Decreased background parenchymal enhancement of the contralateral breast after two cycles of neoadjuvant chemotherapy is associated with tumor response in HER2-positive breast cancer

Abstract

Background

Several recent studies have focused on the association between background parenchymal enhancement (BPE) and tumor response to neoadjuvant chemotherapy (NAC), but early prediction of tumor response based on BPE has yet not been investigated.

Purpose

To retrospectively investigate whether changes in the BPE of the contralateral breast following NAC could help predict tumor response in early stage HER2-positive breast cancer.

Material and Methods

Data from 71 patients who were diagnosed with unilateral HER2 positive breast cancer and then underwent NAC with trastuzumab before surgery were analyzed retrospectively. Two experienced radiologists independently categorized the patients’ levels of BPE of the contralateral breast into four categories (1 = minimal, 2 = mild, 3 = moderate, 4 = marked) at baseline and after the second cycle of NAC. After undergoing surgery, 34 patients achieved pathologic complete response (pCR) and 37 patients had residual disease (non-pCR). The association between BPE and histopathologic tumor response was analyzed.

Result

The level of BPE was higher in premenopausal than post-menopausal women both at baseline and after the second cycle of NAC (P < 0.005). A significant reduction in BPE (P < 0.001) was observed after the second NAC cycle; however, a more obvious decrease in BPE was identified in premenopausal relative to post-menopausal women (P = 0.041). No significant association was identified between pCR and baseline BPE (P = 0.287). However, after the second NAC cycle, decreased BPE was significantly associated with pCR (P = 0.003).

Conclusion

For HER2-positive patients, changes in BPE may serve as an additional imaging biomarker of treatment response at an early stage.

Keywords

Background parenchymal enhancement magnetic resonance imaging (MRI)HER2-positive breast cancer neoadjuvant chemotherapy tumor response

Introduction

Breast parenchymal enhancement (BPE) is defined as the enhancement of normal breast tissue on MR mammography (MRM) after injection of contrast medium (1,2). Recently, BPE categories, including minimal, mild, moderate and marked, have been added to the updated American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) atlas (3).

Previous studies have demonstrated that BPE was influenced by endogenous or exogenous hormones and vascular anatomy (4). Authors have noted that BPE may be lowest during the second week of the menstrual cycle and higher in premenopausal women, lactating women, and women undergoing hormonal replacement therapy (5 –10). The effect of BPE on pretreatment has been further investigated. The presence of moderate or marked BPE surrounding breast tumors has been reported to affect the accuracy of tumor size estimation, possibly resulting in a positive resection margin or the performance of an overly extensive surgery (11,12). The literature also indicates that moderate or marked BPE before treatment was associated with response to neoadjuvant chemotherapy (NAC) (13 –15).

However, most previous studies of BPE have predominantly focused on tumor-induced changes in the surrounding parenchyma. A few studies have even investigated the association between BPE in the contralateral breast and therapy outcome; however, conclusions regarding this association remain uncertain (16 –19). Additionally, previous studies have enrolled patients with various phenotypes of breast cancers, and none of these previous studies have attempted to consider the alteration in BPE that may result from the use of different chemotherapy regimens or various course of NAC. Our study investigated the effect of NAC on BPE in patients with human epidermal growth factor receptor 2 (HER2) positive breast cancers who were undergoing the same chemotherapy regimen. Recently, preoperative neoadjuvant therapy with trastuzumab has resulted in notable tumor reduction during the early stages of NAC and markedly improved patient outcomes (20). Therefore, we focused on BPE of the contralateral normal breast, with a particular emphasis on the early stages of NAC (the second NAC cycle in our study).

In this retrospective study, we compared the level of BPE between HER2-positive breast cancer patients who achieved pathologic complete response (pCR) and those who did not (non-pCR). The aim was to analyze the effects of NAC on BPE of the healthy contralateral breast in HER2-positive breast cancer and to further investigate whether early changes in BPE could help predict tumor response.

Material and Methods

Patient collection

The institutional review board granted a waiver of authorization and waived the need for informed consent due to the retrospective nature of the study, which was conducted in compliance with the Health Insurance Portability and Accountability Act (HIPAA). Between April 2010 and September 2013, 79 patients who were diagnosed with HER2-positive breast cancers and underwent at least six cycles of neoadjuvant chemotherapy with trastuzumab during the study period were eligible for inclusion. The diagnosis of HER2-positive breast cancer was confirmed by core needle biopsy. Of the participants, eight were excluded because of insufficient magnetic resonance imaging (MRI) data (n = 3), interrupted therapy (n = 2), and identification of a contralateral breast tumor (n = 3). Finally, 71 patients (mean age = 47.65 ± 10.10 years; age range = 26–71 years) with unilateral breast cancer (70 invasive ductal carcinomas, one invasive lobular carcinoma) were enrolled in our study. All 71 individuals underwent contrast-enhanced breast MRI before and after NAC. The pre- or postmenopausal status of the patients was also noted. To avoid a delay in treatment, MRI was performed regardless of menstrual cycle, thus preventing an analysis of the effect of menstrual cycle point at MRI examination (19).

MRI study protocol

MR images were obtained using a 1.5-T dedicated spiral breast MRI System (Aurora Imaging Technology, Aurora Systems, Inc., Vancouver, Canada) with a breast coil. The images were acquired while the patients were in the prone position: The following sequences of images were obtained in the axial plane before and four times after a bolus injection of gadopentetate dimeglumine at 2 mL/s (Magnevist, Bayer Schering Pharma, Berlin, Germany) using an automatic injector: a T2-weighted (T2W) fat-suppressed sequence (TR =4008 ms, TE = 68 ms, thickness = 3 mm), and a spiral axial sequence (TR/TE = 5/29 ms, thickness = 1.1 mm, field of view [FOV] = 360 mm, matrix = 360 ×360 × 128). Post-contrast images were obtained 90, 180, 270, and 360 s after injection.

The baseline MRI scans were scheduled prior to the initiation of neoadjuvant therapy. The follow-up MRI scans were usually scheduled after the second, fourth and sixth cycles of NAC. In our study, we aimed to compare the MRI scans obtained before NAC and after the second cycle, the latter of which were performed after completion of the second cycle of neoadjuvant therapy and just before commencing third cycle. The same acquisition parameters were used for baseline and follow-up examinations.

NAC treatment protocol

HER2 protein levels were determined according to the recommendations of the HercepTest (21). The intensities of and patterns in HER2 membrane staining were evaluated using a scale with possible scores in the range of 0–3+. Tumors with immunohistochemical (IHC) scores of 3+ and 2+ were further tested by fluorescence in situ hybridization (FISH). Only those confirmed as HER2-positive by FISH were included herein. For HER2-positive patients, four weekly cycles of 80 mg/m² paclitaxel, carboplatin at an AUC of two in combination, and 2 mg/kg trastuzumab (on days 1, 8, 15, and 22; and a loading dose of 4 mg/kg provided for week 1) were administered.

Pathologic assessment

pCR was defined as the absence of residual invasive cancer upon hematoxylin and eosin evaluation of a complete resected breast specimen and all sampled regional lymph nodes following completion of neoadjuvant systemic therapy (20). Absence of in situ carcinoma was not required to meet the definition of pCR.

Data analysis

All images were independently reviewed by two radiologists with 20 and five years of experience, respectively, specialized in assessing breast imaging using the BI-RADS categories. BPE of the contralateral breast was evaluated on the early post-contrast phase of the dynamic set of images. BPE was categorized into four levels in the range of 1–4, which indicated minimal, mild, moderate, and marked enhancement, respectively. When BPE categorization differed between readers, two readers considered the BPE established upon consensus to be the overall assessment. The change in BPE was calculated as the BPE level identified in the MR image taken after the second cycle of NAC minus the BPE identified in the baseline MR image. Tumor size, corresponding to the maximal diameter of the tumor, was measured on early post-contrast images (90 s after contrast material injection).

Inter-reader agreement

Agreement between readers (inter-reader) for BPE was evaluated using the kappa statistics. The strength of the kappa agreement was categorized as: <0.00 = poor; 0.00–0.20 = slight; 0.21–0.40 = fair; 0.41–0.60 = moderate; 0.61–0.81 = substantial; and 0.81–1.00 = almost perfect.

Statistical analysis

The data were analyzed using the SPSS 16.0 software (SPSS Inc., Chicago, IL, USA. The t test and Pearson chi-squared test were used to compare the baseline characteristics of the pCR and Non-pCR groups. The Wilcoxon test was used to test for changes of BPE categories after NAC. The Mann–Whitney U test was used to test the difference of BPE between pre- and post-menopausal status. Associations were evaluated using binary logistic regression models. P < 0.05 was considered statistically significant.

Results

Patient cohort and tumor characteristics

The patient and tumor characteristics of the pCR and non-pCR groups are summarized in Table 1. The histopathological tumor response was identified as pCR in 34 cases (47.89%) and non-pCR in 37 cases (52.11%). We noted that only one patient who had cancer with a lobular component had residual cancer after NAC. pCR was achieved more frequently in patients with ER/PR-negative tumors (60%) than in those with ER/PR-positive status (32.26%) in HER2-positive breast cancer.

Table 1.

Baseline characteristics of the patient cohort overall and in the pCR and non-pCR groups.

Characteristic	All (n = 71)	pCR (n = 34)	Non-pCR (n = 37)	P value
Age (years)	47.65 ± 10.10	49.79 ± 9.56	45.68 ± 10.32	0.510
Largest tumor diameter on MR (mm)	62.63 ± 32.41	54.35 ± 26.72	70.24 ± 25.54	0.038
Histopathological finding				0.330
IDC	70	34	36
ILC	1	0	1
T				0.297
T1	5	1	4
T2	22	12	10
T3	5	1	4
T4	39	20	19
N				0.369
N0	5	3	2
N1	44	22	22
N2	8	5	3
N3	14	4	10
ER/PR status				0.020
ER/PR (–)	40	24	16
ER/PR (+)	31	10	21
Enhancement pattern of the tumor on MR				0.272
Mass	29	17	12
Non-mass like	30	13	17
Mass and non-mass like	12	4	8

T and N stages were both clinical stage. P value between pCR group and non-pCR group.

IDC, invasive ductal carcinoma; ILC, invasive lobular carcinoma; pCR, pathologic complete response; ER, estrogen receptor; PR, progesterone receptor.

BPE and NAC

A significant reduction in BPE category (Z = 6.434, P < 0.001) was observed in HRE2-positive breast cancer at follow-up. Decreased BPE was observed in 48 cases and no change in BPE were observed in 23 cases. Increased BPE was not observed (Table 2).

Table 2.

Background parenchymal enhancement (BPE) in premenopausal and non-premenopausal patients at baseline and after second NAC.

	BPE
	Minimal	Mild	Moderate	Mark
Baseline
Pre-menopausal	1	14	15	7
Post-menopausal	11	20	2	1
After NAC
Pre-menopausal	23	10	7	1
Post-menopausal	26	3	1	0

BPE and menopausal status

Thirty-seven patients were premenopausal and only nine of them had MRI in the second week of their menstrual cycle. Thirty-four patients were post-menopausal at baseline breast MRI. BPE was categorized as higher in premenopausal than post-menopausal women at baseline (U = 244.00, P < 0.001) and after two cycles of NAC therapy (U = 389.5, P = 0.003) (Table 2). The impact of changes in BPE of premenopausal and post-menopausal patients was also examined. Furthermore, compared with post-menopausal women, premenopausal patients had a significant reduction in BPE, which demonstrated a more obviously decreased BPE (U = 469.00, P = 0.041) (Table 3).

Table 3.

Changes in BPE after the 2^nd NAC cycle in premenopausal and post-menopausal women.

BPE categories of baseline status	Unchanged	Decreased
BPE categories of baseline status	Unchanged	Level 1	Level 2	Level 3
Pre-menopausal
Minimal (n = 1)	1 (100)	0 (0)	0 (0)	0 (0)
Mild (n = 14)	4 (29)	10 (71)	0 (0)	0 (0)
Moderate (n = 15)	3 (20)	6 (40)	6 (40)	0 (0)
Marked (n = 7)	1 (14)	4 (57)	0 (0)	2 (29)
Post-menopausal
Minimal (n = 11)	11 (100)	0 (0)	0 (0)	0 (0)
Mild (n = 20)	3 (15)	17 (85)	0 (0)	0 (0)
Moderate (n = 2)	0 (0)	0 (0)	2 (100)	0 (0)
Marked (n = 1)	0 (0)	1 (100)	0 (0)	0 (0)

Non-numeric data are presented as number of patients (%).

Association analysis

At baseline, no significant association between pCR and BPE was identified (P = 0.287). However, after the second NAC cycle, both the moderate or marked BPE identified after the second NAC as well as the reduction in BPE were significantly associated with pCR (P < 0.001, odds ratio [OR] = 43.3 and P = 0.003, OR = 0.219 respectively) (Table 4; Figs. 1 and 2).

Table 4.

Association between pCR and BPE.

Variables	OR	95% CI	P
BPE at baseline
Minimal/mild	1.00 (Reference)	–	–
Moderate/marked	0.568	0.01–1.61	0.287
BPE after second cycle
Minimal/mild	1.00 (Reference)	–	–
Moderate/marked	43.3	5.34–351.2	<0.001
BPE change
Unchanged	1.00 (Reference)	–	–
Decreased	0.219	0.080–0.602	0.003

Fig. 1.

A 41-year-old woman with invasive ductal carcinoma. The baseline MRI obtained in the early post-contrast phase (a) and acquired maximal intensity projection (MIP) obtained in the subtracted early post-contrast phase (b) both show marked BPE in the contralateral breast and multicentric lesions enhancement involving the right breast.

Fig. 2.

After the second NAC cycle, the MRI obtained in the early post-contrast phase (a) and acquired MIP obtained in the subtracted early post-contrast phase (b) both showed that the BPE decreased to the minimal grade and that the mass enhancement also notable decreased in size. The histopathological tumor response was complete remission (CR).

Inter-reader agreement

Inter-observer agreement regarding BPE categorization was substantial (ϰ = 0.701 at baseline and κ = 0.784 after the second NAC cycle).

Discussion

In our study, of the 71 HER2-positive patients with unilateral breast cancer, the identification of decreased BPE of the contralateral normal breast after the second cycle of NAC was found to be associated with tumor response.

Because BPE may be affected by hormone levels and administration of therapeutic agents, a few recent investigations have found that BPE on MRI may play a clinically significant role in assessing tumor response following NAC. However, whether the changes that occur in BPE at an early stage could predict tumor response at the end of treatment is not yet well known.

This study first investigated whether BPE of the contralateral normal breast could serve as an early predictor of pCR. Considering that different degrees of BPE changes may result from different chemotherapy regimens, HER2-positive breast cancer was used as the long research subject in our study. We analyzed BPE at baseline and after the second NAC cycle. We found that BPE at baseline did not predict tumor response but did find a reduction in BPE after the second NAC to be associated with pCR in HER2-positive breast cancer. Previous studies have concluded that the effect of BPE at baseline remains uncertain. One previous study suggested that higher BPE was observed in premenopausal patients with pCR relative to their non-pCR counterparts (17); however, another study reported the contradictory finding that a higher BPE on pre-NAC MRI was significantly associated with worse outcomes following NAC (15). These studies, however, did not consider the effect of immunohistochemistry (IHC) subtype. In our study, we did not find any association between BPE at baseline and tumor response in HER2-positive breast cancer. Further studies are needed in the future. Furthermore, our study demonstrated that the presence of significant decreased BPE after completion of the second NAC cycle was associated with pCR in HER2-positive breast cancer. As other studies have suggested, this decrease in BPE may represent the effects of chemotherapy agents on blood perfusion suppression, which could be associated with short-term (such as pCR) and long-term outcomes (such as disease-free survival) in patients who undergo NAC (13,14,16,17). Some published studies have aimed to assess alterations in BPE after the completion of NAC, while other studies have ignored this factor to unify BPE after NAC as a single observation point. Therefore, the results of our study not only confirmed the idea that the degree to which BPE decreased after NAC could predict outcome, a result that is in line with those of previous studies, but also revealed a new finding, suggesting that changes in BPE may serve as an early outcome predictor in HER2-positive breast cancer. Even though tumor size has been known to be effective in predicting pCR, the identification of the role of changes in BPE occurring at the early stage may be a promising addition to current risk assessments of NAC in HER2-positive patients.

Similar to the results of previous studies, our study identified significantly decreased BPE after NAC in HER2-positive patients. However, a previous study demonstrated that changes in BPE were only observed in premenopausal women (17). In our study, the effects of NAC on BPE were identified regardless of menopausal status. These findings may be explained as follows. Since BPE is related to the vascular supply and ovarian function, it is affected by endogenous hormone levels. Driven by chemotherapy-induced ovarian suppression, decreased BPE is most often identified in premenopausal women. Another possible reason is the suppression of blood perfusion. Direct damage of vessels in normal tissues may result from the loss of tissue proliferation after NAC, which can cause BPE to decrease in both pre- and post-menopausal women. As others have investigated, we also evaluated the effect of BPE on menopausal status. BPE was significant higher in premenopausal women than post-menopausal women both at baseline and after starting NAC. As BPE may be affected by endogenous hormone levels, it may be sensitive to the hormonal changes that coincide with changes in menopausal status. Our findings are in line with these studies (5,6).

In addition, only 34 (47.89%) of the 71 patients included in our study achieved pCR. The high pCR response rate observed herein was similar to that previously reported in HER2-positive breast cancer (22,23). In our study, women with ER/PR negative tumors were more prone to achieve pCR (60%) than those with ER/PR positive tumors (32.26%). Similar findings have been previously explained by the hypothesis that hormonal receptor status seemed to be an important escape mechanism mediating chemo-resistance in HER2-positive breast cancer (24,25).

There are a few limitations to the present study. First, because all included patients had biopsy-proven breast cancer, MRI was not always performed during the recommended menstrual cycle. Second, another limitation of our study was the qualitative evaluation of BPE, making it potentially prone to inter-observer variability and difficulties in avoiding the effect of MRI noise contamination. While the identified inter-observer agreement was substantial, it is still necessary to confirm our data with quantitative measurements, such as fully automated segmented methods. By using quantitatively analysis, we hope the evaluation of BPE combined with the tumor-induced changes, such as tumor size, ADC value, Ktrans value, will form a new imaging predicting model of NAC. Last, this was a retrospectively study and validation in larger populations is required to further understand BPE changes in different breast cancer subtypes.

In conclusion, our results suggest that for HER2-positive patients undergoing NAC plus trastuzumab, a reduction of BPE of the contralateral normal breast after the second NAC cycle could be an early predictor of tumor response. Changes in BPE also showed potential as an imaging biomarker of treatment response and prognosis. Further studies are needed to investigate the impact of BPE combined with the change of tumor size on different breast cancer subtypes in terms of disease-free survival and overall survival.

Footnotes

Acknowledgements

The authors thank all the patients for their participation in the study. They also thank Xiaoxin Hu, Jian Mao, Lei Yue, and Jie Tang at the Department of Radiology and Wentao Yang at the Department of Pathology for their excellent assistance.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported (in part) by the grants from Shanghai Development Center for Construction Foundation of Clinical Supportive Department (Grant no. SHDC22015030), National Public Welfare Foundation (Grant no. 201402020), and Jiangsu Natural Science Foundation (Grant no. BK20150476).

References

Ikeda

Hylton

Kuhl

et al.

BI-RADS: Magnetic Resonance Imaging. In: D’Orsi

Mendelson

Ikeda

et al. (eds). Breast Imaging Reporting and Data System: ACR BI-RADS, breast imaging atlas, 1st ed. Reston, VA: American College of Radiology, 2003.

Sippo

Warden

Andriole

et al.

Automated extraction of BI-RADS final assessment categories from radiology reports with natural language processing. J Digit Imaging 2013; 26: 989–994.

Morris

Comstock

Lee

et al.

BI-RADS: magnetic resonance imaging 2013. In: D’Orsi

Sickles

Mendelson

et al. (eds). Breast Imaging Reporting and Data System: ACR BI-RADS, breast imaging atlas, Reston, VA: American College of Radiology, 2013, pp. 27–32.

Giess

Yeh

Raza

et al.

Background parenchymal enhancement at breast MR imaging: normal patterns, diagnostic challenges, and potential for false-positive and false-negative interpretation. Radiographics 2014; 34: 234–247.

Kuhl

Bieling

Gieseke

et al.

Healthy premenopausal breast parenchyma in dynamic contrast-enhanced MR imaging of the breast: normal contrast medium enhancement and cyclical-phase dependency. Radiology 1997; 203: 137–144.

King

Kaplan

et al.

Impact of menopausal status on background parenchymal enhancement and fibroglandular tissue on breast MRI. Eur Radiol 2012; 22: 2641–2647.

Ellis

. Optimal timing of breast MRI examinations for premenopausal women who do not have a normal menstrual cycle. Am J Roentgenol 2009; 193: 1738–1740.

Taylor

Lazberger

Ives

et al.

Reducing delay in the diagnosis of pregnancy-associated breast cancer: how imaging can help us. J Med Imaging Radiat Oncol 2011; 55: 33–42.

King

Kaplan

Pike

et al.

Impact of tamoxifen on amount of fibroglandular tissue, background parenchymal enhancement, and cysts on breast magnetic resonance imaging. Breast J 2012; 18: 527–534.

10.

Oksa

Parkkola

Luukkaala

et al.

Breast magnetic resonance imaging findings in women treated with toremifene for premenstrual mastalgia. Acta Radiol 2009; 50: 984–989.

11.

Baek

Kim

Lee

. Background parenchymal enhancement in breast MRIs of breast cancer patients: impact on tumor size estimation. Eur J Radiol 2014; 83: 1356–1362.

12.

Uematsu

Kasami

Watanabe

. Does the degree of background enhancement in breast MRI affect the detection and staging of breast cancer? Eur Radiol 2011; 21: 2261–2267.

13.

Kim

et al.

Enhancement parameters on dynamic contrast enhanced breast MRI: do they correlate with prognostic factors and subtypes of breast cancers?

Magn Reson Imaging 2015; 33: 72–80.

14.

Kim

Cho

Ryu

et al.

Background parenchymal signal enhancement ratio at preoperative MR imaging: association with subsequent local recurrence in patients with ductal carcinoma in situ after breast conservation surgery. Radiology 2014; 270: 699–707.

15.

Choi

et al.

Background parenchymal enhancement on preoperative magnetic resonance imaging: association with recurrence-free survival in breast cancer patients treated with neoadjuvant chemotherapy. Medicine (Baltimore) 2016; 95: e3000–e3000.

16.

van der Velden

Dmitriev

Loo

et al.

Association between parenchymal enhancement of the contralateral breast in dynamic contrast-enhanced MR imaging and outcome of patients with unilateral invasive breast cancer. Radiology 2015; 276: 675–685.

17.

Chen

Hsu

et al.

Background parenchymal enhancement of the contralateral normal breast: association with tumor response in breast cancer patients receiving neoadjuvant chemotherapy. Transl Oncol 2015; 8: 204–209.

18.

Chen

Lin

et al.

Background parenchymal enhancement in the contralateral normal breast of patients undergoing neoadjuvant chemotherapy measured by DCE-MRI. Magn Reson Imaging 2013; 31: 1465–1471.

19.

Preibsch

Wanner

Bahrs

et al.

Background parenchymal enhancement in breast MRI before and after neoadjuvant chemotherapy: correlation with tumour response. Eur Radiol 2016; 26: 1590–1596.

20.

Ogston

Miller

Payne

et al.

A new histological grading system to assess response of breast cancers to primary chemotherapy: prognostic significance and survival. Breast 2003; 12: 320–327.

21.

Wolff

Hammond

Hicks

et al.

Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol 2013; 31: 3997–4013.

22.

Liu

Huang

et al.

Similar prognoses for invasive micropapillary breast carcinoma and pure invasive ductal carcinoma: a retrospectively matched cohort study in China. PLoS One 2014; 9: e106564–e106564.

23.

Cheng

Wang

et al.

18F-fluorodeoxyglucose (FDG) PET/CT after two cycles of neoadjuvant therapy may predict response in HER2-negative, but not in HER2-positive breast cancer. Oncotarget 2015; 6: 29388–29395.

24.

Wang

Morrison

Gillihan

et al.

Different mechanisms for resistance to trastuzumab versus lapatinib in HER2-positive breast cancers–role of estrogen receptor and HER2 reactivation. Breast Cancer Res 2011; 13: R121–R121.

25.

Vaz-Luis

Winer

Lin

. Human epidermal growth factor receptor-2-positive breast cancer: does estrogen receptor status define two distinct subtypes? Ann Oncol 2013; 24: 283–291.