African American women have a lower lifetime incidence of breast cancer than white/Caucasian Americans yet have a higher risk of breast cancer mortality. African American women are also more likely to be diagnosed with breast cancer at young ages, and they have higher risk for the biologically more aggressive triple-negative breast cancers. These features are also more common among women from western, sub-Saharan Africa who share ancestry with African Americans, and this prompts questions regarding an association between African ancestry and inherited susceptibility for certain patterns of mammary carcinogenesis.
Key points
- •
Population-based lifetime breast cancer incidence rates are lower for African American women than for white/Caucasian American women.
- •
Breast cancer mortality rates are higher in African American women than in white/Caucasian American women, and this difference is partially explained by socioeconomic disparities resulting in delayed diagnoses and suboptimal treatment in the African American community.
- •
Women with African ancestry (African American and Africans) are more likely to be diagnosed with early onset/premenopausal and triple-negative breast cancer (TNBC) than women with Caucasian/European background.
- •
Male breast cancer is more frequent in populations with African ancestry.
- •
Genotyping for ancestry informative markers may inform the discussion of how African ancestry may be related to hereditary susceptibility for triple-negative, early onset, and male breast cancer.
Introduction
Differences in breast cancer burden (represented by incidence and mortality rates) between white/Caucasian Americans and African Americans have been recognized and documented in the United States during the past several decades. These race/ethnicity-related differences are summarized as follows:
- 1.
Population-based incidence rates for breast cancer are lower in African American women than in white/Caucasian American women.
- 2.
The stage distribution for breast cancer is more advanced in African American women than in white/Caucasian American women.
- 3.
African American women have a younger age distribution for breast cancer at time of diagnosis than white/Caucasian American women.
- 4.
Compared with white/Caucasian American women, African Americans are more likely to be diagnosed with breast cancers that are negative for the estrogen receptor (ER), progesterone receptor (PR), and HER2/ neu marker.
- 5.
Male breast cancer is more common in African Americans than in white/Caucasian Americans.
Contemporary studies of variation in breast cancer subtypes associated with racial/ethnic identity conclusively demonstrate that individuals with African ancestry (African Americans and sub-Saharan Africans) are more likely to be diagnosed with biologically aggressive patterns of disease, but it is clear that inequities in the socioeconomic structure of the United States also play a significant role in explaining these breast cancer differences, just as they account for increased morbidity and mortality burden among African Americans diagnosed with colorectal cancer, lung cancer, and other nonmalignant medical problems such as diabetes or hypertension. Socioeconomic disadvantages such as poverty and absent/inadequate health care coverage are two to three times higher for the African American community, and these issues clearly account for many of the breast cancer disparities listed earlier, by acting as barriers to effective and optimal access to the health care system. The end results are delays in breast cancer diagnosis, less-efficient delivery of multidisciplinary treatment, and ultimately higher mortality rates.
Introduction
Differences in breast cancer burden (represented by incidence and mortality rates) between white/Caucasian Americans and African Americans have been recognized and documented in the United States during the past several decades. These race/ethnicity-related differences are summarized as follows:
- 1.
Population-based incidence rates for breast cancer are lower in African American women than in white/Caucasian American women.
- 2.
The stage distribution for breast cancer is more advanced in African American women than in white/Caucasian American women.
- 3.
African American women have a younger age distribution for breast cancer at time of diagnosis than white/Caucasian American women.
- 4.
Compared with white/Caucasian American women, African Americans are more likely to be diagnosed with breast cancers that are negative for the estrogen receptor (ER), progesterone receptor (PR), and HER2/ neu marker.
- 5.
Male breast cancer is more common in African Americans than in white/Caucasian Americans.
Contemporary studies of variation in breast cancer subtypes associated with racial/ethnic identity conclusively demonstrate that individuals with African ancestry (African Americans and sub-Saharan Africans) are more likely to be diagnosed with biologically aggressive patterns of disease, but it is clear that inequities in the socioeconomic structure of the United States also play a significant role in explaining these breast cancer differences, just as they account for increased morbidity and mortality burden among African Americans diagnosed with colorectal cancer, lung cancer, and other nonmalignant medical problems such as diabetes or hypertension. Socioeconomic disadvantages such as poverty and absent/inadequate health care coverage are two to three times higher for the African American community, and these issues clearly account for many of the breast cancer disparities listed earlier, by acting as barriers to effective and optimal access to the health care system. The end results are delays in breast cancer diagnosis, less-efficient delivery of multidisciplinary treatment, and ultimately higher mortality rates.
Disentangling socioeconomic resources from racial-ethnic identity in explaining breast cancer disparities
Many investigators have sought to disentangle the confounding effects of socioeconomic disadvantage and race/ethnic identity on cancer survival with interesting and provocative results. In 2004, Ward and colleagues documented that poverty is an adverse oncologic prognostic feature in their landmark study reporting on 5-year survival rates after any cancer diagnosis, stratified by income level. They demonstrated that poverty was associated with worse survival regardless of racial-ethnic identity. As far back as 1994, the National Cancer Institute tried to address this question within the context of breast cancer, with the Black-White Cancer Survival Study, a case-control analysis of 612 African American and 518 white/Caucasian American patients with breast. This comprehensive epidemiologic study used personal interviews and medical record information to account for sociodemographic factors and clinicopathologic cancer features in explaining breast cancer outcomes. Mortality rates were more than twice as high for the African American patients with breast, and approximately 75% of this survival disparity was explained by the various socioeconomic and clinical features studied, leaving 25% of the disparity unexplained and therefore possibly related to some poorly defined, primary race/ethnicity-related factor. Moreover, when the survival data of Ward and colleagues are analyzed from the perspective of outcome stratified by racial/ethnic identity within specific income strata for female cancers (and therefore data largely driven by outcomes from breast cancer), one continues to see incremental decreases in survival for African American patients compared with white American patients. Delving further into adjustments for socioeconomic status in comparisons of outcome between African American and white American patients, Newman and colleagues conducted a meta-analysis of all studies in the medical literature in which a Cox proportional hazard survival analysis had been performed, accounting for racial ethnic identity and some measure of socioeconomic advantage. This pooled analysis yielded a robust sample, more than 13,000 African American patients with breast cancer whose survival was compared with more than 75,000 white American patients, and African American identity was associated with a statistically significant nearly 30% higher mortality rate (mortality hazard, 1.27; 95% confidence rate, 1.18–1.38).
Results from prospective clinical trials in cancer management provide another strategy for disentangling the effects of poverty and racial/ethnic identity on breast cancer survival, because tumor type, treatment, and follow-up are all standardized and regimented through the clinical trials mechanism. The Southwest Oncology Group conducted this type of analysis and reported outcomes in nearly 20,000 patients with cancer participating in 35 prospective randomized clinical trials between 1974 and 2001. These investigators found that equal treatments indeed did result in equal outcomes (regardless of racial ethnic identity) except for patients treated for hormonally driven cancers. African Americans participating in clinical trials for breast, prostate, and ovarian cancers had statistically significant survival disadvantages compared with participants of other racial/ethnic identities.
Epidemiology and risk factors for breast cancer associated with African ancestry
Several of the features describing the breast cancer burden of African American women cannot easily be ascribed to socioeconomic factors. For example, the age-specific patterns of breast cancer incidence do not lend themselves to a sociodemographic explanation. Approximately 20% of white/Caucasian American patients are diagnosed at younger than 50 years, compared with nearly one-third of African American cases. Although the population-based lifetime risk of breast cancer is lower for African American patients than for white/Caucasian American patients, for women younger than age 45 years, the incidence rates are higher among the African Americans. Similarly, the increased risk of male breast cancer, as well as the increased risk of being diagnosed with adverse primary tumor features, such as high-grade histology, hormone-receptor negative, and triple-negative disease, suggests that some primary race/ethnicity-associated factors also contribute to the reported survival differences.
Reproductive history has been proposed as an explanation for ethnicity-associated variations in age distribution. Multiple pregnancies result in diminished lifetime exposure of the breasts to estrogen, yielding a lower breast cancer incidence among multiparous women than in nulliparous women; however, there is a brief increase in breast cancer risk that occurs in the early postpartum period. Because early childbearing is more common among African American women, it is biologically plausible that a dual effect of parity on breast cancer risk might be observed at a population level, with multiple pregnancies at young ages causing an increased risk of premenopausal disease but a lower lifetime incidence. This theory was supported in an analysis of breast cancer risk in the Black Woman’s Health Study, a prospective study of self-reported health and lifestyle issues among African American women subscribers to the magazine Essence . More than 50,000 women completed a follow-up questionnaire, representing more than 214,000 person-years of follow-up between 1995 and 1999, and evaluation of breast cancer events revealed that multiparity was associated with an increased incidence rate ratio (IRR) among women younger than 45 years (IRR, 2.4; 95% confidence interval, 1.1–5.1) but was protective against breast cancer among women aged 45 years and older (IRR, 0.5; 95% confidence interval, 0.3–0.9).
In contrast, investigators from the Women’s Contraceptive and Reproductive Experience (CARE) study found no differences in effect of parity on age-related breast cancer risk. The Women’s CARE study was a population-based, multicenter, case-control study of more than 3000 African American women and nearly 6000 white American women (approximately 50% of both subsets with a history of breast cancer), and this study was specifically designed to analyze the impact of endogenous and exogenous hormonal factors on breast cancer risk in a large, biracial/ethnic dataset. Ursin and colleagues reported that parity was associated with similar degrees of breast cancer risk reduction among younger (age 35–49 years) versus older (age 50–64 years) women in both African American and white American subsets. Risk reduction proportions were 13% and 10% for younger and older white Americans, respectively, compared with 10% and 6% risk reductions for younger and older African Americans.
The increased risk of hormone-receptor-negative disease among African American women and its impact on breast cancer burden is apparent from time trends in population-based incidence and mortality rates. As shown in Fig. 1 , breast cancer incidence rates have fluctuated since the Surveillance, Epidemiology, and End Results data became available in the early 1970s, but the incidence curves for both population subsets always changed in parallel, with incidence rates being lower for the African Americans. Population-based mortality rates, however, demonstrate a different pattern. Mortality rates for African American and white/Caucasian Americans were actually equivalent until the early 1980s, at which point the curves separate, but the mortality disparity is primarily caused by decreasing mortality rates in white/Caucasian American women but largely unchanged mortality rates in African Americans. This pattern is probably explained by advances in systemic therapy for breast cancer, coupled with differences in the race/ethnicity-specific molecular epidemiology of the disease. In the late 1970s, tamoxifen became available as endocrine therapy for breast cancer, but this treatment disproportionately benefited white/Caucasian American patients, in which case frequency of hormone-receptor-positive disease is approximately 2-fold higher than that of African American patients. As shown by Anderson and colleagues, risk of ER-negative breast cancer is higher for African Americans on a population basis, and this pattern persists regardless of age at diagnosis, as well as after stratifying for cases diagnosed as early stage/resectable breast cancer, locally advanced breast cancer, and inflammatory breast cancer. It is likely that outcome disparities have worsened as progress in endocrine therapy for breast cancer has advanced. The armamentarium of endocrine therapies for hormone-receptor-positive disease has expanded to include a variety of aromatase inhibitors for postmenopausal patients, and commercially available genetic profiling analyses (eg, Oncotype Dx Recurrence Score testing) can be performed as a more refined way of identifying ER-positive/node-negative patients with breast cancer who can benefit from adjuvant chemotherapy and endocrine therapy.
In 2007, Carey and colleagues pursued race/ethnicity-related variation further with their landmark report from the Carolina Breast Cancer Study, demonstrating 2-fold higher frequency of tumors that are negative for ER, PR, and HER2/ neu (TNBC) in premenopausal African American patients with breast cancer when compared with others. Others have confirmed this finding using both single-institution tumor registry data and population-based statistics. As with prior studies of ER-negative disease, the increased risk of TNBC is observed among African American women regardless of age at diagnosis and regardless of stage at diagnosis.
Breast cancer in African American women: implications for screening mammography recommendations
The race/ethnicity-associated differences in population-based incidence of TNBC are also relevant to the controversial discussion of breast cancer surveillance and the age at which American women should initiate screening mammography. In November 2009, the US Preventive Services Task Force (USPSTF) published an updated reevaluation of the historic prospective randomized clinical trials conducted 20 to 30 years ago, comparing mammography screening to usual medical care in the absence of screening. The outcomes data from these 7 trials were pooled together and analyzed by a statistical consortium, forming the evidence basis for the USPSTF’s updated recommendations. The statistical experts formulated 2 separate perspectives for interpreting the mammography trials data. The efficiency model studied the number of women that would need to be invited for screening to save 1 life from breast cancer; these age-based numbers were 377 for women aged 60 years and older, 1339 for women aged 50 to 59 years, and 1904 for women aged 40 to 49 years. The longevity model studied the numbers of years of life saved with various mammography screening intervals and demonstrated that more years of life are saved when mammographic screening is initiated at age 40 years. The USPSTF chose to select the efficiency model as being the preferred model for making screening recommendations to American women, and they furthermore opted to choose age 50 years as being the most appropriate age at which to initiate screening, based on their interpretation of the data as suggesting that needing to invite nearly 2000 women to screen to save 1 life is excessive but that needing to invite 1300 women to save 1 life is acceptable. The statistical experts specifically commented in their article that by virtue of their respective study populations, the mammography screening trials almost exclusively reflected the impact of mammography on white/Caucasian women from North America and Europe. These data therefore could not necessarily be generalized to the setting of population subsets such as African American women, in which case the age distribution and disease biology of breast cancer differs. In contrast, the USPSTF article fails to comment on existing disparities in breast cancer burden associated with racial/ethnic identity and the potential influence of delayed mammographic screening on African American women, whereby incidence rates are higher for the younger-aged women and risk of TNBC is disproportionately higher in the younger women as well. As shown by Amirikia and colleagues in a study from the California Cancer Registry, population-based incidence rates of TNBC for African American women in the 40- to 49-year age range is comparable to those of white/Caucasian American women in the 60- to 69-year age range; these data are summarized in Fig. 2 .
