Evaluation of axillary involvement by ultrasound-guided lymph node biopsy: A prospective study

Gyne´cologie Obste´trique & Fertilite´ 43 (2015) 431–436

Available online at

ScienceDirect www.sciencedirect.com

Original article

Evaluation of axillary involvement by ultrasound-guided lymph node biopsy: A prospective study E´valuation de l’envahissement ganglionnaire axillaire par un pre´le`vement e´choguide´ : e´tude prospective C. Jankowski a,*, D. Hudry a, D. Vaillant b, O. Varbedian b, N. Mejean b, F. Guy b, S. Feutray b, C. Coutant a a b

Department of Surgical Oncology, Georges-Franc¸ois-Leclerc Cancer Center, 1, rue du Professeur-Marion, 21000 Dijon, France Radiology Department, Georges-Franc¸ois-Leclerc Cancer Center, 21000 Dijon, France

A R T I C L E I N F O

A B S T R A C T

Article history: Received 4 March 2015 Accepted 17 April 2015 Available online 16 May 2015

Objectives. – The primary aim of this prospective study was to evaluate the accuracy of pre-operative ultrasound (US) alone and associated with a fine needle aspiration cytology (FNAC) or a core needle biopsy (CNB) in the diagnosis of axillary node involvement in patient with breast cancer. The secondary study objective was to determine if this US  FNAC or CNB can lead to the adequate axillary surgery in cN0 and cN1 patient. Methods. – A total of 121 consecutive women with stage cT1 to cT2, cN0/cN1, invasive breast cancer were prospectively identified at our institution between February 2, 2013 and August 30, 2013. The sensitivity, specificity, VPP, NPV were calculated, with confidence intervals, using the definitive histological result of the sentinel node biopsy (SLNB) or axillary lymph node dissection (ALND) as the baseline. Results. – Twenty-seven CNB and 2 FNAC were performed. For the whole series, the sensitivity and the specificity of US alone were 48.7% [36–59%] and 89% [83–94%]. For US  FNAC or CNB, the sensitivity and the specificity were 35.9% [26–38%] and 98.8% [94–100%]. Seven women with cN1 clinical examination had SLNB, which permit to decrease the number of ALND of 16.3%. It would have avoided unnecessary SLNB, prompting immediate ALND in 9 patients with cN0 axillae, which means a reduction of SLNB of 8.6%. US  FNAC or CNB lead to the adequate surgery in 72.7% of cases. Conclusion. – US  CNB or FNAC is also a relatively efficient and safe test and should be considered routinely. It allowed triaging patients to the well axillary surgery (SLNB or ALND). ß 2015 Elsevier Masson SAS. All rights reserved.

Keywords: Breast cancer Ultrasound Core needle biopsy Axillary surgery Lymph node involvement

R E´ S U M E´

Mots cle´s : Cancer du sein E´chographie Microbiopsies Chirurgie axillaire Envahissement ganglionnaire

Objectifs. – L’objectif principal de cette e´tude e´tait d’e´valuer les performances de l’e´chographie axillaire associe´e ou non a` un pre´le`vement axillaire histologique ou cytologique e´choguide´ (PA) dans la stadification ganglionnaire pre´-the´rapeutique chez les patientes ayant un cancer du sein. L’objectif secondaire e´tait de de´terminer si cette e´chographie associe´e ou non a` un PA permettait de de´cider du bon geste axillaire chez les patientes cN0 et cN1. Me´thodes. – Les re´sultats de 121 e´chographies axillaires, chez des patientes cN0/cN1 prises en charge ˆ t 2013. La par chirurgie premie`re, ont e´te´ recense´s de fac¸on prospective entre le 2 fe´vrier et le 30 aou sensibilite´, la spe´cificite´, la VPP et la VPN e´taient calcule´s en utilisant comme crite`re de jugement l’envahissement du creux axillaire e´value´ par un ganglion sentinelle (GS) ou un curage axillaire (CA).

* Corresponding author. E-mail address: [email protected] (C. Jankowski). http://dx.doi.org/10.1016/j.gyobfe.2015.04.007 1297-9589/ß 2015 Elsevier Masson SAS. All rights reserved.

432

C. Jankowski et al. / Gyne´cologie Obste´trique & Fertilite´ 43 (2015) 431–436

Re´sultats. – Vingt-sept PA ont e´te´ re´alise´s (25 microbiopsies et 2 cytoponctions). Sur l’ensemble de la population, la sensibilite´ et la spe´cificite´ e´taient de 48,7 % [36–59 %], et 89 % [83–94 %] pour l’e´chographie seule. La sensibilite´ et la spe´cificite´ e´taient de 35,9 % [26–38 %], et 98,8 % [94–100 %] pour l’e´chographie associe´e a` un PA. Sept patientes cN1 ont pu be´ne´ficier d’une proce´dure du GS, ce qui a diminue´ le nombre de CA de 16,3 %. Neuf patientes cN0 ont pu be´ne´ficier d’un CA leur e´vitant un GS inutile. Le couple e´chographie associe´ ou non a` un PA a permis de re´aliser le bon geste axillaire (GS ou CA) dans 72,7 % des cas. Conclusion. – L’e´chographie axillaire associe´e a` un PA est un examen sensible et tre`s spe´cifique pour de´terminer le statut axillaire et permettre de re´aliser le bon geste axillaire. ß 2015 Elsevier Masson SAS. Tous droits re´serve´s.

1. Introduction Locoregional lymph node involvement is a key element in the staging of patients with invasive breast cancer. Lymph node status represents a major prognostic factor and can influence adjuvant treatment, such as radiotherapy, intra-operative accelerated partial breast radiotherapy, immediate reconstruction and axillary surgery in case of neoadjuvant chemotherapy. Sentinel lymph node biopsy (SLNB) has supplanted axillary node dissection (ALND) as the gold standard for the operative nodal staging of T1-2 N0 breast cancer. The decision to limit axillary lymph node dissection to SLNB is based on the clinical and radiological examination of the axillae. Although SLNB is less invasive and associated with lower morbidity, up to 50% of sentinel nodes will contain tumor deposits and require complete ALND [1]. However, some criteria have changed since the publication of the ACOSOGZ0011 trial especially in patients with micrometastasis [2]. No pre-operative non-invasive exam is sufficiently accurate to determine node status before surgery [3]. Multiple non-surgical methods have been used with varying success to predict lymph node involvement, including physical examination, digital mammography, ultrasonography, computed tomography (CT) scan, positron emission tomographic (PET) imaging and magnetic resonance imaging (MRI) [4–11]. Physical examination, the oldest method of investigation of axillary lymph nodes, has limited reliability, and thus, axillary ultrasound (US) with fine needle cytology (FNAC) or core needle biopsy (CNB) has become the standard for pre-operative evaluation of axillary lymph nodes [12]. When there are no palpable nodes (cN0), identifying axillary node involvement pre-operatively can save the patient and surgeon a SLNB procedure and can reduce operating time. Conversely, when there are palpable nodes (cN1), if the combination of US  FNAC or CNB shows no node involvement, ALND can be avoided and SLNB can be performed. Accurately screening patients to determine whether axillary lymph nodes are suspicious is necessary to refer the patient for the appropriate type of axillary surgery. In this paper, we present a prospective study whose primary objective was to evaluate the accuracy of pre-operative US alone and associated with FNAC or CNB (in terms of sensitivity, specificity, positive (PPV) and negative predictive value (NPV)) for the diagnosis of axillary lymph node involvement in patients with breast cancer. The secondary objective was to determine whether US associated or not with axillary biopsy leads to appropriate axillary surgery in cN0 and cN1 patients.

2. Methods 2.1. Patients This study was approved by the ethics committee and scientific board of the Georges-Franc¸ois-Leclerc Cancer Center. The committee considered that US  FNAC or CNB constituted routine treatment, since this is the treatment recommended by the National

Cancer Institute and, therefore, the need for informed consent was waived. The data were prospectively collected between February and August 2013. Patients undergoing surgery for invasive breast cancer, cN0 or cN1, T1 or T2 were prospectively reviewed. Patients with a clinical finding of nodes fixed to each other, patients with T3 or T4 tumor, patients with ductal or lobular carcinoma in situ and those treated with pre-operative chemotherapy were excluded from the study. The following data are prospectively recorded for all patients: age, cT stage, cN stage, multifocal tumor, US tumor size, tumor type, histologic size, Scarff-Bloom-Richardson (SBR) classification, Ki67, estrogen receptor status, progesterone status, HER2 overexpressed or amplified, immunochemistry classification (luminal A, luminal B, luminal B HER2, triple negative), HER2 overexpressed or amplified, surgical techniques and pathological result (breast conserving surgery, mastectomy, SLNB, number of sentinel nodes (SN), result of frozen section of SN, combination of serial sectioning and immunochemistry, ALND, number of metastatic nodes in ALND). 2.2. Physical examination Physical examination was performed by the radiologist before mammography and US. Physical examination was considered as normal if no axillary lymph nodes were palpable, and considered as abnormal if one or more movable axillary lymph nodes were palpable. 2.3. Ultrasound and final reference standard The US, FNAC and CNB were performed by one of 5 dedicated breast radiologists using a Toshiba Aplio1 or Xario1 machine (Toshiba Medical Systems1, Zoetermeer, Netherlands) and a high frequency (8 MHz) linear array probe at the time of examination of the primary tumor. All patients underwent US imaging of the ipsilateral and contralateral axillary lymph nodes. The radiologist categorized the axillary lymph nodes as normal or suspicious. The radiological criteria for classifying a lymph node as suspicious were a round shape, a hypoechoic cortex, a local or a general thickening of the cortex, and a loss of symmetry compared the other side. If a lymph node was suspected to be pathological, FNAC or CNB was performed, at the discretion of the operator. US-guided CNB was performed using a 14- or 16-G spring-loaded Magnum1 core biopsy device (Bard Biopsy Systems, Tempe, AZ, USA). USguided FNAC was performed with a 21-G needle (BD Mircolance 31) In case of cytologically proven axillary metastasis, the patient underwent primary ALND. If the US was normal or if the result showed normal lymphatic tissue, SLNB using radiocolloid and blue dye injection was performed, followed by ALND for sentinel node positivity. As per the American Joint Committee on Cancer breast cancer stage classification published in 2010, any patients with isolated tumor cells (pN0(i+)) were considered to be node-negative and did not undergo any additional lymph node surgery [13].

C. Jankowski et al. / Gyne´cologie Obste´trique & Fertilite´ 43 (2015) 431–436

2.4. Histopathological evaluation FNAC or CNB were evaluated by pathologists using permanent sectioning. Aspirates of FNAC were rinsed and centrifugated to create a monolayer preparation. Monolayer slides were stained with the Papanicolaou and MGG (May-Gru¨nwald-Giemsa) methods. Sentinel lymph node (SN) processing consisted of taking two parallel sections at three levels, 500 microns apart, stained with haematoxylin-eosin and processed for immunohistochemical studies.

433

intervals) were calculated for axillary US alone and for axillary US  FNAB or CNB. For patients who had both US + FNAC or CNB, the test was considered as positive if the FNAC or CNB were positive and the test was considered as negative if the FNAC or CNB was negative. We calculated the percentage of SLNB that were avoided by studying the patients who were initially candidates for SLNB (single invasive tumor <5 cm and cN0 examination) who had positive FNAC or CNB. We calculated the percentage of patients who underwent appropriate surgery, defined as:  SLNB for patients who had no axillary involvement;  ALND for patients who had axillary involvement.

2.5. Statistical analysis All data were entered into a single spreadsheet (Excel; Microsoft, Redmond, WA, USA), and results were analyzed with descriptive statistical methods. We recorded the following variables: ultrasonographic characteristics of the lymph nodes, findings at histological study of axillary lymph node FNAC or CNB, definitive histologic diagnosis after SLNB and/or ALND. Findings at axillary US  FNAC or CNB were correlated with the histologic result of the lymph node at SLNB and/or ALND; 2  2 contingency tables were prepared and the resulting measures of diagnostic accuracy (sensitivity, specificity, PPV, NPV and their respective confidence

Table 1 Patient demographics and tumor characteristics. Characteristics

n = 121

Age (years  SD)

63.2  13.3

cT stage cT1 cT1a cT1b cT1c cT2

96 (79.3%) 5 (4.1%) 39 (32.2%) 52 (43%) 25 (20.7%)

cN stage cN0 cN1

108 (89.3%) 13 (10.7%)

Multifocal

16 (13.2%)

Ultrasound tumor size (mm  SD)

16.5  10.1

Tumor type Undifferentiated Lobular Other

96 (79.3%) 23 (19%) 4 (3.3%)

Histologic size (mm  SD)

17.8  12.6

Scarff-Bloom-Richardson (SBR) 1 2 3

45 (37.2%) 51 (42.1%) 25 (20.7%)

Ki67 (%, range) <15% 15–30% >30%

44 (36.4%) 42 (34.7%) 32 (26.4%)

Statistical analysis for differences or trends in estimates was performed using the Chi-square test (x2), significance being set at a P < 0.05 level. Data analysis was performed using R version 2.15.2 (R Development Core Team1, 2012). 3. Results 3.1. Patients demographics and tumor characteristics A total of 121 consecutive women with stage cT1 to cT2, cN0/ cN1, invasive breast cancer were prospectively identified at our institution between February 2, 2013 and August 30, 2013. Table 1 details the characteristics of the study population as well as the tumor characteristics. Average age was 63.2  13.3 years. Ninetysix patients had pT1 (79.3%) and 25 patients had pT2 (20.3%) tumors. Sixteen patients (13%) had bifocal tumors. Pathological N staging was cN0 in 108 (89.3%) patients and cN1 in 13 (10.7%) patients. The average size of the tumor at ultrasound was 16.5  10.1 mm. The tumor type was undifferentiated in 94 cases (77.7%), lobular in 23 cases (19%) and other type in 4 cases (3.3%). 3.2. Results of axillary ultrasound findings The results of axillary US are summarized in Table 2. The nodes were round shaped in 4.1% of cases (n = 5). There was a hypoechoic cortex in 5 of the cases (4.1%). There was a cortical thickening in 22.3% of cases (n = 27). There was a loss of symmetry compared to the controlateral side in 19.8% of cases (n = 24). US was considered Table 2 Axillary ultrasound findings. Criteria

n = 121

Shape Reniform Round

116 (95.9%) 5 (4.1%)

Cortex Normal Hypoechoic

116 (95.9%) 5 (4.1%)

109 (90.1%) 12 (9.9%)

Cortical aspect Thin Local thickening General thickening

94 (77.7%) 14 (11.6%) 13 (10.7%)

PR status Positive Negative

88 (72.7%) 33 (27.3%)

Symmetry compared to other side Normal Loss

97 (80.2%) 24 (19.8%)

Her2 overexpressed or amplified

9 (7.4%)

Immunohistochemistry classification Luminal A Luminal B Luminal B HER2 Triple negative HER2 overexpressed or amplified

45 (37.2%) 60 (49.6%) 4 (3.3%) 7 (5.8%) 5 (4.1%)

Ultrasound conclusion Normal Suspicious

93 (76.9%) 28 (23.1%)

No axillary biopsy

94 (77.7%)

Axillary biopsy Needle core biopsy Fine needle aspiration cytology

27 (22.3%) 25 (20.7%) 2 (1.7%)

ER status Positive Negative

ER: estrogen receptor; PR: progesterone receptor; SD: standard deviation.

C. Jankowski et al. / Gyne´cologie Obste´trique & Fertilite´ 43 (2015) 431–436

434 Table 3 Surgical techniques and pathological results.

Table 5 Accuracy of ultrasound  FNAB or NCB.

Characteristics

Values

Breast conserving surgery

94 (77.7%)

Mastectomy

27 (22.3%)

SLNB

95 (78.5%)

Number of sentinel nodes (mean  SD)

2.3  1.2

Results of frozen section of sentinel nodes Positive Negative

9 (9.5%) 86 (90.5%)

Combination of serial sectioning and immunohistochemistry Negative Isolated cells Micrometastasis Macrometastasis

69 (72.6%) 3 (3.2%) 9 (9.5%) 14 (14.7%)

ALND

36 (29.7%)

Number of nodes in ALND (mean  SD)

13.2  5.4

Number of metastasis nodes in ALND (mean  SD)

2.2  3.3

SLNB: sentinel lymph node biopsy; ALND: axillary lymph node dissection; SD: standard deviation.

Sensitivity

Specificity

PPV

NPV

All cN 35.9% [26–38%] 98.8% [94–100%] 93.3% [68–100%] 76.4% [73–77%] 27.3% [17–30%] 98.7% [94–100%] 90% [56–100%] 77.5% [72–77%] cN0 cN1 83.3% [47–83%] 100% [69–100%] 100% [56–100%] 87.5% [60–88%] PPV: positive predictive value; NPV: negative predictive value; []: 95% confidence interval.

group. A cN1 result on examination was significantly associated with a suspicious US (P < 0.001). For the whole series, the sensitivity, specificity, PPV and NPV were respectively 48.7% (95%CI [36–59%]), 89% (95%CI [83–94%]), 67.9% (95%CI [50–82%]) and 78.5% (95%CI [73–83%]). For the cN0 group, the sensitivity, specificity, PPV and NPV were respectively 39.4% (95%CI [26–49%]), 92% (95%CI [86–96%]), 68.4% (95%CI [46–86%]) and 77.5% (95%CI [73–81%]). For the cN1 group, the sensitivity, specificity, PPV and NPV were respectively 100% (95%CI [65–100%]), 57.1% (95%CI [27–57%]), 66.7% (95%CI [43–67%]) and 100% (95%CI [47–100%]). 3.6. Accuracy of US  FNAC or CNB

as suspicious in 28 patients (23.1%). CNB was performed in 25 women, FNAC in 2 women. One patient with a suspicious US did not have CNB or FNAC because ALND was indicated for bifocal tumor and the nodes appeared to be positive. 3.3. Surgical techniques and nodal status The surgical techniques and the nodal status are summarized in Table 3. Mastectomy was performed in 27 (22.3%) patients, whereas conservative surgery was performed in 94 (77.7%). SLNB was performed in 95 patients (78.5%). The mean number of sentinel nodes removed was 2.3  1.2. Thirty-six women underwent ALND (29.7%; 15 had a positive US + FNAC/CNB, 11 SLNB were positive, the others underwent ALND because of bifocal tumors or failed SLNB). The mean number of metastasis nodes was 2.2  3.3. In 39 cases (32.2%), histological examination showed metastases in the SN or ALND. 3.4. Accuracy of clinical examination Among the 13 cN1 patients, 6 had positive lymph nodes. Conversely, 33 patients with cN0 examination had positive lymph nodes. A palpable node was not significantly associated with axillary metastasis (P = 0.41). The sensitivity, the specificity, the PPV and the NPV of the examination were 15.4% (95%CI [7–24%]), 91.5% (95%CI [87–95%]), 46.2% (95%CI [21–73%]), and 69.4% (95%CI [66–73%]). 3.5. Accuracy of US alone The diagnostic performance of axillary US alone is summarized in Table 4. Among the 39 patients with axillary metastases, 19 were recognized as pathological by the radiologist, corresponding to a detection rate by US alone of 48.7%. In the cN0 group, 17.6% (n = 19) patients had a positive US versus 69.2% (n = 9) in the cN1 Table 4 Accuracy of ultrasound alone.

All cN cN0 cN1

Sensibility

Specificity

PPV

NPV

46.3% [34–56%] 46.2% [30–58%] 53.8% [32–70%]

89% [83–93%] 91.4% [86–96%] 75% [51–92%]

67.9% [50–82%] 66.7% [43–84%] 70% [41–91%]

76% [72–81%] 82.1% [77–86%] 60% [41–74%]

PPV: positive predictive value; NPV: negative predictive value; []: 95% confidence interval.

The diagnostic performance of axillary US  FNAC or CNB is summarized in Table 5. Twenty-five CNB and 2 FNAC were performed. There were 1 false positive (0.08%) and 25 false negative results (23.6%). One patient did not undergo any axillary FNAC or CNB because the nodes seemed to be positive at US. The pathologic examination revealed no axillary metastasis. Among the 39 N+ patients, 19 were identified by US with CNB, with a detection rate of 48.7%. Among the 13 negative FNAC or CNB, 8 were confirmed on final pathology, while 5 demonstrated metastases on pathologic node examination. No patient with micrometastasis was detected with US  FNAC or CNB. For the entire population, the sensitivity, specificity, PPV and NPV of US  FNAC or CNB for the detection of axillary metastases were respectively 35.9% (95%CI [26–38%]), 98.8% (95%CI [94–100%]), 93.3% (95%CI [68–100%]), and 76.4% (95%CI [73–77%]). The accuracy was increased by the association of US with FNAC or CNB. In the cN0 group, the sensitivity, specificity, PPV and NPV were respectively 27.3% (95%CI [17–30%]), 98.7% (95%CI [94–100%]), 90% (95%CI [56–100%]), and 77.5% (95%CI [72–77%]). In the cN1 group, the sensitivity, specificity, PPV and NPV were respectively 83.3% (95%CI [47–83%]), 100% (95%CI [69–100%]), 100% (95%CI [56–100%]), and 87.5% (95%CI [60–88%]). 3.7. Influence on the axillary surgery The pre-operative identification rate of axillary metastases was 35.9% (14/39) corresponding to a false negative rate was about 23.6%. A total of 7 women with cN1 clinical examination had SLNB, which made it possible to reduce the number of ALND by 16.3%. The combined US  FNAC or CNB strategy made it possible to avoid unnecessary SLNB by prompting immediate ALND in 9 patients with clinically negative axillae, corresponding to a reduction in SLNB of 8.6%. In the cN0 subset, 9 patients had ALND only, and thus did not undergo unnecessary SLNB. US  FNAC or CNB led to appropriate surgery in 72.7% of cases.

4. Discussion In the present study, we demonstrated that US  FNAC or CNB led to appropriate surgery in 72.7% of cases. We determined the accuracy of pre-operative axillary US  FNAC or CNB in order to

C. Jankowski et al. / Gyne´cologie Obste´trique & Fertilite´ 43 (2015) 431–436

ascertain its value to do the adequate axillary surgery in cN0 and cN1 patients. Axillary node staging of women with primary breast cancer has developed over time, and women should now be triaged to initial SLNB or ALND according to the most accurate pre-operative diagnostic information. Our study shows that axillary US  FNAC or CNB has adequate sensitivity and high specificity, and may be a useful triage test for deciding on which initial nodal surgery procedure to perform. When US was used alone to detect axillary metastases, sensitivity and specificity of 48.7 and 89% respectively were observed. A meta-analysis by Alvarez et al. found that sensitivity varied between 54.7 and 92.3%, and specificity between 80.4 and 97.1% [14]. Taken together, all these data showed that US alone has modest accuracy, and does not provide sufficient accuracy to decide on surgical treatment of the axillae. In particular, it is not sufficiently specific to triage patients directly to ALND or SNLB. When combined with FNAC or CNB in case of suspicious US, the sensitivity and specificity of axillary US were 35.9 and 98.8% respectively. A systematic review by Alvarez et al. reported sensitivity ranging from 54.7 to 92.3%, and specificity ranging from 80.4 to 97.1%. The sensitivity observed in our study is lower, but this can be explained by the fact that we performed FNAC or CNB on lymph nodes that were suspicious at US only. Conversely, some studies performed FNAC or CNB if a lymph node was sonographically visualized [15,16]. In women who had clinically negative nodes, US  FNAC or CNB is moderately sensitive (27.3%) and extremely specific (98.7%). In this group, it will therefore identify about a third of the cases (9/33) with node metastases (who can then proceed to ALND directly). The benefits of forgoing SLNB in node-positive breast cancer patients include cost saving (even if real medico-economic trials have to be done), decreased morbidity, and avoidance of the risk of false negative SLNB [17–19]. SLNB involves some risks and uses hospital resources, so it is important to accurately select the patients who may benefit from this procedure. For CNB, these results are supported by GarciaOrtega et al., who identified 33% of node-positive patients [20]. For FNAC, the number of SLNB procedures avoided by axillary sampling, redirecting patients to ALND, was estimated in nine studies [17,18,20–26] and was between 8 and 50%. In the meta-analysis by Houssami et al., the median proportion of women triaged directly to ALND was 19.8% for both FNAC and CNB together [27]. We performed only 2 FNAC and both were false negative. This can be explained by the absence of a pathologist during the procedure. As Ciatto et al. reported, this technique suffers from false positives and false negatives, requires an experienced cytologist during the procedure, and fails to detect small metastatic deposits in axillary lymph nodes [24]. A good alternative is CNB. This technique allows more thorough histological examination without the need for a pathologist to be present during the procedure. CNB is not widely used, probably due to the fear of complications given the proximity of the axillary vessels and axillary plexus. However, like other authors [20,28], we observed no complications and CNB was well tolerated in our patients. Thus, CNB is convenient, safe and seems to be more accurate than FNAC. However, a prospective randomized study is warranted to confirm these findings. In women with palpable node, sensitivity and specificity increased (83.3 and 100% respectively). Five out of 6 patients with axillary metastases were correctly identified with US + FNAC or CNB. cN1 patients are more easily imaged on ultrasound, which makes patients with clinically node-negative disease more technically challenging for sonographic evaluation. In this study, the pre-operative identification rate of axillary metastases was 35.9% (14/39). Our false negative rate was 23.6%. Diepstraten et al. reviewed 31 studies and showed a pooled false negative rate of 25% [29]. The risk of having false negative results is lower in the cN1 group; accordingly, the rate of false negatives was 22.5% for cN0 patients versus 0% for cN1 patients. This indicates that there is still room for improvement in the field of

435

pre-operative assessment of the axillae and reinforces the view that US  FNAC or CNB cannot be relied upon to stage the axillae. Therefore, a negative or inadequate result of US  FNAC or CNB necessitates further evaluation with SLNB. However, US  FNAC or CNB did make it possible to triage patients to the right form of axillary surgery (SLNB or ALND). The fact that the results of US are operator dependent may have implications for the widespread application of this technique. The false negatives are linked to the size of the metastasis. CNB or FNAC only sample a portion of a lymph node and therefore may miss clusters of tumor cells present elsewhere in the lymph node. In our study, no patient with micrometastasis was detected by US associated with FNAC or CNB. In the study by GarciaOrtega et al., they also failed to detect any micrometastasis with US associated with CNB [20]. In the study by Alkuwari and Auger, sensitivity was 16% in the SLNB group, where the mean size of the metastasis was 0.25 mm, whereas it was 88% in the lymph node dissection group, where the mean size was 15 mm [30]. The false positive rate is an important performance measure for a test that provides staging information. We had just one false positive result, but admittedly, there is a bias since this patient did not undergo either FNAC or CNB as the node seemed to be pathological at US and the ALND was ruled out because of multifocal tumor. To date, the role of US-guided CNB or FNAC for suspicious axillary lymph nodes was to guide node-positive patients to ALND directly. However, the indications for performing ALND have become more limited, particularly since the publication of the randomized ACOSOG-Z0011 trial. This study affirmed that women with a positive SLNB and clinical T1–T2 carcinomas undergoing lumpectomy with breast radiation therapy do not benefit from ALND in terms of local control, disease free survival and overall survival [2]. In this group of patients, the added value of axillary US  CNB or FNAC has become questionable. The utility of US with CNB or FNAC will depend on whether or not surgeons choose to adhere to the recommendations of the aforementioned study, by omitting ALND in patients with minimal sentinel node disease. In conclusion, US  CNB or FNAC improves selection of breast cancer patients for SLNB or ALND with a higher accuracy than triage based on clinical and US evaluation alone. For cN1 patients, there is a huge benefit in terms of morbidity because it could avoid unnecessary ALND. However, our findings also reinforce evidence that US, even with CNB or FNAC, does not have a sufficiently strong negative predictive value to forgo operative axillary staging based on current standard of care. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. References [1] Tanis PJ, van Sandick JW, Nieweg OE, Valde´s Olmos RA, Rutgers EJT, Hoefnagel CA, et al. The hidden sentinel node in breast cancer. Eur J Nucl Med Mol Imaging 2002;29:305–11. [2] Giuliano AE, Hunt KK, Ballman KV, Beitsch PD, Whitworth PW, Blumencranz PW, et al. Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: a randomized clinical trial. JAMA J Am Med Assoc 2011;305:569–75. http://dx.doi.org/10.1001/jama.2011.90. [3] Cody 3rd HS. Current surgical management of breast cancer. Curr Opin Obstet Gynecol 2002;14:45–52. [4] Shetty MK, Carpenter WS. Sonographic evaluation of isolated abnormal axillary lymph nodes identified on mammograms. J Ultrasound Med 2004;23:63–71. [5] Pamilo M, Soiva M, Lavast EM. Real-time ultrasound, axillary mammography, and clinical examination in the detection of axillary lymph node metastases in breast cancer patients. J Ultrasound Med 1989;8:115–20. [6] Nori J, Vanzi E, Bazzocchi M, Bufalini FN, Distante V, Branconi F, et al. Role of axillary ultrasound examination in the selection of breast cancer patients for sentinel node biopsy. Am J Surg 2007;193:16–20. http://dx.doi.org/10.1016/ j.amjsurg.2006.02.021.

436

C. Jankowski et al. / Gyne´cologie Obste´trique & Fertilite´ 43 (2015) 431–436

[7] Lee MC, Eatrides J, Chau A, Han G, Kiluk JV, Khakpour N, et al. Consequences of axillary ultrasound in patients with T2 or greater invasive breast cancers. Ann Surg Oncol 2011;18:72–7. http://dx.doi.org/10.1245/s10434-010-1171-4. [8] March DE, Wechsler RJ, Kurtz AB, Rosenberg AL, Needleman L. CT-pathologic correlation of axillary lymph nodes in breast carcinoma. J Comput Assist Tomogr 1991;15:440–4. [9] Peare R, Staff RT, Heys SD. The use of FDG-PET in assessing axillary lymph node status in breast cancer: a systematic review and meta-analysis of the literature. Breast Cancer Res Treat 2010;123:281–90. http://dx.doi.org/10.1007/ s10549-010-0771-9. [10] Kvistad KA, Rydland J, Smethurst HB, Lundgren S, Fjøsne HE, Haraldseth O. Axillary lymph node metastases in breast cancer: preoperative detection with dynamic contrast-enhanced MRI. Eur Radiol 2000;10:1464–71. [11] Yoshimura, Sakurai, Oura, Suzuma, Tamaki, Umemura. et al. Evaluation of axillary lymph node status in breast cancer with MRI. Breast Cancer 1999;6: 249–58. [12] INCa. Cancer du sein infiltrant non me´tastatique – Questions d’actualite´ – Rapport inte´gral. Boulogne-Billancourt: INCa; 2012. [13] Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 2010;17:1471–4. http://dx.doi.org/10.1245/s10434-010-0985-4. ˜ orbe E, Alcorta P, Lo´pez F, Alonso I, Corte´s J. Role of sonography in [14] Alvarez S, An the diagnosis of axillary lymph node metastases in breast cancer: a systematic review. AJR Am J Roentgenol 2006;186:1342–8. http://dx.doi.org/10.2214/ AJR.05.0936. [15] Bonnema J, van Geel AN, van Ooijen B, Mali SP, Tjiam SL, Henzen-Logmans SC, 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] De Kanter AY, van Eijck CH, van Geel AN, Kruijt RH, Henzen SC, Paul MA, et al. Multicentre study of ultrasonographically guided axillary node biopsy in patients with breast cancer. Br J Surg 1999;86:1459–62. http://dx.doi.org/ 10.1046/j.1365-2168.1999.01243.x. [17] Van Rijk MC, Tanis PJ, Nieweg OE, Olmos RAV, Rutgers EJT, Hoefnagel CA, et al. Clinical implications of sentinel nodes outside the axilla and internal mammary chain in patients with breast cancer. J Surg Oncol 2006;94:281–6. http:// dx.doi.org/10.1002/jso.20574. [18] Baruah BP, Goyal A, Young P, Douglas-Jones AG, Mansel RE. Axillary node staging by ultrasonography and fine-needle aspiration cytology in patients with breast cancer. Br J Surg 2010;97:680–3. http://dx.doi.org/10.1002/ bjs.6964. [19] Caretta-Weyer H, Sisney GA, Beckman C, Burnside ES, Salkowsi LR, Strigel RM, et al. Impact of axillary ultrasound and core needle biopsy on the utility of intraoperative frozen section analysis and treatment decision making in women with invasive breast cancer. Am J Surg 2012;204:308–14. http:// dx.doi.org/10.1016/j.amjsurg.2011.10.016.

[20] Garcia-Ortega MJ, Benito MA, Vahamonde EF, Torres PR, Velasco AB, Paredes MM. Pretreatment axillary ultrasonography and core biopsy in patients with suspected breast cancer: diagnostic accuracy and impact on management. Eur J Radiol 2011;79:64–72. http://dx.doi.org/10.1016/j.ejrad.2009.12.011. [21] Jain A, Haisfield-Wolfe ME, Lange J, Ahuja N, Khouri N, Tsangaris T, et al. The role of ultrasound-guided fine-needle aspiration of axillary nodes in the staging of breast cancer. Ann Surg Oncol 2008;15:462–71. http://dx.doi.org/ 10.1245/s10434-007-9623-1. [22] Podkrajsek M, Music MM, Kadivec M, Zgajnar J, Besic N, Pogacnik A, et al. Role of ultrasound in the preoperative staging of patients with breast cancer. Eur Radiol 2005;15:1044–50. http://dx.doi.org/10.1007/s00330-004-2545-4. [23] Somasundar P, Gass J, Steinhoff M, Koeliker S, Dizon D, Cady B, et al. Role of ultrasound-guided axillary fine-needle aspiration in the management of invasive breast cancer. Am J Surg 2006;192:458–61. http://dx.doi.org/ 10.1016/j.amjsurg.2006.06.029. [24] Ciatto S, Brancato B, Risso G, Ambrogetti D, Bulgaresi P, Maddau C, et al. Accuracy of fine needle aspiration cytology (FNAC) of axillary lymph nodes as a triage test in breast cancer staging. Breast Cancer Res Treat 2007;103:85–91. http://dx.doi.org/10.1007/s10549-006-9355-0. [25] Davis JT, Brill YM, Simmons S, Sachleben BC, Cibull ML, McGrath P, et al. Ultrasound-guided fine-needle aspiration of clinically negative lymph nodes versus sentinel node mapping in patients at high risk for axillary metastasis. Ann Surg Oncol 2006;13:1545–52. http://dx.doi.org/10.1245/s10434-0069095-8. [26] Altomare V, Guerriero G, Carino R, Battista C, Primavera A, Altomare A, et al. Axillary lymph node echo-guided fine-needle aspiration cytology enables breast cancer patients to avoid a sentinel lymph node biopsy. Preliminary experience and a review of the literature. Surg Today 2007;37:735–9. http:// dx.doi.org/10.1007/s00595-006-3366-7. [27] Houssami N, Ciatto S, Turner RM, Cody 3rd HS, Macaskill P. Preoperative ultrasound-guided needle biopsy of axillary nodes in invasive breast cancer: meta-analysis of its accuracy and utility in staging the axilla. Ann Surg 2011;254:243–51. http://dx.doi.org/10.1097/SLA.0b013e31821f1564. [28] Abe H, Schmidt RA, Kulkarni K, Sennett CA, Mueller JS, Newstead GM. Axillary lymph nodes suspicious for breast cancer metastasis: sampling with USguided 14-gauge core-needle biopsy – clinical experience in 100 patients. Radiology 2009;250:41–9. http://dx.doi.org/10.1148/radiol.2493071483. [29] Diepstraten SCE, Sever AR, Buckens CFM, Veldhuis WB, van Dalen T, van den Bosch MAAJ, et al. Value of preoperative ultrasound-guided axillary lymph node biopsy for preventing completion axillary lymph node dissection in breast cancer: a systematic review and meta-analysis. Ann Surg Oncol 2014;21:51–9. http://dx.doi.org/10.1245/s10434-013-3229-6. [30] Alkuwari E, Auger M. Accuracy of fine-needle aspiration cytology of axillary lymph nodes in breast cancer patients: a study of 115 cases with cytologichistologic correlation. Cancer 2008;114:89–93. http://dx.doi.org/10.1002/ cncr.23344.