• breast neoplasms;
  • receptors;
  • estrogen;
  • receptors;
  • progesterone;
  • human epidermal growth factor receptor 2/neu;
  • ethnic groups;
  • health disparities;
  • survival analysis


  1. Top of page
  2. Abstract


Breast cancers that are negative for estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) (triple negative [TN]) have been associated with high-grade histology, aggressive clinical behavior, and poor survival. It has been determined that breast cancers that are negative for ER and PR but positive for HER2 (double negative [DN]) share features with TN breast cancers. In this report, the authors quantified the contribution of HER2 as well as demographic and tumor characteristics to the survival of women with TN tumors, DN tumors, and other breast cancers (OBC).


In total, 61,309 women who were diagnosed with invasive breast cancer between 1999–2004 were identified in the California Cancer Registry. Demographic and tumor characteristics of women with TN tumors were compared with those from women with DN tumors and women with OBC. A compound proportional hazards regression analysis (PHPH) (a generalization of the Cox proportional hazards model) was used to model these characteristics.


Women with TN tumors were younger, African American, Hispanic, and of lower socioeconomic status (SES), whereas women with DN tumors were slightly older; African American, and Asian/Pacific Islander. Women with TN and DN tumors presented with larger, higher grade, and higher stage than women with OBC. Survival among women with TN tumors was poorer compared with that among women with OBC but was nearly the same as that of women with DN tumors. Results of the regression analysis indicated that disease stage, tumor grade, SES, and race/ethnicity were significant risk factors for survival. Negative ER and PR status was associated with an increased risk of death. There was a small but significant difference in both long-term and short-term survival patients who had TN tumors compared with patients who had DN tumors.


Patients with TN tumors shared many clinical, demographic, and tumor features and had survival that was very similar survival to that of patients with DN tumors, and survival for both groups contrasted greatly with survival for patients with OBC. Disease stage, tumor grade, SES, race/ethnicity, negative ER and PR status, rather than negative HER2 status, were risk factors for survival. Cancer 2008. © 2008 American Cancer Society.

Breast cancer is the most common cancer among women in California. Despite decreased mortality, breast cancer remains a significant cause of cancer death.1 Breast cancer incidence and mortality vary greatly according to demographic factors, such as age, race/ethnicity, and socioeconomic status (SES).2–4 These demographic factors produce wide disparities in survival, including lower rates for younger women,5–9 ethnic minority women,10, 11 and less affluent women.12, 13 Breast cancer survival depends on prognostic factors like tumor size, histologic grade, and tumor receptor status, regardless of treatment.14 These prognostic factors are distributed differentially in the population by age and race/ethnicity.15–19

Human epidermal growth factor (HER), estrogen, and progesterone regulate cell growth, apoptosis, and differentiation. Tumor cell expression for receptors of HER2 is considered a prognostic factor in breast cancer19–21 and is one of the risk-stratification features of the St. Gallen International Consensus Guidelines.22 Overexpression of HER2 is associated with worse clinical outcomes in both patients with lymph node-negative cancer and patients with lymph node-positive cancer.23, 24 These patients tend to have more aggressive disease, which leads to shortened overall survival.25 However, it has not been demonstrated that HER2-negative tumors predict a poor prognosis in women with breast cancer in the absence of other tumor marker data.26 Endocrine sensitivity, as assessed by the expression of estrogen receptor (ER) and/or progesterone receptor (PR), is an important prognostic and predictive factor. Patients who are negative for these receptors have a worse prognosis, at least in the first 5 to 10 years after treatment.27, 28 However, these tumor markers are invaluable as predictors of response to therapy: ER and PR predict response to endocrine therapy,27 such as tamoxifen, aromatase inhibitors, and ovarian suppression; and HER2 overexpression predicts for response to targeted anti-HER2 therapy.29, 30

Breast cancer is a heterogeneous disease with a wide spectrum of clinical, pathologic, and molecular features.31–33 Gene-expression profiling studies have identified at least 4 categories of breast cancer.31, 33 These molecular categories correlate with biomarker phenotypes (luminal A is ER-positive and/or PR-positive/HER2-negative, luminal B is ER-positive and/or PR-positive/HER2-positive, HER2 overexpression is ER-negative/PR-negative/HER2-positive, and basal-like is ER-negative/PR-negative/HER2 negative)33, 34 and have distinct differences in disease progression, prognosis, and survival.33 Specifically, the basal-like subtype, known as triple-negative (TN), is associated with aggressive histology, poor clinical outcomes,33, 35 and BRCA1-related breast cancer.36, 37 TN breast cancer is more prevalent among young, premenopausal, African-American,33, 38, 39 and Hispanic women39 and is a strong contributing factor to the poor clinical outcomes in these women.

In an earlier investigation, we observed that patients with ER-negative/PR-negative/HER2-negative (TN) breast tumors or TN breast cancers had demographic, clinical/pathologic features, and survival that were similar to those of patients with ER-negative/PR-negative/HER2-positive breast cancers, that is, the HER2-overexpressed molecular subtype or double negative [DN].40 We observed that overall survival was nearly identical for women with the TN and DN phenotypes, suggesting that HER2 played a minimal role in survival. The current study extends our earlier observations by quantifying the contribution of HER2, along with clinical, demographic, and other tumor characteristics, to the survival of women with TN and DN tumors and of women with other breast cancers (OBC).


  1. Top of page
  2. Abstract

Patient Identification

Patients who were included in the current analyses were identified by using the California Cancer Registry (CCR), a population-based registry composed of 8 regional registries that collect cancer incidence and mortality data for the entire population of California. In 1985, California state law mandated the reporting of all newly diagnosed cancers in California, and statewide implementation began January 1, 1988. Cases are reported to the Cancer Surveillance Branch of the California Department of Health Services from hospitals and any other facilities that provide care or therapy to cancer patients who reside in California.41 For this study, women with primary, invasive breast cancer (International Classification of Diseases for Oncology, 3rd edition [ICDO-3] sites C50.0-C50.9)42 who were diagnosed between January 1, 1999 and December 31, 2004 and were reported to the CCR as of October 2006 were included.

The CCR requires the collection of tumor marker information from the medical record on the ER and PR status of breast cancers diagnosed on or after January 1, 1990 and requires data on the HER2 status of breast cancers diagnosed on or after January 1, 1999.41 Issues associated with the collection and recording of hormone receptor data have been described elsewhere.43 ER and PR status are recorded according to the pathologist's interpretation of the assays. ER and PR are considered negative if immunoperoxidase staining of tumor cell nuclei is <5%. ER and PR status also may be determined by examining cytosol protein. ER is considered negative if there is <3 fmol/mg of cytosol protein, and PR is considered negative if there is <5 fmol/mg of cytosol protein.44 HER2 is assessed through immunohistochemistry (IHC) or fluorescence in situ hybridization (FISH). IHC is scored on a qualitative scale from 0 to 3+, based on interpretation of staining intensity, with 0 and 1+ classified as negative, 2+ classified as borderline, and 3+ classified as positive.45 FISH is scored on a quantitative scale with <2 copies of the HER2 gene classified as negative and with ≥2 copies classified as positive.46

Patients were categorized into distinct groups based on their tumor marker status. Patients who had tumors that were negative for ER, PR, and HER2 were referred to as TN, patients who had tumors that were classified as negative for ER and PR but positive for HER2 were referred to as DN, and patients who had tumors that were classified as neither TN nor DN were referred to as OBC. Patients who had at least 1 unknown tumor marker and those with borderline results were excluded from these analyses.


Race/ethnicity was classified into 4 mutually exclusive categories of Asian-Pacific Islander (API), Hispanic, non-Hispanic black, and non-Hispanic white. Race/ethnicity was based on information obtained from the medical record, which can be derived from patient self-identification, assumptions based on personal appearance, or inferences based on the race/ethnicity of the parents, birthplace, surname, or maiden name. Hispanic ethnicity was based on information from the medical record and computerized comparisons to the 1980 United States Census List of Hispanic Surnames. Patients who were identified as Hispanic on the medical record or patients who were identified as white, black, or of unknown race with a Hispanic surname were classified as Hispanic.47 Patients with unknown race/ethnicity, age, or sex were excluded from these analyses.

SES was assigned based on patient's census block group (2000 United States Census) derived from their address at the time of initial diagnosis as reported in the medical record. This SES variable is an index that uses education, employment characteristics, median household income, proportion of the population living 200% below the Federal Poverty Level, median rent, and median housing value of census tract of residence for case and denominator population. A principal components analysis was used to identify quintiles of SES ranging from 1 (the lowest) to 5 (the highest).48

Stage at diagnosis was collected from the patient's medical record and was coded according to the American Joint Commission on Cancer (AJCC) Cancer Staging Manual, 6th edition.49 The CCR collected Surveillance, Epidemiology, and End Results (SEER) Extent of Disease (EOD) data for breast cancer cases diagnosed from 1988 through December 200350 and, in 2004, began collecting Collaborative Staging data items.51 EOD was converted to AJCC stage at diagnosis by using SEER guidelines.52 For some of these analyses, stages III and IV at diagnosis were combined, and patients who had unknown disease stage at diagnosis were omitted. Tumor grade was collected from the medical record and was coded according to ICDO-3.42

Statistical Analysis

Statistical significance for pairwise comparisons was determined by using the Wilcoxon signed-rank test (to compare medians) and the test for independent proportions.53 Comparisons of survival among patients' tumor marker phenotypic groups and stage at diagnosis were performed by using the 5-year cumulative relative survival of all available years of data. Counts, 5-year cumulative relative survival, and 95% confidence intervals (95% CIs) were calculated by using SEER*Stat software (version 6.2.4; Silver Spring, Md).

Because previous studies indicated that the proportional hazards (PH) model is positioned poorly to reproduce the effects of ER and PR status on breast cancer survival, a more general survival model was selected for the current analysis.54 Survival was modeled using a compound regression analysis method (PHPH) regression model, a generalization of the Cox PH model that was developed for use when data indicate a possibility of long-term survival and nonproportional hazards. Departures from the proportionality assumption indicate that hazard ratios (HRs) vary with time and are introduced in the form of short-term effects.55 A pure short-term effect corresponds to survival curves that converge with time and long-term survivors who have no differences with respect to variables included in the model. The PHPH model is constructed by a composition of 2 separate PH models: long-term hazard and short-term hazard. Both models are defined by using limiting behavior of survival and hazard functions; hence, there is no sharp cutoff point between short-term and long-term survival. The PHPH model55 describes the survival function S(t|z) as follows:

  • equation image(1)

where F(t) is a baseline survival function (“proper” means 1 with zero chance of long-term survival), and θ(z) and η(z) are predictors that depend on the variables of interest z (covariates). In the equation above, η(z) describes the short-term survival effects. Thus θ(z)/θ(0) and η(z) represent the relative risks (RR) for long- and short-term survival, respectively. The diversity of responses reproduced by the PHPH model includes crossing survival curves characterized by counteracting short-term and long-term effects. A universal estimation algorithm, the so-called quasi-EM procedure, has been developed to provide inference for such models.56 It should be stressed that “long-term” survival in the semiparametric model1 is a mathematical term used to define risks operating at the end of the follow-up range. The follow-up range in the current study is 6 years, which, by the standards of breast cancer, cannot be identified with the time when the risk becomes negligible, because many 6-year survivors still are expected to fail.


  1. Top of page
  2. Abstract

We identified 110,163 incident cases of invasive breast cancer from 1999 to 2004 from the CCR. Of these selected cases, 61,309 had definitive results for all 3 tumor markers on record. A comparison of the cases that were included in these analyses with cases that were omitted has been discussed elsewhere. We observed no significant differences in results when omitted cases were added to the analyses.39

Table 1 compares the demographic and tumor characteristics of TN patients with both OBC patients and DN patients. The median age at diagnosis of patients with TN tumors was significantly younger than the median age of OBC patients (54 years vs 60 years; P ≤ .001). Patients with TN tumors were significantly more likely to reside in areas of lower SES than OBC patients (P < .001). A significantly higher proportion of non-Hispanic black and Hispanic patients had tumors that were TN (P ≤ .001 for both). TN tumors were significantly more likely to be larger (P ≤ .001), poorly differentiated (P ≤ .001), or anaplastic (P ≤ .01), and they were significantly more likely to present as stage II or III disease (P ≤ .001 and P ≤ .01, respectively).

Table 1. Selected Demographic and Tumor Characteristics, Female Invasive Breast Cancers by Tumor Marker Status, California, 1999–2004
  1. OBC indicates other breast cancers; DN, double negative (estrogen receptor [ER] negative, progesterone receptor [PR] negative, and HER2 positive); TN, triple negative (ER negative, PR negative, and HER2 negative); SES, socioeconomic status; AJCC, American Joint Committee on Cancer.

  2. Red lines present a comparison of TN and OBC. Blue lines present a comparison of TN and DN.

Total no of patients48,8511008022100 4436100 
Median age at diagnosis, y60 54 ≤.00155  
Age group at diagnosis, y
 1 (Low)46519.5107713.4≤.00159613.4 
 3987420.2168721 93321 
 412,05924.7186623.3 98522.2 
 5 (High)14,87930.5192224≤.001109024.6 
 Non-Hispanic white35,77673.2492261.4≤.001270060.9 
 Non-Hispanic black20944.386010.7≤.0013187.2≤.05
 Asian-Pacific Islander44479.16908.6 56312.7≤.05
 Other/unknown3150.6570.7 250.6 
Median tumor size, mm16 22 ≤.00123 ≤.05
 Histologic grade
 Unknown33686.93834.8 2696.1 
AJCC stage at diagnosis
 IV146833334.2 2796.3 
 Other/unknown196343214 2265.1 

Patients with DN tumors and patients with TN tumors were very similar and had few differences that reached statistical significance. The median age of patients with DN tumors was only slightly older than the median age of patients with TN tumors. There were no significant differences in SES between these 2 groups. Race/ethnicity was distributed similarly, with the exception of API women, who were significantly more likely to have DN tumors (P ≤ .05). Both DN tumors and TN tumors tended to be approximately the same size and tended to be poorly differentiated at diagnosis, and significantly fewer DN tumors presented as grade 3 (P ≤ .001). Although the distribution by stage of diagnosis was similar for DN tumors and TN tumors, a significantly larger proportion of patients with DN tumors presented with stage III disease (P ≤ .01).

The 5-year cumulative relative survival rates by tumor maker status were 76.2% (95% CI, 74.4–78%) for women with TN tumors, 75.9% (95% CI, 73.6–78.3%) for women with DN tumors, and 94.2% (95% CI, 93.6–94.8%) for women with OBC (data not shown). Figure 1 illustrates 5-year cumulative relative survival by tumor marker status and by stage. Relative survival patterns of patients with TN tumors and patients with DN tumors appeared to be similar, and both were in sharp contrast to the survival patterns in patients with OBC. Overall, relative survival among women with TN tumors generally was poorer than that for women with OBC but was nearly the same as that for women with DN tumors. Among patients with stage I disease, the relative survival of patients with TN tumors was significantly shorter than that for patients with OBC (P = .0032) but did not differ significantly from that of DN patients. Among patients with stage II and III/IV disease, the relative survival of patients with TN tumors was significantly shorter than the relative survival both for patients with OBC (P < .001 for both) and for patients with DN tumors (P = .01 and P < 0.001, respectively).

thumbnail image

Figure 1. Five-year relative cumulative survival for invasive female breast cancers by tumor marker phenotype and American Joint Committee on Cancer stage at diagnosis: California, 1999–2004. Double negative (DN) indicates tumors that were estrogen receptor (ER) negative, progesterone receptor (PR) negative, and human epidermal growth factor receptor 2 (HER2) positive; triple negative (TN), tumors that were ER negative, PR negative, and HER2 negative.

Download figure to PowerPoint

The relative contribution of demographic and tumor characteristics were modeled by using PHPH regression (Table 2). In the PHPH models, breast cancer survival data were fit with disease stage; tumor grade; ER, PR, and HER2 status; SES; and race/ethnicity as explanatory variables for both long-term and short-term probability of death while controlling for age at diagnosis. In the full model, not surprisingly, the long-term probability of death increased significantly with progressive stage at diagnosis and tumor histology (P < .0001). High SES was associated with better long-term survival. For SES level 3 versus 1, the long-term difference was characterized by an RR of 0.82 (P = .038); and, for SES level 5 versus 1, the difference was characterized by an RR of 0.75 (P = .003). Similar effects were observed for short-term survival, although only SES level 5 versus 1 demonstrated a significant difference (HR, 0.8; P = .003).

Table 2. Compound Proportional Hazards (Long-term Hazard and Short-term Hazard) Regression Analysis of Survival by Selected Tumor and Demographic Characteristics in Women With Invasive Breast Cancer, California, 1999 to 2004
CharacteristicModel 1Model 2
  • HR indicates hazard ratio; 95% CI, 95% confidence interval; SES, socioeconomic status; NH, non-Hispanic; API, Asian-Pacific Islander; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TN, triple negative (ER negative, PR negative, and HER2 negative); DN, double negative (ER negative, PR negative, and HER2 positive).

  • *

    For all variables combined (stage at diagnosis, histology, SES, race/ethnicity, receptor status, and phenotype), the log (-log) of baseline probability of long-term survival was 0.03; for all subgroups combined, the log (-log) of baseline probability of long-term survival was 0.02.

Long-term risk
 Stage at diagnosis
  Regional vs localized3.272.83–3.78<.00013.312.86–3.83<.0001
  Distant vs localized14.9012.50–17.76<.000115.7313.17–18.78<.0001
 Tumor histology (differentiation)
  Moderate vs well1.971.42–2.73<.00011.921.39–2.66<.0001
  Poor vs well3.692.68–5.06<.00013.762.74–5.16<.0001
  Undifferentiated vs well3.452.36–5.05<.00013.392.32–4.94<.0001
  2 vs 1 (Low)1.030.85–1.25.7451.030.85–1.24.789
  3 vs 1 (Low)0.820.68–0.99.0380.820.68–0.98.031
  4 vs 1 (Low)0.860.71–1.05.1320.880.72–1.07.214
  5 vs 1 (Low)0.750.62–0.91.0030.750.62–0.91.004
  NH black vs NH white1.321.09–1.60.0051.321.09–1.61.005
  Hispanic vs NH white0.950.76–1.18.6300.970.77–1.21.767
  API vs NH white1.110.94–1.30.2161.100.93–1.29.265
 Tumor receptor status
  ER negative vs positive1.481.26–1.73<.0001   
  PR negative vs positive1.201.02–1.41.030   
  HER2 negative vs positive0.980.87–1.10.715   
  DN vs OBC   1.481.26–1.73<.0001
  TN vs OBC   1.741.52–2.00<.0001
  TN vs DN   1.171.00–1.37.045
Short-term risk
 Stage at diagnosis
  Regional vs localized1.311.08–1.58.0061.291.07–1.56.009
  Distant vs localized3.042.41–3.83<.00012.892.28–3.65<.0001
 Tumor histology (differentiation)
  Moderate vs well0.980.65–1.49.9281.050.69–1.59.819
  Poor vs well1.080.72–1.61.7091.160.78–1.71.476
  Undifferentiated vs well1.280.78–2.08.3321.450.89–2.34.132
  2 vs 1 (Low)0.970.76–1.25.8300.970.75–1.25.812
  3 vs 1 (Low)1.060.83–1.35.6411.090.85–1.38.507
  4 vs 1 (Low)0.820.64–1.06.1340.790.61–1.030.078
  5 vs 1 (Low)0.800.62–1.03.0870.800.62–1.03.080
  NH black vs NH white0.980.76–1.26.8650.980.75–1.26.846
  Hispanic vs NH white0.830.62–1.11.2190.820.61–1.10.178
  API vs NH white0.870.71–1.08.2070.890.72–1.10.272
 Tumor receptor status
  ER negative vs positive1.701.38–2.08<.0001   
  PR negative vs positive1.411.14–1.73.001   
  HER2 negative vs positive1.181.01–1.38.038   
  DN vs OBC   1.781.45–2.19<.0001
  TN vs OBC   2.271.90–2.72<.0001
  TN vs DN   1.291.04–1.58.018
Log (-log) of baseline probability of long-term survival*0.04 <.00010.030.02–0.04<.0001

Non-Hispanic black women had worse survival compared with non-Hispanic white women (long-term RR, 1.32; P = .0005). The effect of race followed the PH assumption, indicating no short-term departures (P = .865). Although that may be meaningful to patient survival, differences in many of demographic and tumor characteristics were not statistically significant for short-term survival. The exception was disease stage at diagnosis, which maintained a pattern similar to that observed in long-term survival but with a reduced effect. Most important, negative receptor status for each of the tumor markers (ER, PR, and HER2) was associated with an adverse effect on both long- and short-term survival with the exception of long-term survival on HER2 (RR, 0.98; P = .715). Negative ER status (RR, 1.48; P < .0001) and negative PR status (RR, 1.20; P = .030) were associated with an increase in the long-term risk of death, whereas short-term effects were even stronger (ER-negative tumors: RR, 1.70; P < .0001; PR-negative tumors: RR, 1.41; P = .001). Negative HER2 status had little effect on long-term survival (RR, 0.98; P = .715) but increased the short-term risk of death (RR, 1.18; P = .038).

Next, the phenotypic combinations TN and DN were used as explanatory variables in the full model rather than individual tumor makers, whereas all other variables remained the same. In this model, nearly identical patterns were observed in both long-term and short-term survival for stage at diagnosis, tumor histology, race/ethnicity, and SES, similar to what was observed in the first model. Patients with TN tumors and patients with DN tumors had a clear survival disadvantage in long-term and short-term survival compared with patients who had OBC. Patients who had TN tumors showed 1.74 times higher long-term risk (P < .0001) compared with patients who had OBC; whereas the short-term risk had a greater effect (HR, 2.27; P < .0001). Patients who had DN tumors had 1.48 times higher long-term risk compared with patients who had OBC (P < .0001) and had an HR of 1.78 (P < .0001) for the short-term effect compared with patients who had OBC. TN tumors were compared with DN tumors in the full model, which was rerun with a different coding for the phenotype variable (all other variables remained the same; DN was the assumed baseline category of the phenotype variable). There was a small but significant difference in both long-term survival (HR, 1.17) and short-term survival (HR, 1.29) for patients who had TN tumors compared with patients who had DN tumors.


  1. Top of page
  2. Abstract

In this large, population-based cohort study, we described the demographic, clinical, and tumor characteristics of patients with invasive breast cancer in California who had either TN tumors or DN tumors and compared them with all other invasive breast cancer phenotypes. We observed that patients who had TN tumors and patients who had DN tumors shared many clinical, demographic, and tumor features; had very similar survival; and contrasted greatly with women who had OBC.

In agreement with other studies, we demonstrated that important risk factors for the women with TN tumors were younger age, race/ethnicity (significant differences for non-Hispanic black and Hispanic women), and lower SES.33, 38, 39 Women with TN breast cancers were diagnosed with larger tumors, more aggressive histology, and at a more advanced stage,33, 39 which contributes to faster progression to metastasis and poorer prognosis.57–59 Risk factors for DN tumors and TN tumors were slightly older age and race/ethnicity (significant differences for non-Hispanic black and API); however, it is noteworthy that SES played no role. Women with DN and breast cancer presented with larger tumors, more aggressive histology, and more advanced disease stage.

Patients with TN tumors and patients with DN tumors had very similar survival that was significantly shorter than the survival of patients who had OBC regardless of disease stage at diagnosis.33, 39 Modeling data for individual tumor markers suggested that the shortened survival of women with TN tumors and women with DN tumors was caused primarily by the negative ER and PR status and that receptor negativity affected long-term survival more than short-term survival. The statistically insignificant contribution of HER2 to the lethality of these phenotypes requires further study.

Poor outcomes in women with breast cancer have been associated with negative ER and PR status without regard to HER2 status.28, 60 Some investigators have postulated that all ER-negative tumors carry a relatively poor prognosis, irrespective of the cytokeratin composition or the gene expression signature.26 Tumors that are HER2 positive have been associated for some time with worse clinical outcomes.24, 61 The survival of women with TN tumors is almost identical to the survival of women with DN tumors, as reported previously.26, 32, 33, 62, 63 Jumppanen et al. concluded that, within the ER-negative tumor entity, there was no difference in survival between nonbasal tumors and basal-like tumors as classified by IHC or gene expression.26 Previous studies were focused principally on defining gene expression-based classification of breast cancers32, 62, 63 or specifically on characterizing the basal-like phenotype among young African-American women,33 and they did not discuss the relative importance of HER2, perhaps because small sample sizes may have precluded further investigation.

In the current study, patients who had TN tumors had a small but significantly increased risk for long-term and short-term earlier death compared with patients who had DN tumors. We observed that African-American women consistently were at significantly greater risk of death than women of other ethnicities. Bivariate analyses indicated that Hispanic women were at greater risk of having TN tumors, and API women were at greater risk of having DN tumors. Unlike African-American ethnicity, Hispanic and API ethnicity were not identified as independent contributors to early death in multivariate analyses. Although SES was prominent in the bivariate analysis for patients with TN tumors, it played a limited role in short-term survival; however, it had significant long-term survival effects in multivariate analyses.

Racial, ethnic, and SES disparities in breast cancer incidence and mortality are well documented in the medical literature. Numerous studies have demonstrated that African-American women have a lower incidence of breast cancer but worse survival compared with non-Hispanic white women.2, 60, 64–67 Fewer studies have described the risk for Hispanic women, but those studies indicated that, similar to African-American women, Hispanic women had lower incidence but worse survival.10, 68, 69 It is believed that breast cancer survival among African-American and Hispanic women is compromised, in that these women are younger at the time of diagnosis,5, 70–72 their tumors more often are ER-negative,64, 71, 73 and they present at higher stages,74 perhaps because of issues associated with access to healthcare.12, 75–78 In studies that adjusted for clinical and SES factors, African-American women continued to have slightly but significantly poorer survival compared with white women, whereas the differences in survival among Hispanic and Asian women compared with white women were ameliorated.79–81

The current study included a large number of patients from an ethnically diverse population, thus allowing us to compare the clinical, demographic, and tumor features of women with TN breast cancer, DN breast cancer, and OBC. Nevertheless, this study was not without limitations. The data were taken from a population-based cancer registry and were not supplemented with other clinical data or gene-expression analyses. Population-based cancer registry data derive from many sources. Histologic grading of tumors and tests for ER, PR, and HER2 were performed by a wide variety of laboratories without central review. In addition, almost 50% of the initial study population lacked information about ER, PR, and HER2 status, with the latter constituting the bulk of missing data. In 1999, the CCR began collection of HER2 results; however, >50% of newly diagnosed breast cancers lacked this test result. By 2003, HER2 testing of primary breast cancer was more common, and >70% of patients with breast cancer included HER2 data.43 HER2 testing was not recommended for all women with invasive breast cancer until 200782; thus an increase in testing and improvements in documentation over time are expected. Finally, adjuvant therapy information in the CCR was considered limited, so we did not include it in the survival models. Therefore, we made no attempt to attribute overall survival to any specific form of therapy. The omission of therapies may confound the correlation between overall breast cancer survival and phenotypic group.

In summary, the current results indicate that patients with TN tumors share many clinical, demographic, and tumor features and have survival that is almost identical to that of patients with DN tumors. We also observed that disease stage, tumor grade, SES, race/ethnicity, and negative ER and PR status remained as significant and independent risk factors for long-term survival. In addition, we demonstrated that negative ER and PR status, rather than negative HER2 status, was the predominant factor contributing to poor survival. The TN phenotype, although it is a useful surrogate marker for the identification of basal-like breast cancers, may not explain all of the poor prognostic features of breast cancer in young African-American women. The correlations between race/ethnicity, SES, and breast cancer survival are provocative and difficult to untangle, and this difficulty underscores the need for more precise reporting of patient characteristics and tumor factors in future studies.


  1. Top of page
  2. Abstract
  • 1
    American Cancer Society, California Division and Public Health Institute, California Cancer Registry. California Facts and Figures 2007. Oakland, Calif: American Cancer Society, Division; 2006.
  • 2
    Campleman S,Curtis R. Demographic aspects of breast cancer incidence and mortality in California, 1988–1999. In: MorrisC,KwongSL, eds. Breast Cancer in California, 2003, vol July. Sacramento, Calif: California Department of Health Services, Cancer Surveillance Section; 2004.
  • 3
    Chlebowski RT,Chen Z,Anderson GL, et al. Ethnicity and breast cancer: factors influencing differences in incidence and outcome. J Natl Cancer Inst. 2005; 97: 439448.
  • 4
    Jatoi I,Anderson WF,Rao SR,Devesa SS. Breast cancer trends among black and white women in the United States. J Clin Oncol. 2005; 23: 78367841.
  • 5
    Shavers VL,Harlan LC,Stevens JL. Racial/ethnic variation in clinical presentation, treatment, and survival among breast cancer patients under age 35. Cancer. 2003; 97: 134147.
  • 6
    Jayasinghe UW,Taylor R,Boyages J. Is age at diagnosis an independent prognostic factor for survival following breast cancer? Aust N Z J Surg. 2005; 75: 762767.
  • 7
    Swanson GM,Lin CS. Survival patterns among younger women with breast cancer: the effects of age, race, stage, and treatment. J Natl Cancer Inst Monogr. 1994; 16: 6977.
  • 8
    Hartley MC,McKinley BP,Rogers EA, et al. Differential expression of prognostic factors and effect on survival in young (≤40) breast cancer patients: a case-control study. Am Surg. 2006; 72: 11891194; discussion 1194–1195.
  • 9
    Jmor S,Al-Sayer H,Heys SD,Payne S,Miller I,Ah-See A, et al. Breast cancer in women aged 35 and under: prognosis and survival. J R Coll Surg Edinburgh. 2002; 47: 693699.
  • 10
    Frost F,Tollestrup K,Hunt WC,Gilliland F,Key CR,Urbina CE. Breast cancer survival among New Mexico Hispanic, American Indian, and non-Hispanic white women (1973–1992). Cancer Epidemiol Biomarkers Prev. 1996; 5: 861866.
  • 11
    Campbell JB. Breast cancer-race, ethnicity, and survival: a literature review. Breast Cancer Res Treat. 2002; 74: 187192.
  • 12
    Gordon NH. Socioeconomic factors and breast cancer in black and white Americans. Cancer Metastasis Rev. 2003; 22: 5565.
  • 13
    Field TS,Buist DS,Doubeni C, et al. Disparities and survival among breast cancer patients. J Natl Cancer Inst Monogr. 2005; 35: 8895.
  • 14
    Esteva FJ,Sahin AA,Cristofanilli M,Arun B,Hortobagyi GN. Molecular prognostic factors for breast cancer metastasis and survival. Semin Radiat Oncol. 2002; 12: 319328.
  • 15
    Joslyn SA. Hormone receptors in breast cancer: racial differences in distribution and survival. Breast Cancer Res Treat. 2002; 73: 4559.
  • 16
    Ademuyiwa FO,Olopade OI. Racial differences in genetic factors associated with breast cancer. Cancer Metastasis Rev. 2003; 22: 4753.
  • 17
    Huang HJ,Neven P,Drijkoningen M, et al. Association between HER-2/neu and the progesterone receptor in oestrogen-dependent breast cancer is age-related. Breast Cancer Res Treat. 2005; 91: 8187.
  • 18
    Hlupic L,Jakic-Razumovic J,Bozikov J,Coric M,Belev B,Vrbanec D. Prognostic value of different factors in breast carcinoma. Tumori. 2004; 90: 112119.
  • 19
    Wu Y,Khan H,Chillar R,Vadgama JV. Prognostic value of plasma HER-2/neu in African American and Hispanic women with breast cancer. Int J Oncol. 1999; 14: 10211037.
  • 20
    Tandon AK,Clark GM,Chamness GC,Ullrich A,McGuire WL. HER-2/neu oncogene protein and prognosis in breast cancer. J Clin Oncol. 1989; 7: 11201128.
  • 21
    Kroger N,Milde-Langosch K,Riethdorf S, et al. Prognostic and predictive effects of immunohistochemical factors in high-risk primary breast cancer patients. Clin Cancer Res. 2006; 12: 159168.
  • 22
    Goldhirsch A,Glick JH,Gelber RD,Coates AS,Thurlimann B,Senn HJ. Meeting highlights: International Expert Consensus on the Primary Therapy of Early Breast Cancer 2005. Ann Oncol. 2005; 16: 15691583.
  • 23
    Menard S,Fortis S,Castiglioni F,Agresti R,Balsari A. HER2 as a prognostic factor in breast cancer. Oncology. 2001; 61( suppl 2): 6772.
  • 24
    Luftner D,Jung A,Schmid P, et al. Up-regulation of HER-2/neu by ovarian ablation: results of a randomized trial comparing leuprorelin to CMF as adjuvant therapy in node-positive breast cancer patients. Breast Cancer Res Treat. 2003; 80: 245255.
  • 25
    Dowsett M,Cooke T,Ellis I, et al. Assessment of HER2 status in breast cancer: why, when and how? Eur J Cancer. 2000; 36: 170176.
  • 26
    Jumppanen M,Gruvberger-Saal S,Kauraniemi P, et al. Basal-like phenotype is not associated with patient survival in estrogen-receptor-negative breast cancers [electronic resource]. Breast Cancer Res. 2007; 9: R16.
  • 27
    Early Breast Cancer Trialist's Collaborative Group. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005; 365: 16871717.
  • 28
    Grann VR,Troxel AB,Zojwalla NJ,Jacobson JS,Hershman D,Neugut AI. Hormone receptor status and survival in a population-based cohort of patients with breast carcinoma. Cancer. 2005; 103: 22412251.
  • 29
    Slamon DJ,Leyland-Jones B,Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001; 344: 783792.
  • 30
    Vogel CL,Cobleigh MA,Tripathy D, et al. Efficacy and safety of trastuzumab as a single agent in first-line treatment of HER2-overexpressing metastatic breast cancer. J Clin Oncol. 2002; 20: 719726.
  • 31
    Perou CM,Sorlie T,Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000; 406: 747752.
  • 32
    Sorlie T,Perou CM,Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA. 2001; 98: 1086910874.
  • 33
    Carey LA,Perou CM,Livasy CA, et al. Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA. 2006; 295: 24922502.
  • 34
    Collins LC,Schnitt SJ,Colditz GA, et al. Risk factors for luminal, HER2, and basal-like breast cancer subtypes: results from a tissue microarray-based analysis of invasive breast cancers from women enrolled in the Nurses' Health Study [Abstract 5041]. Breast Cancer Res Treat. 2006; 100( suppl 1): S275.
  • 35
    Rakha EA,El-Rehim DA,Paish C, et al. Basal phenotype identifies a poor prognostic subgroup of breast cancer of clinical importance. Eur J Cancer. 2006; 42: 31493156.
  • 36
    Lakhani SR,Reis-Filho JS,Fulford L, et al. Prediction of BRCA1 status in patients with breast cancer using estrogen receptor and basal phenotype. Clin Cancer Res. 2005; 11: 51755180.
  • 37
    Mountain JL,Risch N. Assessing genetic contributions to phenotypic differences among ‘racial’ and ‘ethnic’ groups. Nat Genet. 2004; 36(11 suppl ): S48S53.
  • 38
    Lund MB,Eley JW,Brawley OW,Liff JM,Coates RJ,Porter PL. Racial differences in triple negative breast tumors among women in Atlanta [abstract]. Proc Am Assoc Cancer Res. 2006; 47; LB-4.
  • 39
    Bauer KR,Brown M,Cress RD,Parise CA,Caggiano V. Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a population-based study from the California Cancer Registry. Cancer. 2007; 109: 17211128.
  • 40
    Caggiano V,Brown M,Bauer KR,Tsodikov A,Parise C. Incidence, risk factors, and survival among triple negative phenotypic (estrogen receptor, progesterone receptor, and HER2 negative) breast cancers in California, 1999–2003 [Abstract 6057]. Breast Cancer Res Treat. 2006; 100( suppl 1): S275.
  • 41
    California Department of Health Services, Cancer Surveillance Section. Cancer Reporting in California: Standards for Automated Reporting. California cancer reporting system standards, vol II. Sacramento, Calif: California Department of Health Services, Cancer Surveillance Section; 2006.
  • 42
    Percy C,Fritz A,Jack A, et al. International Classification of Diseases for Oncology, 3rd ed. Geneva, Switzerland: World Health Organization; 2000.
  • 43
    Bauer KR,Brown M,Creech C,Schlag NC,Caggiano V. Data quality assessment of HER2 in the Sacramento Region of the California Cancer Registry. J Registry Manage. 2007; 34: 47.
  • 44
    Fritz A.Recording Tumor Markers in Collaborative Staging System Site-Specific Factors, vol 2006. 2005. Available at URL: Accessed December 18, 2007.
  • 45
    DAKO. HERCEPTest® Information Web site. Summary of Procedure. Available at URL: downloadpackageinsert.pdf?objectid=114969002. Accessed December 18, 2007.
  • 46
    Vysis, Inc. PathVysion®: HER-2 DNA Probe Kit [package insert]. Downers Grove, Ill: Vysis, Inc.; 2002.
  • 47
    Stewart SL,Swallen KC,Glaser SL,Horn-Ross PL,West DW. Comparison of methods for classifying Hispanic ethnicity in a population-based cancer registry. Am J Epidemiol. 1999; 149: 10631071.
  • 48
    Yost K,Perkins C,Cohen R,Morris C,Wright W. Socioeconomic status and breast cancer incidence in California for different race/ethnic groups. Cancer Causes Control. 2001; 12: 703711.
  • 49
    GreeneFL,PageDL,FlemingID, et al., eds. AJCC Cancer Staging Manual,6th ed. New York, NY: Springer-Verlag; 2002.
  • 50
    FritzA,RiesL, eds. SEER Extent of Disease 1998, Codes and Coding Instructions,3rd ed. Bethesda, Md: National Cancer Institute, National Institutes of Health; 1998.
  • 51
    Collaborative Staging Task Force of the American Joint Committee on Cancer. Collaborative Staging Manual and Coding Instructions, version 01.03.00. Jointly published: Chicago, Ill; American Joint Committee on Cancer;Bethesda, Md,U.S. Department of Health and Human Services; 2004.
  • 52
    Seiffert JE, National Cancer Institute (US). SEER Program. Comparative Staging Guide for Cancer. Bethesda, Md: U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Cancer Institute; 1993.
  • 53
    Fleiss JL,Levin BA,Paik MC. Statistical Methods for Rates and Proportions, 3rd ed. Hoboken, NJ: Wiley-Interscience; 2003.
  • 54
    Broet P,De Rycke Y,Tubert-Bitter P,Lellouch J,Asselain B,Moreau T. A semiparametric approach for the 2-sample comparison of survival times with long-term survivors. Biometrics. 2001; 57: 844852.
  • 55
    Tsodikov A. A proportional hazards model taking account of long-term survivors. Biometrics. 1998; 54: 15081516.
  • 56
    Tsodikov AD,Ibrahim JG,Yakovlev AJ. Estimating cure rates from survival data: an alternative to two-component mixture models. J Am Stat Assoc. 2003; 98: 10631078.
  • 57
    Carter CL,Allen C,Henson DE. Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases. Cancer. 1989; 63: 181187.
  • 58
    Fisher B,Slack NH,Bross ID. Cancer of the breast: size of neoplasm and prognosis. Cancer. 1969; 24: 10711080.
  • 59
    Koscielny S,Tubiana M,Le MG, et al. Breast cancer: relationship between the size of the primary tumour and the probability of metastatic dissemination. Br J Cancer. 1984; 49: 709715.
  • 60
    Cunningham JE,Butler WM. Racial disparities in female breast cancer in South Carolina: clinical evidence for a biological basis. Breast Cancer Res Treat. 2004; 88: 161176.
  • 61
    Menard S,Casalini P,Campiglio M,Pupa S,Agresti R,Tagliabue E. HER2 overexpression in various tumor types, focusing on its relationship to the development of invasive breast cancer. Ann Oncol. 2001; 12( suppl 1): S15S19.
  • 62
    Sorlie T,Tibshirani R,Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci USA. 2003; 100: 84188423.
  • 63
    Sotiriou C,Neo SY,McShane LM, et al. Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci USA. 2003; 100: 1039310398.
  • 64
    Chu KC,Lamar CA,Freeman HP. Racial disparities in breast carcinoma survival rates: separating factors that affect diagnosis from factors that affect treatment. Cancer. 2003; 97: 28532860.
  • 65
    Newman LA. Breast cancer in African-American women. Oncologist. 2005; 10: 114.
  • 66
    Ries L,Eisner M,Kosary M. SEER Cancer Statistics Review, 1975–2000. Bethesda, Md: National Cancer Institute; 2003.
  • 67
    Woods LM,Rachet B,Coleman MP. Origins of socio-economic inequalities in cancer survival: a review. Ann Oncol. 2006; 17: 519.
  • 68
    Elledge RM,Clark GM,Chamness GC,Osborne CK. Tumor biologic factors and breast cancer prognosis among white, Hispanic, and black women in the United States. J Natl Cancer Inst. 1994; 86: 705712.
  • 69
    Boyer-Chammard A,Taylor TH,Anton-Culver H. Survival differences in breast cancer among racial/ethnic groups: a population-based study. Cancer Detect Prev. 1999; 23: 463473.
  • 70
    El-Tamer MB,Wait RB. Age at presentation of African-American and Caucasian breast cancer patients. J Am Coll Surg. 1999; 188: 237240.
  • 71
    Aziz H,Hussain F,Sohn C, et al. Early onset of breast carcinoma in African American women with poor prognostic factors. Am J Clin Oncol. 1999; 22: 436440.
  • 72
    Boyle T,McPadden E. Breast cancer presents at an earlier age in Mexican American women. Breast J. 2004; 10: 462464.
  • 73
    Colon E,Reyes JS,Gonzalez Keelan C,Climent-Peris C. Prevalence of steroid receptors and HER 2/neu in breast cancer biopsies of women living in Puerto Rico. P R Health Sci J. 2002; 21: 299303.
  • 74
    Hsu JL,Glaser SL,West DW. Racial/ethnic differences in breast cancer survival among San Francisco Bay Area women. J Natl Cancer Inst. 1997; 89: 13111312.
  • 75
    Ayanian JZ,Kohler BA,Abe T,Epstein AM. The relation between health insurance coverage and clinical outcomes among women with breast cancer. N Engl J Med. 1993; 329: 326331.
  • 76
    Breen N,Wesley MN,Merrill RM,Johnson K. The relationship of socio-economic status and access to minimum expected therapy among female breast cancer patients in the National Cancer Institute Black-White Cancer Survival Study. Ethn Dis. 1999; 9: 111125.
  • 77
    Bradley CJ,Given CW,Roberts C. Race, socioeconomic status, and breast cancer treatment and survival. J Natl Cancer Inst. 2002; 94: 490496.
  • 78
    Brawley OW. Disaggregating the effects of race and poverty on breast cancer outcomes. J Natl Cancer Inst. 2002; 94: 471473.
  • 79
    O'Malley CD,Le GM,Glaser SL,Shema SJ,West DW. Socioeconomic status and breast carcinoma survival in 4 racial/ethnic groups: a population-based study. Cancer. 2003; 97: 13031311.
  • 80
    Yood MU,Johnson CC,Blount A, et al. Race and differences in breast cancer survival in a managed care population. J Natl Cancer Inst. 1999; 91: 14871491.
  • 81
    Franzini L,Williams AF,Franklin J,Singletary SE,Theriault RL. Effects of race and socioeconomic status on survival of 1,332 black, Hispanic, and white women with breast cancer. Ann Surg Oncol. 1997; 4: 111118.
  • 82
    Wolff AC,Hammond ME,Schwartz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol. 2007; 25: 118145.