Breast Cancer Research and Treatment

, Volume 161, Issue 2, pp 279–287 | Cite as

Triple-negative breast cancer has worse overall survival and cause-specific survival than non-triple-negative breast cancer

  • Xiaoxian Li
  • Jing Yang
  • Limin Peng
  • Aysegul A. Sahin
  • Lei Huo
  • Kevin C. Ward
  • Ruth O’Regan
  • Mylin A. Torres
  • Jane L. Meisel
Preclinical study

Abstract

Purpose

The current American Joint Committee on Cancer (AJCC) staging manual uses tumor size, lymph node, and metastatic status to stage breast cancer across different subtypes. We examined the prognosis of triple-negative breast cancer (TNBC) versus non-TNBC within the same stages and sub-stages to evaluate whether TNBC had worse prognosis than non-TNBC.

Methods

We reviewed the National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) data and identified 158,358 patients diagnosed with breast cancer from 2010 to 2012. The overall survival (OS) time and breast cancer cause-specific survival time were compared between patients with TNBC and non-TNBC in each stage and sub-stages. The results were validated using a dataset of 2049 patients with longer follow-up from our institution.

Results

Compared with patients with non-TNBC, patients with TNBC had worse OS and breast cancer cause-specific survival time in every stage and sub-stage in univariate and multivariate analyses adjusting for age, race, tumor grade, and surgery and radiation treatments in the SEER data. The worse OS time in patients with TNBC was validated in our institutional dataset.

Conclusions

Patients with TNBC have worse survival than patients with non-TNBC. The new AJCC staging manual should consider breast cancer biomarker information.

Keywords

Triple-negative breast cancer TNBC Prognosis Overall survival Cause-specific survival AJCC Cancer staging 

Introduction

Triple-negative breast cancer (TNBC) is a subtype of breast cancer defined by negative expression of the estrogen receptor (ER) and progesterone receptor (PR), and lack of amplification of the human epidermal growth factor-2 (HER2) gene. TNBC accounts for approximately 15–20% of the more than one million breast cancer cases diagnosed worldwide each year [1]. TNBC is more common in women below the age of 40, and in women of black race [2]. Up to 20% of patients with TNBC harbor a breast cancer gene (BRCA) mutation, particularly in BRCA1 [3]. TNBC has different molecular and immunohistochemical characteristics than ER+ or HER2+ breast cancers [4].

TNBCs typically behave more aggressively than other subtypes of breast cancers [5]. In a 2012 study of more than 12,000 women who presented to National Comprehensive Cancer Network (NCCN) centers, patients with TNBC experienced worse breast cancer-specific survival, worse overall survival, and a dramatic increase in death within 2 years of diagnosis when compared with patients with hormone receptor (HR)-positive breast cancer [6]. Unlike endocrine therapy for ER+ and PR+ patients or HER2-directed therapy for HER2+ patients, there are no targeted therapies available to treat TNBCs.

However, despite the more aggressive nature and less favorable outcomes of TNBC, the clinical staging of breast cancer applies uniformly to all breast cancers, regardless of subtype using the American Joint Committee on Cancer and the International Union for Cancer Control (AJCC–UICC) Tumor, Node, Metastases (TNM) staging system [7]. The part of the reason for tumor staging is to determine treatment plans and to give patients the estimates about prognosis. Since TNBC often necessitates more aggressive treatment and prognosis is often worse, there have been some recent discussions within the field about whether a different staging system for breast cancer incorporating biomarker information might be relevant to consider [8, 9].

To further understand whether the current AJCC staging system accurately reflects the nature of different subtypes of breast cancer, we performed an analysis using the National Cancer Institute Surveillance, Epidemiology, and End Result (SEER) data and a companion analysis using the Emory clinic database. We aimed to understand the magnitude of the differences in outcomes between TNBC and non-TNBC breast cancers when compared stage by stage.

Patients and methods

Patient information

We searched the SEER database for all female breast cancer patients diagnosed from 2010 to 2012 and identified 180,996 cases. Patients were excluded from our analysis if their disease was stage 0 and stage NOS or if information on stage was missing. Cases with missing HR or HER2 status were also excluded from analysis. There were 158,358 patients who met study criteria. We also searched the Emory University Hospital clinical database from 2002 to 2013 and identified 2049 cases that met the same inclusion and exclusion criteria.

For both the SEER and Emory cases, we collected the following de-identified information: patient gender, age at diagnosis, ethnicity, overall survival (OS) times, tumor grade, HR and HER2 status, and staging based on the TNM and Roman Numeral Staging system from the American Joint Committee on Cancer guidelines (AJCC). We also collected breast cancer cause-specific survival times, surgery at the primary site (breast), and radiation treatment information from the SEER data. We classified breast carcinomas by HR and HER2 status into four subtypes: HR+/HER2–, HR+/HER2+ , HR–/HER2+ and HR–/HER2–. The HR−/HER2− subtype was referred as the TNBC cancer. Non-TNBC referred to combined cases including HR+/HER2–, HR+/HER2+ and HR–/HER2+ cancers. The follow-up time was up to 35 months (median 15 months) in the SEER database and 144.1 months (median 61.6 months) in the Emory database. The Emory University Institutional Review Board approved the study.

Statistical analysis

The OS times of patients with different cancer subtypes were summarized by Kaplan–Meier survival curves and compared by Cox proportional hazard model in all stages including sub-stages.

Differences between the TNBC and non-TNBC group were assessed by the two-sample t test for continuous variables (age) and by the Chi square test for categorical variables (race, tumor grade, and surgery and radiation treatments). The OS times of patients with TNBC and non-TNBC in all stages and sub-stages were summarized by Kaplan–Meier survival curves and compared by log-rank tests. A univariate Cox proportional hazard model was conducted separately for age, ethnicity, tumor grade, and surgery and radiation treatments. A multivariate Cox regression analyses were further conducted to study the OS time of patients with TNBC and non-TNBC cancers, while adjusting for the significant predictors identified based on univariate analyses. In the SEER dataset, Cox regression was also used for studying the breast cancer cause-specific survival between TNBC versus non-TNBC cases. Because many cases in the Emory dataset were missing information on tumor grade, only age and race were adjusted for when analyzing the Emory data. Hazard ratios and their associated 95% confidence intervals (CIs) were obtained from the Cox regression analysis. A P value below .05 was regarded as statistically significant. In the univariate and multivariate analyses, subjects with missing relevant covariate(s) were excluded. All analyses were performed using the SAS 9.4 software.

Patient age at diagnosis was analyzed as a continuous variable and tumor grade as a categorical variable (grade 1 vs. grade 2 vs. grade 3 and 4; grade 4 was defined by SEER as undifferentiated or anaplastic carcinoma). Surgery and radiation treatments were analyzed as binary variables (yes vs. no). For ethnicity, the categories American Indian/Alaska Native and Asian/Pacific Islander were combined as “Others.”

Results

Clinico-pathologic characteristics of patients

Supplementary Table 1 and Table 1 summarize the demographic information of the SEER data. In the SEER data, the majority of patients were white (80.0%); HR+/HER2− subtype carcinomas accounted for 73.5% of all the cases. Comparing patients with non-TNBC, patients with TNBC presented at younger age (mean age at diagnosis 58.9 vs. 61.8 years, P < .0001) and with higher tumor grade (grade II–IV: 97.8 vs. 75.3%, P < .0001; Table 1). Surgery and radiation treatments were comparable between patients with TNBC and patients with non-TNBC (surgery 91.9 vs. 93.1%; radiation 42.8 vs. 46.7%; Table 1).The Emory data also showed that TNBC occurred more frequently in black patients than non-TNBC (65.2 vs. 45.4%, Table 2).
Table 1

Demographics of triple-negative breast cancer (TNBC) and non-TNBC cases from the National Cancer Institute Surveillance, Epidemiology, and End Result (SEER) data

Predictor

TNBC (n = 18855)

Non-TNBC (n = 139503)

P value

Age at diagnosis (year)

58.9 (14.0)

61.8 (13.5)

<.0001

Race

<.0001

 White

13,724 (72.8%)

112,947 (81.0%)

 

 Black

3714 (19.7%)

13,392 (9.6%)

 

 Others

1332 (7.1%)

12,305 (8.8%)

 

 Unknown

85 (.5%)

859 (.6%)

 

Grade

<.0001

 I

414 (2.2%)

34,486 (24.7%)

 

 II

3263 (17.3%)

63,179 (45.3%)

 

 III

14,128 (74.9%)

35,138 (25.2%)

 

 IV

193 (1.0%)

424 (.3%)

 

 Unknown

857 (4.6%)

6276 (4.5%)

 

Surgery of Primary Site

<.0001

 Yes

17,325 (91.9%)

129,920 (93.1%)

 

 No

1503 (8.0%)

9442 (6.8%)

 

 Unknown

27 (.1%)

141 (.1%)

 

Radiation

<.0001

 Yes

8064 (42.8%)

65,208 (46.7%)

 

 No

10791 (57.2%)

74,295 (53.3%)

 

Stage

<.0001

 IA

6703 (35.6%)

70,264 (50.4%)

 

 IB

235 (1.3%)

3522 (2.5%)

 

 IIA

5419 (28.7%)

28,891 (20.7%)

 

 IIB

2444 (13.0%)

14,622 (10.5%)

 

 IIIA

1399 (7.4%)

8756 (6.3%)

 

 IIIB

711 (3.8%)

2699 (1.9%)

 

 IIIC

762 (4.0%)

3735 (2.7%)

 

 IV

1182 (6.3%)

7014 (5.0%)

 

Mean (SD) was for age at diagnosis and frequency (%) for other predictors

Two-sample t test for age and Chi square test for other predictors. “Unknown” categories were not included in the tests

Table 2

Demographics of triple-negative (TNBC) and non-TNBC cases from the Emory data

 

Non-TNBC (n = 1727)

TNBC (n = 322)

P value

Age at diagnosis (year)

58.9 (13.5)

56.1 (13.3)

<.0001

Race

<.0001

 White

870 (50.4%)

93 (28.9%)

 

 Black

784 (45.4%)

210 (65.2%)

 

 Others

37 (2.1%)

10 (3.1%)

 

 Unknown

35 (2.0%)

9 (2.8%)

 

Grade

 1

3 (.0%)

0 (.0%)

 

 2

3 (.0%)

1 (.0%)

 

 3

310 (18.0%)

219 (68.0%)

 

 Unknown

1419 (82.2%)

105 (32.6%)

 

Stage

 IA

910 (52.7%)

143 (44.4%)

 

 IB

24 (1.4%)

3 (.9%)

 

 IIA

447 (25.9%)

99 (30.8%)

 

 IIB

141 (8.2%)

27 (8.4%)

 

 IIIA

130 (7.5%)

26 (8.1%)

 

 IIIB

26 (1.5%)

14 (4.4%)

 

 IIIC

48 (2.8%)

7 (2.2%)

 

 IV

1 (.1%)

3 (.9%)

 

Mean (SD) was for age at diagnosis and frequency (%) for other predictors

Two-sample t test for age, Chi square test for race without including “unknown” categories

The univariate Cox model showed that patient age, ethnicity, tumor grade, and surgery and radiation treatments significantly affected OS in the SEER data. Factors associated with a worse prognosis included older age, black ethnicity, high tumor grade, and without surgery and without radiation treatments.

Patients with TNBC had worse OS than patients with non-TNBC patients from the SEER data

Table 1 and Supplementary Table 2 summarize the number of cases in each subtype and each sub-stage. Compared to patients with other three subtypes of carcinoma (HR+/HER2−, HR+/HER2+ and HR−/HER2+), patients with TNBC had worse OS time in almost every sub-stage (Supplementary Figure; Table 3). The only non-significant results were TNBC versus HR−/HER2+ carcinomas in stage IA and IB (Table 3).
Table 3

Univariate analysis comparing overall survival times between triple-negative breast cancer (TNBC) and each other subtype from the National Cancer Institute Surveillance, Epidemiology, and End Result (SEER) data

Stage

Hazard ratio

95% CI

P value

IA

 TNBC versus HR+/HER2−

1.62

1.39

1.88

<.0001

 TNBC versus HR+/HER2+ 

1.59

1.26

2.00

<.0001

 TNBC versus HR−/HER2+

1.21

.90

1.63

.1999

IB

 TNBC versus HR+/HER2−

3.46

1.92

6.25

<.0001

 TNBC versus HR+/HER2+ 

2.84

1.15

7.03

.0243

 TNBC versus HR−/HER2+

1.76

.58

5.35

.3182

IIA

 TNBC versus HR+/HER2−

1.67

1.46

1.90

<.0001

 TNBC versus HR+/HER2+ 

1.74

1.41

2.14

<.0001

 TNBC versus HR−/HER2+

1.40

1.07

1.83

.0147

IIB

 TNBC versus HR+/HER2−

2.59

2.21

3.04

<.0001

 TNBC versus HR+/HER2+ 

4.19

3.09

5.67

<.0001

 TNBC versus HR−/HER2+

1.54

1.16

2.06

.0033

IIIA

 TNBC versus HR+/HER2−

3.65

3.02

4.40

<.0001

 TNBC versus HR+/HER2+ 

3.95

2.88

5.42

<.0001

 TNBC versus HR−/HER2+

2.58

1.82

3.66

<.0001

IIIB

 TNBC versus HR+/HER2−

2.39

1.99

2.88

<.0001

 TNBC versus HR+/HER2+ 

3.83

2.77

5.31

<.0001

 TNBC versus HR−/HER2+

2.36

1.73

3.22

<.0001

IIIC

 TNBC versus HR+/HER2−

3.78

3.12

4.58

<.0001

 TNBC versus HR+/HER2+ 

3.75

2.72

5.17

<.0001

 TNBC versus HR−/HER2+

3.26

2.28

4.68

<.0001

IV

 TNBC versus HR+/HER2−

2.58

2.35

2.82

<.0001

 TNBC versus HR+/HER2+ 

2.76

2.42

3.14

<.0001

 TNBC versus HR−/HER2+

1.85

1.60

2.13

<.0001

We then compared the OS times between patients with TNBC and patients with non-TNBC (combined cases of HR+/HER2−, HR+/HER2+ and HR−/HER2+ subtypes) in the SEER data. Patients with TNBC had worse OS times in every stage and sub-stage (Fig. 1; Table 4) in univariate and multivariate analyses after adjusting for age, race, tumor grade, and surgery and radiation treatments (Table 4).
Fig. 1

Kaplan–Meier survival curves of triple-negative breast cancer (TNBC) versus non-TNBC cases in different stages from the National Cancer Institute Surveillance, Epidemiology, and End Result (SEER) data

Table 4

Univariate and multivariate analyses of overall survival time between triple-negative breast cancer (TNBC) and non-TNBC cases by sub-stage from the National Cancer Institute Surveillance, Epidemiology, and End Result (SEER) data

Stage

Univariate analysis

Multivariate analysis

Hazard ratio

95% CI

P value

Hazard ratio

95% CI

P value

I

1.66

1.43

1.91

<.0001

1.79

1.52

2.11

<.0001

II

1.96

1.78

2.16

<.0001

2.02

1.81

2.26

<.0001

III

3.47

3.13

3.85

<.0001

2.74

2.45

3.07

<.0001

IV

2.51

2.31

2.74

<.0001

2.41

2.18

2.67

<.0001

IA

1.60

1.38

1.86

<.0001

1.75

1.48

2.08

<.0001

IB

3.28

1.84

5.86

<.0001

2.98

1.54

5.76

.0012

IIA

1.66

1.46

1.88

<.0001

1.74

1.51

2.01

<.0001

IIB

2.63

2.25

3.07

<.0001

2.70

2.26

3.21

<.0001

IIIA

3.57

2.99

4.27

<.0001

2.45

2.01

2.98

<.0001

IIIB

2.58

2.16

3.07

<.0001

2.79

2.29

3.39

<.0001

IIIC

3.71

3.10

4.45

<.0001

2.95

2.42

3.61

<.0001

Hazard ratio (TNBC vs. Non-TNBC)

Multivariate analysis was adjusted for age, race, tumor grade, radiation, surgery

Patients with TNBC had worse breast cancer cause-specific survival time than patients with non-TNBC

We retrieved breast cancer cause-specific survival data from the SEER database. Comparing patients with non-TNBC, patients with TNBC had worse breast cancer cause-specific survival times in every stage and sub-stages in univariate analysis and multivariate analysis after adjusting for age, race, tumor grade, and surgery and radiation treatments (Table 5).
Table 5

Univariate and multivariate analyses of cause-specific survival between triple-negative breast cancer (TNBC) and non-TNBC cases in each sub-stage from the National Cancer Institute Surveillance, Epidemiology, and End Result (SEER) data

Stage

Univariate analysis

Multivariate analysis

Hazard ratio

95% CI

P value

Hazard ratio

95% CI

P value

I

5.18

3.98

6.73

<.0001

3.44

2.51

4.70

<.0001

II

3.45

2.99

3.98

<.0001

2.56

2.17

3.00

<.0001

III

4.22

3.71

4.8

<.0001

3.00

2.60

3.46

<.0001

IV

2.62

2.38

2.9

<.0001

2.48

2.21

2.78

<.0001

IA

4.90

3.72

6.47

<.0001

3.34

2.39

4.65

<.0001

IB

11.23

4.73

26.66

<.0001

6.04

2.28

15.96

.0003

IIA

2.83

2.32

3.44

<.0001

2.10

1.68

2.62

<.0001

IIB

4.61

3.73

5.69

<.0001

3.60

2.83

4.57

<.0001

IIIA

4.09

3.27

5.10

<.0001

2.44

1.91

3.13

<.0001

IIIB

3.35

2.67

4.20

<.0001

3.22

2.50

4.14

<.0001

IIIC

4.48

3.58

5.61

<.0001

3.41

2.67

4.36

<.0001

Multivariate analysis was adjusted for age, race, tumor grade, radiation, surgery

Emory data verified worse OS in TNBC compared with non-TNBC

We used the Emory data with longer follow-up to validate the analysis from the SEER data. The demographic information and numbers of cases are summarized in Table 2. Analysis for stage IV patients was not performed due to insufficient number of cases. Consistent with the SEER data, patients with TNBC had worse OS in almost every stage and sub-stage than patients with non-TNBC except in stage IIIB and IIIC (Fig. 2; Supplementary Table 3), which had very few cases (14 TNBC cases in stage IIIB and 7 TNBC cases in stage IIIC, Table 2). The worse OS time of patients with TNBC was also seen in multivariate analysis after adjusting for age and race (Supplementary Table 3).
Fig. 2

Kaplan–Meier survival curves of triple-negative breast cancer (TNBC) versus non-TNBC cases in different stages from the Emory data

Discussion

The AJCC TNM tumor staging system describes the anatomic extent of cancer, and its objectives are to aid in treatment planning and prognostication, to assist in the evaluation of the effectiveness of neoadjuvant therapy, and to facilitate the research that allows us to continue improving the effectiveness of our treatments [10]. Since the first edition of the AJCC Cancer Staging Manual was published, it has been updated regularly, with changes in staging criteria made to reflect the most current knowledge in the field. These updates allow us to be more precise in determining prognosis and to choose optimal treatment regimens while minimizing unnecessary side effects and toxicities. For example, treatments for small breast cancers can vary significantly depending on patient and disease characteristics. The American Cancer Society states “If the tumor is smaller than 1 cm (about ½ inch) across, adjuvant chemotherapy (chemo) is not usually needed. Some doctors may suggest chemo if a cancer smaller than 1 cm has any unfavorable features (such as being high-grade, hormone receptor-negative, HER2-positive, or having a high score on a gene panel such as Oncotype Dx)…” (http://www.cancer.org/cancer/breastcancer/detailedguide/breast-cancer-treating-by-stage). These guidelines do not make a strong statement as to whether sub-centimeter TNBC should receive chemotherapy, largely due to lack of information on the prognosis of these tumors. Our data showed that early-stage TNBC had worse prognosis than non-TNBC, which physicians and patients may wish to take into account when making decisions about whether to include or omit chemotherapy in the treatment plan.

It is important to note that although AJCC staging is widely used to determine breast cancer prognosis, patient survival within each stage shows wide variation [7]. Differences in tumor biology likely influence this variation and several important studies have shown that the addition of biologic markers to AJCC staging improves determination of prognosis. Yi et al. identified a cohort of over 3700 breast cancer patients who underwent surgery, and examined the impact of adding tumor grade, lymphovascular invasion, ER and PR status, and HER2 status to the current anatomic staging system. They found that scores based on pathologic stage plus tumor grade and ER status ± PR status significantly refined the assessment of prognosis in patients with breast cancer [8]. Similarly, Mittendorf et al. performed a study looking at over 8000 women (a large cohort treated at the MD Anderson Cancer Center and a cohort from the ACOSOG Z0010 trial) to compare recurrence-free survival, disease-specific survival, and overall survival in patients with stage IA cancers (T1N0M0) and stage IB cancers (T1N1miM0) [9]. They found no differences between stage IA and stage IB patients in any of these three survival parameters studied, suggesting that patients with micro-metastases and negative nodes may have similar outcomes. However, similar to the study by Yi et al., they found that ER status and tumor grade significantly stratified patients with stage I disease. These findings suggest that incorporating biologic factors such as ER status and histologic grade should be considered in the future iterations of the staging system.

In the current study, we included a large number of breast cancer cases from the SEER database to evaluate the impact of hormone receptor and HER2 status on prognosis and used another cohort from our own institution with more extensive follow-up data to confirm the findings from the SEER data. We found that patients with TNBC had significantly worse disease cause-specific survival and overall survival times compared to patients with non-TNBC breast cancers even when adjusted for patient age, race, tumor grade, and surgery and radiation treatments.

This study has a number of strengths including the large number of patients evaluated; the ability to control for multiple patient-related, tumor-related, and treatment-related variables; the contemporary nature of the data (obtained during the era of trastuzumab-based therapy for HER2-positive patients as well as modern endocrine therapy for hormone receptor-positive patients); and the validation of SEER data with a companion evaluation of our institutional data.

One weakness of our analysis is that our datasets did not incorporate chemotherapy information, but arguably, because there is a much lower threshold to give chemotherapy to TNBC at earlier stage and to be more aggressive about the choice of regimens, patients with TNBC would likely have received more aggressive chemotherapy than their HR-positive or HER2-positive counterparts. Great strides have been made in the last decade in the form of targeted therapy for HER2-positive [11] and HR-positive breast cancers [12] that have improved outcomes, which has undoubtedly widened the already present gap between TNBC patients and those with HR-positive and/or HER2-positive disease. Treatment information is not incorporated in the current AJCC tumor staging system, which evaluates the overall survival of a specific type of tumor regardless of regimen chosen. The second weakness is the relative short follow-up time of the SEER data. However, because information on HER2 status was only incorporated into SEER data starting in 2010, it would be nearly impossible to have longer follow-up at this time point without leaving out that important variable. We tried to account for this by validating the SEER analysis with a companion analysis of our Emory data, which had a much longer duration of follow-up. The dramatic difference in prognosis between TNBC and non-TNBC in every stage and sub-stage was validated in this setting.

A growing body of data suggests that because of the role of BRCA1 in DNA damage response and cell cycle checkpoint control, TNBCs associated with BRCA1 mutations, such as the high-grade serous ovarian cancers associated with this mutation, may display greater sensitivity to platinum-based chemotherapy regimens [13, 14]. The androgen receptor is another area of interest. There are some early data to suggest that positive expression of the androgen receptor in TNBC may represent a distinct subgroup of TNBCs, which might respond to androgen receptor blockade [15, 16]. The recently identified claudin-low subtype, including metaplastic carcinoma, shows breast cancer stem cell signatures and may be resistant to conventional chemotherapies [17, 18]. Finally, high levels of tumor-infiltrating lymphocytes (TILs) have been shown to be both predictive of improved response to neoadjuvant chemotherapy and prognostic for improved survival in TNBC [19, 20, 21, 22]. TNBC with prominent TILs may be more responsive to immunotherapy. A more detailed staging system may be needed as our understanding of the various subtypes of TNBC grows. Lehmann et al. identified 6 defined subtypes of TNBC and an unstable subtype based on different gene expression profiles [23]. These different subtypes may respond differently to therapies and have different prognoses [24, 25]. We have targeted therapies for non-TNBC breast cancers that improve their prognosis. There is still much research that needs to be done in TNBC in order to refine the multiple subtypes of this disease entity and to better understand which subtypes are likely to respond best to specific treatments. We need to continue to better understand the so-called “TNBCs” to find their specific targets and improve their treatment and prognosis.

The field is advancing rapidly, and better treatments for TNBC may be more than a distant dream. This study included breast cancer cases diagnosed and treated in the modern era and shows that regardless of tumor grade, patient’s age, or patient’s ethnicity, TNBC is more aggressive at every stage than ER+ and/or HER2+ disease. Our results suggest that a modification of the current AJCC staging system to incorporate biological information may be warranted. Such a modification would allow physicians to make more accurate assessments of prognosis and give patients a better understanding of the gravity of their disease, the importance of adherence to therapy, and the critical nature of close follow-up.

Notes

Authors Contributions

Conception and design: XL, LP, LH, JLM. Analysis and interpretation of data: All authors. Manuscript drafting and reviewing: All authors. XL is responsible for the overall content.

Compliance with ethical standards

Conflict of interest

All authors declare no conflict of interest.

Human and animal rights

This article does not contain any studies with human participants or animals performed by any of the authors.

Supplementary material

10549_2016_4059_MOESM1_ESM.pdf (20 kb)
Supplementary FigureKaplan-Meier survival curves of all four subtype breast cancers in different stages from the National Cancer Institute Surveillance, Epidemiology and End Result (SEER) data. Triple negative breast cancer had worse overall survival times than other subtypes in all sub-stages except for in stage IA and IB when compared with the hormonal receptor posive/HER2- breast cancer. Hazard ratio and P values are summerized in table 5 (PDF 20 kb)
10549_2016_4059_MOESM2_ESM.docx (20 kb)
Supplementary Tables(DOCX 18 kb)

References

  1. 1.
    Ismail-Khan R, Bui MM (2010) A review of triple-negative breast cancer. Cancer Control 17(3):173–176PubMedGoogle Scholar
  2. 2.
    Millikan RC, Newman B, Tse CK, Moorman PG, Conway K, Dressler LG, Smith LV, Labbok MH, Geradts J, Bensen JT et al (2008) Epidemiology of basal-like breast cancer. Breast Cancer Res Treat 109(1):123–139CrossRefPubMedGoogle Scholar
  3. 3.
    Gonzalez-Angulo AM, Timms KM, Liu S, Chen H, Litton JK, Potter J, Lanchbury JS, Stemke-Hale K, Hennessy BT, Arun BK et al (2011) Incidence and outcome of BRCA mutations in unselected patients with triple receptor-negative breast cancer. Clin Cancer Res 17(5):1082–1089CrossRefPubMedPubMedCentralGoogle Scholar
  4. 4.
    Davis DG, Siddiqui MT, Oprea-Ilies G, Stevens K, Osunkoya AO, Cohen C, Li XB (2016) GATA-3 and FOXA1 expression is useful to differentiate breast carcinoma from other carcinomas. Hum Pathol 47(1):26–31CrossRefPubMedGoogle Scholar
  5. 5.
    Foulkes WD, Smith IE, Reis-Filho JS (2010) Triple-negative breast cancer. New Engl J Med 363(20):1938–1948CrossRefPubMedGoogle Scholar
  6. 6.
    Lin NU, Vanderplas A, Hughes ME, Theriault RL, Edge SB, Wong YN, Blayney DW, Niland JC, Winer EP, Weeks JC (2012) Clinicopathologic features, patterns of recurrence, and survival among women with triple-negative breast cancer in the National Comprehensive Cancer Network. Cancer 118(22):5463–5472CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Edge SB (2010) American Joint Committee on cancer: AJCC cancer staging manual, 7th edn. Springer, New YorkGoogle Scholar
  8. 8.
    Yi M, Mittendorf EA, Cormier JN, Buchholz TA, Bilimoria K, Sahin AA, Hortobagyi GN, Gonzalez-Angulo AM, Luo S, Buzdar AU et al (2011) Novel staging system for predicting disease-specific survival in patients with breast cancer treated with surgery as the first intervention: time to modify the current American Joint Committee on Cancer staging system. J Clin Oncol 29(35):4654–4661CrossRefPubMedPubMedCentralGoogle Scholar
  9. 9.
    Mittendorf EA, Ballman KV, McCall LM, Yi M, Sahin AA, Bedrosian I, Hansen N, Gabram S, Hurd T, Giuliano AE et al (2015) Evaluation of the stage IB designation of the American Joint Committee on Cancer staging system in breast cancer. J Clin Oncol 33(10):1119–1127CrossRefPubMedGoogle Scholar
  10. 10.
    Sobin LH (2001) TNM: principles, history, and relation to other prognostic factors. Cancer 91(8 Suppl):1589–1592CrossRefPubMedGoogle Scholar
  11. 11.
    Perez EA, Romond EH, Suman VJ, Jeong JH, Sledge G, Geyer CE Jr, Martino S, Rastogi P, Gralow J, Swain SM et al (2014) Trastuzumab plus adjuvant chemotherapy for human epidermal growth factor receptor 2-positive breast cancer: planned joint analysis of overall survival from NSABP B-31 and NCCTG N9831. J Clin Oncol 32(33):3744–3752CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Davies C, Godwin J, Gray R, Clarke M, Cutter D, Darby S, McGale P, Pan HC, Taylor C, Wang YC et al (2011) Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet 378(9793):771–784CrossRefPubMedGoogle Scholar
  13. 13.
    Byrski T, Gronwald J, Huzarski T, Grzybowska E, Budryk M, Stawicka M, Mierzwa T, Szwiec M, Wisniowski R, Siolek M et al (2010) Pathologic complete response rates in young women with BRCA1-positive breast cancers after neoadjuvant chemotherapy. J Clin Oncol 28(3):375–379CrossRefPubMedGoogle Scholar
  14. 14.
    Silver DP, Richardson AL, Eklund AC, Wang ZC, Szallasi Z, Li Q, Juul N, Leong CO, Calogrias D, Buraimoh A et al (2010) Efficacy of neoadjuvant Cisplatin in triple-negative breast cancer. J Clin Oncol 28(7):1145–1153CrossRefPubMedPubMedCentralGoogle Scholar
  15. 15.
    Choi JE, Kang SH, Lee SJ, Bae YK (2015) Androgen receptor expression predicts decreased survival in early stage triple-negative breast cancer. Ann Surg Oncol 22(1):82–89CrossRefPubMedGoogle Scholar
  16. 16.
    Gucalp A, Traina TA (2010) Triple-negative breast cancer: role of the androgen receptor. Cancer J 16(1):62–65CrossRefPubMedGoogle Scholar
  17. 17.
    Creighton CJ, Li X, Landis M, Dixon JM, Neumeister VM, Sjolund A, Rimm DL, Wong H, Rodriguez A, Herschkowitz JI et al (2009) Residual breast cancers after conventional therapy display mesenchymal as well as tumor-initiating features. Proc Natl Acad Sci USA 106(33):13820–13825CrossRefPubMedPubMedCentralGoogle Scholar
  18. 18.
    Li X, Lewis MT, Huang J, Gutierrez C, Osborne CK, Wu MF, Hilsenbeck SG, Pavlick A, Zhang X, Chamness GC et al (2008) Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. J Natl Cancer Inst 100(9):672–679CrossRefPubMedGoogle Scholar
  19. 19.
    Denkert C, Loibl S, Noske A, Roller M, Muller BM, Komor M, Budczies J, Darb-Esfahani S, Kronenwett R, Hanusch C et al (2010) Tumor-associated lymphocytes as an independent predictor of response to neoadjuvant chemotherapy in breast cancer. J Clin Oncol 28(1):105–113CrossRefPubMedGoogle Scholar
  20. 20.
    Pardoll DM (2012) The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 12(4):252–264CrossRefPubMedPubMedCentralGoogle Scholar
  21. 21.
    Li XB, Krishnamurti U, Bhattarai S, Klimov S, Reid MD, O’Regan R, Aneja R (2016) Biomarkers predicting pathologic complete response to neoadjuvant chemotherapy in breast cancer. Am J Clin Pathol 145(6):871–878CrossRefPubMedGoogle Scholar
  22. 22.
    Li X, Wetherilt CS, Krishnamurti U, Yang J, Ma Y, Styblo TM, Meisel JL, Peng L, Siddiqui MT, Cohen C et al (2016) Stromal PD-L1 expression is associated with better disease-free survival in triple-negative breast cancer. Am J Clin Pathol 146(4):496–502CrossRefPubMedGoogle Scholar
  23. 23.
    Lehmann BD, Bauer JA, Chen X, Sanders ME, Chakravarthy AB, Shyr Y, Pietenpol JA (2011) Identification of human triple-negative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Investig 121(7):2750–2767CrossRefPubMedPubMedCentralGoogle Scholar
  24. 24.
    Lehmann BD, Pietenpol JA (2015) Clinical implications of molecular heterogeneity in triple negative breast cancer. Breast 24(Suppl 2):S36–S40CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Dent R, Trudeau M, Pritchard KI, Hanna WM, Kahn HK, Sawka CA, Lickley LA, Rawlinson E, Sun P, Narod SA (2007) Triple-negative breast cancer: clinical features and patterns of recurrence. Clinical Cancer Res 13(15 Pt 1):4429–4434CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  1. 1.Department of Pathology and Laboratory Medicine, Glenn Family Breast Center, Winship Cancer InstituteEmory UniversityAtlantaUSA
  2. 2.Department of Biostatistics and BioinformaticsEmory UniversityAtlantaUSA
  3. 3.Department of PathologyThe University of Texas MD Anderson Cancer CenterHoustonUSA
  4. 4.Department of MedicineUniversity of WisconsinMadisonUSA
  5. 5.Department of Radiation Oncology, Glenn Family Breast Center, Winship Cancer InstituteEmory UniversityAtlantaUSA
  6. 6.Department of Hematology and Medical Oncology, Glenn Family Breast Center, Winship Cancer InstituteEmory UniversityAtlantaUSA

Personalised recommendations