Breast Cancer in Black Women

Methods

The MEDLINE database (1966 to 1995) was searched to identify all English-language articles related to breast cancer in black women; the following medical subject headings were used: African American, black, negro, breast neoplasms, breast diseases, cancer screening, and mammography. In addition, the reference sections of all identified articles were reviewed for additional pertinent sources of information. In the review of the collected literature, results from large cohort and case–control studies were emphasized whenever such studies were available. Information from case series and case reports was used only when other data was not available.

Epidemiology and Risk Factors

The age-adjusted incidence of breast cancer in U.S. women has steadily increased over the past 50 years, and a particularly sharp increase was seen in the mid-1980s, partially because of increased use of mammography. During this period, white women have had a consistently higher risk for developing breast cancer, with an age-adjusted incidence in 1990 of 112.7 new cases per 100 000 women. The age-adjusted incidence for black women in 1990 was 95.8 new cases per 100 000 women [1]. Closer examination of incidence statistics shows that the 20% higher incidence rate in white women is not uniform among all age groups but has followed a crossover pattern in which the risk for breast cancer in young women is slightly but consistently greater in the black population and the risk of developing breast cancer in the middle-aged and elderly populations is substantially higher among white women (Figure 1). As the incidence of breast cancer in young black women has increased faster than the incidence in young white women, the age at which the crossover in incidence occurs has gradually increased over the past 20 years and is now between 45 and 49 years of age. By reviewing the acknowledged risk factors for the development of breast cancer, investigators have proposed several hypotheses to explain the racial variation in breast cancer incidence.

View larger version:

• In this window
• In a new window

Figure 1. Data obtained from Miller and colleagues . Age-specific breast cancer incidence rates, 1986-1990.[1]

Large epidemiologic studies of the U.S. population have identified various potential risk factors for breast cancer. The factors most consistently found to alter the risk for breast cancer development include various fertility and reproductive measurements, family history of cancer, presence or absence of certain benign breast diseases, and several socioeconomic indicators.

Three reports [3-5] have been published that examined risk factors for the development of breast cancer in black women (Table 1). These case–control studies have evaluated 1) factors that are known to contribute to the development of breast cancer in the general population and 2) various other conditions hypothesized to contribute to the differences in incidence rates seen between various population groups. Overall, risk factors for breast cancer in black women appear to be similar to those in the general population: early age at menarche, late age at menopause, nulliparity, late age at first full-term pregnancy, history of breast cancer in first-degree relatives, and a history of benign breast disease. In addition, the possibility that prolonged (> 10 years) use of oral contraceptives may adversely affect breast cancer risk in black women was raised by one large population-based study [5]. Higher level of education, higher socioeconomic status, and higher body mass index—characteristics often identified as associated with elevated risk for breast cancer—have not been as consistently documented to alter risk in black women [6-8].

These well-established risk factors have been examined for their possible contribution to the differences in rates of breast cancer incidence seen between black and white women. Early age at menarche, late age at menopause, late age at first full-term pregnancy, and nulliparity all appear to contribute to a higher risk for breast cancer [9]. The median age at menarche is slightly lower in black than in white girls: 12.5 years compared with 12.8 years [10]. Black women also have an earlier median age of natural menopause, 49.3 years compared with 50.0 years in white women, and are more likely than white women to have had surgical menopause [11]. Finally, the age at first full-term pregnancy is consistently about 2 years less in black women [12]. Although these differences in reproductive values seem small, it has been estimated that they can alter the risk for breast cancer enough to be clinically evident in large populations [13]. It is also interesting to note that the combination of reproductive factors seen in black women results in an expected risk pattern for breast cancer development similar to that documented in the actual incidence data. Early menarche increases the risk for cancer throughout a woman's life; early pregnancy results in a transient (10- to 15-year) increase in the risk for breast cancer, followed by a significant reduction in breast cancer risk [14]; and early menopause contributes to a decreased risk for breast cancer in older women. These factors contribute to higher than expected rates in young women and substantially lower rates after menopause.

The use of oral contraceptives, particularly at a young age and for prolonged periods of time, is a less firmly established reproductive factor affecting the risk for breast cancer [15, 16]. Some studies have also suggested that use before the first pregnancy is a risk factor for breast cancer development. Many characteristics of oral contraceptive use in black women are similar to those seen in white women. Approximately 80% of women of childbearing age in either race have used oral contraceptives at some time in their reproductive life [17]. The average duration of use is similar in black and white women (5 years), and the proportion of long-term (≥ 10 years) users is also similar (15% to 20%). The only notable differences in usage patterns are a consistently earlier age at the start of oral contraceptive use by black women and a lower incidence of use before the first pregnancy by black women [17]. These two characteristics would have potentially opposite effects on the development of breast cancer. Currently, the actual influence of oral contraceptive use on the risk for breast cancer in black women compared with white women awaits further epidemiologic studies that more clearly define the importance of specific usage patterns on cancer risk.

Hormone replacement therapy has also been associated with an increased risk for the development of breast cancer, especially in long-term users [18-20]. Demographic data on the patterns of use of postmenopausal hormone replacement therapy are not available, making it impossible to comment on racial differences in hormone use and their possible contribution to breast cancer risk.

History of breast cancer in a first-degree relative increases a woman's risk for breast cancer 1.5- to 3.0- fold. This excess risk has been determined by studying groups of predominantly white women. Two case–control studies [21, 22] that analyzed black women with breast cancer (identified through the Surveillance, Epidemiology, and End Results program of the National Cancer Institute) found an approximately twofold increase in the risk for breast cancer in black women who had a first-degree relative with breast cancer. This finding was consistent with what would be predicted in a similar group of white women. No further studies have been reported, and no racial or demographic information is yet available on the prevalence of germline mutations in the recently discovered genes associated with familial breast cancer (BRCA1 and p53) in black women [23, 24], although one black family with a high prevalence of breast cancer has been associated through linkage analysis with a probable mutation of the BRCA1 gene [25].

Finally, the presence of certain benign breast lesions has also been associated with an increased risk for the development of breast cancer in the general population. Specifically, proliferative disease without atypia is generally associated with a relative risk of 1.5 to 1.9 compared with the risk of control populations, and proliferative disease with atypical hyperplasia is associated with a relative risk of 2.5 to 4.5 [26, 27]. Unfortunately, little is known about the pattern of benign breast disease in black women. Fibroadenomas are diagnosed more frequently in black women and at an earlier age (15 to 30 years of age); the frequency of proliferative and nonproliferative benign breast disease in black women is unknown [28-31].

Early Detection

The early detection of breast cancer has traditionally involved three components of an age-specific examination: monthly self-examination of the breast, yearly clinical examination of the breast by a health care provider, and yearly mammography. In recent years, several large studies have examined the demographic characteristics of breast cancer screening participation, including knowledge of appropriate screening behavior, use of screening tests, and barriers to receiving the recommended breast cancer screening. These studies have helped to identify the particular needs of the black community in the United States.

Monthly breast self-examination has been recommended for all women 20 years of age and older, although the benefit of this on mortality from breast cancer has never been documented in a randomized trial [32]. Reported rates of monthly breast self-examination range from 25% to 70%; the lower numbers come from reports in the early 1980s, and the higher figures come from more recent studies [33-35]. The use of breast self-examination by black women appears to be at least as high as that in white women, and in some studies it is higher (70% compared with 62% in the National Health Interview Survey from 1987 [34]). None of these studies has assessed knowledge of appropriate technique or training in breast self-examination; both of these factors are clearly related to the accuracy of this intervention [36].

Clinical breast examination by a health care professional has also not been studied as a single intervention in a randomized trial. It has, however, been a component of many of the mammographic screening trials that have documented a survival benefit associated with the use of breast cancer screening and has thus become an important part of the recommended cancer screening guidelines. Recent studies assessing the use of yearly clinical breast examination by women older than 40 years of age have found that 40% to 70% of women interviewed comply with this guideline [34, 35, 37-40]. Race has not had an independent effect on the likelihood of obtaining a clinical breast examination, but women with lower incomes and less education have consistently had lower screening rates. This effect is particularly important for black women, because they may be disproportionately represented in these groups.

Yearly mammography in women older than 50 years of age is the cornerstone of the breast cancer screening effort in the United States. Mammography use has increased substantially during the last decade; the most recent surveys document an increase among both black and white U.S. women. Six studies [34, 35, 37, 41-43] in the last 7 years have assessed the effect of race on the use of mammography. Four of the six found no difference in the use of mammography between black and white women. The two reports that did find lower use of mammography among black women did not control for income level, which is a clear discriminating factor in most studies of mammography use. Thus, the lower rates of use identified in black women may result from financial barriers more than from societal or health care barriers.

Demographic characteristics that predict low rates of mammography use include increasing age, low income level, low level of education, rural residence, and lack of health insurance [37, 39-41]. In addition to these documented demographic findings, surveys of the target population for mammography have identified several important barriers to the regular use of mammography. In black women, the most commonly cited reasons for not having mammography are a belief that mammography is not needed if no clinical breast problems are present, the failure of a physician to recommend the procedure, the cost of the procedure, and a knowledge deficit about the increasing risk for breast cancer with age [34, 42-46]. These barriers are the same as those mentioned by white women, although black women more frequently cite lack of physician recommendation as an important reason for not obtaining yearly mammography.

Two clinical trials designed to increase breast cancer screening use have specifically targeted black women. Mandelblatt and colleagues [47] used a nurse practitioner to recruit all women attending a public general internal medicine clinic who were 65 years of age or older to participate in breast, cervical, and colorectal cancer screening during their normally scheduled clinic visit. Seventy-one percent of the women approached agreed to participate, and two thirds of these women had mammography. Because only 8% of the eligible population had received a recent mammographic examination, this showed a substantial increase in recommended screening behavior. In a second study by Skinner and colleagues [48], women attending a family practice clinic were interviewed about their breast cancer screening behavior and beliefs. A letter individually tailored to their responses was then sent to half of the women; the other half received a standard form letter describing the importance of mammography. Although the tailored letter did not have a greater effect than the standard letter on the group as a whole, subgroup analysis found that black women and women with lower incomes responded positively to the tailored message. These two studies suggest that intervention by health care providers can significantly improve short-term compliance with breast cancer screening guidelines by black women. Studies examining adherence to repeated screening at the recommended intervals have not been reported.

Clinical Presentation

Large population-based studies have repeatedly shown that black women have breast tumors at a more advanced stage at the time of diagnosis. Black women have larger primary tumors, a higher incidence of spread to the axillary lymph nodes, and more distant metastatic disease than white women [1, 49-55] (Figure 2). The most striking difference is in the proportion of tumors that are node negative—53% of white and 42% of black women in the United States present with disease confined to the breast [1]. Because lymph node involvement is the most important indicator of prognosis, this discrepancy becomes particularly important when the previously mentioned differences in survival rates are considered. It is important to note, however, that the absolute number of axillary lymph nodes involved with cancer in women with regional disease is not clearly different in black and white women [51, 53, 56, 57]. Although these findings have been consistent over time, a trend can be seen in the last 15 years toward the diagnosis of disease at an earlier stage in black women, possibly as a result of increased screening activity. The incidence of breast tumors that are not invasive (carcinoma in situ) or that are invasive but small (< 2 cm) and confined to the breast, has increased by 100% to 200% in both blacks and whites during this time, although white women still present more frequently with early-stage disease [57-59].

View larger version:

• In this window
• In a new window

Figure 2. Differences in trend and all individual differences in stage of disease are significant ( < 0.001). Data obtained from Miller and colleagues . Stage distribution of breast cancer, 1983-1987.P[1]

Breast cancer histology is similar in black and white women; most cases consist of infiltrating ductal carcinoma. Several studies [31, 51, 60-62] have found a slightly higher incidence of medullary carcinoma in black women than in white women (7% compared with 3%), but this has not been a consistent finding. Most comparative studies have also found a higher incidence of poorly differentiated tumors in black women and an increased frequency of nuclear atypia, higher mitotic activity, and tumor necrosis [50, 62].

The higher frequency of poorly differentiated tumors in black women is compatible with the finding of a higher frequency of hormone receptor-negative tumors in black women, both in the United States and in South Africa [51, 54-56, 63-66]. Most investigators find that 60% to 80% of white women with breast cancer have estrogen receptor-positive tumors; the corresponding number for black women is 40% to 60%. Similarly, black women are less likely to have progesterone receptor-positive breast tumors, although this characteristic has been less widely studied [31, 55, 66]. Because postmenopausal women are more likely to have hormone receptor-positive tumors, some of the reported racial differences in estrogen and progesterone receptor levels could be the result of the younger median age of black women with breast cancer. Examination of this factor shows a persistent difference in hormone receptor levels between black and white women when they are separated by menopausal status [51, 67]. Both premenopausal and postmenopausal black women have a lower frequency of estrogen receptor-positive tumors than do corresponding white women.

Elledge and colleagues [55] have examined some of the more recently identified markers of breast tumor biology, including DNA ploidy, S-phase fraction (an index of tumor proliferation), HER2/neu protein levels (an oncogene product), and p53 protein accumulation (the product of a tumor suppressor gene). Their retrospective study involved 6678 white, black, and Hispanic women and found that the only difference between black and white women was in the S-phase fraction. White women had a significantly lower S-phase fraction than either the black or Hispanic women. This finding is not unexpected given the higher frequency of poorly differentiated tumors in black women.

Attempts to explain the above-described differences in clinical presentation and pathology of breast tumors have centered on an exploration of the effect of socioeconomic status on tumor stage at diagnosis and on tumor biology. It has been documented that women of lower socioeconomic standing have more advanced breast cancer at the time of diagnosis [54, 68-71]. Uninsured women and women who rely on public assistance to finance their medical care appear to have greater barriers to accessing that care for nonemergent problems [72]. As a result, some investigators have found that poorer women with breast cancer have a longer symptomatic period before seeking medical attention [73, 74]. Because black women are disproportionately represented in the lower socioeconomic strata, economic factors may explain the advanced disease in this population. Unfortunately, few studies have attempted to address this issue scientifically, and the results of these studies have generally conflicted [54, 68, 69]. It does appear, however, that black women seek medical attention for breast symptoms later than white women, even when socioeconomic status is considered [74]. This difference in symptom duration between black and white women is small (2 days) and seems unlikely to explain the substantial difference in stage at diagnosis that is seen between black and white women. Finally, socioeconomic status, through its effect on nutrition and environmental exposures, may also affect breast tumor histology. Chen and colleagues [62], who examined this issue in black women, found that socioeconomic or lifestyle factors could not explain the higher incidence of poorly differentiated breast cancers in black women.

Treatment

A few studies have evaluated the treatment of breast cancer in black women. Several studies from the 1980s [52, 75] suggested that black women were less likely than white women to have surgical therapy for local or regional disease. These studies failed to adequately consider local tumor characteristics, such as skin involvement and direct extension of the tumor outside of the breast, that could have contributed to treatment decisions. More recently, the Black/White Cancer Survival Study Group [76] reported that among women with equivalent cancer stage, black women were just as likely as white women to have surgical therapy as part of their primary treatment plan. The group found that black women were less likely to have breast-conserving surgery and more likely to have a modified radical mastectomy. However, when tumor size and comorbid conditions such as diabetes mellitus and hypertension were considered, race was not found to be a significant predictor of breast-conserving surgery. The results of treatment with conservative surgery followed by radiation therapy in black women appear to be similar to those seen in the population as a whole. No difference has been documented in local failure rates between black and white women (5% and 8%, respectively), and cosmetic results have been considered good to excellent in more than 80% of women of both races [77].

The use of systemic adjuvant therapy, either chemotherapy or endocrine therapy, has generally not been found to vary significantly according to race, although the data in this area are limited [50, 55, 76, 78]. Even less information is available about the efficacy of systemic therapy in preventing relapse or improving survival rates in black women with breast cancer. One study [79], presented only in abstract form, suggested that black women enrolled in Eastern Cooperative Oncology Group chemotherapy studies for breast cancer had worse survival than matched controls, but not enough information was presented to adequately analyze the reported findings. Similarly, the Piedmont Oncology Group [80] found that although the response of black women with metastatic breast cancer to chemotherapy was similar to that of white controls, the survival rate of black women was significantly shorter. Comorbid conditions and socioeconomic variables that may have affected survival were not analyzed.

Outcome

The relative 5-year survival of black women with breast cancer in 1990 was 64%; the corresponding rate for white women was 80% [1]. Although a steady improvement in the 5-year survival rate has been documented in both races since 1960, the difference in the proportion of black and white women surviving has remained relatively stable. This substantially higher case-mortality rate for black women has resulted in a higher age-adjusted breast cancer mortality rate for black women in the United States, despite the lower incidence of cancer in this population (Figure 3). More advanced stage at diagnosis, differences in tumor biology, treatment differences, sociodemographic issues, and the presence of comorbid illnesses have all been suggested as factors contributing to the poorer survival of black women with breast cancer.

View larger version:

• In this window
• In a new window

Figure 3. Data obtained from Miller and colleagues . Age-adjusted breast cancer mortality rates.[1]

Tumor stage is the most important determinant of outcome in women with breast cancer, and it is well documented that advanced stage disease is more common in black women. Therefore, many investigators have felt that the poorer survival rates seen in black women result from the difference in the stage distribution of breast tumors at the time of diagnosis. Adjustment for stage of disease generally narrows the survival difference but frequently does not eliminate it [49, 50, 55, 81, 82]: A disparity in survival rates tends to persist when women of the same stage are compared (Figure 4). Several groups [50, 55, 71, 81] have found that black and white women with disease in a very early stage (lymph nodes not involved with cancer) and women with metastatic disease have similar survival; those with intermediate prognosis disease (local or regional disease and lymph nodes involved with cancer) continue to show a disparity in survival. This raises the possibility that treatment differences may also play a role in the outcome disparity, because adjuvant treatment has its greatest absolute effect on women with local or regional disease and lymph nodes involved with cancer. As discussed earlier, little information is available about treatment patterns in black women, especially about the use of adjuvant chemotherapy and the efficacy of such treatment in this population. In one large population-based study [82], treatment variation did not substantially contribute to the survival differences seen between black and white women.

View larger version:

• In this window
• In a new window

Figure 4. Measurements in centimeters refer to the size of the primary tumor. N+ = lymph nodes involved with cancer; N− = lymph nodes not involved with cancer. Data obtained from Swanson and colleagues . Five-year relative survival rate by cancer stage at diagnosis.[57]

The previously described differences in breast tumor biology seen in black women could also contribute to the disparity in survival. Women with poorly differentiated tumors, hormone receptor-negative tumors, or tumors with a higher S-phase fraction have all been shown to have a worse prognosis than women with tumors without these characteristics [83-90]. The Black/White Cancer Survival Study [82], the most comprehensive study of racial survival differences in breast cancer to date, found that tumor biological characteristics (tumor grade and hormone receptor status) were second only to tumor stage in contributing to the survival difference. In addition, several groups [56, 91] have found that estrogen receptor status is a particularly strong prognostic indicator in black women, especially postmenopausal black women. The few studies [55, 92-94] that have evaluated more recently identified markers of tumor biology that may affect prognosis have reported conflicting results, and further information is clearly needed before the importance of these markers as prognostic indicators in black women can be assessed.

Underlying medical conditions could adversely contribute to overall health or prevent the delivery of optimal therapy for breast cancer, and they have also been implicated in the worse outcome of black women with breast cancer. Several investigators [50, 95, 96] have found an unusually high risk for death from other causes in black persons or uninsured persons with cancer. The Black/White Cancer Survival Study Group found that black women with breast cancer were significantly more likely to have serious underlying medical problems (diabetes, hypertension, heart disease, lung disease, or kidney disease) than a matched population of white women. Sixty-seven percent of white women were free of underlying medical problems compared with 44% of black women. Black women were also more likely to be overweight and to have markers of poor nutritional status (high body mass index and low serum albumin levels, hemoglobin levels, and lymphocyte counts) [53]. Most investigators have found that such comorbid conditions contribute modestly to the worse prognosis for survival in black women with breast cancer [50, 53, 82].

The most difficult issue to clarify is the contribution of socioeconomic status to the worse survival seen in black women with breast cancer. It is not hard to understand why economically disadvantaged women may present with disease at a more advanced stage, may be at higher risk for receiving suboptimal therapy, are more likely to have important underlying comorbid medical conditions, and may even have some differences in tumor biology. General conclusions that can be drawn from the existing data include a strong association between race and socioeconomic status and an apparent correlation of socioeconomic status with worse survival in persons with cancer [53, 56, 71, 82, 95, 97, 98]. Of the five studies that have tried to directly measure the effect of socioeconomic factors on the worse survival of black women with breast cancer, two [56, 97] have found that socioeconomic factors combined with tumor stage completely explain the racial difference in survival, and the other three [73, 82, 98] found that although social and economic factors contributed to survival differences, they could not completely explain them. It appears that the worse prognosis of black women with breast cancer cannot be explained by any single factor but results from a complex interaction of many issues, including tumor stage, tumor biology, comorbid conditions, and socioeconomic variables.

Conclusion

Breast cancer has a major effect on the health of black women in the United States. The incidence of breast cancer is lower in black women than in white women, predominantly because of a lower risk in the elderly that may be explained by variations in reproductive history. However, once cancer develops, black women have a dramatically worse prognosis; their 5-year survival rate is 64%, and that of white women is 80%. The poor prognosis may result because of the more advanced cancer at diagnosis in black women, because of the differences in tumor biology between black women and white women, because of socioeconomic factors, and possibly because of differences in treatment.

To close the survival gap between black and white women, a combination of further research and focused intervention is necessary. In the research arena, it is particularly important to develop a better understanding of the treatment options available to all black women, the choices these women make about therapy, and the efficacy of therapy in black women and white women. Because the tumors of black women are more histologically aggressive and are diagnosed at a more advanced stage, appropriate and intensive adjuvant therapy will be vital in attempts to improve survival in these patients. The methods used to effectively deliver this therapy to economically disadvantaged women and women with serious comorbid conditions also needs more attention.

Continued efforts to encourage early detection and treatment are important, especially those that focus on poor women, the elderly, and women without a consistent source of health care—all groups that have been recognized as less likely to receive yearly mammography. Black women are disproportionately affected by the economic barriers to early detection and should also be a focus of early detection interventions. It is hoped that, with a combination of early detection programs, more effective ways to deliver high-quality care to the economically disadvantaged, and a better understanding of the determinants of breast cancer biology, the next century will see substantial gains in the survival of black women with breast cancer.