the National ALS Registry (“Registry”), in the ALS state and metropolitan surveillance project (“Surveillance”), both of which estimate the population prevalence of ALS by race/ethnicity. Also plotted is enrollment in U.S.‐based ALS trials since 2010. Both the Registry and Surveillance programs approximate 25% non‐white prevalence of ALS, while U.S.‐based ALS trial enrollment is less than 10% for non‐white participants. Race categories published for the CDC Registry include only White, Black and Other races/multiracial. The Surveillance Project includes White, Black, Asian, and other races/multiracial. Reporting categories for US ALS Trials are more complete.

3.2. U.S. Prevalence Estimates

The estimated proportion of women with ALS was 46% in the Registry and 44% in the Surveilance project. The published Registry and Surveillance project data had similar prevalences of race, notably approximately 75% white participants and fewer than 10% black (Figure 1). In the Surveillance project, the proportional ALS prevalence by ethnicity was 77% non‐Hispanic, 11% Hispanic, and 12% Unknown.

3.3. Participation‐Prevalence Ratios

Compared to the Registry, White participants were overrepresented in U.S. clinical trials, while Black, multiracial participants, and participants of other races were underrepresented (Table 1 and Figure 2A). Female participants showed a statistical trend toward underrepresentation (Table 1 and Figure 2A).

TABLE 1.

Participant to prevalence ratio by sex, race and ethnicity for all trials relative to the CDC National ALS Registry and the 2014 State Surveillance Project.

	Registry PPR	Registry 95% CI		Surveillance PPR	Surveillance 95% CI
Sex				Sex
Female	0.761	0.627–0.895	Female	0.810	0.667–0.953
Race				Race
White	1.263	1.235–1.291	White	1.270	1.243–1.297
Black	0.339	0.165–0.512	Black	0.234	0.114–0.354
Asian	*	*	Asian	0.418	0.693–0.142
Other races/multiracial**	0.353	0.143–0.562	Other races/multiracial	0.680	0.013–1.347
Ethnicity				Ethnicity
Hispanic	*	*	Hispanic	0.442	0.346–0.538

Note: 95% CI = 95% confidence interval.

Abbreviation: PPR, Participation to prevalence ratio.

^*Data not reported in the CDC Registry.

^**Inclusive of Asian.

FIGURE 2.

Participation to prevalence ratios for ALS trials. Participation to prevalence ratios (PPRs) by sex and race are presented. PPR between 0.8 and 1.2 indicates that a subgroup participates in a trial at the same rate as the population prevalence of the disease. Less than 0.8 suggests participation at a rate less than expected based on disease prevalence; more than 1.2 suggests participation rates higher than expected based on disease prevalence. ALS prevalence is estimated based on two sources a) CDC National ALS Registry (“Registry”) data and b) the Surveillance Project (“Surveillance”) [8]. The shaded gray areas represent the conventional 0.8–1.2 range within which a PPR is considered to adequately reflect the general population of interest. Error bars represent the 95% confidence interval of each PPR.

Compared to the Surveillance project, White participants were again overrepresented in U.S. clinical trials, while Black, Asian, and Hispanic participants were underrepresented (Table 1 and Figure 2B). Other Race/Multiracial participants showed a statistical trend toward underrepresentation (Table 1 and Figure 2B). Female participants showed a less pronounced trend toward underrepresentation (Table 1 and Figure 2B).

3.4. Subset Analysis—Comparison of Largest Trials to the Registries

The subset analyses comparing the four trials with the largest number of enrolling sites to the National Registry and Surveillance projects also demonstrated underenrollment of Black participants (Registry PPR 0.45, 95% CI 0.13–0.77; Surveillance PPR 0.31, 95% CI 0.09–0.53).

3.5. Time Trend Analysis—Comparison of Pre‐2020 to Post‐2020 Trials

Compared to the Registry, there was no significant difference between the PPRs for race of participants enrolled in trials from 2020 to 2023 compared to before 2020 (White: t = 1.44, p = 0.22; Black: t = −0.99, p = 0.37; Other races/multiracial: t = −1.50, p = 0.21).

4. Discussion

Our study demonstrated a non‐significant trend toward a slight underrepresentation of women in U.S. ALS trials. Prior reports did not demonstrate an underrepresentation of women in global ALS trials [2, 3]. Importantly, we found that White participants were statistically significantly overrepresented and Black and Asian participants were underrepresented in U.S.‐based ALS trials. Other Race/Multiracial participants and Hispanic participants showed statistical trends toward underrepresentation. These findings held true both in the comparison of trial populations to prevalence estimates from the CDC National ALS Registry, which is large but reports fewer racial categories, and the state and metropolitan Surveillance project, which is smaller but reports race and ethnicity in more detail and collected cases from more racially diverse parts of the country.

Racial diversity in trials is important because it may improve generalizability and the trial's ability to predict safety in clinical practice. Several widely used pharmaceuticals, including riluzole, have variable pharmacokinetics, safety, or tolerability by race [12]. Additionally, for diseases like ALS, enrolling broadly across demographics can increase the rate of enrollment and hasten drug development.

There are likely multifactorial and complex reasons for the underrepresentation of non‐White participants in U.S.‐based ALS trials. This could be partially attributed to the location of ALS trial centers and the racial makeup of the surrounding community, and including sites in multiracial regions might boost racial diversity in ALS trials, though the subgroup analysis suggests this is not the major driver of representation in U.S.‐based ALS trials. Differences in healthcare access may lead to delayed or missed ALS diagnoses or reduced referral rates to ALS Centers amongst non‐White people living with ALS, limiting access to clinical trials [13, 14]. Or, differences in trial enrollment patterns between people of different races who are cared for at ALS centers might also be a factor driving the underrepresentation of non‐White participants.

Under‐enrollment of women and non‐White participants in ALS trials need not be an inevitability in the modern landscape of healthcare provision and novel trial designs. Further studies of barriers to research enrollment and clinical trial access could point to evenhanded enrollment strategies and expedite broad enrollment and trial availability. Such studies could quantify the relative contributions of factors such as site geography, access of patients to ALS clinics, and access/enrollment of ALS clinic patients of different races into ALS trials. This quantification may, in turn, suggest effective actions to improve overall racial representation in ALS trials. For example, targeted social media and search engine ads have improved racial and geographic enrollment diversity in a Parkinson's disease registry [15]. Community‐based participatory research (CBPR) can engage whole communities in research and improve accessibility [16]. Funding research stipends for participants can also increase research access across socioeconomic strata [17].

Our study has potential limitations. First, the CDC registry is an incomplete and imperfect estimate of U.S. ALS demographics. It does not include ethnicity and combines many races into a single “Other Races/Multiracial” category. It may undercount non‐White and uninsured people with ALS due to methodology for case identification. If non‐White races are undercounted in the CDC prevalence estimates, then the trial PPR for those groups would be even lower, meaning our study has underestimated the disparity. The 2014 ALS Surveillance Project was smaller but provided more robust reporting on race and ethnicity, specifically in more racially diverse cities and regions.

A second limitation is that our study compares ALS trial enrollment to prevalence estimates across the U.S. If there is a substantial geographic mismatch between the Registry and Surveillance projects compared to the enrolling sites for ALS trials, then the differences in racial makeup of the trials may simply be due to variation in racial makeup by geography. Our substudy of the four trials with the largest number of enrolling sites suggests this is not the main driver of the underrepresentation we saw, but more focused research should be conducted to quantify the relative contribution of trial site geography on the racial makeup of ALS trial participants. Thirdly, we were able to include only a relatively small number of ALS trials due to the rarity of the disease, our focus on U.S.‐based trials, and incomplete reporting of race prior to 2010. However, the racial enrollment data were consistent across trials, bolstering our confidence in the results. We noted that there were two trials in the Healey Platform that enrolled non‐White participants at a slightly higher rate than the other two Healey Platform trials. Because the 4 trials all recruited among the same sites, this makes differences in recruitment catchment unlikely to account for the difference. However, it is possible that sites used recruitment techniques better suited to a broader group of participants or perhaps there was something about those investigational therapies that had broader cultural acceptance or understanding.

In summary, our study provides data demonstrating that non‐White people with ALS were significantly under‐enrolled in U.S. ALS trials, and there was a non‐significant trend toward slight under‐enrollment of women in US‐based ALS trials. The current analysis provides a starting point for further quantification of the underlying contributors and development of effective approaches to broaden enrollment across all groups of people with ALS.

Author Contributions

M.P.L. was involved in the design of the study, data acquisition, analysis and interpretation of the data, and drafting and revision of the manuscript. S.Y.C. was involved in the conception and design of the study, data acquisition and revisions to the manuscript. K.H.Q. was involved in the design of the study, analysis and interpretation of the data, and revision of the manuscript. J.C. was involved in the conception and design of the study, interpretation of the data, and revision of the manuscript. S.B. was involved in the design of the study, analysis and interpretation of the data, and revision of the manuscript. S.P. was involved in the conception and design of the study, acquisition, analysis and interpretation of the data, and revision of the manuscript. J.D.B. was involved in the conception and design of the study, acquisition, analysis and interpretation of the data, and revision of the manuscript. All authors reviewed and approved the final version and are accountable for the integrity of their respective contributions.

Ethics Statement

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Conflicts of Interest

Suma Babu reports research grants (to institution) from NINDS, Biogen, Ionis, Novartis, OrphAI Therapeutics, and Denali; institutional consulting fees from Uniqure and Marvelbiome; and honoraria for educational talks at annual meetings from the American Association of Neuromuscular and Electrodiagnostic Medicine and the American Academy of Neurology. Dr. Babu reported serving as an appointed member of the congressionally directed National Academies of Science, Engineering, and Medicine working group on Amyotrophic Lateral Sclerosis and coauthored the report living with ALS. Sabrina Paganoni reports research grants from Amylyx Therapeutics, Revalesio Corporation, Eledon, Alector, UCB, Biohaven, Clene Nanomedicine, Prilenia, Seelos, Calico, Denali, NIH, CDC, DoD, ALS Association, Muscular Dystrophy Association, Tambourine and reports consulting fees from Amylyx, Arrowhead, BMS, Clene, Iris, Eikonizo, Sola, Prilenia, Revalesio, Merck, Biogen, Janssen, and Cytokinetics. She has been a paid educational speaker for PeerView, i3Health, and Medscape. James D. Berry reports research grants from Biogen, Brainstorm Cell Therapeutics, ProJenX, MT Pharma of America, Alexion, Rapa Therapeutics, the ALS Association, the Muscular Dystrophy Association, ALS One, ALS Finding a Cure, Tambourine, and DoD, NINDS, and reports personal consulting fees from Biogen, Clene Nanomedicine, MT Pharma of America, Janssen, Projects in Knowledge, Roon, and has been a paid DSMB member for Sanofi. The other authors have no conflicts of interest.

Supporting information

Table S1: List of ALS trials in our analysis with total enrollment and percentage of non‐White participants reported.

Levine M. P., Chung S. Y., Quigg K. H., et al., “Participation in U.S.‐Based ALS Clinical Trials by Sex and Race,” Muscle & Nerve 73, no. 1 (2026): 50–55, 10.1002/mus.70050.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.