Abstract
Purpose
Breast cancer (BC) survival favors White versus Black Americans despite advances in screening and treatment. We hypothesized that these differences were dependent upon quality of care by analyzing long-term outcomes of 3139 early BC patients at our quaternary care center where uniform access and management of BC is provided to women irrespective of race.
Methods
Prospectively collected data for clinical stage I-II BC patients from our quaternary care cancer center were analyzed, focusing on disease-specific survival (DSS). Subgroup analyses included the overall cohort, triple-negative BC (TNBC), non-TNBC and HER2/neu positive patients. Multivariable analyses to evaluate associations of variables with DSS were performed for each subgroup.
Results
The overall cohort consisted of 3139 BC patients (1159 Black, 1980 White). Black and White patients did not differ by most baseline variables. Black patients had higher rates of TNBC (18% versus 10%, p < 0.0001). Kaplan-Meier analysis of all subgroups (overall, TNBC, non-TNBC, HER2/neu positive) did not reveal DSS differences between Black and White patients. Multivariable analysis of subgroups also did not find race to be associated with DSS.
Conclusion
In this large, carefully controlled, long term, single-institution prospective cohort study DSS in Black and White early BC patients with equal access to high quality care, did not differ. While BC patients with adverse molecular markers did slightly worse than those with more favorable markers, there is no observable difference between Black and White women with the same markers. These observations support the conclusion that equal access to, and quality, of BC care abolishes racial disparities in DSS.
Introduction
Breast cancer (BC) survival has improved over the past several decades due to advances in diagnosis and treatments.(1) However, there remains a clear survival disparity between Black and White American BC patients, with significantly improved survival in White patients.(2–6) The likely causes of these disparities have been a frequent subject of debate, with conflicting conclusions favoring underlying comorbidities, access to timely diagnosis, later stage of presentation in Black women, inadequate treatment in Black patients, molecular subtypes of BC, socio-economic factors, and, more recently, a more refined definition of racial subtypes.(2–4, 6–20) It is likely that a combination of these factors plays a role. To determine which of these factors is the most important in improving BC outcome in Black women would require a large study that includes all these competing variables. Most of these factors cannot be changed. However, there is compelling evidence that disparate access to quality of care drives BC survival differences in the United States.(2, 3, 6, 12, 13, 15, 16) This is a variable that could be changed in a positive way by providing equal access and quality of care to all women, irrespective of race, socio-economic status, or geographic location.
Black people in the United States have been subject to inequality in all walks of life, including access to quality medical care, and poor access in the context of BC causes disparate outcomes in two major ways: more advanced stage at diagnosis, and inferior treatment quality.(2, 3, 6, 12, 13, 15, 16) Because of the structure and style of practice in our large health care system, in Detroit, a city with a high proportion of Black patients, and extensive clinical coverage of a wide swathe of mixed race suburban populations in South East Michigan, we were able to more directly assess whether equal access to, and quality of care in Black and White American women helped overcome disparate BC outcomes. We did so by performing a prospective cohort study of over three thousand Black and White patients screened, treated, and followed for up to twenty-four years at a single quaternary care cancer center where access to care of the treated patients was uniformly equal at every stage of management.
Methods
Data was prospectively collected from patients with BC treated within our quaternary-care hospital system from 1995 through 2019. All patients included were diagnosed and treated by a highly specialized, guideline-controlled, multidisciplinary team of BC physicians and ancillary staff in a vertically integrated academic health system. Data has been continually collected and consistently audited by a single collection team. Data collection has been approved on a yearly basis by the health system institutional review board.(7)
Race was determined from a direct patient-incorporated questionnaire completed at the initial appointment within the health system. No attempt was made to subclassify ‘Black African American’ patients based upon the concentration of their African heritage. A total of 374 patients identified as races other than Black or White American, or had unknown race (American Indian = 15, Asian = 60, Black/White = 5, Hispanic = 1, Middle Eastern = 1, Native Hawaiian/Pacific Islander = 2, declined to answer = 151, unsure = 27, other = 56, missing = 56). These patients were excluded from the analysis due to comparatively small sample size (Figure 1).(11)
Figure 1:
CONSORT flow diagram of study cohort.
The data base was established to include all breast cancer patients with clinically negative ipsilateral axillae and no initial evidence of systemic disease – those who were eligible for sentinel lymph node biopsy (SLNB). A total of 90 patients either had two surgeries, or bilateral procedures. If a patient had bilateral surgery on the same date, the side with the positive SLN or with SLN data collected (if no positive SLN found) was used. In patients undergoing lumpectomy for ductal carcinoma in situ axillary node surgery was initially omitted but done at a later date if the final pathology review showed invasive cancer. If a patient had two surgeries on different dates, lymph node data from the second surgery was used. Patients with pathology of ‘DCIS only’ were excluded from the analysis (N=253) even if they had SLNB performed at the discretion of the surgeon. Patients with missing data for SLN and axillary dissection were also excluded (N = 22). Due to the nature of the treated population (eligible for surgery), no patients with metastatic disease at initial presentation are included in this database (Figure 1).
The Henry Ford Breast Cancer Services, officially approved by site visit every three years by the American College of Surgeons, had oversight committees that tracked data and reported to a multidisciplinary breast committee, and members attended twice weekly breast cancer tumor boards and clinics where management decisions for every patient were discussed. A team of fellowship-trained breast pathologists reviewed each case as a team and reports were signed out only after uniform agreement. All patients were treated according to National Comprehensive Center Network (NCCN) guidelines in four facilities by the same full-time salaried board-certified surgeons, and medical oncologists, who treated patients in those facilities by traveling from the main downtown hospital to suburban facilities that all were part of the same health system (Henry Ford). While systemic therapy, in the form of chemotherapy, hormonal manipulation, targeted therapy, and immunotherapy have changed over the years of observation, there were no major differences in the form of therapy for any of the patients for any particular time period.(7) All patients, regardless of race, socio-economic or insurance status, received comparable treatments.(7) The radiation oncologists, stationed at three different sites, had the same equipment, maintained by the same radiation physicists, and all under the same administrative and clinical leadership, reinforcing uniform standards of care.
Included patients were Black or White American and grouped as such. Differences between groups were compared with χ2 tests of independence in terms of age, age group, hormone receptor status (estrogen receptor (ER) and progesterone receptor (PR)), HER2/neu expressional status, triple negative breast cancer status (TNBC; combination of hormone receptor and HER2/neu variables), lympho-vascular invasion, tumor (T) and lymph node (N) stage at diagnosis, overall stage at diagnosis, whether they underwent mastectomy, whether they received chemotherapy, and whether they received radiation. Disease-specific survival (DSS) was the primary outcome of interest.
Multiple analyses of different groups and subgroups were performed that included group variable difference comparison, Kaplan-Meier estimation of DSS, and univariate and multivariable estimates of variable associations with DSS. The primary comparison variable was race, and subgroup analyses were of among TNBC, non-TNBC, and HER2/neu positive patients. Multivariable analyses used a forced model which included the variables listed above with specific control for redundancy depending on the model. Significance was established at p < 0.05. All statistical analyses were performed in SAS 9.4 (SAS Institute, Cary, NC, USA).
Results
Overall cohort
A total of 3139 patients were included in the analysis (median follow-up 5.15 years), 1980 (63%) of whom were White (median follow-up 5.29 years) and 1159 (37%) of whom were Black (median follow-up 4,83 years). Black and White patients were comparable in terms of age group at diagnosis, tumor (T) stage, lymph node (N) stage, overall stage, lympho-vascular invasion (LVI), HER2/neu status, type of surgery (mastectomy or breast conserving surgery) and how often they received radiation. Black patients were slightly younger (p = 0.0326), had lower rates of ER positive tumors (p < 0.0001), PR positive tumors (p < 0.0001), and higher rates of TNBC (p < 0.0001) and received more chemotherapy (p = 0.0008) (Table 1). There was no significant difference in DSS between races (p = 0.1767), with median survival not reached at 21 years for either race (Figure 2).
Table 1:
Univariate comparisons of racial groups in overall cohort.
| Variable | Response | White/(N=1980) | Black/(N=1159) | p-value |
|---|---|---|---|---|
| age | Mean (SD) | 62.8 (12.7) | 61.8 (12.8) | 0.0326 |
| age group | < 50 | 333 (17%) | 203 (18%) | 0.6165 |
| >=50 | 1647 (83%) | 956 (82%) | ||
| estrogen receptor status | NEGATIVE | 293 (15%) | 291 (25%) | <0.0001 |
| POSITIVE | 1687 (85%) | 868 (75%) | ||
| progesterone receptor status | NEGATIVE | 421 (21%) | 386 (33%) | <0.0001 |
| POSITIVE | 1559 (79%) | 773 (67%) | ||
| HER2/neu status | NEGATIVE | 1635 (83%) | 929 (80%) | 0.0907 |
| POSITIVE | 345 (17%) | 230 (20%) | ||
| triple negative | NO | 1789 (90%) | 953 (82%) | <0.0001 |
| YES | 191 (10%) | 206 (18%) | ||
| lymphovascular invasion | NEGATIVE | 1646 (85%) | 964 (85%) | 0.8943 |
| POSITIVE | 284 (15%) | 164 (15%) | ||
| tumor (T) stage | T1a | 196 (10%) | 123 (11%) | 0.2225 |
| T1b | 477 (24%) | 286 (25%) | ||
| T1c | 813 (41%) | 437 (38%) | ||
| T2 | 438 (22%) | 287 (25%) | ||
| T3 | 56 (3%) | 26 (2%) | ||
| lymph node (N) stage | N0 | 1526 (77%) | 907 (78%) | 0.5123 |
| N1 | 376 (19%) | 217 (19%) | ||
| N2 | 53 (3%) | 26 (2%) | ||
| N3 | 25 (1%) | 9 (1%) | ||
| overall stage | IA | 617 (31%) | 341 (29%) | 0.4084 |
| IIA | 452 (23%) | 268 (23%) | ||
| IIB | 150 (8%) | 103 (9%) | ||
| IIIA | 736 (37%) | 438 (38%) | ||
| IIIC | 25 (1%) | 9 (1%) | ||
| mastectomy | NO | 1450 (73%) | 846 (73%) | 0.8683 |
| YES | 524 (27%) | 310 (27%) | ||
| chemotherapy | NO | 1037 (56%) | 538 (49%) | 0.0008 |
| YES | 821 (44%) | 550 (51%) | ||
| radiation | NO | 443 (24%) | 267 (24%) | 0.7900 |
| YES | 1422 (76%) | 837 (76%) |
Figure 2:
Disease specific survival of overall cohort by race.
TNBC subgroup
There were 397 patients in the TNBC subgroup (median follow-up 5.51 years), 191 (48%) of whom were White (median follow-up 6.17 years) and 206 (52%) of whom were Black (median follow-up 4.43 years). Black and White patients in this subgroup were comparable in terms of age group at diagnosis, tumor (T) stage, lymph node (N) stage, overall stage, lympho-vascular invasion (LVI), type of surgery (mastectomy or breast conserving surgery) and how often they received radiation. (Table 2). There was no significant difference in DSS between races in this TNBC subgroup (p = 0.6161), with median survival not reached at 21 years for either race (Figure 3).
Table 2:
Univariate comparisons of racial groups in TNBC subgroup.
| Variable | Response | White/(N=191) | Black/(N=206) | p-value |
|---|---|---|---|---|
| age | N Mean (SD) | 191 58.7 (13.4) | 206 60.3 (12.7) | 0.2171 |
| age group | < 50 | 51 (27%) | 43 (21%) | 0.1723 |
| >=50 | 140 (73%) | 163 (79%) | ||
| lymphovascular invasion | NEGATIVE | 151 (83%) | 170 (85%) | 0.6692 |
| POSITIVE | 31 (17%) | 31 (15%) | ||
| tumor (T) stage | T1a | 15 (8%) | 25 (12%) | 0.5961 |
| T1b | 41 (21%) | 45 (22%) | ||
| T1c | 76 (40%) | 74 (36%) | ||
| T2 | 53 (28%) | 58 (28%) | ||
| T3 | 6 (3%) | 4 (2%) | ||
| lymph node (N) stage | N0 | 147 (77%) | 176 (85%) | 0.3465 |
| N1 | 37 (19%) | 24 (12%) | ||
| N2 | 5 (3%) | 2 (1%) | ||
| N3 | 2 (1%) | 4 (2%) | ||
| overall stage | IA | 60 (31%) | 65 (32%) | 0.8328 |
| IIA | 47 (25%) | 50 (24%) | ||
| IIB | 19 (10%) | 15 (7%) | ||
| IIIA | 63 (33%) | 72 (35%) | ||
| IIIC | 2 (1%) | 4 (2%) | ||
| mastectomy | NO | 131 (69%) | 154 (75%) | 0.1479 |
| YES | 60 (31%) | 51 (25%) | ||
| chemotherapy | NO | 42 (24%) | 49 (25%) | 0.7324 |
| YES | 135 (76%) | 145 (75%) | ||
| radiation | NO | 51 (29%) | 47 (24%) | 0.3169 |
| YES | 126 (71%) | 147 (76%) |
Figure 3:
Disease specific survival of TNBC subgroup by race.
Non-TNBC subgroup
There were a total of 2741 patients in the non-TNBC subgroup (median follow-up 5.14 years), 1789 (65%) of whom were White (median follow-up 5.22 years) and 953 (35%) of whom were Black (median follow-up 4.92 years). Black and White patients in this subgroup were comparable in terms of age group at diagnosis, tumor (T) stage, lymph node (N) stage, overall stage, lympho-vascular invasion (LVI), type of surgery (mastectomy or breast conserving surgery) and how often they received radiation. Black patients were slightly younger (p = 0.0266) and had higher rates of receiving chemotherapy (p = 0.0281) (Supplementary Table 1). There was no significant difference in DSS between races in this non-TNBC subgroup (p = 0.4001), with median survival not reached at 21 years for either race (Supplementary Figure 1).
HER2/neu positive subgroup
There were 575 patients in the HER2/neu positive subgroup (median follow-up 4.82 years), 345 (60%) of whom were White (median follow-up 4.86 years) and 230 (40%) of whom were Black (median follow-up 4.70 years). Black and White patients in this subgroup were comparable in terms of age group at diagnosis, tumor (T) stage, overall stage, lympho-vascular invasion (LVI), type of surgery (mastectomy or breast conserving surgery) and how often they received radiation. Black patients were slightly younger (p = 0.0412), had lower rates of ER positive tumors (p = 0.0200), and lower lymph node (N) stage (p = 0.0363) (Supplementary Table 2). There was no significant difference in DSS between races in this non-TNBC subgroup (p = 0.8593), with median survival not reached at 21 years for either race (Supplementary Figure 2).
Multivariable disease-specific survival analysis
Variables were assessed for univariate and multivariable associations with DSS in the overall cohort and various subgroups (TNBC, non-TNBC, HER2/neu positive). By univariate or multivariable analysis, Black vs White race was not found to be significantly associated with DSS in the overall cohort or any of the subgroups (Supplementary Tables 3–10).
Discussion
In this study we have shown that Black BC patients receiving the same care compared to White BC patients had equivalent disease specific survival (DSS) outcomes. This was true overall, among TNBC patients, non-TNBC patients, and HER2/neu positive patients. We additionally confirmed TNBC was more common among Black African American women, which is concordant with current literature, but the similar survival rates between Black and White TNBC patients indicate that similar diagnostic quality and availability, and unvarying appropriate treatment may have more importance in determining outcomes at this juncture than racially determined biological differences between tumors.(2, 4, 8–12, 15, 17) All of this was made possible by standardized diagnostic availability plus local, regional and systemic therapies.
There is abundant evidence that survival disparities between Black and White BC patients are largely due to disparate access to quality care.(2–6, 8–10, 12, 13, 15–18) Our study reinforces this fact by showing long-term similarity in survival outcomes when all patients are provided the same quality care; this has been previously demonstrated in other types of cancer and other patient populations.(21) Though we do not dismiss the importance of the growing body of BC biologic subtype literature, it seems likely that disparity in access to quality care is the single biggest area in need of improvement in the United States, and which could make the most meaningful difference in terms of improving survival outcomes for Black BC patients.(20) Equalizing the care of Black and White women with early BC should be the goal of everyone that treats BC. This should obviously not be done by decreasing the quality and access of care of White women but by improving those parameters for Black women.
It is important to note several potential detracting factors in the design of this study. Patients followed in our study all presented to our surgery clinics with localized BC without clinical evidence of metastasis to regional lymph nodes or to systemic sites. Since Black patients may present initially with more advanced stages of the disease than White patients, possibly because of access or cultural issues, those higher stage patients would not be amongst the ones followed in this study.(19) If we had included all BC patients, our data would probably have identified the same racial disparities outcomes as have been reported by many others. Our study, by providing long term detailed follow-up of early-stage BC patients of both races, shows no significant difference in survival decades after initial presentation of clinically lymph-node negative BC. This makes it likely that differences in metastatic disease rate at presentation are also attributable to disparities in access to screening. Early diagnosis is only feasible when patients are appropriately educated on the importance of screening and have convenient access to screening facilities. Both of these factors remain a significant area of disparity between Black and White BC patients in the United States.(3, 8, 12–17) The original design of the data base did not question distinctions between screen-detected and clinically detected presentations although we did not recognize any obvious differences between black and white patients. We also did not separately interrogate neoadjuvant systemic therapy as a factor. We did not look at comorbidities in this study but, because we look specifically at death from breast cancer and not from other disease processes, this is not likely to be an issue that could explain lack of racial disparities in DSS in our study.
Comparing racial groups for survival disparity research purposes has, until recently, been entirely dependent upon patient-reported phenotypes. Our study, entirely dependent upon self-reported racial groupings, could not evaluate a more exact phenotype that might be possible in the future by examining detailed genotypes. An analysis of genetic and transcriptomic profiles that identify West African, East African, or African American women with BC may assist in determining the relative contribution of racial biology to survival outcomes, given uniform BC management.(11, 20)
In summary, there is growing agreement that disparate access to care is driving survival differences between Black and White BC patients. Our study, by careful follow-up of BC patients for over two decades, reinforces this fact by controlling for easy access to high quality care and equality of management by a team of dedicated BC clinicians, pathologists, and researchers. It accentuates the need for improved access to quality care for Black BC patients by showing what is possible if quality of care is as good for Black women as it is for White women. While research of the mechanisms of BC metastasis and biologic differences in tumors between races should continue, disparate BC outcomes may be better addressed by first improving access to quality care, a process that requires focused education and public health efforts for all women.
Supplementary Material
Funding:
The Nathanson/Rands Chair in Breast Cancer Research
Footnotes
Disclosures: The authors have no disclosures to report.
References
- 1.American Cancer Society. Cancer Facts & Figures 2021. Atlanta: American Cancer Society; 2021. [Google Scholar]
- 2.Curtis E, Quale C, Haggstrom D, Smith-Bindman R. Racial and ethnic differences in breast cancer survival: how much is explained by screening, tumor severity, biology, treatment, comorbidities, and demographics? Cancer. 2008;112(1):171–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Jatoi I, Sung H, Jemal A. The Emergence of the Racial Disparity in U.S. Breast-Cancer Mortality. The New England Journal of Medicine. 2022. [DOI] [PubMed] [Google Scholar]
- 4.Ko D, Andreopoulou E, Moo T-A. Is Race a Prognostically Significant Factor in Survival Outcomes for Breast Cancer Among Caucasian and African-American Women? Clinics in Surgery. 2016. [Google Scholar]
- 5.Li CI, Malone KE, Daling JR. Differences in Breast Cancer Stage, Treatment, and Survival by Race and Ethnicity. JAMA Internal Medicine. 2003. [DOI] [PubMed] [Google Scholar]
- 6.Silber JH, Rosenbaum PR, Clark AS, Giantonio BJ, Ross RN, Teng Y, et al. Characteristics associated with differences in survival among black and white women with breast cancer. JAMA. 2013;310(4):389–97. [DOI] [PubMed] [Google Scholar]
- 7.Chen Y, Susick L, Davis M, Bensenhaver J, Nathanson SD, Burns J, et al. Evaluation of Triple-Negative Breast Cancer Early Detection via Mammography Screening and Outcomes in African American and White American Patients. JAMA Surgery. 2020;155(5):440–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Deyrup A, Graves J, J. L Racial Biology and Medical Misconceptions. The New England Journal of Medicine. 2022. [DOI] [PubMed] [Google Scholar]
- 9.Dietze EC, Sistrunk C, Miranda-Carboni G, O’Regan R, Seewaldt VL. Triple-negative breast cancer in African-American women: disparities versus biology. Nature Reviews Cancer. 2015;15(4):248–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Doepker MP, Holt SD, Durkin MW, Chu CH, Nottingham JM. Triple-Negative Breast Cancer: A Comparison of Race and Survival. The American Surgeon. 2018. [PubMed] [Google Scholar]
- 11.Flanagin A, Frey T, Christiansen ST, Bachner H. The Reporting of Race and Ethnicity in Medical and Science Journals. JAMA. 2021. [DOI] [PubMed] [Google Scholar]
- 12.Hossain F, Danos D, Prakash O, Gilliland A, Ferguson TF, Simonsen N, et al. Neighborhood Social Determinants of Triple Negative Breast Cancer. Frontiers in Public Health. 2019;7:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ko NY, Hong S, Winn RA, Calip GS. Association of Insurance Status and Racial Disparities With the Detection of Early-Stage Breast Cancer. JAMA Oncology. 2020;6(3):385–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Maskarinec G, Sen C, Koga K, Conroy SM. Ethnic differences in breast cancer survival: status and determinants. Women’s Health. 2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Newman LA, Kaljee LM. Health Disparities and Triple-Negative Breast Cancer in African American Women: A Review. JAMA Surgery. 2017;152(5):485–93. [DOI] [PubMed] [Google Scholar]
- 16.Obeng-Gyasi S, Asad S, Fisher JL, Rahurkar S, Stover DG. Socioeconomic and Surgical Disparities are Associated with Rapid Relapse in Patients with Triple-Negative Breast Cancer. Annals of Surgical Oncology. 2021;28(11):6500–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Siddharth S, Sharma D. Racial Disparity and Triple-Negative Breast Cancer in African-American Women: A Multifaceted Affair between Obesity, Biology, and Socioeconomic Determinants. Cancers. 2018;10(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Tammemagi CM, Nerenz D, Neslund-Dudas C, Feldkamp C, Nathanson D. Comorbidity and Survival Disparities Among Black and White Patients With Breast Cancer. JAMA. 2005. [DOI] [PubMed] [Google Scholar]
- 19.Zhang C, Zhang C, Wang Q, Li Z, Lin J, Wang H. Differences in Stage of Cancer at Diagnosis, Treatment, and Survival by Race and Ethnicity Among Leading Cancer Types. JAMA Open. 2020;3(4):e202950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hoehn RS, Rieser CJ, Winters S, Stitt L, Hogg ME, Bartlett DL, et al. A Pancreatic Cancer Multidisciplinary Clinic Eliminates Socioeconomic Disparities in Treatment and Improves Survival. Annals of Surgical Oncology. 2021;28(5):2438–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.