Race and age disparities in randomized trials of acute myeloid leukemia: a systematic review and meta-analysis.

Introduction

Significant racial and ethnic disparities in the incidence and survival of patients diagnosed with hematologic malignancies have been reported. 1 , 2 , 3 , 4 , 5 Black patients with acute myeloid leukemia (AML) have been shown to have an increased risk of death compared with White patients, as have Hispanic patients compared with non-Hispanic (NH) White patients. 1 , 2 Black patients with multiple myeloma have had reduced survival benefits compared with White patients 3 and worse survival rates after unrelated donor hematopoietic cell transplantation. 4 Hispanic patients undergoing hematopoietic cell transplant have a higher risk of treatment failure and mortality. 5

These outcome disparities may be in part due to poor representation and access of racial and ethnic minorities to clinical trials. A systematic review that focused on trials leading to US Food and Drug Administration oncology drug approvals from 2008 to 2018 found that race was reported in only 63% of trials, and Black and Hispanic patients were underrepresented. 6 An observational study assessing reporting and enrollment data for acute lymphocytic leukemia (ALL) and AML trials in the United States found that there were significantly lower enrollment proportions of NH-Black, NH–Native American, NH-Asian, and Hispanic patients compared with NH-White patients. 7 In a retrospective analysis of 106 patients with intermediate-risk AML, investigators found that there was a notably increased rate of clinical trial enrollment in NH-White patients compared with other groups. 8

Older age has also not been extensively evaluated in the enrollment of patients in randomized trials of hematologic malignancies. A study analyzing data from select clinical trials for breast, colorectal, lung, and prostate cancer found that there was a disparity in age enrollment in these trials with threefold younger participants represented than those aged ≥65 years. 9 Given the lower survival rates in older adults with hematologic malignancies, 10 this is an important concern.

Although outcome disparities have been reported, the representation of ethnic and racial minorities as well as older adults in hematologic malignancies, has not yet been comprehensively studied. Understanding the discrepancies in trial enrollment is vital for the better understanding of access to care and clinical trial conduct. We, therefore, sought to systematically review the literature on enrollment of racial and ethnic minorities and older adults into randomized controlled trials (RCTs) of AML, provide enrollment estimates, and compare these with population characteristics.

Methods

Search strategy and study selection

Ovid MEDLINE was searched from inception through June 2023 using medical subject headings and free-text words. This review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. 11 The search strategy is provided as a supplemental Table 1 , Search Strategy. The inclusion criteria were phase 2 and 3 RCTs of disease-modifying therapy for AML in adults (≥18 years) reporting efficacy and safety outcomes. We excluded single-arm trials, trials with unpublished results, conference abstracts, follow-up reports, subgroup/post hoc/exploratory analyses, and supportive care trials, except those evaluating disease-modifying therapies and reporting clinical efficacy. Pairs of reviewers independently screened titles, abstracts, and full texts of records. Disagreements were resolved by consensus or in conjunction with a third reviewer. Covidence 21 was used to facilitate the study selection.

Data collection and data items

Data were extracted from each article by 2 reviewers (N.L. and O.K.). A standardized form was pilot tested and used to extract data related to trial characteristics. These included the following: the years during which a trial was performed; the total population for each trial; whether racial and ethnic subgroup data were reported for each subgroup (African American/Black, Asian, American/Alaska Native, Native Hawaiian or other Pacific Islander, White, and Hispanic), and if so, the number of participants from each subgroup; how age was recorded (if age ≥65 years was recorded, if studies dichotomized age, or if studies used a range of ages); and the continents or countries where the trials were performed.

Statistical analysis

Trial proportions

We calculated trial proportions to describe the representation of different categories within each subgroup of race or ethnicity. We pooled the numerators and denominators of trial proportions from each trial to determine the overall proportions of each race and ethnicity. We then performed a subgroup analysis, stratifying trial proportions by decade and the geographic location of the trials. We compared the proportion of trials that reported on race or ethnicity in the last 10 years (2014 and after) with those before the last 10 years (before 2014). We further stratified by trial phase, comparing phase 2 trials with phase 3 trials. We also assessed the age description used in the trials, comparing trials that used a distinct age cutoff with those that used age >65 years and with those that used no age cutoff. We compared studies completed before and after 2014 in terms of how they reported age.

Acquisition of population-based estimates

We acquired US population–based estimates of AML incidence for each racial and ethnic category from the National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) program database (SEER 21: 2000-2018 and SEER 13:1992-2018). 12 , 13 The acquisition of the US population–based estimates was tailored for each trial depending on the years during which the trial was performed.

Estimation of EIRs

Enrollment incidence ratios (EIRs) were used to assess the disparity in the enrollment of each racial or ethnic subgroup compared with the US AML population. EIRs are the ratio of trial proportions of members of a racial and ethnic subgroup divided by the US population–based incidence in the corresponding racial and ethnic subgroup in the SEER database. We calculated EIRs for each race (Black, White, Asian, and Hispanic) for each of the 4 US trials that included racial data, using SEER data in each trial from when the trial was conducted (ie, for a trial conducted 1995-1997, SEER data were extracted in the years 1995, 1996, and 1997). 14

Meta-analysis

We conducted a random-effects meta-analysis to pool EIRs. Log-transformed EIRs and the associated standard error estimated from individual trials were pooled using an inverse variance approach. Standard error was calculated using the formula √((1/a)+(1/b)), in which “a” is the number of participants of a specific race (Black, White, Hispanic, or Asian) and “b” is the total number of participants in a trial. 15 A random-effects meta-analysis was then conducted on JASP (version 0.18.1) to summarize the overall enrollment disparity for racial and ethnic subgroups.

Results

The search yielded 7759 titles and abstracts. After title and abstract screening, 157 full texts were screened, and 90 studies were included ( Figure 1 ). Fourteen of these studies (15%) were US trials. Overall, there were 21 of 90 trials (23.3%) that reported race or ethnicity and 26 of 90 (28.9%) that reported enrollment proportion of those aged ≥65 years. Of the 21 trials with data on race or ethnicity, 15 (71.4%) had data on African American/Blacks, 21 (100%) on Whites, 14 (66.7%) on Asian or Pacific Islanders, 2 (9.52%) on Native American/Alaska Native, and 4 (19.0%) on Hispanics ( Table 1 ). Of the 26 trials reporting on age, 27.5% of those conducted before 2014 reported and dichotomized by age ≥65 years, compared with 30.8% after 2014. However, 38.5% of the trials after 2014 reported on age using another cutoff, which was an increase compared with trials before 2014 in which only 21.7% did so ( Table 2 ).

Of the 21 trials reporting on race, 176 participants (4.7%) were African American/Black, 360 (9.8%) were Asian or Pacific Islander, 5 (0.47%) were Native American/Alaska Native, 3888 (80.8%) were White, and 39 (3.4%) were Hispanic ( Table 3 ). In the subgroup analysis stratifying trial proportions by decade, we found that race data were reported in trials performed in the 1990s (1 trial), 2000s (6 trials), and 2010s (13 trials). We found that in the last decade (2014 onward), there was an increase in the proportion of trials reporting race (46.2%) compared with before 2014 (19.5%). The largest increase was in the proportion of trials reporting White patients (from 19.5% to 46.2%), with a modest increase in the reporting of Asian/Pacific Islander patients (from 14.3% to 23.0%); there was little difference in the reporting of African American patients (16.9% before 2014 vs 15.4% after 2014), as indicated in Table 1 . In terms of enrollment proportion, there was an increase over time in the proportion of Asian and Pacific Islander participants (0.9% in 1990s to 15.7% in 2010s) and an increase in the proportion of Hispanic participants (3% in 1990s to 9.3% in 2010s; Figure 2 ). When comparing phase 2 and phase 3 trials, we found that phase 3 trials had more reporting of race and ethnicity both before and after 2014, and there was an overall increase in reporting in phase 3 trials after 2014 compared with before 2014 ( Table 4 ).

In the subgroup analysis stratifying trial proportions by trial location, race data were reported in trials performed in Canada and the United States (4 trials), trials performed in the United States only (4 trials), and trials performed in multiple continents (13 trials; Figure 3 ). The proportion of Black, White, and Hispanic patients was similar in trials performed in Canada and the United States, the United States only, and multiple continents; however, there was a higher proportion of Asian and Pacific Islander patients in trials performed in multiple continents (14.7%) than Canada and the United States (4.1%) and the United States only (1.3%). There was a slightly higher proportion of Native American/Alaska Native patients in trials performed in the United States (0.8%) than in multiple continents (0.19%). When we assessed the countries included in trials performed in multiple continents, we found that most included countries in Europe and North America (88% and 97%, respectively), whereas 39% included countries in Asia, 36% in Australia/New Zealand, and 24% the Middle East ( Table 5 ).

Of the 14 US-only trials, 4 (28.6%) reported on race or ethnicity, and 4 (28.6%) reported enrollment proportion of older adults. The random-effects meta-analysis of EIRs, performed using the 4 US trials reporting on race and ethnicity, showed that Hispanic patients (EIR, 0.28; 95% confidence interval [CI], 0.15-0.50; I 2 = 46.9%) and Asian patients (EIR, 0.16; 95% CI, 0.09-0.28; I 2 = 0%) were significantly underrepresented, whereas White patients (EIR, 1.23; 95% CI, 1.13-1.35; I 2 = 0%) were significantly overrepresented. Although Black patients were represented (EIR, 0.95; 95% CI, 0.43-2.09; I 2 = 87.4%), the CIs were wide, demonstrating significant variability in their enrollment in the 4 US trials ( Table 6 ; Figure 4 ).

Discussion

Main findings

Our systematic review revealed that the majority of published RCTs assessing disease-modifying therapies for patients with AML did not report data on race and ethnicity and on the enrollment proportion of participants with clearly dichotomized age. Within trials in which race and ethnicity were reported, only a small proportion of participants enrolled were from racial and ethnic minority groups. This, unfortunately, was not unexpected. Our meta-analysis showed that Hispanic and Asian patients were significantly underrepresented in RCTs, whereas White patients were overrepresented when considering US AML population statistics. We did, however, find that, in more recent years, there was an increase in both the reporting of race and ethnicity data and the proportion of participants from racial and ethnic minority groups. We found higher rates of reporting on race and ethnicity in phase 3 trials than phase 2 trials. We also found that in the reporting of age, although only 29% of trials reported an enrollment proportion of participants aged ≥65 years, there was an increase in the use of clear age cutoffs in the last 10 years.

Strengths and limitations

A strength of our study was that we performed a comprehensive and broad search, identifying relevant randomized trials from inception through June 2023, resulting in the inclusion of 90 full-text articles. For a list of the 90 randomized controlled trials, see supplemental Table 2 . We also performed a comparative meta-analysis with population-based data to demonstrate the significant differences in the enrollment populations and the US populations, assessing for both race and ethnicity as well as age, which are important prognostic factors in patients with AML. The use of EIRs allowed us to capture disparities in the enrollment rates of each racial or ethnic subgroup compared with the US AML population, thereby highlighting the overrepresentation or underrepresentation of racial and ethnic subgroups in the existing literature.

Limitations included the fact that few randomized trials included race or ethnicity data. In fact, only 4 trials performed in the United States included race and ethnicity data. Although the EIR for Black participants is close to 1.0, CIs for this EIR (95% CI, 0.43-2.09) were very wide. Another limitation is the publication bias of the studies included in the systematic review, in which more studies conducted in countries with a majority White population would be published.

Comparison with other studies

To our knowledge, our study is the first systematic review of racial and ethnic minority and age enrollment disparities in RCTs of patients with AML. In a previous analysis by Loree et al, 6 63% of 230 trials that led to US Food and Drug Administration oncology drug approval reported on at least 1 race, and 7.8% reported on all 4 races (White, Asian, Black, and Hispanic); although Black (EIR, 0.22) and Hispanic patients (EIR, 0.44) were underrepresented, White (EIR, 0.98), and Asian patients (EIR, 4.38) were overrepresented. This study included international trials and compared them with US-based population data to calculate the EIRs, possibly explaining why Asian patients were overrepresented in Loree et al analysis. Our study was restricted to AML, whereas in their analysis 63% of trials were on solid tumor malignancies, and only 32% were on hematologic neoplasms. Our finding of Black patients not being underrepresented was likely due to the low number of trials in our analysis.

Another study by Hantel et al, 7 based on searches within ClinicalTrials.gov alone, assessed reporting and enrollment data for ALL and AML trials in the United States performed between 2002 and 2017. They found that in AML trials, 44% of trials reported race, and 30% reported ethnicity, which was slightly higher than our study, which found that 23% of trials reported on race/ethnicity. This could be explained by the fact that our study included trials from earlier decades in which race/ethnicity reporting was not common, and we included international trials rather than those performed in the United States only. When Hantel et al 7 stratified by year (2005-2008 compared with 2011-2017), they found a significant increase over time in reporting race in AML trials (39%-57%). Our stratification by the last 10 years similarly demonstrated an increase from 19.5% to 46.2%. Significantly lower enrollment proportions of NH-Black, NH–Native American, NH-Asian, and Hispanic patients compared with NH-White patients were noted in their study, as well as significantly lower odds of enrollment of all race categories than White patients (with the exception of NH-Black patients with ALL enrolled in pediatric trials). This was similar to the findings of our analysis. Their study also collected age data and found that only 105 of 223 AML trials (47%) and 63 of 97 ALL trials (65%) reported age.

A retrospective analysis by Abraham et al 8 to assess patient characteristics, treatment patterns, and outcomes in 106 patients with intermediate-risk AML reported an increased rate of clinical trial enrollment in NH-White patients compared with other groups: 20 NH-White patients were treated on a clinical trial protocol vs only 9 patients of all other racial and ethnic subgroups (NH-Black, Hispanic, and Other; P = .041).

Explanation of findings

One possible explanation for poor enrollment equality is that restrictive eligibility requirements for clinical trials and for specific treatments such as stem cell transplantation could prevent non-White patient groups from accessing enrollment in clinical trials. In a study analyzing rates and reasons for trial ineligibility by race and ethnicity in AML clinical trials, non-White patients were more likely to be excluded from clinical trials (26.7% White patients, 29.4% Black patients, and 35.9% Asian patients). 15 The reason for ineligibility in Black and Asian participants was most frequently “lack of relevant disease mutation.” 15

A study assessing eligibility for stem cell transplantation, intense chemotherapy treatment, or treatment with best supportive care in TP53-mutated patients with AML was stratified by race. 16 They found that a higher proportion of NH-White patients than NH-Black patients received intensive chemotherapy, including allogeneic hematopoietic stem cell transplantation, whereas a higher proportion of NH-Black patients than NH-White patients were not treated with induction chemotherapy but were treated with best supportive care only. 16

Similarly, another study found that in patients with AML, although time to referral and proportion of patients receiving intensive chemotherapy did not differ between Black and White patients, there were differences in clinical trial participation (43% in Black patients and 54% in White patients) and hematopoietic stem cell transplantation (17% in Black patients vs 35% in White patients). 17

Another possible explanation for poor enrollment equality is geographic discrimination. Many clinical trials are conducted out of larger, more specialized hospitals rather than rural or community settings. This geographic selection bias may limit members of certain racial groups from enrolling in clinical trials.

A challenge in explaining the limited reporting on race, ethnicity, and age in published trials is the lack of information about their enrollment processes; in our study, few trials provided information about the process of participant enrollment. As proposed by Casey et al, 18 if reporting practices and requirements were revised to mandate complete reporting of participant demographic characteristics, this could motivate investigators to attempt to enroll participants from a greater variety of racial groups.

Our finding of increased reporting of enrollment race and ethnicity in trials performed in the most recent 10 years can likely be explained by clinical trial regulations made in the late 2000s to improve the reporting of demographic data. The increased reporting of race and ethnicity in phase 3 trials compared with phase 2 trials is uncertain, but it may be because phase 3 trials, in addition to determining efficacy, seek generalizability of interventions, so reporting of race and ethnicity may be more valued in these trials. Section 801 of the Food and Drug Administration Amendments Act of 2007 (FDAAA801) required the reporting of demographic enrollment data to ClinicalTrials.gov within 12 months of study completion, with financial consequences, and updates were made to the Declaration of Helsinki that considered public registration of trials and disclosure of enrolled demographic characteristics obligatory research practices. 19

However, although the FDAAA801 requires the reporting of demographic enrollment data, including race and ethnicity, obtained during the trial, it does not mandate the collection of race and ethnicity data; if they are not collected, investigators are simply required to indicate such information was not collected. 19

In a study assessing whether race and race-ethnicity reporting increased following the implementation of FDAAA801 and the Declaration of Helsinki, Hantel et al 7 found that trials reporting race and race-ethnicity did increase (only race significantly). Notably, the relative enrollment of NH-Black and Hispanic patients decreased after the new requirements, suggesting that these populations are actually less represented than previously thought and increased reporting has revealed worse disparities in enrollment. Our study demonstrated a decrease in the proportion of Black participants in the 1990s compared with 2010s (from 7% to 4%), as shown in Figure 2 ; however, there was an increase in the proportion of Hispanic patients (from 3% to 9%) and Asian participants (from 2% to 16%), which differs from the findings of Hantel et al. 7 This could possibly be explained by our inclusion of trials from the 1990s, whereas Hantel et al 7 included trials from 2002 to 2017. It is unclear what factors may have been different in these decades that contributed to the fluctuations in enrollment. Although the US federal requirements were legislated in the late 2000s, it is possible that there was already a growing awareness of enrollment discrepancies, and investigators may have tried to ameliorate this, leading to the increase in Hispanic participants; the subsequent introduction of legislation may have then unmasked worse disparities. Overall, our comparison of the 1990s with 2010s demonstrates that there was a net improvement over 20 years.

It appears that stricter requirements necessitating the collection of racial and ethnicity data may be an effective strategy to improve enrollment reporting. It should be noted that the reporting of racial and ethnicity data is not equivalent to the representative enrollment of participants.

In our analysis, the proportion of trials reporting on age by using an age cutoff of 65 years was similar before and after 2014; however, we did find a substantial increase in the proportion of trials reporting on age by using a different cutoff, demonstrating an increased awareness of standard age reporting methods in RCTs. It is possible that this change is attributable to the clinical trial regulations made in the late 2000s. There were limited trials in the existing literature that analyzed age enrollment. A meta-analysis of prostate cancer clinical trials found that only 26.2% of the trials included (75/286) reported patient numbers by prespecified age categories and found an increase over the years from 1995 to 2020 of the trials reporting age (from just under 10% of trials to 40% of trials). 20 We found no leukemia trials that outlined or reported age enrollment.

Conclusion

Most trials did not report data on race and ethnicity or on enrollment proportion of participants aged ≥65 years. In those trials that did report data on race and ethnicity, only a small proportion of trial participants were from racial and ethnic minority groups. Hispanic and Asian patients were significantly underrepresented, whereas White patients were overrepresented. There has been an increase in the last 10 years in the proportion of trials reporting on race and an overall increase in the last 20 years on the enrollment proportions of participants from racial and ethnic minorities (specifically in the Asian and Pacific Islander and Hispanic populations). We found that phase 3 trials were more likely to report on race and ethnicity than phase 2 trials. Finally, most trials did not report on enrollment proportion of participants aged ≥65 years; however, there was an increase in the proportion of trials in the last 10 years of trials reporting on age using a different age cutoff. Stricter requirements that necessitate the collection of data on race and ethnicity should be pursued. Efforts should continue to include patients from underrepresented minority groups in clinical trials to prevent racial disparities in research.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Acknowledgments

Authorship

Contribution: N.L. and A.G.-H. designed the study and drafted the manuscript; N.L., O.K., A.J., P.T., and D.T. screened titles, abstracts, and full texts; N.L. and O.K. extracted data; and N.L. performed data analysis.

Footnotes

Presented in abstract form at the 65th annual meeting of the American Society of Hematology, San Diego, CA, 10 December 2023.

The data that support the findings of this study are available on reasonable request from the corresponding author, Nathalie Loeb ( nathalie.loeb@medportal.ca ).

The full-text version of this article contains a data supplement.

Supplementary Material

Supplemental Table 1

Supplemental Table 2

References

1. Kirtane K, Lee SJ. Racial and ethnic disparities in hematologic malignancies. Blood. 2017;130(15):1699–1705. doi: 10.1182/blood-2017-04-778225.

2. Patel MI, Ma Y, Mitchell BS, Rhoads KF. Understanding disparities in leukemia: a national study. Cancer Causes Control. 2012;23(11):1831–1837. doi: 10.1007/s10552-012-0062-3.

3. Fiala MA, Wildes TM. Racial disparities in treatment use for multiple myeloma. Cancer. 2017;123(9):1590–1596. doi: 10.1002/cncr.30526.

4. Baker KS, Davies SM, Majhail NS, et al. Race and socioeconomic status influence outcomes of unrelated donor hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2009;15(12):1543–1554. doi: 10.1016/j.bbmt.2009.07.023.

5. Baker KS, Loberiza FR, Jr., Yu H, et al. Outcome of ethnic minorities with acute or chronic leukemia treated with hematopoietic stem-cell transplantation in the United States. J Clin Oncol. 2005;23(28):7032–7042. doi: 10.1200/JCO.2005.01.7269.

6. Loree JM, Anand S, Dasari A, et al. Disparity of race reporting and representation in clinical trials leading to cancer drug approvals from 2008 to 2018. JAMA Oncol. 2019;5(10) doi: 10.1001/jamaoncol.2019.1870.

7. Hantel A, Luskin MR, Garcia JS, Stock W, DeAngelo DJ, Abel GA. Racial and ethnic enrollment disparities and demographic reporting requirements in acute leukemia clinical trials. Blood Adv. 2021;5(21):4352–4360. doi: 10.1182/bloodadvances.2021005148.

8. Abraham IE, Patel AA, Wang H, et al. Impact of race on outcomes in intermediate-risk acute myeloid leukemia. Cancer Causes Control. 2021;32(7):705–712. doi: 10.1007/s10552-021-01422-4.

9. Murthy VH, Krumholz HM, Gross CP. Participation in cancer clinical trials: race-sex-and age-based disparities. JAMA. 2004;291(22):2720–2726. doi: 10.1001/jama.291.22.2720.

10. Pulte D, Jansen L, Castro FA, Brenner H. Changes in the survival of older patients with hematologic malignancies in the early 21st century. Cancer. 2016;122(13):2031–2040. doi: 10.1002/cncr.30003.

11. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71.

12. National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) Program. SEER∗stat database: incidence—SEER 13 regs research data. https://seer.cancer.gov/ Accessed 5 July 2023.

13. National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) Program. SEER∗stat database: incidence—SEER 21 regs research data. https://seer.cancer.gov/ Accessed 5 July 2023.

14. Riaz IB, Islam M, Ikram W, et al. Disparities in the inclusion of racial and ethnic minority groups and older adults in prostate cancer clinical trials: a meta-analysis. JAMA Oncol. 2023;9(2):180–187. doi: 10.1001/jamaoncol.2022.5511.

15. Pulte D, Fernandes L, Wei G, et al. FDA analysis of ineligibility for acute myeloid leukemia clinical trials by race and ethnicity. Clin Lymphoma Myeloma Leuk. 2023;23(6):463–470.e1. doi: 10.1016/j.clml.2023.03.012.

16. Badar T, Litzow MR, Shallis RM, et al. Disparities in receiving disease-directed therapy, allogeneic stem cell transplantation in non-Hispanic Black patients with TP53-mutated acute myeloid leukemia. Cancer. 2023;129(6):934–945. doi: 10.1002/cncr.34604.

17. Bierenbaum J, Davidoff AJ, Ning Y, Tidwell ML, Gojo I, Baer MR. Racial differences in presentation, referral and treatment patterns and survival in adult patients with acute myeloid leukemia: a single-institution experience. Leuk Res. 2012;36(2):140–145. doi: 10.1016/j.leukres.2011.10.018.

18. Casey M, Odhiambo L, Aggarwal N, Shoukier M, Islam KM, Cortes J. Representation of the population in need for pivotal clinical trials in lymphomas. Blood. 2023;142(9):846–855. doi: 10.1182/blood.2023020052.

19. National Institutes of Health, Department of Health and Human Services Clinical trials registration and results information submission. https://www.federalregister.gov/documents/2016/09/21/2016-22129/clinical-trials-registration-and-results-information-submission#sectno-reference-%E2%80%8911.42

20. Riaz IB, Islam M, Ikram W, et al. Disparities in the inclusion of racial and ethnic minority groups and older adults in prostate cancer clinical trials: a meta-analysis. JAMA Oncol. 2023;9(2):180–187. doi: 10.1001/jamaoncol.2022.5511.

21. Covidence systematic review software Veritas Health Innovation, Melbourne, Australia. www.covidence.org Available at.

Associated Data

Supplementary Materials

Supplemental Table 1

Supplemental Table 2