Ann Surg Oncol

DOI 10.1245/s10434-012-2609-7

ORIGINAL ARTICLE – BREAST ONCOLOGY

Preoperative Axillary Ultrasound and Fine-needle Aspiration

Biopsy in the Diagnosis of Axillary Metastases in Patients
with Breast Cancer: Predictors of Accuracy and Future
Implications
Jonathan Cools-Lartigue, MD1, Alison Sinclair, MD, PhD2, Nora Trabulsi, MD1, Ari Meguerditchian, MD1,
Benoit Mesurolle, MD3, Rebecca Fuhrer, PhD2, and Sarkis Meterissian, MD1

1
Department of Surgery, McGill University Health Centre, Montreal, QC, Canada; 2Department of Epidemiology,
Biostatistics, and Occupational Health, McGill University, Montreal, QC, Canada; 3Department of Diagnostic Radiology,
McGill University Health Centre, Montreal, QC, Canada

ABSTRACT nodal metastases or a greater proportion of gross extranodal

Background. The utility of axillary lymph node dissection disease compared to patients not subjected to FNAB.
after sentinel lymph node biopsy has been called into Conclusions. Axillary ultrasound with FNAB has an
question. We sought to determine the sensitivity, specific- accuracy of [70% in this series. It is easily performed and
ity, and accuracy of axillary ultrasound and fine-needle may avoid unnecessary sentinel lymph node biopsy in a
aspiration biopsy (FNAB) in the identification of axillary significant number of patients.
nodal metastasis in early breast cancer patients.
Methods. Data of patients with stage I and II breast cancer
who underwent surgery and staging were reviewed. Axil-
Sentinel lymph node biopsy (SLNB) has supplanted
lary ultrasound findings were assessed and lymph node
axillary node dissection (ALND) as the gold standard in the
status recorded after axillary dissection. The data were
operative nodal staging of carcinoma of the breast. This
cross-tabulated, and test characteristics were calculated.
technique, which has been validated prospectively in mul-
Results. Of 235 patients, none demonstrated more than 2
tiple studies, relies on the observation that the breast
positive sentinel lymph nodes. Ductal carcinoma was present
behaves as a single lymphatic unit, with drainage occurring
in 68 %, estrogen and progesterone receptors were positive in
first to one or several sentinel lymph nodes (SLNs).1–3
81 and 64 %, respectively, Her-2/neu was positive in 10 %,
Currently, patients identified by SLNB with node-positive
and 36 % were axillary node positive. The sensitivity and
disease frequently undergo a complete ALND and further
specificity of ultrasound alone were 55 and 88 %, respec-
surgical and medical management as indicated by their
tively. Predictors of abnormal ultrasound included size of
pathology.1,4–7 With the publication of the American Col-
metastasis, estrogen receptor and Her-2 status, tumor grade,
lege of Surgeons Oncology Group (ACOSOG) Z0011 trial,
and presence of lymphovascular invasion. Addition of FNAB
it has been suggested that early stage (clinical T1/T2N0M0)
increased the sensitivity and specificity to 69 and 100 %. In
breast cancer patients with node-positive disease identified
conjunction with FNAB, the positive and negative predictive
by SLNB may be appropriately treated with whole-breast
values were 100 and 54 %, respectively. Ten percent of
irradiation and systemic chemotherapy in lieu of ALND.
patients with nodal metastases demonstrated a positive
Furthermore, in such patients with a low burden of axillary
FNAB. Patients with a positive FNAB did not harbor more
disease, ALND has little impact on local recurrence and
overall survival. In conjunction with the ongoing European
Organization for Research and Treatment of Cancer
Ó Society of Surgical Oncology 2012 (EORTC) After Mapping of the Axilla: Radiotherapy or
First Received: 1 February 2012 Surgery (AMAROS) trial, which aims to determine whether
J. Cools-Lartigue, MD
node-positive patients as determined by SLNB can be
e-mail: jonathan.cools-lartigue@mail.mcgill.ca treated with axillary radiotherapy versus ALND, ACOSOG
 J. Cools-Lartigue et al.

Z0011 may help redefine the role of axillary surgery in the
treatment of breast cancer and highlights the importance of
the accurate diagnosis of axillary nodal involvement.
The prediction of axillary nodal status, which is the
single most important prognostic factor in breast cancer,
has yet to be accomplished without surgical intervention at
an acceptable rate. To date, no modality that can boast an
accuracy exceeding 90–95 % has been validated prospec-
tively.1 Consequently, many patients with node-positive
disease are subject to SLNB. Although this is currently
unavoidable at the level of institutional policy, the reality is
that thousands of patients are subjected to an unnecessary
surgical procedure, with the burden of cost bearing heavily
on health care systems.8 Thus, the ability to accurately
predict which patients are likely to have involved axillary
nodes would be of great benefit.
Preoperative axillary ultrasound (US) is a frequently
performed procedure in patients diagnosed with primary
breast cancer. This technique has demonstrated good sen-
sitivity and specificity, particularly in conjunction with
fine-needle aspiration biopsy (FNAB)/core biopsy, in the
preoperative diagnosis of nodal metastasis in breast can-
cer.9 It has been used to identify patients who can forego
SLNB and proceed directly to ALND. Although it results
in considerably less morbidity than ALND, SLNB is not
without its disadvantages. This procedure carries a 4–14 %
rate of complications, including lymphedema, paresthesia,
chronic pain, and immobility.10 In addition, it places
financial demands on resources of a system in which
operating costs are already high.8,11,12
At our institution, axillary US is routinely performed in
all patients referred for evaluation of malignant breast
disease. We sought to determine the sensitivity, specificity,
and accuracy of axillary US in the detection of nodal
metastases with or without the addition of FNAB. Fur-
thermore, we sought to characterize the axillary disease
burden in patients with nodal metastasis identified by US
and FNAB versus SLNB and determine the proportion of
patients who can be spared an unnecessary SLNB and
proceed directly to ALND.
preoperatively identified axillary nodal involvement pro-
ceeded directly to ALND. Patients with sonographically
suspicious nodes who were not subject to FNAB were still
subject to SLNB at the time of surgery with ALND in the
event of a positive SLN. The goal of the present study was to
determine the sensitivity, specificity, and accuracy of
axillary US in the detection of nodal metastases with or
without the addition of FNAB, and to compare nodal
disease burden in patients identified by FNAB versus
SLNB. Figure 1 depicts the institutional algorithm of care
described.
Patients
Patients undergoing axillary US evaluation for primary
carcinoma of the breast were identified from a prospective
database at the McGill University Health Center between
2005 and 2007. This was done in compliance with the insti-
tutional policies regarding patient confidentiality. Patient
charts were reviewed for patient demographics, year of
diagnosis, primary tumor histology, grade, stage, hormone
status, Her-2/neu status, results of axillary US, number of
sentinel nodes retrieved (if sentinel node examination was
conducted), number of positive nodes by histological exami-
nation, number of nodes retrieved with axillary dissection,
number of positive nodes by histological examination, size of
the largest metastatic deposit, presence of lymphovascular

MATERIALS AND METHODS

Management Algorithm

It is institutional policy for all patients with a new diagnosis

of breast carcinoma and clinically negative axillae to undergo
axillary US. Patients with sonographically negative nodes are
subjected to SLNB at the time of their surgery. Conversely,
FIG. 1 Institutional algorithm of care in the diagnosis and early
patients with axillary lymph nodes suspicious for metastatic treatment of invasive carcinoma of the breast. ‘‘Suspicious’’ is defined
involvement undergo FNAB at the discretion of the sonog- as absence of a fatty nodal hilum, eccentric cortical thickening, round
rapher. Over the time period of the study, patients with hypoechoic node, and multiple enlarged ([1 cm) nodes
Diagnosis of Axillary Metastases in Breast Cancer
invasion (LVI) or extracapsular extension (ECE), and results
of FNAB and/or core biopsy, if performed. Hormone status of
the primary tumor was recorded as positive or negative for
estrogen receptor (ER) and progesterone receptor (PR), and
positive, negative, or equivocal for Her-2/neu on the basis of
the laboratory report. Data were entered into a Windows Excel
spreadsheet, Microsoft Office version 2007 (Microsoft,
Redmond, CA).
Ultrasound
Axillary US was performed by a dedicated axillary
sonographer at the university hospital breast center. The
equipment used to perform sonography included 2 high-
resolution scanners with high-frequency linear-array 10–14-
MHz transducers (15L8w broadband transducer, Sequoia,
Siemens Medical Solutions, Acuson; high-frequency Matrix
transducer PLT1204AX, Aplio, Toshiba Medical Systems,
Tustin, CA). Lymph nodes were identified as abnormal
according to sonographic criteria including absence of a fatty
nodal hilum, eccentric cortical thickening, and a round
hypoechoic node as previously described.9 Axillae with
multiple enlarged ([1 cm) nodes were also identified as
abnormal. US-guided FNAB was performed on patients at
the discretion of the ultrasonographer: while the patient was
under local anesthesia, a 22-gauge needle attached to a
10-mL syringe was used to obtain specimens for cytologic
examination. Two samples were routinely obtained from the
selected lymph node.
Data Analysis
The data analysis was conducted by SAS software, version
9.2 of the SAS System for Windows (SAS, Cary, NC). Con-
tinuous and categorical summaries and comparisons used
PROC UNIVARIATE and PROC FREQ, respectively, and
univariate and multivariate logistic regression used PROC
LOGISTIC. US results were dichotomized as normal versus
abnormal, as described by the radiologist. A patient was con-
sidered to have a positive pathology result if one or more
retrieved lymph nodes were recorded as positive and/or if she
had a positive FNAB. A patient was considered to have a
negative pathology result if she had at least some nodes
retrieved (total number of nodes not zero) and if the total
number of positive nodes equaled 0 with a negative FNAB, if
one was reported. One patient had no nodes recorded as
retrieved and no FNAB result reported and was therefore
excluded from the analysis. The size of the largest metastatic
deposit observed on histopathological examination of excised
nodes was categorized as none, isolated tumor cells (ITC;
\0.2 mm), micrometastases (0.2–2 mm), or macrometastases
([2 mm). Data were summarized as mean, median, standard
deviation, minimum and maximum for continuous variables,
and by number and percentage for categorical variables.
Sensitivity, specificity, positive and negative predictive
value, and accuracy of axillary US were calculated, with
confidence intervals, using the pathology result as the gold
standard. Univariate regression analysis used a binary logit
model with optimization by Fisher’s scoring, and testing of
the global null hypothesis (regression slope equals zero) by
likelihood ratio, Score, and Wald test. The Wald Chi-
square p value was reported for analysis of effects. The
dependent variable was US normal/abnormal. Independent
variables were: any lymph node metastases, size of lymph
nodes, number of positive sentinel, nonsentinel and total
lymph nodes, American Joint Committee on Cancer
(AJCC) class, age at diagnosis, age at diagnosis dichoto-
mized around the age of 50 years, primary tumor histology,
primary tumor grade, ER, PR, Her-2/neu, LVI, and ECE.
Multivariate analysis used a binary logit model, as above,
with US normal/abnormal for the dependent variable.
Lymph node status was modeled in two different ways: by
presence of nodal metastases (binary variable), and by size
of metastasis (four-category variable). Prognostic variables
(AJCC, ER, PR, Her-2/neu, LVI, grade, and age at diag-
nosis) were added to the model. The Wald Chi-square
p value was reported for analysis of effects.
RESULTS
Lymph Node Metastases and Primary Tumor Pathology
Ultrasonography was performed in 235 patients with
primary carcinoma of the breast of stage I or stage II.
Patient demographics and tumor characteristics are shown
in Table 1. The histology of the primary tumor was inva-
sive ductal carcinoma in 160 patients (68 %), invasive
lobular carcinoma in 15 patients (6.4 %), invasive ductal
and lobular carcinoma in 35 patients (14.9 %), and other
histologic types in 25 patients (10.6 %). There were 120
patients with stage I disease (51 %) and 115 patients with
stage II disease (49 %).
Overall, a mean of 7.47 nodes were excised per patient in
patients who underwent SLNB and ALND (range 1–36
nodes, median 6 nodes). Lymph nodes were negative for
metastatic disease in 144 patients (63.7 %) (both histopa-
thology and FNAB, if available, were negative) and positive
in 91 patients (38.7 %) (positive finding on histopathology or
FNAB, or both). In these 91 patients, macrometastases were
identified in 58 patients (25.7 %), micrometastases in 16
patients (7.1 %), and ITC in 8 patients (3.5 %); nine patients
did not have tumor size reported.
With regard to SLN, a total of 346 nodes were excised,
with a mean of 2.11 nodes per patient (range 1–6 nodes,
 J. Cools-Lartigue et al.

TABLE 1 Patient demographics and tumor characteristics of 235 TABLE 2 Presence of nodal metastases and pathologic features of
patients the primary tumors and nodal metastatic foci
Characteristic Value Characteristic Value

Age No. of nodes excised, mean (range) 7.47 (1.0–36.0)

a
Mean ± SD 57.8 ± 13.1 Nodal metastases
Range 22–97 No metastasis 144 (61.3 %)
B50 years 73 (31.1 %) ITC 8 (3.4 %)
[50 years 162 (68.9 %) Micrometastasis 16 (6.8 %)
Primary tumor histology Macrometastasis 58 (24.7 %)
Ductal 160 (68.1 %) Missing 9 (3.8 %)
Lobular 15 (6.4 %) Pathologic prognosticatorsb
Mixed 35 (14.9 %) ER
Other 25 (10.6 %) Positive 160 (68.1 %)
Primary tumor gradea Negative 37 (15.7 %)
1 32 (13.9 %) Missing 38 (16.2 %)
2 134 (58.0 %) PR
3 65 (28.1 %) Positive 127 (54.0 %)
AJCC stage Negative 70 (29.8 %)
1 120 (51.1 %) Missing 38 (16.2 %)
2 115 (48.9 %) Her-2/neu
AJCC American Joint Committee on Cancer Positive 18 (7.7 %)
a
Primary tumor grade was missing for 4 patients Negative 158 (67.2 %)
Missing 59 (25.1 %)
median 2 nodes). No patient demonstrated more than 2 LVI
positive SLN. Nodal characteristics and additional patho- Positive 43 (18.3 %)
logic features of the primary tumors are presented in Negative 129 (54.9 %)
Table 2. Missing 63 (26.8 %)
ECE
US Findings and Accuracy Positive 33 (14.0 %)
Negative 36 (15.3 %)
Axillary US was abnormal in 68 patients (28.9 %) and Missing 166 (70.6 %)
normal in 167 patients (71.1 %). Forty-nine patients had
available FNAB results, with 25 demonstrating a metastatic ITC isolated tumor cells, ER estrogen receptor, PR progesterone
receptor, LVI lymphovascular invasion, ECE extracapsular extension
focus. Of the 25 positive FNABs, 22 were confirmed on a
Size of nodal metastases was quantified in 226 of 235 patients
histopathology of excised lymph nodes (1 falsely positive; b
Pathologic features of the primary tumor or lymph node metastasis
pathology was unavailable for 2 axillae). The one false- (ECE) were recorded
positive FNAB was subsequently excluded from the final
analysis because this patient received neoadjuvant chemo- In this study, with an axillary metastasis prevalence of
therapy (NACT) before ALND. Of the 24 negative FNABs, 38.7 %, the sensitivity, specificity, positive predictive
13 were confirmed on final pathology, while 12 demon- value, negative predictive value, and accuracy of axillary
strated metastases on pathologic lymph node examination. US alone in detecting axillary node metastases were 55, 88,
Four FNAB procedures were performed on axillae charac- 74, 75, and 75 %, respectively. In the subset of patients with
terized as not suspicious. The reasons likely reflect the available FNAB results, the sensitivity, specificity, positive
definition of a suspicious node used in this study. In par- predictive value, negative predictive value, and accuracy of
ticular, a single enlarged node without additional high-risk axillary US were 69, 100, 100, 54, and 76 %, respectively
features was not considered to be suspicious but may still (Table 3).
have been sampled at the discretion of the attending radi- To determine whether patients with a positive FNAB
ologist. None of these nodes was positive after FNAB, with were similar in terms of axillary disease burden to patients
one patient identified as node positive after SLNB. The with positive SLNB, they were compared with respect to
results of ultrasonography with or without FNAB, com- primary tumor histology, grade, number of positive lymph
pared to presence/absence of metastatic disease, are nodes, size of nodal metastases, presence of ECE, ER, PR,
summarized in Fig. 2. and Her-2/neu status, presence of LVI, and AJCC disease
Diagnosis of Axillary Metastases in Breast Cancer

FIG. 2 Ultrasound and fine-needle aspiration biopsy findings in 235 patients with invasive carcinoma of the breast

TABLE 3 Performance characteristics of axillary US with and with a positive SLNB and no FNAB than in patients with a
without FNAB positive FNAB. These results are outlined in Table 4.
Characteristic All US Abnormal US
(n = 235) with FNABa Predictors of an Abnormal US
(n = 49)
Predictors of an abnormal axillary US were identified by
Truly positive 50 25
univariate and multivariate regression, testing the null
Truly negative 126 4
hypothesis of regression slope of zero using the Wald Chi-
Falsely positive 18 0
square test at the 95 % significance level. Univariate anal-
Falsely negative 41 20
ysis predictors included presence of nodal metastases,
Sensitivity 55 % 69 %
primary tumor grade, size of nodal metastasis, ER status,
Specificityb 88 % 100 %
Her-2/neu positivity, and the presence of LVI, as shown in
Positive predictive valuec 74 % 100 %
Table 5. Number of positive lymph nodes and size of
Negative predictive value 75 % 54 %
metastatic focus also predicted an abnormal US, but number
Accuracy 75 % 76 %
of positive SLNs did not (p = 0.2066). In multivariate
US ultrasound, FNAB fine-needle aspiration biopsy, NACT neoadju- models, using either presence of nodal metastases or size of
vant chemotherapy nodal metastases to model node status in addition to the
a
Performance of FNAB as sole diagnostic test presence and size of nodal metastases, only ER remained as
b
Specificity for US ? FNAB defined as: (truly negative/truly neg- an independently significant predictive factor, as shown in
ative) ? falsely positive (4/4 ? 0) = 1; a single false-positive
finding was excluded because of prior NACT
Table 6.
c
Positive predictive value for US ? FNAB defined as: (truly posi-
tive/truly positive) ? falsely positive (25/25 ? 0) = 1; a single false-
positive finding was excluded because of prior NACT DISCUSSION

In the present study, we sought to determine the accuracy

stage. No significant difference was observed between
FNAB-positive and SLNB-positive patients with regard to
primary tumor histology, primary tumor grade, number of
involved lymph nodes, ER or PR status, presence of ECE,
or presence of LVI. Patients differed significantly with
respect to size of nodal metastases and Her-2/neu status.
FNAB-positive patients demonstrated a significantly higher
proportion of macrometastases to overall nodal metastases
compared to patients with a positive SLNB. Significantly
more nodal metastases were Her-2/neu negative in patients
of preoperative axillary US, with or without FNAB, in an
attempt to ascertain its value as a preoperative staging tool.
Accordingly, 235 women with early stage breast cancer
underwent preoperative sonographic axillary examination
over the time course of this study. We demonstrated an
overall prevalence of nodal metastases of 38.7 % with a
significant proportion, 10.2 %, harboring micrometastatic
foci (6.8 %) or ITC (3.4 %). When axillary US was used
alone to detect axillary nodal metastases, a sensitivity,
specificity, negative predictive value, positive predictive
 J. Cools-Lartigue et al.

TABLE 4 Summary of comparison of patients with positive FNAB analyses (Tables 5 ,6). Significant predictors of an abnor-
and SLNB mal US as determined by univariate analysis appeared to
Pathologic feature FNAB SLNB pa reflect disease burden and propensity for aggressive
positive (n = 25) positive behavior and included tumor size, presence and size of
(n = 51) nodal metastases, ER status, Her-2/neu status, and LVI.
Histology Multivariate analysis supported this hypothesis demon-
Ductal 17 (68 %) 36 (70.5 %) 0.4516 strating number and size of nodal metastases as well as ER
Lobular 1 (4 %) 2 (3.9 %) positivity as significant predictors of an abnormal US.
Mixed 3 (12 %) 10 (19.6 %) A review of the literature reveals that axillary US alone
Other 4 (16 %) 3 (5.8 %)
and in conjunction with FNAB represents a sensitive and
Grade
specific modality in the preoperative staging of patients
with invasive breast cancer.13,14 Studies have demonstrated
1 3 (12 %) 4 (7.8 %) 0.7167
that high-resolution US can detect axillary metastasis in
2 12 (48 %) 29 (56.9 %)
breast cancer with sensitivity, specificity, and overall
3 10 (40 %) 18 (35.3 %)
accuracy ranging 50–92 %, 90–97.2 %, and 76–92.1 %,
AJCC
respectively.15–19 When combined with FNAB, the litera-
1 0 3 (5.9 %) 0.5466
ture demonstrates increasing sensitivity and specificity in
2 25 (100 %) 48 (94.1 %)
the detection of nodal metastases, with a specificity of
Size of nodal metastasis
100 % in some series.8 Thus, current evidence highlights
ITC 0 8 (16.7 %) 0.007
the efficacy of preoperative axillary US in conjunction with
Micro 1 (5 %) 12 (25.0 %)
FNAB in the preoperative diagnosis of nodal disease in
Macro 19 (95 %) 28 (58.3 %)
patients with primary breast cancer.
ER
All patients with a new diagnosis of carcinoma of the
Positive 18 (81.8 %) 26 (85.7 %) 0.7257 breast at the McGill University Health Center had an
Negative 4 (18.2 %) 6 (14.3 %) axillary US regardless of disease stage. All patients
PR included in the study ultimately had AJCC stage I or II
Positive 14 (63.6 %) 38 (80.9 %) 0.1428 disease. Sixty-three percent of patients were ultimately
Pathologic features of the primary tumor or lymph node metastasis found to be node negative, while 10.2 % of patients with
were recorded. Missing values are not included, so columns may not positive nodes were ultimately found to harbor either
add up to total micrometastatic disease or ITC. Thus, the population
FNAB fine-needle aspiration biopsy, SLNB sentinel lymph node studied encompassed a significant group of patients early in
biopsy, AJCC American Joint Committee on Cancer, ITC isolated
tumor cell, ER estrogen receptor, PR progesterone receptor the natural history of their neoplasm, with a small burden
a
Fisher’s exact test of metastatic disease. Significant predictors of lymph node
positivity on US by multivariate analysis included ER
value, and accuracy of 55, 88, 74, 75, and 75 % were status and size of metastasis. These findings are congruent
observed, respectively. When combined with FNAB, the with the notion that more aggressive, advanced-sage
sensitivity, specificity, positive predictive value, negative tumors are more likely to have metastasized at the time of
predictive value, and accuracy of axillary US were 69, 100, diagnosis compared to less aggressive or early tumors.
100, 54, and 76 %, respectively. In this study population, The specificity of axillary US and FNAB is 100 % in the
the use of preoperative axillary US in conjunction with present study, with a positive predictive value of 100 %.
FNAB could have appropriately identified 10.6 % of This finding is corroborated by multiple studies and sup-
patients with axillary lymph node metastases without any ports the combined use of these modalities in the
axillary surgical intervention, sparing them an unnecessary preoperative identification of breast cancer patients with
SLNB. This result supports the use of US and FNAB in node-positive disease.9 ,20–23 One patient was subjected to
streamlining the management of axillary disease in breast FNAB; she subsequently underwent NACT and demon-
cancer and in appropriately identifying patients with axil- strated a false-positive result after surgical lymph node
lary disease who can proceed directly to ALND. US in excision. Several studies suggest that post-NACT axillary
conjunction with FNAB has been demonstrated to be a cost- US has good specificity in identifying node-positive dis-
effective modality related to one-step axillary surgery after ease, and when combined with FNAB, the sensitivity and
identification of node-positive patients without SLNB. specificity have been reported as 100 % in a single study.23
We similarly identified pathologic predictors of an However, the bulk of current evidence argues against the
abnormal US using both univariate and multivariate use of axillary US alone to identify patients with node-
Diagnosis of Axillary Metastases in Breast Cancer

TABLE 5 Significant predictors of abnormal US by unadjusted univariate analysis

Variable n Total, n (%)a Abnormal US, n (%)b pc

Presence of nodal metastases 235 \0.0001

Yes 91 (38.7) 50 (55.0)
No 144 (61.3) 18 (12.5)
Size of nodal metastases 226 \0.001
No metastasis 144 (63.7) 19 (13.2)
ITC 8 (3.5) 4 (50.0)
Micrometastasis 16 (7.1) 5 (31.2)
Macrometastasis 58 (25.7) 34 (68.1)
ER 197 0.0016
Positive 160 (81.2) 40 (25.0)
Negative 37 (18.8) 19 (51.3)
Her-2/neu 176 0.0268
Positive 18 (10.2) 9 (50.0)
Negative 158 (89.8) 40 (25.3)
Lymphovascular invasion 172 0.0010
Positive 43 (25.0) 21 (48.8)
Negative 129 (75.0) 29 (22.5)
Primary tumor grade 231 \0.0057
1 32 (13.9) 5 (15.6)
2 134 (58.0) 33 (24.6)
3 65 (28.1) 28 (43.1)
US ultrasound, ITC isolated tumor cell, ER estrogen receptor
a
Percentage of total number of patients with available data (maximum 235 patients)
b
Percentage of patients in each given category
c
Wald Chi-square test; analyses were unadjusted

negative disease after NACT who may potentially forgo
ALND.24–26
In an attempt to ascertain whether the patients in this
study with a positive FNAB were comparable to those with
a positive SLNB, they were compared in terms of primary
tumor histology, primary tumor grade, LVI, the total
number, size, and pathologic features of their lymph node
metastasis including, ER, PR, and Her-2/neu positivity, and
ECE, as well as final pathologic stage. Eighty-four percent
of patients harbored only 1 metastatic SLN, with 16 %
harboring 2 positive SLN. Patients with a positive FNAB
demonstrated a higher proportion of macrometastases than
did SLNB-positive patients. This result is not surprising
because a larger metastatic focus is more like to be sam-
pled during FNAB than a smaller one.
We observed a greater propensity for patients in the
SLNB group to harbor Her-2/neu-negative tumors com-
pared to the positive FNAB group. This may represent the
propensity of Her-2/neu-positive tumors to metastasize and
thus be overrepresented in the FNAB-positive group.27
Furthermore, although not statistically significant, there was
a trend to higher rates of ECE in patients with a positive
FNAB compared to positive SLNB. This is significant as
patients in the ACOSOG Z0011 trial were excluded if they
demonstrated evidence of gross extranodal disease.
These results suggest that patients with early stage
breast cancer subjected to US with FNAB with positive
lymph nodes are unlikely to harbor a heavy axillary met-
astatic disease burden. Furthermore, in patients with
clinical early stage breast cancer, axillary disease as
defined by a positive US, and FNAB does not appear to be
significantly more advanced than in those patients identi-
fied as node positive by SLNB. These results are of value
from the standpoint of hypothesis generation as contem-
porary studies in the management of node-positive breast
cancer are focusing on restricting surgical management of
the axilla to the minimum number of patients for whom it
is required.
The ACOSOG Z0011 trial examined patients with 2 or
fewer positive SLN, and suggests that it may be safe to
omit ALND completely. A total of 89.4 % of patients
harbored 2 or fewer nodes, with 71.1 % harboring only 1
positive SLN. At the conclusion of the study, no difference
in overall survival or disease-free survival at 5 years was
observed in patients with 2 or fewer SLN subject to ALND
versus whole-breast irradiation and systemic chemotherapy
 J. Cools-Lartigue et al.

TABLE 6 Multivariate predictors of US result (normal/abnormal), metastases as ‘‘any positive LN’’

Variable in model Missing Any positive LN AJCC, ER Her-2/ LVI, p Grade, Age at
data, n p neu, p p diagnosis,
OR 95 % CI p OR 95 % CI p p

Single variable – – \0.0001 \0.0001 – – 0.0022 0.0324 0.0013 0.0070 0.2525

(univariate logistic
regression)
Any positive LN, AJCC 0 10.3 3.3–32.5 \0.0001 0.6816 – – – – – – –
Any positive LN, AJCC, 38 13.2 3.8–45.6 \0.0001 0.6042 0.226 0.093–0.551 0.0011 – – – –
ER
Any positive LN, AJCC, 59 22.2 4.5–109.1 0.0001 0.1837 0.322 0.126–0.820 0.0175 0.2484 – – –
ER, Her-2/neu
Any positive LN, AJCC, 84 25.4 2.9–219.7 0.0033 0.1788 0.294 0.105–0.825 0.0200 0.9933 0.4968 – –
ER, Her-2/neu, LVI
Any positive LN, AJCC, 85 24.7 2.9–212.8 0.0035 0.1882 0.273 0.091–0.818 0.0204 0.9686 0.5112 0.9088 –
ER, Her-2/neu, LVI,
grade
Any positive LN, AJCC, 85 21.1 2.48–184.89 0.0059 0.2496 0.291 0.094–0.899 0.0320 0.9297 0.4664 0.8822 0.1613
ER, Her-2/neu, LVI,
grade, age at diagnosis
US ultrasound, LN lymph node, OR odds ratio, CI confidence interval, AJCC American Joint Committee on Cancer, ER estrogen receptor, LVI
lymphovascular invasion
Multivariate analysis demonstrates that ER status remains a significant independent predictor of abnormal US in addition to the size of nodal
metastases. At the time of analysis, not all pathologic data were available for all patients. Patients with incomplete data were omitted from
multivariate analysis. The number of omitted patients is represented as missing data

(ALND 91.8 % vs. SLNB 92.5 % and ALND 82.2 % vs.
SLNB 83.9 %, respectively).
Additional studies are underway that may serve to
broaden the population of patients who may forego axillary
surgery after diagnosis of axillary disease by SLNB. The
recently presented MA-20 study demonstrated a survival
benefit in women with 1–3 positive SLN treated with
breast-conserving therapy and whole-breast irradiation who
received additional regional node irradiation in the axilla.
These results demonstrated an approximate 2 % improve-
ment in locoregional disease-free survival and an
approximate 5 % improvement in distant disease-free sur-
vival, with a trend to increased overall survival with
axillary lymph node irradiation in patients who met the
inclusion criteria for the ACOSOG Z0011 trial.28 The
EORTC AMAROS trial is an ongoing study, which ran-
domizes patients with positive SLNB to receive either
ALND or axillary radiotherapy alone.29
An extension of results of the ACOSOG Z0011 trial has
profound clinical implications—namely, preoperative
identification of axillary lymph node metastases may 1 day
obviate the need for axillary surgery altogether. Con-
versely, SLNB could be reserved for patients who cannot
be diagnosed as node positive in the preoperative period.30
Although it cannot be assumed that observation of the
axilla in such patients would lead to the same outcomes
after SLNB as observed in ACOSOG Z0011, the question
as to whether patients with a positive FNAB required
additional axillary surgery as opposed to whole-breast
irradiation and chemotherapy logically arises.30,31
In conclusion, patients with early stage breast cancer
and clinically negative axillae are unlikely to demonstrate
heavy axillary disease burden. In these patients, axillary
US in conjunction with FNAB represents an effective
modality for the preoperative detection of nodal metasta-
ses. The current treatment of invasive carcinoma of the
breast indicates that the majority of patients undergo a
SLNB in order to determine axillary lymph node status
with subsequent ALND in the case of node-positive dis-
ease. This approach to the assessment of the axilla
increases costs associated with the use of operating room
time, radioisotope, and surgeon and pathologist fees.
Given the high specificity of axillary US when combined
with FNAB, a significant number of patients—10 % in the
present study—can be preoperatively identified as node
positive and potentially be spared unnecessary SLNB.
Furthermore, as the role of ALND in the treatment of node-
positive breast cancer evolves, the possibility of avoiding
axillary surgery altogether in selected patients arises. This
study offers an analysis of the nodal characteristics in
patients with axillary metastases identified by US and
SLNB, and suggests that total disease burden and patho-
logic features are similar in these two groups. It is important
to note, however, that in this study, the comparison of
pathologic features in FNAB and SLNB specimens revealed
an increased percentage of macrometastases and Her-2/neu
Diagnosis of Axillary Metastases in Breast Cancer
positivity in patients with positive FNAB. The clinical
implications of these findings remain uncertain. Accord-
ingly, additional prospective studies examining larger
groups of patients are required in order to substantiate these
findings.
REFERENCES
1. Cody HS 3rd. Current surgical management of breast cancer.
Curr Opin Obstet Gynecol. 2002;14:45–52.
2. Mansel RE, Fallowfield L, Kissin M, et al. Randomized multi-
center trial of sentinel node biopsy versus standard axillary
treatment in operable breast cancer: the ALMANAC Trial. J Natl
Cancer Inst. 2006;98:599–609.
3. Veronesi U, Viale G, Paganelli G, et al. Sentinel lymph node
biopsy in breast cancer: ten-year results of a randomized con-
trolled study. Ann Surg. 2010;251:595–600.
4. Orr RK. The impact of prophylactic axillary node dissection on
breast cancer survival—a Bayesian meta-analysis. Ann Surg
Oncol. 1999;6:109–16.
5. Sakorafas GH, Safioleas M. Breast cancer surgery: an historical
narrative. Part III. From the sunset of the 19th to the dawn of the
21st century. Eur J Cancer Care (Engl). 2010;19:145–66.
6. Fisher B, Bauer M, Wickerham DL, et al. Relation of number of
positive axillary nodes to the prognosis of patients with primary
breast cancer. An NSABP update. Cancer. 1983;52:1551–7.
7. Gipponi M, Bassetti C, Canavese G, et al. Sentinel lymph node as
a new marker for therapeutic planning in breast cancer patients.
J Surg Oncol. 2004;85:102–11.
8. Genta F, Zanon E, Camanni M, et al. Cost/accuracy ratio analysis
in breast cancer patients undergoing ultrasound-guided fine-nee-
dle aspiration cytology, sentinel node biopsy, and frozen section
of node. World J Surg. 2007;31:1155–63.
9. Alvarez S, Añorbe E, Alcorta P, López F, Alonso I, Cortés J. Role
of sonography in the diagnosis of axillary lymph node metastases
in breast cancer: a systematic review. AJR Am J Roentgenol.
2006;186:1342–8.
10. Temple LK, Baron R, Cody HS 3rd, et al. Sensory morbidity after
sentinel lymph node biopsy and axillary dissection: a prospective
study of 233 women. Ann Surg Oncol. 2002;9:654–62.
11. Boughey JC, Moriarty JP, Degnim AC, Gregg MS, Egginton JS,
Long KH. Cost modeling of preoperative axillary ultrasound and
fine-needle aspiration to guide surgery for invasive breast cancer.
Ann Surg Oncol. 2010;17:953–8.
12. Tahir M, Osman KA, Shabbir J, et al. Preoperative axillary
staging in breast cancer—saving time and resources. Breast J.
2008;14:369–71.
13. Swinson C, Ravichandran D, Nayagam M, Allen S. Ultrasound
and fine needle aspiration cytology of the axilla in the pre-
operative identification of axillary nodal involvement in breast
cancer. Eur J Surg Oncol. 2009;35:1152–7.
14. Susini T, Nori J, Olivieri S, et al. Predicting the status of axillary
lymph nodes in breast cancer: a multiparameter approach
including axillary ultrasound scanning. Breast. 2009;18:103–8.
15. Bonnema J, van Geel AN, van Ooijen B, et al. Ultrasound-guided
aspiration biopsy for detection of nonpalpable axillary node
metastases in breast cancer patients: new diagnostic method.
World J Surg. 1997;21:270–4.
16. Motomura K, Inaji H, Komoike Y, et al. Gamma probe and ultra-
sonographically-guided fine-needle aspiration biopsy of sentinel
lymph nodes in breast cancer patients. Eur J Surg Oncol.
2001;27:141–5.
17. Yang WT, Ahuja A, Tang A, Suen M, King W, Metreweli C.
High resolution sonographic detection of axillary lymph node
metastases in breast cancer. J Ultrasound Med. 1996;15:241–6.
18. Vaidya JS, Vyas JJ, Thakur MH, Khandelwal KC, Mittra I. Role
of ultrasonography to detect axillary node involvement in oper-
able breast cancer. Eur J Surg Oncol. 1996;22:140–3.
19. Verbanck J, Vandewiele I, De Winter H, Tytgat J, Van Aelst F,
Tanghe W. Value of axillary ultrasonography and sonographi-
cally guided puncture of axillary nodes: a prospective study in
144 consecutive patients. J Clin Ultrasound. 1997;25:53–6.
20. de Kanter AY, van Eijck CH, van Geel AN, et al. Multicentre
study of ultrasonographically guided axillary node biopsy in
patients with breast cancer. Br J Surg. 1999;86:1459–62.
21. Krishnamurthy S, Sneige N, Bedi DG, et al. Role of ultrasound-
guided fine-needle aspiration of indeterminate and suspicious
axillary lymph nodes in the initial staging of breast carcinoma.
Cancer. 2002;95:982–8.
22. Sapino A, Cassoni P, Zanon E, et al. Ultrasonographically-guided
fine-needle aspiration of axillary lymph nodes: role in breast
cancer management. Br J Cancer. 2003;88:702–6.
23. Boughey JC, Middleton LP, Harker L, et al. Utility of ultrasound
and fine-needle aspiration biopsy of the axilla in the assessment
of invasive lobular carcinoma of the breast. Am J Surg. 2007;
194:450–5.
24. Kuerer HM, Newman LA, Fornage BD, et al. Role of axillary
lymph node dissection after tumor downstaging with induction
chemotherapy for locally advanced breast cancer. Ann Surg
Oncol. 1998;5:673–80.
25. Klauber-Demore N, Kuzmiak C, Rager EL, et al. High-resolution
axillary ultrasound is a poor prognostic test for determining
pathologic lymph node status in patients undergoing neoadjuvant
chemotherapy for locally advanced breast cancer. Am J Surg.
2004;188:386–9.
26. Vlastos G, Fornage BD, Mirza NQ, et al. The correlation of
axillary ultrasonography with histologic breast cancer down-
staging after induction chemotherapy. Am J Surg. 2000;179:
446–52.
27. Tiwari RK, Borgen PI, Wong GY, Cordon-Cardo C, Osborne
MP. HER-2/neu amplification and overexpression in primary
human breast cancer is associated with early metastasis. Anti-
cancer Res. 1992;12:419–25.
28. Latosinsky S, Berrang TS, Cutter CS, et al. CAGS and ACS
Evidence Based Reviews in Surgery. 40. Axillary dissection
versus no axillary dissection in women with invasive breast
cancer and sentinel node metastasis. Can J Surg. 2012;55:66–9.
29. Hurkmans CW, Borger JH, Rutgers EJ, et al. Quality assurance of
axillary radiotherapy in the EORTC AMAROS trial 10981/22023:
the dummy run. Radiother Oncol. 2003;68:233–40.
30. Solon JG, Power C, Al-Azawi D, Duke D, Hill AD. Ultrasound-
guided core biopsy: an effective method of detecting axillary
nodal metastases. J Am Coll Surg. 2010;214:12–7.
31. Giuliano AE, Hunt KK, Ballman KV, et al. Axillary dissection
vs. no axillary dissection in women with invasive breast cancer
and sentinel node metastasis: a randomized clinical trial. JAMA.
2011;305:569–75.