In order to accurately detect and prevent racial disparities, self-reported race (SRR) and ethnicity remain valuable tools; however, inaccurate capture of patient identity and broad aggregation of minoritized race groups present challenges for data interpretation. Also, although SRR is a proxy for shared social/cultural experience, it is not an accurate representation of shared endogenous factors. Biological investigations into cancer disparities, particularly those involving genetic features, should be framed in the context of genetic background or ancestry, as these are heritable aspects of population health. In reality, both genetics and environment work in concert to influence cancer risk and clinical outcomes. The best opportunity to define actionable means for reducing health disparities is in rigorous and comprehensive generation of rich data sets that characterize environmental, biological, and genetic components of disparate disease burden. To translate this pivotal disparities research into clinical tools and improved policies, we describe a diversity, equity, inclusion, and accessibility (DEIA) framework, which will increase participation from diverse backgrounds, reexamine previous research with a rigorous evaluation of appropriate SRR groupings, and engage community leaders to ensure that future research addresses the needs of communities at increased risk. On this path forward, we may finally end cancer disparities.

As we continue to make pivotal strides in cancer disparities awareness, the translation of the collective work is still impeded by barriers that limit impact. From the underrepresentation of minorities to really knowing how to define who they are, the next steps in our journey will define how quickly we can make progress to end racial disparities. The next best step is prioritizing rigor and standardized best practices. All cancer research should be done with the goal of addressing a diverse patient population and assessing the appropriate context of this ethnic diversity. We begin with respecting who patients are, and from there, we will be on the right track toward ending disparities.

Revealing and characterizing racial disparities in cancer incidence and outcomes is the first step in overcoming these unnecessary biases. Reports that periodically surmise these biases, such as the American Association for Cancer Research (AACR) reports on cancer health disparities, thoroughly outline the various origins of racial disparities. Such research relies first and foremost upon the disclosure of patient identity, which is mainly captured as social constructs, such as self-reported race/ethnicity (SRR). The instruments that capture these data are typically clinic intake forms or patient surveys at best, but at their worst may also be imputed from natural language processing of medical records or provider-reported clinical chart notes (Fig. 1). In either case, there are several limiting factors in utilizing SRR annotations for statistical models, including creation of categories that are aggregate groupings to fortify categorical numbers and allowances of only one selection. Consequently, real-world SRR data can generally be only a vague representation of a patient's true ethnic/cultural identity (1, 2). When given unacceptable groups to choose from, patients will either omit SRR or choose what encompasses all the remaining unlisted choices—“other” (2, 3). The use of this category in disparities research actually propagates the implicit bias issue of othering, where individuals who do not identify as part of the so-called majority groups are relegated to an ill-defined catchall category of “other” (4). To avoid the othering of peoples/populations with mixed or underrepresented SRR, we should expand choices and allow patients to either select multiple groups or write in their own ethnic/cultural identity to reflect their family and social history. Notably, the mere action of making this selection of identity and the consideration clinics give toward providing appropriate choices are a reflection of the attitudes toward diversity inclusion and can affect patients’ perceptions of the care being given (5).

Figure 1.

Collection of SRR and ethnicity data in the U.S. Census. Created with BioRender. Information from “What Census Calls Us.” Pew Research Center, Washington, DC (Feb 6, 2020). Available from: https://www.pewresearch.org/interactives/what-census-calls-us/.

Figure 1.

Collection of SRR and ethnicity data in the U.S. Census. Created with BioRender. Information from “What Census Calls Us.” Pew Research Center, Washington, DC (Feb 6, 2020). Available from: https://www.pewresearch.org/interactives/what-census-calls-us/.

Close modal

From a research perspective, studies that investigate causes of racial disparities often rely on SRR groupings with multiple race and ethnicity labels. Inconsistent practices of SRR data collection and coding require best practices to (i) standardize race and Hispanic ethnicity coding in data models and (ii) harmonize multiple SRR identities for individuals of mixed backgrounds. Stratifying all indicated SRR categories usually means dividing the already underrepresented numbers (6), resulting in dilution of statistical power in analysis models. To mitigate power limitations, the categories may be combined into more generalized groups with the good intention of creating rigorous statistical models. Unfortunately, the reality of the resulting models of aggregated groups is that they do not accurately reflect the patient diversity, and effectively these miscategorizations could lead to skewed interpretation of the data, even masking distinct risk factors within certain groups (2, 3).

Therefore, we emphasize not only inclusion of diverse patient populations but also population disaggregation with two key examples to highlight the importance. Native Hawaiians (NH) and other Pacific Islanders (PI) have historically been aggregated with Asian Americans (AsAm) as an AsAm/NHPI SRR grouping, despite the geographic, anthropologic, and genetic distinctions among them all. This aggregation was a consequence of othering when the representation of any given group was too low to utilize in population statistics or could be used to deduce who deidentified patients were and reidentify them. Recent reevaluation of the disaggregated groups reveals worse breast cancer survival outcomes among NHPI compared with AsAm (7). This new perspective of disaggregation also reveals increased risk of other chronic conditions, such as diabetes and obesity among NH, which are known to affect cancer risk, broadly (8). Additionally, disaggregation of U.S. Hispanic populations revealed that, among all Hispanics, Puerto Ricans have the highest overall cancer mortality rates, Mexican males have higher stomach cancer mortality, Cuban males have higher lung cancer mortality, and Cuban females have higher breast cancer mortality (9). Similarly, half of the world's gastric cancer cases and the highest gastric cancer mortality rates are observed in Eastern Asia (10). Varying levels of self-reported hypertension have also been shown among disaggregated Hispanic populations in the United States from National Health and Nutrition Examination Survey (NHANES) data (11). Additionally, evidence to the role of disparity in African populations can be obtained in a number of studies, and one such work is described in our recent publication on molecular characterization of breast cancer (12). These differences that correlate with genetic ancestry could also be a reflection of African ancestry (13) or other diasporas sharing genetic predispositions, which would be confounded if patient groups were not disaggregated. In fact, prior to these studies, there was a reported decreased risk of each of these cancers among AsAm and Hispanics compared with white or European American individuals.

This also reemphasizes the importance of inclusion of diverse populations (14) in studies of cancer and other diseases, globally, in order to capture the broad range of ancestry in SRR groups.

In well-defined experiments, researchers carefully select their tools and variables to best answer their research questions to test their hypotheses. The same care should be taken to determine when it is most appropriate to utilize patient SRR versus using ancestry-based annotations. Beyond general risk statistics of incidence and mortality rates of cancer lies disease plasticity in tumor biology and distinct patho­logy. Differing incidence of tumor phenotypes among SRR groups correlates with unique population disease burden and implicates biological mechanisms that are altered within these groups. Although patient SRR plays an important role in first revealing population-level differences, SRR should not be used as the sole instrument/proxy for understanding shared biology (1, 4). Social determinants of health (SDOH) are linked to racial constructs that are the scaffold for systemic racism and classism. Studies that characterize the impact of social determinants are richly benefited by use of SRR to frame the study around a patient's identity and lived experiences. However, biological determinants that could be driven by genetic factors require data that capture shared genetic traits among patient groups, including quantified genetic ancestry (QGA). Evaluating the shared genetic admixture across SRR groups resolves the ambiguity of othering and/or missingness of SRR and, most importantly, allows us to identify genetic modifications that are relevant to the shared tumor biology among patient groups. Genetic admixture can range vastly from individual to individual, particularly in populations across the African and Latinx diaspora. For example, Black or African American (AfAm) individuals show broad range of African and European ancestry, as well as tremendous heterogeneity of African origin, which has been shown to differ regionally (15, 16), correlated with contemporary migration and social/anthropologic histories (Fig. 2AC). Recent studies that utilize ancestry, rather than SRR, still resort to a single categorization of individuals into generalized groupings of African, European, South or East Asian, and American Native (AMR) or admixed (17).

Figure 2.

Incorporating SRR and ancestry in multiomic approaches in disparities research. A, Regional African ancestry estimations using ancestral informative markers show distinct proportions of West or East African ancestry among African Americans, East Africans, and West Africans in the International Breast Registry (data from Martini et al.; ref. 12). ESN, Esan in Nigeria; GWD, Gambian in Western Divisions in the Gambia; LWK, Luhya in Webuye, Kenya; MSL, Mende in Sierra Leone; YRI, Yoruba in Ibadan, Nigeria. B, Schematic depicting the overlap of differentially expressed genes (DEG) associated with regional African ancestry across the continent. Created with BioRender. C, Proteomics platform of imaging mass cytometry allows for the identification of distinct cell phenotypes, including immune cell phenotypes between SRR groups (unpublished data). The orange circle highlights suppressive immune cell populations, and the black circle highlights activating immune cell populations. AA, African American; EA, European American.

Figure 2.

Incorporating SRR and ancestry in multiomic approaches in disparities research. A, Regional African ancestry estimations using ancestral informative markers show distinct proportions of West or East African ancestry among African Americans, East Africans, and West Africans in the International Breast Registry (data from Martini et al.; ref. 12). ESN, Esan in Nigeria; GWD, Gambian in Western Divisions in the Gambia; LWK, Luhya in Webuye, Kenya; MSL, Mende in Sierra Leone; YRI, Yoruba in Ibadan, Nigeria. B, Schematic depicting the overlap of differentially expressed genes (DEG) associated with regional African ancestry across the continent. Created with BioRender. C, Proteomics platform of imaging mass cytometry allows for the identification of distinct cell phenotypes, including immune cell phenotypes between SRR groups (unpublished data). The orange circle highlights suppressive immune cell populations, and the black circle highlights activating immune cell populations. AA, African American; EA, European American.

Close modal

For many individuals, global ancestry categorization is straightforward, as the clear majority of their ancestry will be from a singular source. However, for individuals with significant admixture where no one ancestry region reaches a predominant majority, determining a singular ancestry group “call” will be problematic, much in the way a singular SRR group can be problematic. In a recent study utilizing ancestry across The Cancer Genome Atlas pan-cancer cohort to identify ancestry-associated gene signatures, a secondary ancestry threshold of 20% was determined to categorize patients as admixed (17); however, a lower ancestry threshold could also be relevant in ancestry-associated gene expression drivers, depending upon the genomic loci of the ancestry. For example, a patient with 62% European ancestry and 30% AMR ancestry that is designated as a single European category oversimplifies the patient's true genetic background and may confound the ability to identify a genetic driver locus that is among their AMR genetic regions. This confounding variation within the SRR groups signals a need to be consistent with how genetic ancestry is measured and utilized. Alternatively, we can create gene association models that incorporate the localized ancestry in specific genomic regions, such as Bayesian models, which will allow retention of secondary and even tertiary ancestry across admixed cohorts. This approach's impact will be similar to population disaggregation approaches that unmask risk factors of disparate outcomes that are strong enough to observe across SRR groups but have no clear molecular driver because of complex heterogeneity in genomic background/ancestry. We see the functional relevance of these details when we use QGA compared with SRR in transcriptome studies (Fig. 2B). We find that some QGA-associated gene signatures are distinct from SRR-associated gene signatures, which are enriched for biological pathways under regulation of ancestry-specific (i.e., population-private) alleles (12, 17), such as the Duffy-null allele of the DARC/ACKR1 gene (18, 19).

Aside from resolution of ambiguous SRR categories, QGA may not always correspond with the SRR of an individual (e.g., non-Hispanic white patients with 60% African ancestry; ref. 20). In these circumstances, combined tools of social and genetic annotations can be used, noting whether differences in tumor biology or gene networks are influenced by SRR or ancestry, which can discern whether biological differences are under genetic or environmental control. In the context of systemic racism translating to biological impact, having both SRR and ancestry will be key to deconvolution of risk factors of incidence, disease progression, and survival. In fact, to fully characterize all plausible factors that influence disparate outcomes among diverse ethnic populations, integration or convergence of tools measuring biological and social determinants is imperative and should be considered the most rigorous approach (21). The ancestry–genetics approaches identify inherited risk factors as well as consequential somatic genomic landscapes (e.g., epigenetic modifications or structural variation) that occur in response to an interaction between genetic predisposition and environmental exposures such as allostatic load (22) or nutritional and/or neighborhood deprivation (23, 24). These all ultimately converge to translate as tumor biology that is linked to SRR groups. This convergence of social and biological factors not only gives insight into societal factors that affect cancer biology but also allows us to quantify this impact as a way to benchmark prevention and survival risk mitigation efforts. Defining when and where to use these variables requires a larger conversation, such as within working groups or special interest groups among professional organizations that build task forces to comprehensively set guidelines to develop best practices for researchers.

In the past decade, research characterizing health disparities and mitigation strategies was heavily endowed with social determinants and accessibility to equitable clinical care as its focus (Fig. 3), but has often theorized the intersection of genetic, societal, and environmental influences on disease incidence, progression, and pathology of tumors (25–27). In support of these hypotheses, studies from African cohorts show how genetic factors and population substructure could affect disease susceptibility and tumor biology through evolutionary adaptation (Fig. 2C; refs. 25, 28, 29). However, such findings are typically shown separate from social factors, though biological underpinnings are dynamically derived from both endogenous and exogenous influences and are not mutually exclusive.

Figure 3.

Increase in disparities research efforts from 2011 to 2016 to 2021. Numbers are from PubMed searches of search terms indicated above the bubbles. Publication numbers denoted with an asterisk (*) were number of publications when “racial disparities” was included in the search (in place of “disparities”). Created with BioRender. SES, socioeconomic status.

Figure 3.

Increase in disparities research efforts from 2011 to 2016 to 2021. Numbers are from PubMed searches of search terms indicated above the bubbles. Publication numbers denoted with an asterisk (*) were number of publications when “racial disparities” was included in the search (in place of “disparities”). Created with BioRender. SES, socioeconomic status.

Close modal

Now we have come to an exciting time where numerous collaboration opportunities to share cultural knowledge, genomic data, real-world data, and precision analytical tools can allow the full characterization of the impact of admixed endogenous genetic influences to fully elucidate the impact of the exogenous cultural influences. And whether defined by the geographic placement of where patients live (where they are economically marginalized to live), as well as their lifestyle and cultural choices, the exogenous cultural factors are complex and require expertise beyond epidemiology. Engaging anthropologists, social scientists, ethnologists, and other disciplines that bring expertise of distinct peoples and cultures could greatly assist with designing appropriate instruments for novel perspectives that enrich disparities research, as well as the appropriate interpretation and application of this work.

Research efforts in precision medicine have provided models for risk of recurrence and therapeutic response, which have improved cancer survival by tailoring combinatorial treatments and better monitoring during survivorship. However, patient groups with certain clinical status and/or tumor phenotypes do not benefit from the established standard-of-care treatment paradigms due to either resistance or intolerable adverse events (30). When the incidence of resistant tumor types and adverse outcomes correlates with SRR, racial disparities unfold. These disparities are not an issue of patient SDOH or treatment accessibility. Rather, because the initial studies that developed clinical tools lacked patient diversity, this translates to a gap in clinical utility. Key examples of this gap include genetic risk allele discrepancies from genome-wide association studies (GWAS), polygenic risk scores derived from GWAS, and recurrence scores (31, 32), which all fail to perform as accurately outside of the European ancestry reference and training set populations (31, 33, 34). For instance, the landmark TAILORx clinical trial investigated the utility of a gene expression–based recurrence score tool, which established clinically relevant score values, setting thresholds of tiered risk among women with hormone receptor–positive, node-negative breast cancer. The study showed that the treatment standard of adjuvant chemotherapy was not necessary for long-term recurrence-free survival in moderate to lower tier risk patients (35). However, studies of patient cohorts with more diversity than the initial trials indicate that the performance of the recurrence score shows lower accuracy among AfAm women. Given that AfAm women tend to have scores in the reported “top-tier risk,” potential implications may include overtreatment and higher occurrence of adverse events leading to poorer survival and longevity in this group than in others (36). The American Society of Clinical Oncology (ASCO) had updated its clinical guidelines regarding the use of adjuvant chemotherapy, incorporating the TAILORx trial findings (37); however, the underperformance of the recurrence score among AfAm women and increased risk of mortality in the adjuvant setting should also be taken into consideration regarding treatment for this population of women. Perhaps the variability of the risk score performance should be further investigated, possibly including additional ancestry-associated gene signatures or consideration for tumor heterogeneity, so that a more accurate assessment can be established.

Beyond differences in risk scores among ancestry groups is consideration of disparities in response to therapeutic options. Historically, clinical trials have lacked diversity in patient enrollment, and therefore ancestry-specific and/or SRR differences in therapeutic responses have not been fully explored. Specifically, ancestry-specific genetic variants that affect the function or regulation of mechanisms of drug targets or drug metabolism may lead to differences in treatment response or overall outcome. A key example is the higher occurrence of treatment-induced heart failure and peripheral neuropathy in AfAm patients linked to sensitivity of chemotherapy that is associated with African ancestry–associated gene variants (38, 39). Additionally, lack of diversity in clinical trials limits our understanding of the impact of co-occurring comorbidities that are more prevalent in certain patient groups, such as diabetes in AfAm. Significant differences among SRR groups in insulin sensitivity/response and Hb1Ac levels suggest that typical insulin regimens may not be optimized across diverse patient populations and result in divergent responses (40). Emerging evidence even suggests significant differences, including an increased benefit or better response to certain therapeutic agents, for specific SRR groups with the same cancer diagnoses. For example, a real-world data analysis of patients with advanced non–small cell lung carcinoma showed that AfAm had more favorable outcomes with anti–PD-1/PD-L1 therapy, with longer time-to-treatment discontinuation and overall survival than their white counterparts (41). However, in prostate cancer, although recent findings indicate that AfAm have increased PD-1 expression in treatment-naive prostate tumors, clinical trials of anti–PD-1/PD-L1 therapy among patients with prostate cancer have lacked significant enrollment of individuals of African ancestry, and have shown limited success in this disease, which may be due to lack of diverse inclusion rather than lack of efficacy (42, 43). This lack of diversity seems shocking given that the incidence rates of prostate cancer are higher in AfAm than in white people and that the tumor biology is significantly different between these two SRR groups (43). Interestingly, given recent findings of higher incidence of immune-suppressive phenotypes in patients of African descent across several cancer types (Fig. 2C), immune inhibitor therapy benefits in AfAm patients may be tumor-type agnostic with potential to transform several cancer disparities (44).

The challenge that remains for researchers and clinicians is how to translate diversity and equity findings into relevant clinical guidelines to optimize patient treatments and improve disparate outcomes. The translational data are steadily forthcoming, and the opportunities to implement clinical trials are increasing. So how can we ensure these studies will have more impact than previous research, with systematic strategies that address the remaining gaps and limitations? Here we outline potential steps toward a culturally appropriate and data-driven set of diversity, equity, inclusion, and accessibility (DEIA) actions (Fig. 4).

Figure 4.

Actionable measures toward DEIA. Created with BioRender.

Figure 4.

Actionable measures toward DEIA. Created with BioRender.

Close modal

Step 1: Diversity

Prospective recruitment of larger numbers to improve the diversity of data sets will assist in the limitations of small numbers that lead to generalized aggregation. Broad SRR categories that combine distinct ethnic groups may seem to be a simple manner of tracking demographics in clinical operations. However, it is clear that disaggregation of these categories will be pivotal to truly understanding risk factors that impede positive health outcomes. Therefore, with concerted efforts to accrue appropriate n for statistical power, we will be able to utilize the accurate SRR chosen by patients. In addition, we should do the due diligence of including accurate assessments of genetic backgrounds to inform genetic associations; currently, this means QGA that reaches into depths of poly-ancestry and admixture.

Step 2: Equity

To truly reach equity, we must evaluate the inequity present across all underrepresented groups. Reevaluation of studies with race–group comparisons, utilizing the disaggregated patient groups, will uncover the reality of inequity to empower mitigation efforts. In fact, disaggregation studies emerging in the past year are in effect the correction of an error, as these aggregated patient groups have always identified as being distinct from one another, and of note, investigators who are members of these groups are leading these important efforts. Forced recategorization into broader categories is a disservice to the science; though the intention was to increase rigor, the result was confounding. In the context of analyzing genomic data, there are benefits and limitations of using SRR and/or QGA to define risk factor associations across comparison groups. Additionally, depending on which variables are chosen for an analysis, careful interpretation of the data in the context of SRR or QGA is necessary. Culturally appropriate science should respect the heritage and lived experiences of all populations distinctly. Only then will we come closer to achieving equity.

Step 3: Inclusion and Accessibility

Once the first two steps are in place, the next step includes designing studies that address relevant questions, which should reflect the needs and interests of the members of the communities being studied. Investigators from these communities are increasing in numbers as a result of strategic workforce diversity initiatives. However, their participation is still limited by systemic barriers from limited resources at minority-serving institutions and structural racism in scientific institutions. Efforts such as the UNITE initiative are working to identify and mitigate structural racism, including offering funding opportunities for health disparities and health equity research, developing policies, and providing more opportunities to engage individuals from diverse backgrounds to the biomedical research community (https://www.nih.gov/ending-structural-racism/unite). This is an im­por­tant aspect of the science, given that the interpretation and application of disparities research rest with scientists who often have no connection to the communities being studied. The consequences to inappropriate use and interpretation of health disparities data are prolonged disparities, prorogated by lack of action despite awareness of the factors. Characterizing risk across diverse and distinct population groups, led by members of these groups, allows us to better identify those at increased risk for disease and to bring more focus and resources for higher impact and implementation to eliminate disparate health outcomes.

M.B. Davis reports grants from the NIH/NCI during the conduct of the study; grants from Genentech and Susan G. Komen, and grants and nonfinancial support from the New York Genome Center outside the submitted work; and a patent for the DARC diagnostic tool pending. No disclosures were reported by the other authors.

1.
Mersha
TB
,
Abebe
T
.
Self-reported race/ethnicity in the age of genomic research: its potential impact on understanding health disparities
.
Hum Genomics
2015
;
9
:
1
.
2.
Lopez
MM
,
Bevans
M
,
Wehrlen
L
,
Yang
L
,
Wallen
GR
.
Discrepancies in race and ethnicity documentation: a potential barrier in identifying racial and ethnic disparities
.
J Racial Ethn Health Disparities
2017
;
4
:
812
8
.
3.
Flores
G
.
Language barriers and hospitalized children: are we overlooking the most important risk factor for adverse events?
JAMA Pediatr
2020
;
174
:
e203238
.
4.
Flanagin
A
,
Frey
T
,
Christiansen
SL
,
Bauchner
H
.
The reporting of race and ethnicity in medical and science journals: comments invited
.
JAMA
2021
;
325
:
1049
52
.
5.
Lin
OM
,
Reid
HW
,
Fabbro
RL
,
Johnson
KS
,
Batch
BC
,
Olsen
MK
, et al
.
Association of provider perspectives on race and racial health care disparities with patient perceptions of care and health outcomes
.
Health Equity
2021
;
5
:
466
75
.
6.
Ross
PT
,
Hart-Johnson
T
,
Santen
SA
,
Zaidi
NLB
.
Considerations for using race and ethnicity as quantitative variables in medical education research
.
Perspect Med Educ
2020
;
9
:
318
23
.
7.
Taparra
K
,
Dee
EC
,
Dao
D
,
Patel
R
,
Santos
PMG
,
Chino
F
.
Disaggre­gating Pacific Islanders and major Asian subpopulations to reveal hidden breast cancer disparities
.
J Clin Oncol
39
,
2021
(
suppl 28; abstr
80
).
8.
Taparra
K
.
Pacific islanders searching for inclusion in medicine
.
JAMA Health Forum
2021
;
2
:
e210153
.
9.
Zamora
SM
,
Pinheiro
PS
,
Gomez
SL
,
Hastings
KG
,
Palaniappan
LP
,
Hu
J
, et al
.
Disaggregating Hispanic American cancer mortality burden by detailed ethnicity
.
Cancer Epidemiol Biomarkers Prev
2019
;
28
:
1353
63
.
10.
Park
B
,
Yang
S
,
Lee
J
,
Woo
HD
,
Choi
IJ
,
Kim
YW
, et al
.
Genome-wide association of genetic variation in the PSCA gene with gastric cancer susceptibility in a Korean population
.
Cancer Res Treat
2019
;
51
:
748
57
.
11.
Borrell
LN
,
Crawford
ND
.
Disparities in self-reported hypertension in Hispanic subgroups, non-Hispanic black and non-Hispanic white adults: the national health interview survey
.
Ann Epidemiol
2008
;
18
:
803
12
.
12.
Martini
R
,
Delpe
P
,
Chu
TR
,
Arora
K
,
Lord
B
,
Verma
A
, et al
.
African-ancestry associated gene expression signatures and pathways in triple-negative breast cancer, a comparison across women of African descent
.
medRxiv
2022.01.27.22269747 [Preprint].
2022
.
Available from:
https://doi.org/10.1101/2022.01.27.22269747.
13.
Davis
MB
,
Newman
LA
.
Oncologic anthropology: an interdisciplinary approach to understanding the association between genetically defined African ancestry and susceptibility for triple negative breast cancer
.
Current Breast Cancer Reports
2021
;
13
:
247
58
.
14.
Davis
MB
.
Genomics and cancer disparities: the justice and power of inclusion
.
Cancer Discov
2021
;
11
:
805
9
.
15.
Lord
BD
,
Martini
RN
,
Davis
MB
.
Understanding how genetic ancestry may influence cancer development
.
Trends Cancer
2022
;
8
:
276
9
.
16.
Bryc
K
,
Durand
EY
,
Macpherson
JM
,
Reich
D
,
Mountain
JL
.
The genetic ancestry of African Americans, Latinos, and European Americans across the United States
.
Am J Hum Genet
2015
;
96
:
37
53
.
17.
Carrot-Zhang
J
,
Chambwe
N
,
Damrauer
JS
,
Knijnenburg
TA
,
Robertson
AG
,
Yau
C
, et al
.
Comprehensive analysis of genetic ancestry and its molecular correlates in cancer
.
Cancer Cell
2020
;
37
:
639
54
.
18.
Howes
RE
,
Patil
AP
,
Piel
FB
,
Nyangiri
OA
,
Kabaria
CW
,
Gething
PW
, et al
.
The global distribution of the Duffy blood group
.
Nat Commun
2011
;
2
:
266
.
19.
Jenkins
BD
,
Martini
RN
,
Hire
R
,
Brown
A
,
Bennett
B
,
Brown
I
, et al
.
Atypical chemokine receptor 1 (DARC/ACKR1) in breast tumors is associated with survival, circulating chemokines, tumor-infiltrating immune cells, and African ancestry
.
Cancer Epidemiol Biomarkers Prev
2019
;
28
:
690
700
.
20.
Davis
M
,
Martini
R
,
Newman
L
,
Elemento
O
,
White
J
,
Verma
A
, et al
.
Identification of distinct heterogenic subtypes and molecular signatures associated with African ancestry in triple negative breast cancer using quantified genetic ancestry models in admixed race populations
.
Cancers
2020
;
12
:
1220
.
21.
Taliun
D
,
Harris
DN
,
Kessler
MD
,
Carlson
J
,
Szpiech
ZA
,
Torres
R
, et al
.
Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program
.
Nature
2021
;
590
:
290
9
.
22.
Van Dyke
ME
,
Baumhofer
NK
,
Slopen
N
,
Mujahid
MS
,
Clark
CR
,
Williams
DR
, et al
.
Pervasive discrimination and allostatic load in African American and White adults
.
Psychosom Med
2020
;
82
:
316
23
.
23.
Dietze
EC
,
Sistrunk
C
,
Miranda-Carboni
G
,
O'Regan
R
,
Seewaldt
VL
.
Triple-negative breast cancer in African-American women: disparities versus biology
.
Nat Rev Cancer
2015
;
15
:
248
54
.
24.
Beyer
KM
,
Zhou
Y
,
Matthews
K
,
Bemanian
A
,
Laud
PW
,
Nattinger
AB
.
New spatially continuous indices of redlining and racial bias in mortgage lending: links to survival after breast cancer diagnosis and implications for health disparities research
.
Health Place
2016
;
40
:
34
43
.
25.
Yeyeodu
ST
,
Kidd
LR
,
Kimbro
KS
.
Protective innate immune variants in racial/ethnic disparities of breast and prostate cancer
.
Cancer Immunol Res
2019
;
7
:
1384
9
.
26.
Zavala
VA
,
Bracci
PM
,
Carethers
JM
,
Carvajal-Carmona
L
,
Coggins
NB
,
Cruz-Correa
MR
, et al
.
Cancer health disparities in racial/ethnic minorities in the United States
.
Br J Cancer
2021
;
124
:
315
32
.
27.
Obeng-Gyasi
S
,
Tarver
W
,
Carlos
RC
,
Andersen
BL
.
Allostatic load: a framework to understand breast cancer outcomes in Black women
.
NPJ Breast Cancer
2021
;
7
:
100
.
28.
Campbell
MC
,
Tishkoff
SA
.
African genetic diversity: implications for human demographic history, modern human origins, and complex disease mapping
.
Annu Rev Genomics Hum Genet
2008
;
9
:
403
33
.
29.
Gomez
F
,
Hirbo
J
,
Tishkoff
SA
.
Genetic variation and adaptation in Africa: implications for human evolution and disease
.
Cold Spring Harb Perspect Biol
2014
;
6
:
a008524
.
30.
Wang
X
,
Zhang
H
,
Chen
X
.
Drug resistance and combating drug resistance in cancer
.
Cancer Drug Resist
2019
;
2
:
141
60
.
31.
Fritsche
LG
,
Ma
Y
,
Zhang
D
,
Salvatore
M
,
Lee
S
,
Zhou
X
, et al
.
On cross-ancestry cancer polygenic risk scores
.
PLoS Genet
2021
;
17
:
e1009670
.
32.
Popejoy
AB
,
Fullerton
SM
.
Genomics is failing on diversity
.
Nature
2016
;
538
:
161
4
.
33.
Martin
AR
,
Kanai
M
,
Kamatani
Y
,
Okada
Y
,
Neale
BM
,
Daly
MJ
.
Clinical use of current polygenic risk scores may exacerbate health disparities
.
Nat Genet
2019
;
51
:
584
91
.
34.
Shieh
Y
,
Fejerman
L
,
Lott
PC
,
Marker
K
,
Sawyer
SD
,
Hu
D
, et al
.
A polygenic risk score for breast cancer in US Latinas and Latin American women
.
J Natl Cancer Inst
2020
;
112
:
590
8
.
35.
Sparano
JA
,
Gray
RJ
,
Ravdin
PM
,
Makower
DF
,
Pritchard
KI
,
Albain
KS
, et al
.
Clinical and genomic risk to guide the use of adjuvant therapy for breast cancer
.
N Engl J Med
2019
;
380
:
2395
405
.
36.
Hoskins
KF
,
Danciu
OC
,
Ko
NY
,
Calip
GS
.
Association of race/ethnicity and the 21-gene recurrence score with breast cancer-specific mortality among US women
.
JAMA Oncol
2021
;
7
:
370
8
.
37.
Andre
F
,
Ismaila
N
,
Henry
NL
,
Somerfield
MR
,
Bast
RC
,
Barlow
W
, et al
.
Use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive breast cancer: ASCO clinical practice guideline update-integration of results from TAILORx
.
J Clin Oncol
2019
;
37
:
1956
64
.
38.
Taylor
MR
,
Sun
AY
,
Davis
G
,
Fiuzat
M
,
Liggett
SB
,
Bristow
MR
.
Race, common genetic variation, and therapeutic response disparities in heart failure
.
JACC Heart Fail
2014
;
2
:
561
72
.
39.
Schneider
BP
,
Li
L
,
Radovich
M
,
Shen
F
,
Miller
KD
,
Flockhart
DA
, et al
.
Genome-wide association studies for taxane-induced peripheral neuropathy in ECOG-5103 and ECOG-1199
.
Clin Cancer Res
2015
;
21
:
5082
91
.
40.
do Vale Moreira
NC
,
Ceriello
A
,
Basit
A
,
Balde
N
,
Mohan
V
,
Gupta
R
, et al
.
Race/ethnicity and challenges for optimal insulin therapy
.
Diabetes Res Clin Pract
2021
;
175
:
108823
.
41.
Ayers
KL
,
Mullaney
T
,
Zhou
X
,
Liu
JJ
,
Lee
K
,
Ma
M
, et al
.
Analysis of real-world data to investigate the impact of race and ethnicity on response to programmed cell death-1 and programmed cell death-ligand 1 inhibitors in advanced non-small cell lung cancers
.
Oncologist
2021
;
26
:
e1226
e39
.
42.
Calagua
C
,
Russo
J
,
Sun
Y
,
Schaefer
R
,
Lis
R
,
Zhang
Z
, et al
.
Expression of PD-L1 in hormone-naive and treated prostate cancer patients receiving neoadjuvant abiraterone acetate plus prednisone and leuprolide
.
Clin Cancer Res
2017
;
23
:
6812
22
.
43.
Kiely
M
,
Ambs
S
.
Immune inflammation pathways as therapeutic targets to reduce lethal prostate cancer in African American men
.
Cancers
2021
;
13
:
2874
.
44.
Kakarla
M
,
ChallaSivaKanaka
S
,
Hayward
SW
,
Franco
OE
.
Race as a contributor to stromal modulation of tumor progression
.
Cancers
2021
;
13
:
2656
.