Surgical Oncology 27 (2018) 185e191

Contents lists available at ScienceDirect

Surgical Oncology
journal homepage: www.elsevier.com/locate/suronc

Clinical significance of internal mammary lymph node metastasis for

breast cancer: Analysis of 337 breast cancer patients
Xiao-wei Qi a, Jun-ze Du a, Peng Tang a, Xue Liu b, Qing-qing He a, Ling Zhong a,
Xiu-wu Bian b, **, Jun Jiang a, *
a
Breast Disease Center, Southwest Hospital, Third Military Medical University, Chongqing 400038, China
b
Institute of Pathology and Southwest Cancer Center, Southwest Hospital, Third Military Medical University, Chongqing 400038, China

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: Internal mammary nodes (IMNs) is a major pathway of lymphatic drainage for breast cancer,
Received 1 February 2018 apart from axillary lymph node (ALN). However, owing to lack of a feasible and safe biopsy method,
Received in revised form management of IMNs is still controversial in breast surgery.
7 March 2018
Methods: From 2005 to 2009, a total of 337 consecutive breast cancer women patients were recruited. All
Accepted 26 March 2018
patients underwent IMNs biopsy through intercostal space or endoscopic lymphatic chain resection. The
ER, PR and HER-2 status were retested according to the current ASCO/CAP guidelines. We analyzed the
Keywords:
relationship between clinical pathological parameters and IMNs metastasis and investigated the high
Internal mammary nodes
Biopsy
risk factors and prognostic values of IMNs metastasis in breast cancer.
Risk factor Results: Among 337 patients, 314 patients underwent intercostal space IMNs biopsy and 23 patients
Prognosis underwent endoscopic lymphatic chain resection. A total of 63 (18.69%) patients were pathologically
Breast cancer diagnosed with IMNs metastasis. Among them, 28 (44.44%) patients changed the pathological lymph
node staging, and 15 cases (23.81%) changed the postoperative comprehensive treatment program and
accepted extended postoperative radiotherapy. Multivariate analysis showed that compared with no ALN
involvement, the risk of IMNs metastasis was significantly increased in patients with 1e3 ALN
involvement (OR ¼ 42.097, 95% CI ¼ 5.225e339.178; P ¼ 0.0004) and 4 ALN involvement (OR ¼ 82.429,
95%CI ¼ 10.134e670.496; P < 0.0001). The risk of IMNs metastasis in HER-2 positive patients was
significantly higher than that in negative patients (OR ¼ 5.452, 95% CI ¼ 2.353e12.634; P < 0.0001).
However, we did not find IMNs involvement was an independent indicator for both overall survival and
disease-free survival.
Conclusions: Our clinical practice and data indicated that IMNs biopsy through intercostal space and
endoscopic lymphatic chain resection are effective and minimally invasive methods to detect the IMNs
status, which may be helpful for accurate tumor staging, risk assessment and option of chemotherapy or
radiotherapy to improve the patients' survival.
© 2018 Elsevier Ltd. All rights reserved.

1. Background cancer having metastases to the IMN, the management of IMN in

Regional lymph node management is an important part of
breast cancer surgery. The internal mammary nodes (IMN) drain
about 25% of all lymphatics of the breast [1,2]. Although current
evidence suggests that the IMN chain is a major pathway of
lymphatic drainage, with 18%e33% of patients with operable breast
* Corresponding author.
** Corresponding author.
E-mail addresses: bianxiuwu@263.net (X.-w. Bian), jcbd@medmail.com.cn
(J. Jiang).
breast cancer is still controversial [3e6].
The IMN chain is located behind the intercostal muscles and
costal cartilages and lies close to the internal mammary vein and
artery [7]. In the 1950s it became customary to remove these nodes
during radical mastectomy [8]. Initial studies showed that about
one-third of breast cancer patients had IMN involvement [9e11]
and that patients with IMN metastasis had worse outcome. It was
suggested that metastases to axillary lymph nodes (ALN) or the
IMN had the same prognostic relevance for breast cancer patients
[12e19]. However, randomized trials [20e24] later showed that
overall survival (OS) and disease-free survival (DFS) in breast

https://doi.org/10.1016/j.suronc.2018.03.006
0960-7404/© 2018 Elsevier Ltd. All rights reserved.
186 X.-w. Qi et al. / Surgical Oncology 27 (2018) 185e191
cancer patients were comparable with extensive radical mastec-
tomy and with radical mastectomy, while the quality of life of pa-
tients was worse and surgical complications more with the former
surgery. As a result, extensive radical mastectomy was gradually
abandoned.
IMN sampling can be useful for nodal staging and can help guide
therapy [25e35]. For example, addition of local radiotherapy to the
treatment regimen has been shown to benefit patients with IMN
metastases [36,37]. Therefore, surgical sampling remains crucial for
management of breast cancer [38,39]. However, there is no safe
method for biopsy of IMN, and the management of IMN remains a
problem in breast cancer surgery.
The purpose of this study was to identify the risk factors asso-
ciated with metastasis to the IMN, and to determine the relation-
ship between IMN metastasis and prognosis.
2. Patients and methods
2.1. Patient selection
A total of 337 consecutive breast cancer women patients treated
at the Southwest Hospital, Third Military Medical University,
Chongqing, China, during 2005e2009 were included in this retro-
spective study. Patients with newly diagnosed, histopathologically
confirmed breast cancer were eligible for inclusion, with no re-
strictions of age or pathological types. The exclusion criteria were
pregnancy, presence of distant metastasis, unwillingness to un-
dergo biopsy/surgical procedures, congestive heart failure,
ischemic heart disease, other malignancies, severe hepatic or renal
dysfunction, and altered mental status.
This study was approved by the Clinic Ethics Review Committee
of Southwest Hospital, Third Military Medical University. Written
informed consent was obtained from all participants.
2.2. Surgical procedure
2.2.1. Intercostal space IMN biopsy
The details of the procedure have been reported previously.
Briefly, a retractor was used to expose the intercostal space, and the
thoracic transverse fascia was incised. The internal thoracic vessels
were identified by careful dissection. The internal mammary lymph
nodes, which are usually located close to the internal thoracic
vessels, were readily identified anterior, lateral, or medial to the
vessels. If the lymph nodes were not found in the intercostal space,
the costal cartilage was sought and the two costal cartilages above
and below the intercostal space were resected. Care was taken to
avoid injury to the perforating branches of the internal mammary
artery to the intercostal space. After biopsy of the nodes, fine silk
suture was used to close the separated pectoralis major muscle.
2.2.2. Endoscopic lymphatic chain resection
The details of the procedure have been reported previously.
Briefly, trocars were introduced into thoracic cavity through three
<15 mm incisions placed at the third, fifth, and seventh intercostal
spaces along the midaxillary line. A 10-mm rigid 30 thoracoscope
was used to visualize the thoracic cavity and identify and expose
the internal thoracic vessels and the IMN chain. Using an ultra-
sonically activated scalpel, the distal parts of the internal thoracic
vessels were sectioned between two vascular clips, and all collat-
eral branches issuing from the internal thoracic vessels were
transected. This allowed complete dissection of the IMN chain from
the first to the sixth rib. The resected specimen was removed
through the 10-mm thoracoscope Because the IMN chain lies close
to the parietal pleura, special care was taken to avoid injury to the
pleura during intercostal space IMN biopsy or endoscopic
lymphatic chain resection. Small pleural tears were either left alone
or sutured with nonabsorbable material.
2.3. Pathologic and immunohistochemical analysis
All breast tumor and lymph node specimens were sent for
pathologic evaluation to the Institute of Pathology and Southwest
Cancer Center, Southwest Hospital, Third Military Medical Univer-
sity, Chongqing, China. Immunostaining for estrogen receptor (ER),
progesterone receptor (PR), and human epidermal growth factor
receptor 2 (HER-2) protein was re-performed on consecutive tissue
sections before the current data analysis, considering the changes
in the criteria for diagnosis of ER, PR, and HER-2 positivity as well as
technological advances since the time of the operations. The tumor
was regarded as positive for ER and PR if  1% of the cells showed
nuclear staining. HER-2 positivity was defined as intense and
complete membrane staining in >10% of the tumor cells or HER-2
gene amplification by fluorescence in situ hybridization assay.
2.4. Postoperative treatment and follow up
Postoperative radiotherapy and adjuvant treatments were
applied according to the NCCN Clinical Practice Guidelines in
Oncology. All patients were followed up in the outpatient depart-
ment at 3 monthly intervals. Median follow-up duration was 56
months (range, 13e107 months).
2.5. Statistical analysis
Differences in clinical and pathological characteristics between
IMN positive and IMN negative groups were evaluated by the chi-
square test or t-test. The endpoints of prognosis were OS (defined
as the time from surgery until the date of death due to any cause)
and DFS (defined as the time from surgery to the date of local
relapse or distant metastasis). OS and DFS were analyzed by the
KaplaneMeier method, and the log-rank test was used to assess
differences between groups.
Multivariate analysis to identify the factors independently
associated with prognosis was performed using the Cox propor-
tional hazards model. The variables entered into the model were
age (as a continuous variable); side of tumor (right, left); site of
tumor (upper inner quadrant, lower inner quadrant, upper outer
quadrant, lower outer quadrant, and central); T stage (<2 cm,
2 cm); number of involved ALN (0, 1e3, 4); type of IMN oper-
ation (dissection, biopsy); pathological type (invasive ductal car-
cinoma, others); ER status (ERþ, ER); PR status (PRþ, PR); HER2
status (HER-2þ, HER-2); and IMN status (IMNþ, IMN). The odds
ratio (OR) and hazard ratio (HR) and the corresponding 95% con-
fidence intervals (CI) were calculated.
GraphPad Prism, version 6.0 (GraphPad Software Inc., La Jolla,
CA, USA) was used for drawing the KaplaneMeier curves and for
the log-rank test. All other analyses were performed using SAS
software, version 9.4 (SAS Institute, Cary, NC, USA). All tests were
two-sided. P < 0.05 was considered statistically significant.
3. Results
3.1. Overview of IMN metastasis in the study cohort
The mean age of the 337 patients was 46.17 ± 9.68 years (range,
23e79 years). The tumor was located on the left side in 189 patients
and on the right side in 148 patients. Of the 337 patients, 314 un-
derwent intercostal space IMN biopsy (Fig. 1), with the mean
number of IMN removed being 1.89 ± 1.39. The remaining 23 pa-
tients underwent endoscopic lymphatic chain resection (Fig. 2),
 X.-w. Qi et al. / Surgical Oncology 27 (2018) 185e191 187

Table 1
Univariate analysis of IMNs metastasis and clinicalpathological factors in breast
cancer patients.

IMN-(N ¼ 274) IMNþ(N ¼ 63) P

Age (year)
50 180(65.69) 41(65.08) 0.9263
>50 94(34.31) 22(34.92)
Site
upper inner 47(17.15) 15(23.81) 0.6046
lower inner 18(6.57) 2(3.17)
upper outer 138(50.36) 29(46.03)
lower outer 30(10.95) 6(9.52)
central 41(14.96) 11(17.46)
Tumor diameter
<2 cm 106(38.69) 14(22.22) 0.0139
2 cm 168(61.31) 49(77.78)
ALN positive number
0 142(51.82) 1(1.59) <.0001
Fig. 1. Biopsy of internal mammary lymph nodes through intercostal space. 1e3 74(27.01) 20(31.75)
4 58(21.17) 42(66.67)
Pathological type
infiltrating ductal carcinoma 212(77.37) 48(76.19) 0.3036
other 62(22.63) 15(23.81)
ER
þ 137(50) 30(47.62) 0.7332
e 137(50) 33(52.38)
PR
þ 145(52.92) 25(39.68) 0.0581
e 129(47.08) 38(60.32)
HER-2
þ 237(86.5) 36(57.14) <.0001
e 37(13.5) 27(42.86)

ALN: axillary lymph node, ER: estrogen receptor, PR: progesterone, HER-2: human
epidermal growth factor receptor 2.

Table 2
Multivariate analysis of IMNs metastasis and clinicalpathological factors in breast
cancer patients.
Fig. 2. Resection of internal mammary lymph nodes through endoscopy. OR 95%CI P

UL LL

with the mean number of IMN removed being 2.62 ± 1.61. Among Agea 0.985 0.948 1.023 0.4253
the 337 patients, 63 (18.69%) had pathologically confirmed IMN Side
right vs. left 0.600 0.283 1.271 0.1820
metastasis.
Site
upper inner reference
lower inner 1.190 0.159 8.892 0.8656
3.2. Risk factors for IMN metastasis in breast cancer
upper outer 0.517 0.200 1.332 0.1717
lower outer 0.297 0.078 1.126 0.0742
Univariate analysis (Table 1) showed that IMN metastasis was central 0.717 0.225 2.291 0.5752
significantly associated with tumor size >2 cm (P ¼ 0.0139), Tumor diameter
increased number of involved ALN (P < 0.0001), and HER-2 positive 2 cm vs.<2 cm 0.919 0.414 2.040 0.8363
ALN positive number
status (P < 0.0001). Age, tumor location, pathological type, and ER 0 reference
and PR status (P > 0.05) were not associated with IMN metastasis. 1e3 42.097 5.225 339.178 0.0004
Multivariate analysis (Table 2) showed that, compared with 4 82.429 10.134 670.496 <.0001
patients without ALN involvement, the risk of IMN metastasis was IMNs operation
dissection vs. biopsy 2.021 0.580 7.045 0.2695
significantly higher in patients with 1e3 involved ALN
Pathological type
(OR ¼ 42.097, 95% CI: 5.225e339.178; P ¼ 0.0004) and 4 involved IDC vs other 0.599 0.255 1.411 0.2412
ALN (OR ¼ 82.429, 95% CI: 10.134e670.496; P < 0.0001). The risk of ER
IMN metastasis was significantly higher with HER-2 positive tu- þ vs. - 0.684 0.295 1.587 0.3768
mors than with HER-2 negative tumors (OR ¼ 5.452, 95% CI: PR
þ vs. - 2.374 0.953 5.912 0.0632
2.353e12.634; P < 0.0001). HER2
þ vs. - 5.452 2.353 12.634 <.0001

3.3. Multivariate analyses of impact of IMN metastasis on prognosis a: continuous variable; ALN: axillary lymph node, IMN: internal mammary nodes,
(DFS and OS) IDC: invasive ductal carcinoma ER: estrogen receptor, PR: progesterone, HER-2:
human epidermal growth factor receptor 2.

As Tables 3 and 4 show, only the number of involved ALN (1e3

involved ALN and 4 involved ALN) were independent prognostic DFS (HR ¼ 2.538, 95% CI: 1.282e5.023; P ¼ 0.0075; and HR ¼ 2.934,
factors for OS (HR ¼ 4.327, 95% CI: 1.833e10.215; P ¼ 0.0008; and 95% CI: 1.373e6.268; P ¼ 0.0055; respectively). IMN involvement
HR ¼ 5.929, 95% CI: 2.297e15.308; P ¼ 0.0002; respectively) and was not an independent prognostic factor for either OS
188 X.-w. Qi et al. / Surgical Oncology 27 (2018) 185e191

Table 3 (HR ¼ 0.482, 95% CI: 0.210e1.109; P ¼ 0.0861) or DFS (HR ¼ 0.678,
Multivariate analysis of IMNs metastasis and disease-free survival (DFS) in breast 95% CI: 0.349e1.316; P ¼ 0.2512).
cancer patients.

HR 95%CI P
4. Discussion
UL LL

Agea 1.001 0.976 1.027 0.9337 In the present analysis, intercostal space and endoscopic
Side
right vs. left 0.728 0.431 1.230 0.2357
lymphatic chain resection were used to detect the IMN in 337
Site breast cancer patients. IMN metastasis was detected in 18.69% of
upper inner reference the patients. ALN involvement and HER-2 positive status were
lower inner 0.371 0.047 2.952 0.3487 associated with increased risk of IMN metastasis. However, IMN
upper outer 1.031 0.504 2.106 0.9342
metastasis was not an independent prognostic factor for DFS and
lower outer 1.474 0.608 3.578 0.3907
central 1.345 0.549 3.297 0.5172 OS, which was probably because all patients with IMN metastasis in
Tumor diameter this cohort received tailored treatment, including local radiology
2 cm vs.<2 cm 1.152 0.665 1.997 0.6132 and systemic chemotherapy.
ALN positive number Breast lymphatic drainage is mainly to the ALN. However, 9%e
0 reference
1e3 2.538 1.282 5.023 0.0075
45% of breast lymphatics also drain into the IMN chain, and 1%e5%
4 2.934 1.373 6.268 0.0055 drain only to the IMN chain [40]. Despite this, routine sampling of
IMNs operation IMN is not part of current standard breast cancer surgery man-
dissection vs. biopsy 2.076 0.975 4.421 0.0582 agement, primarily because there is no effective and feasible
Pathological type
method to identify IMN metastasis; moreover, there is little evi-
IDC vs. other 1.121 0.597 2.106 0.7217
ER dence to show that IMN metastasis per se is associated with poor
þ vs. - 0.541 0.289 1.013 0.0548 prognosis. The rate of IMN metastasis in breast cancer is reported to
PR be ~30% [9,10]. Recently, a combined analysis of six prospective
þ vs. - 0.841 0.440 1.609 0.6008 studies that used lymphoscintigraphy showed that 20.4% of 3876
HER2
þ vs. - 1.394 0.750 2.590 0.2938
breast cancers had lymphatic drainage to the IMN chain [41]. A
IMN meta-analysis of 15 studies (with a total of 4248 patients) found
þ vs. - 0.678 0.349 1.316 0.2512 metastasis to the IMN in 23% of patients [42]. Our review of liter-
a: continuous variable; ALN: axillary lymph node, IMN: internal mammary nodes, ature showed that metastasis to the IMN ranges from 2.9% to 32.7%
IDC: invasive ductal carcinoma ER: estrogen receptor, PR: progesterone, HER-2: (Table 5). In the present study, 63 patients (18.69%) were diagnosed
human epidermal growth factor receptor 2. with IMN metastases, which is consistent with previous reports.

Table 5
Frequency of IMNs metastasis in breast cancer patients based on literature review.
Table 4 Author Year No. of patients IMN þ patients (%)
Multivariate analysis of IMNs metastasis and overall survival (OS) in breast cancer
patients. Margotti et al. 1949 110 22.7
Handley and Thackray 1954 150 32.7
HR 95%CI P Hutchinson et al. 1956 81 25.8
Caceres et al. 1959 250 21.6
UL LL
Bucalossi et al. 1971 1213 22
Agea 1 0.97 1.031 0.9826 Donegan et al. 1977 113 22.1
Side Veronesi et al. 1985 1119 19
right vs. left 0.719 0.386 1.34 0.2987 Horino et al. 1991 410 18.5
Site Cody and Urban 1995 195 24
upper inner reference U. Veronesi et al. 1999 342 20.5
lower inner 0.392 0.047 3.26 0.386 Sugg et al. 2000 286 25.2
upper outer 0.618 0.257 1.485 0.2824 Dupont et al. 2001 41 26.8
lower outer 0.874 0.297 2.577 0.8076 vander Ent et al. 2001 30 16.7
central 1.624 0.602 4.383 0.3385 Arnez et al. 2005 11 9.1
Tumor diameter Hofer et al. 2005 13 23.4
2 cm vs.<2 cm 1.483 0.749 2.938 0.2583 Madsen et al. 2007 85 23.5
ALN positive number Huang et al. 2008 1679 15.5
0 reference Knight et al. 2008 26 19.2
1e3 4.327 1.833 10.215 0.0008 Heuts et al. 2009 115 24
4 5.929 2.297 15.308 0.0002 Yu et al. 2011 43 14.0
IMNs operation Long et al. 2011 50 40.0
dissection vs. biopsy 0.846 0.298 2.398 0.7526 Andree et al. 2012 195 3.1
Pathological type Elif Hindie et al. 2012 644 17.2
IDC vs. other 0.833 0.395 1.757 0.631 Schaverien et al. 2013 63 20.6
ER Wong et al. 2013 86 12.8
þ vs. - 0.496 0.235 1.049 0.0667 Gnerlich et al. 2014 122 10
PR Caudle et al. 2014 71 15
þ vs. - 0.541 0.244 1.2 0.1307 Gant et al. 2014 717 6.9
HER2 Hellman et al. 2014 32 12.5
þ vs. - 1.24 0.59 2.606 0.5701 Madsen et al. 2015 754 21.3
IMN Ozmen et al. 2015 72 14
þ vs. - 0.482 0.21 1.109 0.0861 Wright et al. 2016 204 2.9
Huang et al. 2016 95 18.1
a: continuous variable; ALN: axillary lymph node, IMN: internal mammary nodes,
García et al. 2016 63 9.5
IDC: invasive ductal carcinoma ER: estrogen receptor, PR: progesterone, HER-2:
Current study 2018 337 18.7
human epidermal growth factor receptor 2.
 X.-w. Qi et al. / Surgical Oncology 27 (2018) 185e191
Thus, the evidence suggests that the IMN chain is also a major site
of lymphatic metastasis in breast cancer patients, and therefore
management of IMN should not be ignored in breast surgery.
In this study, the IMN positivity rate in T1 and N0 stage patients
was 11.67% (14/120) and 0.70% (1/143), respectively. This shows
that IMN metastases may be present even in early-stage breast
cancer. Previous studies have demonstrated that evidence of IMN
metastasis can change the pathologic stage of the cancer and
thereby the management approach [27,31,32,43,44]. In our study
cohort, the biopsy findings changed the pathological N stage in 28/
63 (44.44%) patients, and resulted in addition of local radiotherapy
to the treatment program in 15/63 (23.81%) patients. Thus, IMN
biopsy is necessary for accurate risk assessment and choice of the
treatment approach [42]. Lack of information on IMN metastatic
status may lead to inaccurate staging and over-treatment or under-
treatment.
Currently, there is no reliable and widely accepted method for
determining IMN metastatic status. Although ultrasound, CT, MRI,
or PET-CT have been reported to be useful [45e50], these modal-
ities have low sensitivity for diagnosis of IMN metastasis, mainly
due to the deep anatomical location of the IMN chain and the small
size of the nodes [5,51]. According to the NCCN guidelines, PET-CT
could be considered for preoperative breast cancer examinations,
especially for some patients with advanced breast cancer. However,
due to the high cost (about 2000 USD) of PET-CT and exclusion from
reimbursement of medical insurance in China, only 5 patients
received PET-CT examination before biopsy, which was not
adequate for statistical analysis. Therefore we did not report the
relevant data in the study. Among the 5 patients, 3 got suspicious
INM image through PET-CT, including one patient confirmed with
metastasis. There are few reports on the relationship between PET
avidity and positivity of pathological IMN involvement. In a clinical
trial published in 2014 [52], 62 of 216 patients with operable breast
cancer were found to have suspicious INM through PET-CT. Among
the 62 patients, 31 were confirmed with INM by biopsy. Among the
31 patients, 27 had metastatic IMN. Therefore, the positive pre-
dictive value for INM of PET-CT was 88.7%. In another study pub-
lished in 2017 [53], 67 patients with suspected INM were found by
PET-CT in 114 patients with operable breast cancer. Among the 67
patients, 37 (55%) were diagnosed with INM metastasis by pa-
thology. The above findings suggest that PET-CT has potential
clinical value and significance in the diagnosis of INM metastasis,
but whether it can be routinely used in clinical practice still needs
further research. Recently, lymphoscintigraphy was used to deter-
mine drainage to the IMN [3]. However, it has not proved to be
useful for assessment of nodal involvement [38]. Coombs et al.
attempted to apply lymphoscintigraphy for identification of IMN
metastasis in 490 breast cancer patients, and reported that the
while drainage into the IMN chain was detected in 90 patients, only
20 of these patients (i.e., 4% of the total) had pathologically
confirmed metastasis [54]. Some authors have used lympho-
scintigraphy evidence of drainage to the IMN as indicative of
metastasis [3,28,55e57]; however, this cannot be assumed to be a
reliable method. Therefore, surgical procedures is still crucial for
determining IMN metastatic status. Several clinical studies
[7,14,17,33,50] and one cadaveric study [58] have shown that IMN
metastases are mostly located in the first and fourth intercostal
spaces or posterior to the costal cartilages. In our clinical practice,
we use conventional surgical methods for biopsy of IMNs in the
intercostal space [59], and endoscopic biopsy for IMNs situated
posterior to the costal cartilage [60,61]. The detection efficiency and
IMN positivity rate with our approach were comparable with that
achieved by routine or opportunistic biopsy of IMN during mini-
mally invasive plastic surgery for breast cancer [44,62e65].
Because no guidelines are available for IMN biopsy in routine
189
clinical practice, it could be useful to be able to identify the sub-
group of patients who are likely to have IMN metastasis. The
strongest predictor of IMN involvement has been reported to be
ALN positivity [66], and this was confirmed in our study. In addi-
tion, consistent with a recent study [67], we found that HER-2
positive status also increased the likelihood of IMN metastases.
We found no association between IMN involvement and tumor
location. However, this finding was contrary to the evidence from a
recent meta-analysis and therefore warrants further investigation
[42].
The impact of IMN involvement on prognosis of breast cancer
patients is controversial. Initial studies showed IMN involvement
had the same prognostic relevance as ALN involvement in breast
cancer patients [12e19]. However, results from subsequent ran-
domized trials did not confirm this. For example, Veronesi et al.
found that removal of the IMN chain did not improve the survival of
patients with breast carcinoma [23]. Similar results were reported
by some recent studies [30,33,68]. In the current study, multivariate
analyses did not reveal any impact of IMN involvement on prog-
nosis of breast cancer patients. The likely reason might well be that
our patients received tailored systemic treatments (including
adjuvant endocrine treatments, chemotherapy, and anti-HER-2
therapy) and possibly also radiotherapy to the IMN chain [33,69].
Interestingly, a recent study [53] reported that patients with fine-
needle aspiration (FNA)eproven IMN metastasis have marginally
lower regional recurrenceefree survival than FNA () patients in
cN2b/N3b breast cancer. However, it should be noted that all pa-
tients in that study received external radiotherapy, including to the
IMN compartment, which suggests that some patients may have
been overtreated and therefore the assessment of prognosis may
not be reliable.
In conclusion, the results of this study indicate that IMN biopsy
through the intercostal space and endoscopic lymphatic chain
resection are effective minimally invasive methods for detecting
IMN status in breast cancer patients. Determination of IMN status
allows accurate tumor staging and risk assessment. Anatomic
staging combined with biological profiles could further improve
the precise treatment of breast cancer in the era of ACOSOG Z0011
[70,71] advocating lesser surgery and using gene expression
signature (MINDACT [72] and TAILORx [73] etc) to stratify recur-
rence risk.
Competing financial interests
The authors declare no competing financial interests.
Funding
This study was supported by grants from the Clinical Research
Founding of Southwest Hospital (No. SWH2005B007,
SWH2016BZGFKJ-30), Technological Innovation for Intelligent
Medicine of Southwest Hospital (No. SWH2016ZDCX440 3), and
Talents Training Program of Third Military Medical University (No.
2017MPRC-18).
Acknowledgments
We would like to extend our sincere gratitude and appreciation
to Ying Zhang from University of San Francisco, San Francisco, CA,
USA, for proof-reading.
References
[1] F. Hassiotou, D. Geddes, Anatomy of the human mammary gland: current
status of knowledge, Clin. Anat. 26 (2013) 29e48.
190 X.-w. Qi et al. / Surgical Oncology 27 (2018) 185e191
[2] H. Ellis, G.L. Colborn, J.E. Skandalakis, Surgical embryology and anatomy of the
breast and its related anatomic structures, Surg. Clin. 73 (1993) 611e632.
[3] R.C. Chen, N.U. Lin, M. Golshan, J.R. Harris, J.R. Bellon, Internal mammary
nodes in breast cancer: diagnosis and implications for patient management e
a systematic review, J. Clin. Oncol. 26 (2008) 4981e4989.
[4] N. Klauber-DeMore, J.L. Bevilacqua, K.J. Van Zee, P. Borgen, H.S. Cody 3rd,
Comprehensive review of the management of internal mammary lymph node
metastases in breast cancer, J. Am. Coll. Surg. 193 (2001) 547e555.
[5] D. Vrana, J. Gatek, K. Cwiertka, L. Lukesova, P. Koranda, Internal mammary
node management in breast cancer. A review, Biomed. Pap. Med. Fac. Univ.
Palacky Olomouc Czech Repub. 157 (2013) 261e265.
[6] P.F. Qiu, Y.B. Liu, Y.S. Wang, Internal mammary sentinel lymph node biopsy:
abandon or persist? OncoTargets Ther. 9 (2016) 3879e3882.
[7] W.D. Kaplan, J.W. Andersen, R.L. Siddon, B.T. Connolly, C.A. McCormick,
S.M. Laffin, E.M. Rosenbaum, C.A. Jennings, A. Recht, J.R. Harris, The three-
dimensional localization of internal mammary lymph nodes by radionuclide
lymphoscintigraphy, J. Nucl. Med. 29 (1988) 473e478.
[8] M. Margottini, Recent developments in the surgical treatment of breast car-
cinoma, Acta Unio Int. Contra Cancrum 8 (1952) 176e178.
[9] R.S. Handley, A.C. Thackray, Invasion of internal mammary lymph nodes in
carcinoma of the breast, Br. Med. J. 1 (1954) 61e63.
[10] R.S. Handley, D.H. Patey, B.H. Hand, Excision of the internal mammary chain in
radical mastectomy; results in 57 cases, Lancet 270 (1956) 457e461.
[11] E. Caceres, Incidence of metastasis in the internal mammary chain in operable
cancer of the breast, Surg. Gynecol. Obstet. 108 (1959) 715e720.
[12] J.A. Urban, M.A. Marjani, Significance of internal mammary lymph node me-
tastases in breast cancer, Am. J. Roentgenol. Radium Ther. Nucl. Med. 111
(1971) 130e136.
[13] U. Veronesi, N. Cascinelli, R. Bufalino, A. Morabito, M. Greco, D. Galluzzo,
V. Delle Donne, R. De Lellis, P. Piotti, V. Sacchini, et al., Risk of internal
mammary lymph node metastases and its relevance on prognosis of breast
cancer patients, Ann. Surg. 198 (1983) 681e684.
[14] U. Veronesi, N. Cascinelli, M. Greco, R. Bufalino, A. Morabito, D. Galluzzo,
R. Conti, R. De Lellis, V. Delle Donne, P. Piotti, et al., Prognosis of breast cancer
patients after mastectomy and dissection of internal mammary nodes, Ann.
Surg. 202 (1985) 702e707.
[15] S.L. Sugg, D.J. Ferguson, M.C. Posner, R. Heimann, Should internal mammary
nodes be sampled in the sentinel lymph node era? Ann. Surg Oncol. 7 (2000)
188e192.
[16] P. Bucalossi, U. Veronesi, Long-term results of radical mastectomy with
removal of internal mammary chain, Acta Unio Int. Contra Cancrum 15 (1959)
1052e1055.
[17] S.F. Livingston, M. Arlen, The extended extrapleural radical mastectomy: its
role in the treatment of carcinoma of the breast, Ann. Surg. 179 (1974)
260e265.
[18] P. Bucalossi, U. Veronesi, L. Zingo, C. Cantu, Enlarged mastectomy for breast
cancer. Review of 1,213 cases, Am. J. Roentgenol. Radium Ther. Nucl. Med. 111
(1971) 119e122.
[19] M.G. Le, R. Arriagada, F. de Vathaire, J. Dewar, F. Fontaine, J. Lacour,
G. Contesso, M. Tubiana, Can internal mammary chain treatment decrease the
risk of death for patients with medial breast cancers and positive axillary
lymph nodes? Cancer 66 (1990) 2313e2318.
[20] J. Lacour, P. Bucalossi, E. Cacers, G. Jacobelli, T. Koszarowski, M. Le, C. Rumeau-
Rouquette, U. Veronesi, Radical mastectomy versus radical mastectomy plus
internal mammary dissection. Five-year results of an international coopera-
tive study, Cancer 37 (1976) 206e214.
[21] J. Lacour, M. Le, E. Caceres, T. Koszarowski, U. Veronesi, C. Hill, Radical mas-
tectomy versus radical mastectomy plus internal mammary dissection. Ten
year results of an international cooperative trial in breast cancer, Cancer 51
(1983) 1941e1943.
[22] P. Meier, D.J. Ferguson, T. Karrison, A controlled trial of extended radical
versus radical mastectomy. Ten-year results, Cancer 63 (1989) 188e195.
[23] U. Veronesi, E. Marubini, L. Mariani, P. Valagussa, R. Zucali, The dissection of
internal mammary nodes does not improve the survival of breast cancer
patients. 30-year results of a randomised trial, Eur. J. Canc. 35 (1999)
1320e1325.
[24] R.F. Uren, R. Howman-Giles, J.F. Thompson, Metastatic disease in the internal
mammary lymph nodes has an adverse effect on patient prognosis and sur-
vival, regardless of axillary node status, Ann. Surg Oncol. 7 (2000) 790.
[25] F.W. van der Ent, R.A. Kengen, H.A. van der Pol, J.A. Povel, H.J. Stroeken,
A.G. Hoofwijk, Halsted revisited: internal mammary sentinel lymph node
biopsy in breast cancer, Ann. Surg. 234 (2001) 79e84.
[26] R.F. Uren, R.B. Howman-Giles, J.F. Thompson, D. Malouf, G. Ramsey-Stewart,
F.W. Niesche, S.B. Renwick, Mammary lymphoscintigraphy in breast cancer,
J. Nucl. Med. 36 (1995) 1775e1780.
[27] P. Carcoforo, D. Sortini, L. Feggi, C.V. Feo, G. Soliani, S. Panareo, S. Corcione,
P. Querzoli, K. Maravegias, S. Lanzara, A. Liboni, Clinical and therapeutic
importance of sentinel node biopsy of the internal mammary chain in patients
with breast cancer: a single-center study with long-term follow-up, Ann. Surg
Oncol. 13 (2006) 1338e1343.
[28] M.S. Yao, B.F. Kurland, A.H. Smith, E.K. Schubert, L.K. Dunnwald, D.R. Byrd,
D.A. Mankoff, Internal mammary nodal chain drainage is a prognostic indi-
cator in axillary node-positive breast cancer, Ann. Surg Oncol. 14 (2007)
2985e2993.
[29] N.S. Nikolaevich, K.S. Vasilevich, Why do we need irradiation of internal
mammary lymph nodes in patients with breast cancer: analysis of lymph flow
and radiotherapy studies, Rep. Practical Oncol. Radiother. 22 (2017) 37e41.
[30] E.M. Heuts, F.W. van der Ent, M.F. von Meyenfeldt, A.C. Voogd, Internal
mammary lymph drainage and sentinel node biopsy in breast cancer - a study
on 1008 patients, Eur. J. Surg. Oncol. 35 (2009) 252e257.
[31] J.L. Gnerlich, J.C. Barreto-Andrade, T. Czechura, J.R. John, M.A. Turk,
T.J. Kennedy, D.J. Winchester, Accurate staging with internal mammary chain
sentinel node biopsy for breast cancer, Ann. Surg Oncol. 21 (2014) 368e374.
[32] A.S. Caudle, M. Yi, K.E. Hoffman, E.A. Mittendorf, G.V. Babiera, R.F. Hwang,
F. Meric-Bernstam, A.A. Sahin, K.K. Hunt, Impact of identification of internal
mammary sentinel lymph node metastasis in breast cancer patients, Ann.
Surg Oncol. 21 (2014) 60e65.
[33] S. Dellapasqua, V. Bagnardi, A. Balduzzi, M. Iorfida, N. Rotmensz, B. Santillo,
G. Viale, R. Ghisini, P. Veronesi, A. Luini, A. Morra, A. Goldhirsch, M. Colleoni,
Outcomes of patients with breast cancer who present with ipsilateral
supraclavicular or internal mammary lymph node metastases, Clin. Breast
Canc. 14 (2014) 53e60.
[34] V. Ozmen, B. Ozcinar, A. Bozdogan, Y. Eralp, E. Yavuz, M. Dincer, The effect of
internal mammary lymph node biopsy on the therapeutic decision and sur-
vival of patients with breast cancer, Eur. J. Surg. Oncol. 41 (2015) 1368e1372.
[35] K. Ferreres Garcia, J. Siegrist Ridruejo, P. Rincon Olbes, M.S. Luque Molina,
M.I. Almoguera Arias, Therapeutic attitude towards internal mammary chain
drainage in patients with breast cancer, Breast 30 (2016) 1e4.
[36] P.M. Poortmans, S. Collette, C. Kirkove, E. Van Limbergen, V. Budach,
H. Struikmans, L. Collette, A. Fourquet, P. Maingon, M. Valli, K. De Winter,
S. Marnitz, I. Barillot, L. Scandolaro, E. Vonk, C. Rodenhuis, H. Marsiglia,
N. Weidner, G. van Tienhoven, C. Glanzmann, A. Kuten, R. Arriagada,
H. Bartelink, W. Van den Bogaert, E.R. Oncology, G. Breast Cancer, Internal
mammary and medial supraclavicular irradiation in breast cancer, N. Engl. J.
Med. 373 (2015) 317e327.
[37] L.B. Thorsen, B.V. Offersen, H. Dano, M. Berg, I. Jensen, A.N. Pedersen,
S.J. Zimmermann, H.J. Brodersen, M. Overgaard, J. Overgaard, DBCG-IMN: a
population-based cohort study on the effect of internal mammary node
irradiation in early node-positive breast cancer, J. Clin. Oncol. 34 (2016)
314e320.
[38] L. De Cicco, A. Starace, E. Petazzi, S. Buttignol, A. Lanceni, S. Bracelli, Internal
mammary node drainage in patients with early breast cancer, Tumori 102
(2016) e6.
[39] D. Vrana, J. Gatek, Internal mammary lymph node management - further di-
rection, OncoTargets Ther. 10 (2017) 1111e1114.
[40] S. Cranenbroek, M.J. van der Sangen, G.P. Kuijt, A.C. Voogd, Diagnosis, treat-
ment and prognosis of internal mammary lymph node recurrence in breast
cancer patients, Breast Canc. Res. Treat. 89 (2005) 271e275.
[41] E. Hindie, D. Groheux, C. Hennequin, P. Zanotti-Fregonara, L. Vercellino,
N. Berenger, M.E. Toubert, C. Maylin, J.R. Vilcoq, M. Espie, Lymphoscintigraphy
can select breast cancer patients for internal mammary chain radiotherapy,
Int. J. Radiat. Oncol. Biol. Phys. 83 (2012) 1081e1088.
[42] Z. Li, X. Gu, J. Tong, B. Liu, L. Sun, X. Gao, X. Jiang, A meta-analysis of internal
mammary lymph node metastasis in breast cancer patients, Onkologie 36
(2013) 747e752.
[43] E. Avisar, M.A. Molina, M. Scarlata, F.L. Moffat, Internal mammary sentinel
node biopsy for breast cancer, Am. J. Surg. 196 (2008) 490e494.
[44] J.J. Huang, C.W. Wu, R.J. Allen Jr., S.C. Shen, C.C. Yu, M.H. Cheng, Routine
sampling of internal mammary lymph nodes during microsurgical breast
reconstruction-Experience based on 524 microsurgical breast reconstructions,
J. Surg. Oncol. 114 (2016) 133e139.
[45] M.S. Jochelson, L. Lebron, S.S. Jacobs, J. Zheng, C.S. Moskowitz, S.N. Powell,
V. Sacchini, G.A. Ulaner, E.A. Morris, D.D. Dershaw, Detection of internal
mammary adenopathy in patients with breast cancer by PET/CT and MRI, Am.
J. Roentgenol. 205 (2015) 899e904.
[46] H. Cheon, H.J. Kim, S.W. Lee, D.H. Kim, C.H. Lee, S.H. Cho, K.M. Shin, S.M. Lee,
G.C. Kim, W.H. Kim, Internal mammary node adenopathy on breast MRI and
PET/CT for initial staging in patients with operable breast cancer: prevalence
and associated factors, Breast Canc. Res. Treat. 160 (2016) 523e530.
[47] T. Davidson, M. Ben-David, S. Galper, T. Haskin, M. Howes, R. Scaife, N. Kanana,
U. Amit, N. Weizman, B. Chikman, E. Goshen, S. Ben-Haim, Z. Symon,
J. Goldstein, Use of (18)F-FDG PET-CT imaging to determine internal mam-
mary lymph node location for radiation therapy treatment planning in breast
cancer patients, Pract. Radiat. Oncol. 7 (2017) 373e381.
[48] M. Asadi, D. Krag, Internal mammary sentinel lymph node biopsy in clinical
practice, Int. J. Surg. 36 (2016) 332e334.
[49] Y.Y. An, S.H. Kim, B.J. Kang, A.W. Lee, Comparisons of positron emission to-
mography/computed tomography and ultrasound imaging for detection of
internal mammary lymph node metastases in patients with breast cancer and
pathologic correlation by ultrasound-guided biopsy procedures, J. Ultrasound
Med. 34 (2015) 1385e1394.
[50] C.J. Jeong, J.M. Bae, S.Y. Park, B.K. Han, E.S. Ko, E.Y. Ko, Metastatic internal
mammary lymph nodes or mimickers on parasternal ultrasonography:
focusing on their distribution and depth, Acta Radiol. 59 (2018) 34e40.
[51] B.B. Cong, X.S. Cao, P.F. Qiu, J.M. Yu, Y.S. Wang, Internal mammary sentinel
lymph node biopsy: an effective way to search benefit patients and guide
internal mammary chain irradiation, Breast 33 (2017) 204e205.
[52] M.J. Seo, J.J. Lee, H.O. Kim, S.Y. Chae, S.H. Park, J.S. Ryu, S.H. Ahn, J.W. Lee,
B.H. Son, G.Y. Gong, D.H. Moon, Detection of internal mammary lymph node
metastasis with (18)F-fluorodeoxyglucose positron emission tomography/
 X.-w. Qi et al. / Surgical Oncology 27 (2018) 185e191
computed tomography in patients with stage III breast cancer, Eur. J. Nucl.
Med. Mol. Imag. 41 (2014) 438e445.
[53] J.H. Joo, S.S. Kim, S.D. Ahn, E.K. Choi, J.H. Jung, Y. Jeong, S.H. Ahn, B.H. Son,
J.W. Lee, H.J. Kim, B.S. Go, H.H. Kim, J.H. Cha, H.J. Shin, E.Y. Chae, Impact of
pathologic diagnosis of internal mammary lymph node metastasis in clinical
N2b and N3b breast cancer patients, Breast Canc. Res. Treat. 166 (2017)
511e518.
[54] N.J. Coombs, J. Boyages, J.R. French, O.A. Ung, Internal mammary sentinel
nodes: ignore, irradiate or operate? Eur. J. Canc. 45 (2009) 789e794.
[55] L. Lukesova, D. Vrana, I. Svach, N. Zlamalova, J. Gatek, Z. Vlachova,
M. Matzenauer, P. Koranda, K. Hruzova, T. Tichy, B. Melichar, Prognostic in-
fluence of internal mammary node drainage in patients with early-stage
breast cancer, Anticancer Res. 36 (2016) 6641e6646.
[56] A.L. Kong, W. Tereffe, K.K. Hunt, M. Yi, T. Kang, K. Weatherspoon,
E.A. Mittendorf, I. Bedrosian, R.F. Hwang, G.V. Babiera, T.A. Buchholz, F. Meric-
Bernstam, Impact of internal mammary lymph node drainage identified by
preoperative lymphoscintigraphy on outcomes in patients with stage I to III
breast cancer, Cancer 118 (2012) 6287e6296.
[57] M.Y. Koo, S.K. Lee, S.Y. Bae, M.-Y. Choi, D.H. Cho, S. Kim, J.E. Lee, S.J. Nam, J.-
H. Yang, Long-Term outcome of internal mammary lymph node detected by
lymphoscintigraphy in early breast cancer, J. Breast Cancer 15 (2012) 98.
[58] A.C. Barros, L.J. Mori, D. Nishimura, A.L. Jacomo, Surgical anatomy of the in-
ternal thoracic lymph nodes in fresh human cadavers: basis for sentinel node
biopsy, World J. Surg. Oncol. 14 (2016) 135.
[59] E.F. Cox, R.W. Buxton, Internal mammary lymph node biopsy as a guide to
postmastectomy radiation therapy in breast carcinoma, Am. Surg. 26 (1960)
335e340.
[60] Q. He, J. Jiang, X. Yang, L. Fan, M. Guo, Y. Zhang, A pilot study on thoracoscopic
internal mammary lymphatic chain dissection for breast cancer, Breast 17
(2008) 568e573.
[61] H. Long, Z. Lin, D. Situ, G. Ma, Y. Zheng, T. Rong, Stage migration and therapy
modification after thoracoscopic internal mammary lymph node dissection in
breast cancer patients, Breast 20 (2011) 129e133.
[62] A.L. Grant, K.H. Lutz, C.F. Temple-Oberle, Incidental internal mammary nodes
during recipient vessel dissection in breast reconstruction: are they signifi-
cant? Plast. Reconstr. Surg. Glob. Open 2 (2014) e276.
[63] X. Zhang, S. Jaffer, I.J. Bleiweiss, A. Nayak, The clinical significance of internal
mammary lymph node (IMLN) biopsy during autologous reconstruction in
breast cancer patients, Breast Canc. Res. Treat. 153 (2015) 565e572.
[64] E.J. Wright, A. Momeni, U.M. Kraneburg, L.R. Otake, A. Echo, T. Lee,
E.P. Buchanan, G.K. Lee, Clinical significance of internal mammary lymph node
biopsy during microsurgical breast reconstruction: review of 264 cases, Plast.
Reconstr. Surg. 137 (2016) 917ee922e.
[65] K.Y. Hong, H.B. Lee, S. Yim, J. Lee, T.Y. Kim, W. Han, U.S. Jin, Opportunistic
191
biopsy of internal mammary lymph nodes during immediate breast recon-
struction after mastectomy for breast malignancies, Ann. Surg Oncol. 24
(2017) 1881e1888.
[66] O. Huang, L. Wang, K. Shen, H. Lin, Z. Hu, G. Liu, J. Wu, J. Lu, Z. Shao, Q. Han,
Z. Shen, Breast cancer subpopulation with high risk of internal mammary
lymph nodes metastasis: analysis of 2,269 Chinese breast cancer patients
treated with extended radical mastectomy, Breast Canc. Res. Treat. 107 (2007)
379e387.
[67] L. Lukesova, D. Vrana, J. Gatek, P. Koranda, K. Cwiertka, L. Radova, B. Melichar,
Z. Prouzova, V. Sramek, I. Svach, Predictive parameters for internal mammary
node drainage in patients with early breast cancer, Tumori 100 (2014)
254e258.
[68] C. Tan, R. Caragata, I. Bennett, Is sentinel node biopsy of the internal mam-
mary lymph nodes relevant in the management of breast cancer? Breast J. 23
(2017) 410e414.
[69] E.M. Heuts, F.W. van der Ent, K.W. Hulsewe, M.F. von Meyenfeldt, A.C. Voogd,
Results of tailored treatment for breast cancer patients with internal mam-
mary lymph node metastases, Breast 18 (2009) 254e258.
[70] A.E. Giuliano, K.K. Hunt, K.V. Ballman, P.D. Beitsch, P.W. Whitworth,
P.W. Blumencranz, A.M. Leitch, S. Saha, L.M. McCall, M. Morrow, Axillary
dissection vs no axillary dissection in women with invasive breast cancer and
sentinel node metastasis: a randomized clinical trial, JAMA 305 (2011)
569e575.
[71] A.E. Giuliano, K.V. Ballman, L. McCall, P.D. Beitsch, M.B. Brennan, P.R. Kelemen,
D.W. Ollila, N.M. Hansen, P.W. Whitworth, P.W. Blumencranz, A.M. Leitch,
S. Saha, K.K. Hunt, M. Morrow, Effect of axillary dissection vs No axillary
dissection on 10-year overall survival Among women with invasive breast
cancer and sentinel node metastasis: the ACOSOG Z0011 (Alliance) random-
ized clinical trial, JAMA 318 (2017) 918e926.
[72] F. Cardoso, L.J. van't Veer, J. Bogaerts, L. Slaets, G. Viale, S. Delaloge, J.Y. Pierga,
E. Brain, S. Causeret, M. DeLorenzi, A.M. Glas, V. Golfinopoulos, T. Goulioti,
S. Knox, E. Matos, B. Meulemans, P.A. Neijenhuis, U. Nitz, R. Passalacqua,
P. Ravdin, I.T. Rubio, M. Saghatchian, T.J. Smilde, C. Sotiriou, L. Stork,
C. Straehle, G. Thomas, A.M. Thompson, J.M. van der Hoeven, P. Vuylsteke,
R. Bernards, K. Tryfonidis, E. Rutgers, M. Piccart, M. Investigators, 70-Gene
signature as an aid to treatment decisions in early-stage breast cancer,
N. Engl. J. Med. 375 (2016), 717e429.
[73] J.A. Sparano, R.J. Gray, D.F. Makower, K.I. Pritchard, K.S. Albain, D.F. Hayes,
C.E. Geyer Jr., E.C. Dees, E.A. Perez, J.A. Olson Jr., J. Zujewski, T. Lively,
S.S. Badve, T.J. Saphner, L.I. Wagner, T.J. Whelan, M.J. Ellis, S. Paik, W.C. Wood,
P. Ravdin, M.M. Keane, H.L. Gomez Moreno, P.S. Reddy, T.F. Goggins,
I.A. Mayer, A.M. Brufsky, D.L. Toppmeyer, V.G. Kaklamani, J.N. Atkins,
J.L. Berenberg, G.W. Sledge, Prospective validation of a 21-gene expression
assay in breast cancer, N. Engl. J. Med. 373 (2015) 2005e2014.