Lisa A. Newman, MD, MPH
"WE ABSOLUTELY have an obligation to evaluate all of the features describing our patients with cancer when we are trying to figure out why some patients do better than others,” Lisa A. Newman, MD, MPH, reminded the nearly 700 participants at the 2018 Lynn Sage Breast Cancer Symposium, hosted by the Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago.1 These features include differences in tumor biology and disparities in outcomes between black and white American women diagnosed with breast cancer. Dr. Newman reviewed the history of breast cancer disparities research.
Suggestions that studies describing these differences and disparities might be construed as examples of racial profiling is “concerning, because racial profiling has several negative connotations from its practice in social and legal settings,” Dr. Newman said. “However, when it comes to looking at cancer outcomes, racial profiling is basically epidemiology.”
In an interview with The ASCO Post after her symposium presentation, Dr. Newman said, “I do believe there is some persistent reluctance among many investigators to evaluate the biology of these inequalities, and those types of concerns hold us back in terms of comprehensive cancer research.” Dr. Newman is Chief of the Section of Breast Surgery, Weill Cornell Medicine Cancer Center, and of the Breast Surgical Oncology Program, NewYork-Presbyterian, Weill Cornell Medicine Cancer Network, New York.
Similar Mammography Rates Despite Socioeconomic Differences
“CONVERSATIONS REGARDING health-care disparities associated with racial/ethnic identity inevitably evolve into a discussion of the unequal distribution of wealth in the United States,” Dr. Newman said. “Rates of poverty and unemployment are notably higher in blacks compared with white Americans.” She presented data showing that the percentage of blacks living below the poverty line is more than double that of white Americans as reported by the U.S. Census Bureau.
“We absolutely have an obligation to evaluate all of the features describing our patients with cancer when we are trying to figure out why some patients do better than others.”— Lisa A. Newman, MD, MPH
Tweet this quote
“We therefore reasonably assume that these socioeconomic disadvantages will lead to inadequate health-care access, resulting in more advanced-stage distribution and higher mortality rates documented in our black than in white American patients with breast cancer,” Dr. Newman said. “Interestingly, however, several studies such as the National Health Interview Survey and the Multi-Ethnic Cohort have reported that mammography utilization rates are actually similar between black and white American women. Screening practices, therefore, cannot be assumed to completely explain the differences that we see in stage distribution.”
Conducted by Dr. Newman and colleagues more than a decade ago, a meta-analysis of 20 studies reporting survival data from black and white American patients with breast cancer found that “black identity is a significant and independent predictor of poor outcome from breast cancer, even after accounting for socioeconomic status by conventional measures.”2
“The black patients continued to have a statistically significant nearly 30% higher mortality risk,” Dr. Newman said.
Tumor Biology and Genetics
“IF WE SIMPLY look at the basic features that describe the breast cancer burden of blacks,” Dr. Newman said, “we see several characteristics that cannot be easily attributed to socioeconomic status, such as the younger age distribution for breast cancer in black women as well as the increased frequency of adverse tumor features,” including higher rates of aneuploid, estrogen receptor–negative and triple-negative tumors, higher incidence of inflammatory breast cancer, and higher incidence of male breast cancer. “All of these characteristics underscore the importance of looking at tumor biology and genetics and how these factors contribute to breast cancer differences.”
“You might assume that looking at clinical trials data would be the perfect way to disentangle socioeconomic status from racial ethnic identity, when trying to figure out what drives differences in outcomes from cancer. In the context of participating in a clinical trial, the type, stage of disease, treatment, and outcome should all be very tightly regulated,” Dr. Newman said.
“This was the rationale for a terrific study that was published several years ago by Kathy Albain, MD, and colleagues in the Journal of the National Cancer Institute.3 They basically pooled together the results of the adjuvant therapy trials conducted by SWOG for a whole host of different tumors to see whether equal treatment delivered in the context of participating in a clinical trial would result in equal outcomes,” Dr. Newman reported. “For most cancers, outcomes were the same, regardless of racial ethnic identity,” she noted. “However, when it came to clinical trials of hormonally driven cancers, such as prostate and breast cancers, the black participants continued to have a worse outcome.” This study “was definitely a landmark publication,” Dr. Newman said.
Data from the Surveillance, Epidemiology, and End Results (SEER) Program for 1973 to 2007 showed that incidence rates for breast cancer were usually lower for black women, although the mortality rates were the same until the early 1980s. At that time, the mortality rates declined for white American women but were largely unchanged for black women.
“I would argue that this was a result of unmasking of some of the differences in tumor biology between blacks and whites,” Dr. Newman said. “Tamoxifen was approved for management of this disease in the late 1970s, and by the 1980s, we start seeing a decline in the breast cancer mortality rates with this endocrine therapy. However, since black women have notably higher rates of estrogen receptor–negative cancers, they simply have not benefited from this powerful advance in targeted systemic therapy to the same degree as white American women have benefited.”
Mostly Basal Subtypes
CHARACTERIZING BREAST cancer by gene-expression profiling has “demonstrated the diversity of breast cancer as a disease, and the basal intrinsic subtype tends to have a particularly unfavorable prognosis,” Dr. Newman said. “Triple-negative breast cancers aren’t a perfect surrogate for the basal subtype, but they do account for more than 80% of these aggressive tumors.”
These findings were corroborated in another presentation at the 2018 Lynn Sage Breast Cancer Symposium by Lisa Carey, MD,4 Jacobs Preyer Distinguished Professor of Breast Cancer Research, University of North Carolina Lineberger Comprehensive Cancer Center, Chapel Hill. “Most triple-negative breast cancers are basal-like, and most, but not all, basal-like breast cancers are triple-negative,” Dr. Carey noted. “There is a significant component of basal-like tumors that are hormone receptor–positive or HER2-positive and a significant subset of triple-negative breast cancers that are of luminal or HER2-enriched subtypes.”
The Carolina Breast Cancer Study, a population-based, case-controlled study, found that “young women and women of color have a greater proportion of triple-negative breast cancer if they develop breast cancer, compared with older women and white women,” Dr. Carey said. “It is quite a large disparity.” The rate for triple-negative breast cancer was 27% for black premenopausal women and 16% for black postmenopausal women vs 14% for white premenopausal women and 9% for white postmenopausal women.5
“The basal-like subtype occurred in almost 40% of young black women compared with about 10% to 15% of older white women.”— Lisa Carey, MD
Tweet this quote
In phase III of the Carolina Breast Cancer Study, which looked at racial differences in intrinsic subtypes, the differences between black and white and younger and older women “were even bigger than we thought,” Dr. Carey reported. The basal-like subtype occurred in almost 40% of young black women compared with about 10% to 15% of older white women.6 This study replaced the older and more inaccurate means of identifying subtype by immunohistochemistry and replaced it with RNA expression analyses for molecularly defined instrinsic subtype.
Outcome Disparities
TRIPLE-NEGATIVE breast cancers “are typically twice as common in black patients with breast cancer compared with white American patients with breast cancer,” Dr. Newman said. And the increased risk occurs regardless of age and stage at diagnosis. “Over the past several years, black women have had a disproportionate increase in the incidence rates of breast cancer to the point where now those incidence rates are the same for black and white individuals. This higher incidence rate of disease in blacks, coupled with the higher rates of triple-negative tumors, is already resulting in the widening of the mortality gap, which now sits at 40% higher for black women than white women with breast cancer.”
Several studies have looked at whether outcome disparities persist when patients are stratified for disease phenotype. Most of those studies “demonstrated that within the category of triple-negative tumors, black women and white women, as long as they are treated similarly, have similar outcomes,” Dr. Newman said. “With estrogen receptor–positive disease, however, several studies have shown that black women continue to have a survival disadvantage,” although the reason remains unknown.
In a study at the Henry Ford Health System in Detroit, Dr. Newman’s previous affiliation, among women with triple-negative breast cancers, “as with other investigators, we’ve found similar recurrence and distant organ failure rates for our black and white American patients with triple-negative breast cancer,” Dr. Newman reported.
“We are now understanding that there is tremendous diversity in triple-negative tumors,” she said. “Some of the work being pioneered at Vanderbilt University in triple-negative breast cancer subtyping may also be relevant prognostically and in predicting response to treatment, because the luminal androgen receptor subtype is less likely to respond to neoadjuvant chemotherapy. It is also a subtype that potentially can be manipulated with antiandrogen therapy,” Dr. Newman said. “Unfortunately, however, there are few data specifically addressing triple-negative subtyping in African ancestry population subsets.”
Global Differences
LOOKING AT African ancestry populations in other countries, specifically England, Switzerland, and Brazil, “investigators have demonstrated higher rates of estrogen receptor–negative and triple-negative tumors in their African ancestry population subsets compared with their non-African ancestry population,” Dr. Newman said. “There are several parallels between the breast cancers that we see in African ancestry populations compared with women who have hereditary disease,” Dr. Newman added. They include younger age as well as higher rates of estrogen receptor–negative and aneuploid tumors. These findings have “motivated us to try to evaluate the breast cancer burden of people in the countries of Africa. Hopefully, this work will allow us to gain insights regarding the pathogenesis of this phenotype, triple-negative breast cancer.”
“Our international work is based upon this premise that breast cancer is a heterogeneous disease, with subtypes that vary in prognosis as well as in frequency within different population subsets.”— Lisa A. Newman, MD, MPH
Tweet this quote
Working in Ghana for about 16 years, the International Center for the Study of Breast Cancer Subtypes has found that most of the patients with breast cancer being treated at Komfo Anoyke Teaching Hospital in Kumasi have triple-negative disease. Similar findings were reported at another hospital in Ghana. “The women tend to have a younger age at diagnosis, but of course this is also influenced by the average shorter longevity of individuals in Africa,” Dr. Newman observed.
“Over the past 5 or 6 years, we have also been working with hospitals in Ethiopia,” Dr. Newman shared. “Interestingly, we found that the frequency of triple-negative breast cancers was equally as low in the Ethiopian patients as in white Americans and Europeans,” representing approximately 15% of all tumors. Ghanaian women had the highest rate, 53.2%, and black women had an intermediate rate of 27.8%.
“We think this is further evidence in support of some hereditary component associated with African ancestry that predisposes these women to certain patterns of disease,” declared Dr. Newman. She pointed out that the slave trade of the colonial era mostly brought Africans from the west sub-Saharan coast, where Ghana is located, across the ocean to populate the colonies, but not from Ethiopia, which is located on the east coast. “So, blacks have more shared ancestry with western sub-Saharan Africans, and it does make sense that we are more likely to have breast cancer patterns similar to women in Ghana in contrast with Ethiopian women.” Looking at other markers, such as androgen receptors and aldehyde dehydrogenase 1, “we continue to find that the tumors of blacks and Ghanaians are more similar to each other, compared with the tumors of white Americans and Ethiopians,” she added.
“Another very interesting study by Dr. Ahmedin Jemal from the American Cancer Society,” Dr. Newman noted, looked at the data from the SEER registry to evaluate estrogen receptor–negative breast cancers in white Americans, blacks, and women who were born in either west or east Africa but had their cancers diagnosed in the United States.7 “Similar to what we see in our international patterns, the highest rates of estrogen receptor– negative tumors were in blacks and west African–born patients [39% and 40%]. The lowest rates of estrogen receptor–negative tumors were in white American and East African–born patients [21% and 22%].”
Utilizing New Technology
“STUDIES OF the genetics of certain high-risk phenotypes, such as triple-negative breast cancer, will be important to women of all racial/ethnic backgrounds,” Dr. Newman noted. “Being more scientific about how we define racial/ethnic ancestry is quite important, especially as genetic admixture increases in our population.” Although a woman’s self-reported identification by a single racial-ethnic category “may well have some relevance because it can be linked to particular environmental and socioeconomic factors, being able to characterize and quantify racial/ethnic ancestry by certain genetic markers, is probably going to be much more meaningful in understanding hereditary susceptibility for specific phenotypes of breast cancer.”
“Our international work is based upon this premise that breast cancer is a heterogeneous disease, with subtypes that vary in prognosis as well as in frequency within different population subsets,” Dr. Newman said.
Dr. Newman and colleagues have been able to establish immunohistochemistry programs in the hospitals where they work and core biopsy training programs. They interact with their colleagues via telemedicine programs regularly and started conducting Breast Cancer in Africa symposia on an annual basis.
Dr. Newman noted that she and a colleague, John Carpten, PhD, of the University of Southern California, Los Angeles, “have been challenging our oncology colleagues to look more rigorously and scientifically at this concept of racial/ethnic identity by utilizing genotype technology, as advocated in a viewpoint article published earlier this year in JAMA Surgery.8 “The general public is actually way ahead of us, because commercially available products to quantify racial/ethnic outcome (eg, Ancestry. com and 23andMe) are being purchased by the millions, and we need to catch up and utilize this technology appropriately in oncology research.” ■
DISCLOSURE: Dr. Newman reported no conflicts of interest. Dr. Carey has received research funding from Genentech/Roche, Innocrin Pharma, and Novartis. She also has a royalty-sharing agreement and investorship interest in licensed intellectual property to the start-up company Falcon Therapeutics, which is developing neural stem-cell based therapy.
REFERENCES
1. Newman LA: African ancestry and high-risk/triple-negative breast cancer. 2018 Lynn Sage Breast Cancer Symposium. Presented October 11, 2018.
2. Newman LA, Griffith KA, Jatoi I, et al: Meta-analysis of survival in black and white American patients with breast cancer: Ethnicity compared with socioeconomic status. J Clin Oncol 24:1342-1349, 2006.
3. Albain KS, Unger JM, Crowley JJ, et al: Racial disparities in cancer survival among randomized clinical trials patients of the Southwest Oncology Group. J Natl Cancer Inst 101:984-992, 2009.
4. Carey L: Biologic studies in triple-negative breast cancer. 2018 Lynn Sage Breast Cancer Symposium. Presented October 12, 2018.
5. Carey LA, Perou CM, Livasy CA, et al: Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 295:2492-2502, 2006.
6. Parada H Jr, Sun X, Fleming JM, et al: Race-associated biological differences among luminal A and basal-like breast cancers in the Carolina Breast Cancer Study. Breast Cancer Res 19:131, 2017.
7. Jemal A, Fedewa SA: Is the prevalence of ER-negative breast cancer in the US higher among Africa-born than US-born black women? Breast Cancer Res Treat 135:867-873, 2012.
8. Newman LA, Carpten J: Integrating the genetics of race and ethnicity into cancer research: Trailing Jane and John Q. Public. JAMA Surg 153:299-300, 2018.
