Subscribe to eTOC

Why Should Clinical Trials Include More Diverse Populations?
Consider These Data from Recent Alzheimer's Disease Trials

Article In Brief

Two recent studies showed that the neuropathology of Alzheimer's disease and outcomes were different among black and Hispanic patients compared with whites. These differences underscore the need to ensure greater inclusion of heterogeneous populations in clinical trials, several trialists agreed.

Most of what researchers know about Alzheimer's disease (AD) has been based on studies of mostly white populations, one leading researcher told Neurology Today, in response to the question: Should clinical studies be required to include more heterogeneous populations?

“This can lead to incorrectly assuming that the mechanisms of dementia in non-Hispanic whites also apply to different ethnic groups,” said Dan Mungas, PhD, professor of neurology at the UC-Davis School of Medicine in Sacramento. “Research with diverse groups is needed to test these assumptions.”

Two papers published online in February in two different Alzheimer's disease journals offer that case in point. In one, a retrospective study of the Florida Autopsied Multi-Ethnic (FLAME) cohort published in the February 18 online edition of Alzheimer's Disease and Dementia, investigators reported that Hispanics had an earlier onset and longer duration of Alzheimer's disease than either African-Americans or non-Hispanic whites.

In the other, published online February 11 in the Journal of Alzheimer's Disease, Hispanic and African-American decedents were more likely than non-Hispanic whites with dementia to show neuropathology suggestive of cerebrovascular disease as well as Alzheimer's disease brain changes. The study included 35 deceased African-Americans, 28 Hispanics, and 360 non-Hispanic whites.

The FLAME Cohort

“Because Hispanics and African-Americans are at greater risk of developing the disease (1.5 times and two times respectively) than non-Hispanic whites, we were surprised that Hispanics appear to withstand the disease longer than both non-Hispanic whites and African Americans,” said the lead investigator of the FLAME cohort study, Melissa Murray, PhD, assistant professor of neuroscience at Mayo Clinic in Jacksonville, FL.

This raises the question of whether Hispanics have protective factors that enable them to withstand the disease longer, such as unknown neurobiologic factors or enhanced socialization within tight family networks, said Dr. Murray.

Dr. Murray was cautious about interpreting the results, given that only 19 African-Americans and 67 Hispanics participated in the study compared with 1539 whites. “It will be important moving forward to continue to investigate topographic differences in neurodegenerative lesions to learn how these may differentially influence clinical presentation.”

What's the Value?

Referring to the postmortem study conducted by investigators at the UC Davis, Dr. Mungas, a co-investigator, said: “The increased prevalence of vascular co-morbidity may be an opportunity for interventions that improve vascular health in Hispanics and African-Americans across the lifespan.”

“Although these post-mortem studies were limited by small sample sizes of Hispanics and African-Americans and by convenience sampling, they represent an important step toward understanding the similarities and differences in the neuropathology of dementia in patients from different ethnoracial backgrounds,” said Dr. Mungas.

For example, he said, studies have shown that apolipoprotein E4 (APOE4), which is the strongest genetic risk factor for Alzheimer's, may not carry the same risk for the disease in Hispanics as it does in whites.

Indeed, “population-based research comparing Mexican-Americans to whites with Alzheimer's in north Texas, showed that they had a lower presence of APOE4 but an earlier onset of Alzheimer's disease and mild cognitive impairment,” said Sid O'Bryant, PhD, professor of pharmacology and neuroscience and executive director at the Institute for Translational Research at the University of North Texas Health Sciences Center in Fort Worth. Dr. O'Bryant was a coauthor of the study published in 2013 in The Journal of Alzheimer's Disease.

“How is that possible if the primary genetic risk factor is less common? This underscores that the presence of APOE4 varies by racial groups. If we don't know the biological risk factors for Alzheimer's across diverse groups, then we can't create appropriate therapies,” said Dr. O'Bryant.

Dr. O'Bryant is the lead investigator of a five-year National Institute on Aging-funded study that is examining how biological mechanisms are impacted by health disparities—such as higher rates of depression and diabetes—among Mexican-Americans. Previous research has shown that Mexican-Americans have a more metabolically driven profile than whites, which may be related to diabetes. “We know that diabetes is a risk factor for late-life Alzheimer's and may be a more prominent driver of memory loss during the aging process among Mexican-Americans,” said Dr. O'Bryant.

“The value of inclusion is it will enable us to understand more broadly how this disease impacts populations believed to be at greater risk,” said Lisa L. Barnes, PhD, the Alla V. and Solomon Jesmer Professor of Gerontology and Geriatric Medicine at the Rush Alzheimer's Disease Center in Chicago.

Otherwise, it's possible to draw the wrong conclusions about causes and treatment, said Dr. Barnes.

For example, her 2015 study in Neurology found that African-Americans were much more likely to have mixed disease pathology than whites, which she explained can include infarcts, Lewy bodies and TDP-43, in addition to plaques and tangles.

However, because African-Americans tend to have more vascular disease, “we expected to see more infarcts mixed with Alzheimer's changes. Instead, we found more Alzheimer's pathology mixed with Lewy bodies, the pathology associated with Lewy Body disease,” said Dr. Barnes.

From the diagnostic perspective, it is urgent to test the framework for diagnosing Alzheimer's disease based solely on measures of plaques and tangles, William T. Hu, MD, PhD, associate professor of neurology at Emory University School of Medicine in Atlanta, told Neurology Today.

“In our work, we found differences in key spinal fluid Alzheimer's biomarkers between African-Americans and whites that could lead to the disease being under-diagnosed in African-Americans. We need more studies involving diverse populations to understand how this standard model of Alzheimer's applies to a broader population,” said Dr. Hu.

Bias in Cognition Research

Dr. Mungas who has been investigating how to effectively measure cognitive functioning in the elderly for the past 30 years, said, “There is a concern among clinicians and researchers that neuropsychological tests used to evaluate people for dementia might be biased against certain racial or ethnic groups and that this can lead to misdiagnosing some people with dementia.”

Although whites tend to perform better than other ethnic groups on cognitive tests it doesn't mean they have healthier brains. “Brain variables account for at most 50 percent of the variance in test scores. Other factors such as educational and life experiences contribute to test scores. The challenge is to separate effects of the brain diseases on test results from the non-brain effects. This can be done in a single cross-sectional evaluation by employing good normative data that allows the neuropsychologist to compare the observed test scores with what would be expected based upon the demographic features of each individual,” said Dr. Mungas.

In addition, repeated assessment over time is especially important because cognitive decline is much more closely tied to brain changes than to demographic characteristics. To optimally measure cognitive decline in ethnoracially diverse older adults, Dr. Mungas and his colleagues have developed a battery of cognitive tests for use in the three largest ethnic groups: whites, Hispanics and African-Americans. These tests are both in English and Spanish. “We did a lot of work to ensure that these tests measure cognition the same way in these different populations and have carefully validated them against clinical criteria and brain changes,” said Dr. Mungas

Dr. O'Bryant has examined bias in commonly used neuropsychological assessments mainly among Mexican-Americans. “We, and others, have found in numerous studies that if you use the wrong instruments or normative references, you will get biased results and assign inaccurate diagnoses.”

To correct that, Dr. O'Bryant and his colleagues have developed cultural and linguistically appropriate normative references for Mexican-Americans. In the Texas Mexican American Normative Studies, his group generated education- and age-adjusted normative scores for commonly utilized neuropsychological instruments, such as the Repeatable Battery for the Assessment of Neuropsychological Status and Wechsler Memory Scale Logical Memory.

“This approach allows us to compare our findings to those of others rather than generating new tests that are not included in other studies.”

Requiring Diverse Research Samples

The National Institutes of Health policy for awarding grants states that “clinical research should address and/or include the populations at risk for the disease or condition under study and ensure that the distribution of study participants by sex/gender, race and ethnicity reflects the population needed to accomplish the scientific goals of the study.”

“I don't think this goal has been fully realized, however, due to the amount of effort and funds needed to recruit the samples,” Dr. Barnes said. “To advance the field, I think that investigators with access to diverse communities should be required to include representative samples in their research and more money and effort should go into ensuring this.”

She encouraged researchers to think of diversity beyond solely race and ethnicity to include people from different socioeconomic levels, rural areas, and LGBTQ communities.

“This is imperative for public health planning, especially as the older population is growing and our nation is becoming much more diverse as far as race, ethnicity, geography, and lifestyle.”

Other researchers noted that it may be impractical to recruit minority participants if scientists live in racially homogenous areas. “I prefer guidance about what is needed and how to get there. For example, providing resources and collaboration opportunities with scientists who are recruiting diverse ethnic groups from their communities instead of neurology clinics, which is current practice,” said Dr. O'Bryant.

He also advocated for greater accountability in peer-reviewed grants and journal articles. “In research papers, there should always be a description of the racial/ethnic composition of the population studied in addition to gender, age, and other demographics.”

Moreover, institutions with peer-reviewed grants should be stricter about requiring investigators to recruit minorities, especially upon grant renewal. “If they have been notified and given recruitment guidance and still not actively recruited minorities, their grant should be terminated,” said Dr. O'Bryant.


“To advance the field, I think that investigators with access to diverse communities should be required to include representative samples in their research and more money and effort should go into ensuring this.”


Future Directions

Despite rapid progress, there are still a lot of unanswered questions as to why minority populations appear to be at greater risk, Dr. Barnes. Among those questions, she said, are these: Are biomarkers for the disease the same across different populations? Do risk factors operate the same across different populations? And are the pathologic changes associated with the disease the same or different across populations?

It will also be important to understand how the pathologic markers of the disease develop in people without the clinical signs of dementia, “like we do in our studies where we follow the elderly (who initially enroll without dementia) over time until death and we measure genetics, lifestyle factors, and other characteristics while they are living and then correlate those findings with what we see in their brains postmortem,” said Dr. Barnes.

Dr. Mungas and his colleagues at UC-Davis are involved with the longitudinal LIFE AFTER 90 study, a collaboration with Kaiser Permanent Northern California, with participants from diverse ethnoracial groups including whites, Asians, African-Americans, and Hispanics. “We're hoping to obtain brain donations so that we can directly compare results among these different ethnoracial populations in northern California.”

He noted that this study will have relatively large numbers of autopsies from individuals from different ethnoracial groups in comparison with studies to date, and all cases will be drawn from a defined population of older individuals who receive their health care through Kaiser Permanente.

“Prospective longitudinal studies are occurring in different regions of US, but the challenge is applying a deep dive (i.e., more detailed biomarker analysis than just clinical diagnosis) reflecting the current biological understanding of Alzheimer's disease to all these cohorts,” said Dr. Hu.

“For example, conducting research with serial non-standard MRIs or fluid biomarker measurements with matching genetic information in large pools of people is not easy to operationalize. However, we need to first have small cohorts for the deep dive and then a larger cohort to validate promising findings from the smaller cohorts,” said Dr. Hu.


Drs. Mungas, Murray, Barnes, and Hu had no competing interests.

Link Up for More Information

•. Santos OA, Pedraza O, Lucas JA, et al. Ethnoracial differences in Alzheimer's disease from the Florida Autopsied Multi-Ethnic (FLAME) cohort Alzheimers Dement 2019; Epub 2019 Feb 18.
    •. Filshtein TJ, Dugger BN, Jin LW, et al. Neuropathological diagnoses of demented Hispanic, Black, and non-Hispanic white decedents seen at an Alzheimer's Disease Center J Alzheimers Dis 2019; Epub 2019 Feb 11.
      •. O'Bryant SE, Johnson L, Balldin V, et al. Characterization of Mexican Americans with mild cognitive impairment and Alzheimer's disease J Alzheimers Dis 2013;33(2):373–379.
      •. Barnes LL, Leurgans S, Aggarwal NT, et al. Mixed pathology is more likely in black than white decedents with Alzheimer dementia; Neurology 2015; 85(6):528–534.
      •. Howell JC, Watts KD, Parker MW, et al. Race modifies the relationship between cognition and Alzheimer's disease cerebrospinal fluid biomarkers Alzheimers Res Ther 2017; 9(1): 88.
        •. National Institutes of Health Office of Extramural Research Grants and Funding. Inclusion of Women and Minorities as Participants in Research Involving Human Subjects - Policy Implementation Page.