- Email: [email protected]
Ultrasound-Guided Fine-Needle Aspiration of Non-palpable and Suspicious Axillary Lymph Nodes with Subsequent Removal after Tattooing: False-Negative Results and Concordance with Sentinel Lymph Nodes
Ultrasound in Med. & Biol., Vol. -, No. -, pp. 1–6, 2017 Ó 2017 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$ - see front matter
http://dx.doi.org/10.1016/j.ultrasmedbio.2017.07.011
d
Original Contribution ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION OF NON-PALPABLE AND SUSPICIOUS AXILLARY LYMPH NODES WITH SUBSEQUENT REMOVAL AFTER TATTOOING: FALSE-NEGATIVE RESULTS AND CONCORDANCE WITH SENTINEL LYMPH NODES WON HWA KIM,* HYE JUNG KIM,* JIN HYANG JUNG,y HO YONG PARK,y JEEYEON LEE,y WAN WOOK KIM,y JI YOUNG PARK,z HYEJIN CHEON,* SO MI LEE,* SEUNG HYUN CHO,* KYUNG MIN SHIN,* and GAB CHUL KIM* * Department of Radiology, Kyungpook National University Medical Center, Daegu, Korea; y Department of Surgery, Kyungpook National University Medical Center, Daegu, Korea; and z Department of Pathology, Kyungpook National University Medical Center, Daegu, Korea (Received 10 May 2017; revised 21 June 2017; in final form 14 July 2017)
Abstract—Ultrasonography-guided fine-needle aspiration (US-guided FNA) for axillary lymph nodes (ALNs) is currently used with various techniques for the initial staging of breast cancer and tagging of ALNs. With the implementation of the tattooing of biopsied ALNs, the rate of false-negative results of US-guided FNA for non-palpable and suspicious ALNs and concordance with sentinel lymph nodes were determined by node-to node analyses. A total of 61 patients with breast cancer had negative results for metastasis on US-guided FNA of their non-palpable and suspicious ALNs. The biopsied ALNs were tattooed with an injection of 1–3 mL Charcotrace (Phebra, Lane Cove West, Australia) ink and removed during sentinel lymph node biopsy or axillary dissection. We determined the rate of false-negative results and concordance with the sentinel lymph nodes by a retrospective review of surgical and pathologic findings. The association of false-negative results with clinical and imaging factors was evaluated using logistic regression. Of the 61 ALNs with negative results for USguided FNA, 13 (21%) had metastases on final pathology. In 56 of 61 ALNs (92%), tattooed ALNs corresponded to the sentinel lymph nodes. Among the 5 patients (8%) without correspondence, 1 patient (2%) had 2 metastatic ALNs of 1 tattooed node and 1 sentinel lymph node. In multivariate analysis, atypical cells on FNA results (odds ratio 5 20.7, p 5 0.040) was independently associated with false-negative FNA results. False-negative ALNs after US-guided FNA occur at a rate of 21% and most of the tattooed ALNs showed concordance with sentinel lymph nodes. (E-mail: [email protected]) Ó 2017 World Federation for Ultrasound in Medicine & Biology. Key Words: Fine-needle aspiration, Axillary lymph node, False-negative results, Breast carcinoma, Tattooing.
(Lucci et al. 2007; Veronesi et al. 1997). For patients expected to have a high risk of advanced nodal disease, a two-stage surgical procedure of SLNB with subsequent ALND is associated with an increased cost of lymphoscintigraphy and surgery time. SLNB is also an invasive procedure with several complications, which have been documented through large prospective trials (Lucci et al. 2007; Wilke et al. 2006). Thus, the preliminary staging of the axilla before surgery to help decide whether to perform SLNB has been of interest to date (Hyun et al. 2016; Tucker et al. 2016). For preliminary staging of the axilla, physical examination alone is neither sensitive nor specific, because metastatic ALNs are often non-palpable, and palpable ALNs can be reactive (de Freitas et al. 1991; Pamilo
INTRODUCTION Surgical staging of the axilla is crucial in patients with breast cancer, because the presence of metastatic axillary lymph nodes (ALNs) is a key prognostic and decisive factor to tailor the treatment (Krag et al. 1998). Sentinel lymph node biopsy (SLNB) has become a standard method of axillary staging in patients with clinically node-negative breast cancer, because SLNB achieves similar clinical outcomes with a lower morbidity compared with axillary lymph node dissection (ALND)
Address correspondence to: Hye Jung Kim, Department of Radiology, Kyungpook National University Medical Center, 807 Hoguk-ro, Buk-gu, Daegu 41404, Republic of Korea. E-mail: [email protected] 1
2
Ultrasound in Medicine and Biology
et al. 1989; Sacre 1986). Ultrasonography (US) of the axilla has been increasingly performed to overcome the limitation of physical examination. In a systematic review, US for the diagnosis of non-palpable axillary metastases is moderately sensitive (49%–87%) and fairly specific (56%–97%) (Alvarez et al. 2006). Nonetheless, because the signs of US for metastatic ALNs overlap with those of benign reactive change, US-guided fineneedle aspiration (FNA) is often added (Cho et al. 2009; Ciatto et al. 2007). However, although the addition of US-guided FNA increases the specificity, the sensitivity might be lowered by 10%–30% as a result of the false-negatives from the US-guided FNA (Baruah et al. 2010; Park et al. 2011). Furthermore, SLNB is not a perfect method to evaluate the status of the axilla, and the rates of false-negative SLNB have varied at a range of 0%–19% in previous studies (Kelley et al. 2004; Martin et al. 2005; Veronesi et al. 1997). Therefore, to compensate for the false-negatives from US-guided FNA and SLNB, suspicious ALNs with negative results for metastasis on US-guided FNA were localized through pre-operative tattooing and consecutively removed along with SLNB. This procedure allowed node-to-node analysis in which the suspicious ALNs on US were exactly matched to the ALNs with final pathologic results, in contrast to previous studies that have demonstrated the diagnostic performance of US or US-guided FNA based on a patient-based analysis. In addition, the concordance of suspicious ALNs on US to the sentinel nodes has rarely been evaluated. In this study, we performed node-to-node analyses in the two aspects as follows: (i) correlation of the histologic results on US-guided FNA to the final pathologic results for a suspicious node identified at US, and (ii) correlation of the biopsied node with the sentinel node. Therefore, the purpose of the present study was to determine the rate of false-negative results of USguided FNA for suspicious ALNs and to evaluate the concordance of suspicious ALNs with sentinel lymph nodes by node-to-node analyses. MATERIALS AND METHODS Patients The institutional review board of Kyungpook National University Medical Center approved this retrospective review of data, and the requirement of an informed consent was waived. A total of 67 patients with breast cancer had negative results for metastasis on US-guided FNA for their non-palpable and suspicious ALNs June 2013–October 2016. Nodes were defined as suspicious ALNs when 1 of the following characteristics was observed, on the basis of previous studies: cortical thickness .2 mm, eccentric cortical thickening, loss of fatty
Volume -, Number -, 2017
hilum and round shape (Cho et al. 2009; Park et al. 2013; You et al. 2015). Of the 67 patients, we excluded those with excisional breast biopsy (n 5 1) before USguided FNA and those who underwent neoadjuvant chemotherapy (n 5 5). A total of 61 patients with 61 ALNs were included in this study. US-guided FNA and tattooing FNA and tattooing were performed by 1 of 3 attending radiologists who have 5, 10 and 18 y of experience in breast imaging and intervention using real-time visualization with US. For FNA, a 23-gauge needle was inserted into the cortex of the ALN using a to-and-fro method with manual aspiration. Tattooing of the aspirated ALNs was performed through the injection of 1–3 mL of Charcotrace black ink (Phebra, Lane Cove West, Australia) into the cortex of ALNs and the adjacent soft tissue after local anesthesia (Fig. 1). Tattooing was generally performed on the same day or 1 d before surgery by the same radiologist who performed the FNA. The US-guided tattooing took approximately 5 min per patient. The radiologist marked the location of the ALNs on the skin with an oil-based pen to guide the surgical incision. All patients underwent FNA of the most suspicious-appearing ALN with subsequent tattooing. SLNB and intra-operative evaluation To identify sentinel lymph nodes, a dual method of radioisotopes and blue dye was used (n 5 56). However, in 5 patients, only a radioisotope (n 5 4) or blue dye (n 5 1) was used because of the unavailability of the agent or lymphoscintigraphy. The radioisotope (technetium-99 m phytate) or blue dye was injected before or at the time of surgery. ALNs identified by a gamma probe or ALNs containing blue dye were regarded as sentinel nodes and removed. All tattooed ALNs were identified and subsequently removed through inspection of the axilla during the SLNB. We reviewed the surgical records for the correspondence between the tattooed ALNs and the sentinel lymph nodes. All sentinel lymph nodes and tattooed ALNs were submitted for intra-operative frozen sections. If 3 or more ALNs were positive for metastasis on the frozen section, then axillary lymph node dissection was performed immediately after SLNB. Three attending breast surgeons and trainees performed all the operative procedures. Pathologic evaluation Acquired cells obtained from FNA were stained according to the classic Papanicolaou method. The cytologic results were classified into inadequate sampling, benign, atypical, suspicious for malignancy or definitely malignant (Goyal et al. 2013). A negative result for
False-negative results of fine-needle aspiration for axillary lymph nodes d W. H. KIM et al.
3
Fig. 1. A 45-y-old woman diagnosed with left breast cancer underwent tattooing for her non-palpable and suspicious ALN with the result of atypical cells on FNA. This ALN was removed during the axillary surgery and confirmed as metastasis at final pathology. Ultrasonographic image shows an ALN (arrow) before (a) and after (b) injection of charcoal-based ink. The echogenic ink material is identified in and around (arrowheads) the cortex of the ALN. ALN 5 axillary lymph node; FNA 5 fine-needle aspiration.
metastasis was defined as benign or atypical, and a positive result for metastasis was suspicious for malignancy or definitely malignant. For intra-operative frozen sections, the ALN was bisected and a single 5-mm-thick section was stained with hematoxylin and eosin (H&E). The ALNs after the frozen section were fixed in formalin, embedded in paraffin and sectioned for routine H&E staining. Each ALN was finally classified as negative or positive for metastasis and the numbers of ALNs sampled and with metastasis were recorded. ALNs with micrometastasis were classified as positive for metastasis, and ALNs with isolated tumor cells were classified as negative for metastasis. All cytologic and pathologic evaluation was performed by a pathologist with 18 y of experience in breast pathology. Data collection and statistical analysis The clinical data collected included age at cancer diagnosis, clinical manifestation (symptomatic vs. asymptomatic) and menopausal status. The following histopathological information was included in the study: FNA results; histologic tumor characteristics; tumor size; tumor grade; estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor (HER2) status; and lymphovascular invasion. A false-negative result was defined as a pathologically proven metastatic ALN in which US-guided FNA yielded negative results including benign or atypical. The factors associated with false-negative ALNs were evaluated using the c2 test and a t-test as appropriate. The odds ratio (OR) and 95% confidence interval (CI) for false-negative ALNs were calculated with a univariate logistic regression analysis, and variables with a value of p , 0.2 were selected for the final multivariate model. All statistical analyses were performed with statistical
software (SPSS software, v. 24.0; Chicago, IL, USA), and p , 0.05 was considered to indicate a statistically significant difference. RESULTS The mean age of the patients was 49 y (range, 31– 71 y). The median tumor size of 56 invasive cancers was 21 mm (range, 1–55 mm), and the median size of 5 in situ cancers was 34 mm (range, 10–60 mm). On cytologic report of FNA, no patients had inadequate sampling, 6 patients had atypical cells and the remaining 55 patients had benign results. The median time interval between FNA and surgery was 15 d (ranges, 3–40 d). The median number of retrieved sentinel lymph nodes was 2 (range, 1–6). No adverse reactions were associated with the tattooing procedure. Intra-operatively, all the tattooed ALNs were identified easily with visual inspection, and they were differentiated from the ALNs containing blue dye by verification of the black ink in the perinodal soft tissue. Regarding the node-to-node analysis, among the 61 suspicious ALNs with negative results on US-guided FNA, 12 ALNs (20%; 95% CI, 10%–34%) had metastases in the frozen sections. One ALN that was negative for metastasis on the frozen section had micrometastasis on the final pathology. Ultimately, 13 of 61 ALNs had false-negative results (21%; 95% CI, 11%–36%). Two ALNs that were tattooed nodes had isolated tumor cells. With respect to the patient-based analysis, 14 of 61 patients had false-negative results (23%; 95% CI, 13%– 39%). One patient (2%; 95% CI, 0%–9%) in whom the tattooed ALN was negative for metastasis had 2 metastatic ALNs of 1 sentinel and 1 non-sentinel lymph node. Among the 14 patients, 12 patients (20%; 95% CI, 6%– 21%) had a single (n 5 8) or 2 (n 5 4) metastatic ALNs (pN1), and the remaining 2 patients (3%; 95%
4
Ultrasound in Medicine and Biology
CI, 0.2%–7%) had 12 metastatic ALNs each (pN3). Of note, patients with advanced nodal disease (pN3) had atypical cells on their FNA results. In 56 of 61 patients (92%; 95% CI, 69%–119%), there was a correspondence between the tattooed ALNs and sentinel lymph nodes. Among the 5 patients (8%; 95% CI, 3%–19%) in whom the correspondence was not made, 1 patient (2%; 95% CI, 0%–9%) had 2 metastatic ALNs of 1 tattooed node and 1 sentinel lymph node. In another patient (2%; 95% CI, 0%–9%), the tattooed ALN was negative for metastasis; however, 2 metastatic ALNs of sentinel and non-sentinel lymph nodes were finally confirmed. The remaining 3 patients had no metastatic ALNs in either tattooed or sentinel lymph nodes. The imaging characteristics of the included ALNs are shown in Table 1. No significant differences were found in the imaging characteristics of ALNs. We observed a higher proportion of the presence of eccentric cortical thickening in false-negative ALNs than in true-negative ALNs, but this failed to reach statistical significance (p 5 0.192). In terms of patients and tumor characteristics (Table 2), we found significantly higher proportions of false-negative ALNs in patients who were asymptomatic (p 5 0.032) and had atypical cells on FNA results (p 5 0.005), in comparisons with those who were symptomatic and had no atypical cells on FNA results. In multivariate analysis, atypical cells on FNA results (OR 5 20.7 [95% CI, 1.1–373.7], p 5 0.040) was independently associated with false-negative FNA results. DISCUSSION The sensitivity or specificity of US-guided FNA for preliminary staging of ALNs has been evaluated in multiple studies. However, those studies could not precisely Table 1. Imaging characteristics of true-negative and false-negative ALNs on US-guided FNA* Imaging characteristics Cortical thickness #2 mm .2 mm Loss of fatty hilum Absence Presence Eccentric cortical thickening Absence Presence Shape Oval (L/S ratio $1.5) Round (L/S ratio ,1.5)
True-negative False-negative ALNs (n 5 48) ALNs (n 5 13) p value 0.343 9 (18.8) 39 (81.2)
1 (7.7) 12 (92.3)
44 (91.7) 4 (8.3)
12 (92.3) 1 (7.7)
35 (72.9) 13 (27.1)
7 (53.8) 6 (46.2)
40 (83.3) 8 (16.7)
11 (84.6) 2 (15.4)
0.941 0.192 0.913
ALNs 5 axillary lymph nodes; L/S 5 ratio of long- to short-axis diameter; US-guided FNA 5 ultrasonography-guided fine-needle aspiration. * Data are numbers of ALNs, with percentages in parentheses.
Volume -, Number -, 2017
Table 2. Patients and primary tumor characteristics of true-negative and false-negative axillary ALNs on US-guided FNA* Clinical and tumor characteristics Age (y) Clinical manifestation Symptomatic Asymptomatic Menopausal status Premenopausal Postmenopausal Atypical cell on FNA result Absence Presence Histologic tumor characteristics Ductal Lobular Othersy Tumor size (mm) Tumor gradez Low Intermediate High Hormone receptor status Negative Positive HER2 status Negative Positive Lymphovascular invasion Absence Presence
True-negative False-negative ALNs (n 5 48) ALNs (n 5 13) p value 48.1
53.0
37 (77.1) 11 (22.9)
6 (46.2) 7 (53.8)
30 (62.5) 18 (37.5)
5 (38.5) 8 (61.5)
46 (95.8) 2 (4.2)
9 (69.2) 4 (30.8)
0.360 0.032 0.123 0.005 0.447
38 (79.2) 5 (10.4) 5 (10.4) 22.1 (12.4)
11 (84.6) 0 (0.0) 2 (15.4) 23.0 (8.1)
6 (14.3) 10 (23.8) 26 (61.9)
1 (7.7) 6 (46.2) 6 (46.2)
25 (52.1) 23 (47.9)
4 (30.8) 9 (69.2)
35 (72.9) 13 (27.1)
9 (69.2) 4 (30.8)
30 (69.8) 13 (30.2)
6 (46.2) 7 (53.8)
0.795 0.291 0.176 0.794 0.123
ALNs 5 axillary lymph nodes; US-guided FNA 5 ultrasonographyguided fine needle aspiration. * Data are numbers of patients, with percentages in parentheses. HER2 5 human epidermal growth factor receptor 2. y Others include mucinous (n 5 4), metaplastic (n 5 1), and encapsulated papillary carcinoma (n 5 1). z Limited to patients (n 5 55) with available tumor grade results.
determine false-negative results because node-to-node analysis was unavailable. Furthermore, it remains to be answered whether SLNB is sufficient without the excision of biopsied but suspicious ALNs with negative results on FNA, because the sensitivity of US might be lowered by adding FNA. In our study, even with a negative FNA result, the probability of metastasis (1-negative predictive value [NPV]) was still 21% (13/61) by nodeto-node analysis and 23% (14/61) by patient-based analysis. This rate is comparable or slightly lower than the range of rate (28%–66%) reported by previous studies involving NPV (Alkuwari and Auger 2008; Ciatto et al. 2007; Kuenen-Boumeester et al. 2003; Park et al. 2013). The wide range of the NPV is most likely dependent on the following: palpable versus nonpalpable ALNs, morphologic criteria for determination of suspicious ALNs, radiologists versus clinicians performing FNA, characteristics of primary tumor and method of guidance (US guidance vs. with palpation only). This, together with our results, suggests that
False-negative results of fine-needle aspiration for axillary lymph nodes d W. H. KIM et al.
surgical staging of the axilla by SLNB or excision of biopsied ALNs remains essential. One of the major strengths of the present study is that the correlation between suspicious ALNs on US and sentinel lymph nodes was provided in the context of the feasibility of targeted ALN excision after tattooing. In our study, all the tattooed ALNs were visualized in the operation field and 92% (56/61) of the tattooed ALNs corresponded to sentinel lymph nodes. Furthermore, 1 tattooed ALN, which was not a sentinel lymph node, showed metastasis. To date, the result of pre-operative tattooing of ALNs has been shown in only 1 study with a limited population, and that study revealed a 96% correspondence rate between the tattooed ALNs and sentinel lymph nodes (Choy et al. 2015). Targeted ALN excision using other techniques, such as use of a metallic marker and radioactive seeds, has been described recently (Caudle et al. 2016; Shin et al. 2016). In comparison with these techniques, the tattoo technique has the advantages of a lower cost and higher convenience without the use of additional radioactive materials or a localization procedure to ascertain the location of the marker. The reason that most of the suspicious ALNs on US were the sentinel lymph nodes with a higher concordance rate is unclear. Sentinel lymph nodes are hypothetical lymph nodes or groups of nodes draining a cancer. Metastatic cells traveling in the afferent lymphatic channels are generally embedded in the marginal sinus at the cortex of a lymph node; thus, US classification of ALNs is based on their cortical thickness (Bedi et al. 2008). Hence, it appears to be a natural assumption that the most suspicious ALNs having the most thickened cortices are thus sentinel lymph nodes in many cases, even with benign ALNs. In the present study, we found that the atypical cells on FNA were associated with false-negative FNA results. Park et al. (2013) described that positive hormone receptor status and lymphovascular invasion (LVI) were associated with false-negative ALNs in US and US-guided FNA in their univariate analysis. In their multivariate analysis, high T-stage and LVI were associated with false-negative ALNs. In our study, positive hormone receptor status and LVI were also more frequently found in false-negative ALNs than in true-negative ALNs; however, those did not reach a statistical significance, probably because our selected population had negative FNA results whereas Park et al. (2013) included patients without FNA results. Furthermore, our findings indicated that the FNA result of atypical cells is a strong predictor of falsenegative results; atypical cells on FNA had an OR of 20.7 for metastasis in the multivariate analysis. Among the 6 ALNs with atypical cells on FNA, 4 ALNs (67%)
5
had metastasis in the final pathology, including 2 metastatic ALNs (33%) accompanied by advanced nodal disease. These findings are in line with results of previous studies showing that cytologic atypia is a high-risk feature for malignancy in various organs (Kelman et al. 2001; Olson et al. 2011; Pitman et al. 2013; Widra et al. 1995). Doubts may exist about the classification of the ALNs with atypical cells (n 5 6) on FNA as negative results for metastasis. In our institution, patients who have ALNs with atypical cells undergo SLNB with the tattooing technique because we believe a chance exists that they will have final negative pathologic results; in these cases, ALND is unnecessary. Indeed, about one third of the patients who had ALNs with atypical cells have negative pathologic results after surgery. Even if the negative results of ALNs were confined to only the benign results on US-guided FNA, the rate of falsenegative ALNs (20%) was similar to the results (21%) shown in our current analysis. Although we observed a considerable number of false-negative results for US-guided FNA, US-guided FNA is useful to guide axillary surgery or to tailor systemic treatment. US-guided FNA allows us to find a subgroup eligible for one-stage axillary surgery. If a patient has a positive FNA result for metastasis, direct axillary dissection can be performed without SLNB. In addition, if a patient were proven to have a metastatic ALN at initial staging, the patient would have the opportunity to have neoadjuvant chemotherapy before curative surgery. Our study has several limitations. First, the determination of suspicious ALNs on US during the preliminary staging of the axilla was subjective, potential variability among radiologists can occur, and inter-reader agreement was not calculated. US-guided FNA is an operatordependent procedure for obtaining the desired tissue. Second, because we focused on a node-to-node analysis of the suspicious ALNs with negative FNA results, the overall sensitivity and specificity of US-guided FNA could not be determined. The node-to-node analysis was unfeasible for patients who had positive results for metastasis on US-guided FNA because they usually undergo neoadjuvant chemotherapy or ALND without the tattooing procedure. Third, although the inadequate sampling of cytologic reports of US-guided FNA causes higher false-negative rates, the ALNs included in the present study did not have the results of inadequate sampling. In our institution, the rate of inadequate sampling is low, at approximately 2.6%. This might be in part a result of the following at our institution: relatively slim patients without abundant axillary fat, radiologists experienced in breast imaging and an FNA protocol with obligatory acquisition of at least two samples for each ALN. Finally, this study is limited to a single center in Korea. The characteristics of Asian patients can be
6
Ultrasound in Medicine and Biology
different from those from other countries (e.g., younger age, lower body mass index and different genetics) (Leong et al. 2010). In conclusion, the present study confirms that the false-negative ALNs after US-guided FNA occur at a rate of 21% and are associated with atypical cells on FNA results. Non-palpable and suspicious ALNs with negative FNA results can be excised after tattooing with an injection of the charcoal-based ink, and most of them corresponded to the sentinel lymph nodes. A technique of tattooing is a useful tool for the tagging of ALNs and thereby may potentially enable targeted ALN excision. Further investigations in a large number of patients are warranted. Acknowledgments—This work was supported by Biomedical Research Institute grant, Kyungpook National University Hospital (2017).
REFERENCES Alkuwari E, Auger M. Accuracy of fine-needle aspiration cytology of axillary lymph nodes in breast cancer patients: A study of 115 cases with cytologic-histologic correlation. Cancer 2008;114:89–93. Alvarez S, Anorbe E, Alcorta P, Lopez F, Alonso I, Cortes J. Role of sonography in the diagnosis of axillary lymph node metastases in breast cancer: A systematic review. AJR Am J Roentgenol 2006; 186:1342–1348. 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;97:680–683. Bedi DG, Krishnamurthy R, Krishnamurthy S, Edeiken BS, Le-Petross H, Fornage BD, Bassett RL Jr, Hunt KK. Cortical morphologic features of axillary lymph nodes as a predictor of metastasis in breast cancer: In vitro sonographic study. AJR Am J Roentgenol 2008;191:646–652. Caudle AS, Yang WT, Krishnamurthy S, Mittendorf EA, Black DM, Gilcrease MZ, Bedrosian I, Hobbs BP, DeSnyder SM, Hwang RF, Adrada BE, Shaitelman SF, Chavez-MacGregor M, Smith BD, Candelaria RP, Babiera GV, Dogan BE, Santiago L, Hunt KK, Kuerer HM. Improved axillary evaluation following neoadjuvant therapy for patients with node-positive breast cancer using selective evaluation of clipped nodes: Implementation of targeted axillary dissection. J Clin Oncol 2016;34:1072–1078. Cho N, Moon WK, Han W, Park IA, Cho J, Noh DY. Preoperative sonographic classification of axillary lymph nodes in patients with breast cancer: Node-to-node correlation with surgical histology and sentinel node biopsy results. AJR Am J Roentgenol 2009;193:1731–1737. Choy N, Lipson J, Porter C, Ozawa M, Kieryn A, Pal S, Kao J, Trinh L, Wheeler A, Ikeda D, Jensen K, Allison K, Wapnir I. Initial results with preoperative tattooing of biopsied axillary lymph nodes and correlation to sentinel lymph nodes in breast cancer patients. Ann Surg Oncol 2015;22:377–382. Ciatto S, Brancato B, Risso G, Ambrogetti D, Bulgaresi P, Maddau C, Turco P, Houssami N. 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. de Freitas R Jr, Costa MV, Schneider SV, Nicolau MA, Marussi E. Accuracy of ultrasound and clinical examination in the diagnosis of axillary lymph node metastases in breast cancer. Eur J Surg Oncol 1991;17:240–244. Goyal P, Sehgal S, Ghosh S, Aggarwal D, Shukla P, Kumar A, Gupta R, Singh S. Histopathological correlation of atypical (c3) and suspicious (c4) categories in fine needle aspiration cytology of the breast. Int J Breast Cancer 2013;2013:965498. Hyun SJ, Kim EK, Moon HJ, Yoon JH, Kim MJ. Preoperative axillary lymph node evaluation in breast cancer patients by breast magnetic resonance imaging (MRI): Can breast MRI exclude advanced nodal disease? Eur Radiol 2016;26:3865–3873.
Volume -, Number -, 2017 Kelley MC, Hansen N, McMasters KM. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Am J Surg 2004;188:49–61. Kelman AS, Rathan A, Leibowitz J, Burstein DE, Haber RS. Thyroid cytology and the risk of malignancy in thyroid nodules: Importance of nuclear atypia in indeterminate specimens. Thyroid 2001;11:271–277. Krag D, Weaver D, Ashikaga T, Moffat F, Klimberg VS, Shriver C, Feldman S, Kusminsky R, Gadd M, Kuhn J, Harlow S, Beitsch P. The sentinel node in breast cancer—A multicenter validation study. N Engl J Med 1998;339:941–946. Kuenen-Boumeester V, Menke-Pluymers M, de Kanter AY, Obdeijn IM, Urich D, Van Der Kwast TH. Ultrasound-guided fine needle aspiration cytology of axillary lymph nodes in breast cancer patients. A preoperative staging procedure. Eur J Cancer 2003;39:170–174. Leong SP, Shen ZZ, Liu TJ, Agarwal G, Tajima T, Paik NS, Sandelin K, Derossis A, Cody H, Foulkes WD. Is breast cancer the same disease in Asian and Western countries? World J Surg 2010;34:2308–2324. Lucci A, McCall LM, Beitsch PD, Whitworth PW, Reintgen DS, Blumencranz PW, Leitch AM, Saha S, Hunt KK, Giuliano AE, American College of Surgeons Oncology Group. Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J Clin Oncol 2007;25:3657–3663. Martin RC 2nd, Chagpar A, Scoggins CR, Edwards MJ, Hagendoorn L, Stromberg AJ, McMasters KM, University of Louisville Breast Cancer Sentinel Lymph Node Study. Clinicopathologic factors associated with false-negative sentinel lymph-node biopsy in breast cancer. Ann Surg 2005;241:1005–1012. discussion 1012–1015. Olson MT, Clark DP, Erozan YS, Ali SZ. Spectrum of risk of malignancy in subcategories of ‘atypia of undetermined significance.’. Acta Cytlo 2011;55:518–525. 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;8:115–120. Park SH, Kim EK, Park BW, Kim SI, Moon HJ, Kim MJ. False negative results in axillary lymph nodes by ultrasonography and ultrasonography-guided fine-needle aspiration in patients with invasive ductal carcinoma. Ultrashall Med 2013;34:559–567. Park SH, Kim MJ, Park BW, Moon HJ, Kwak JY, Kim EK. Impact of preoperative ultrasonography and fine-needle aspiration of axillary lymph nodes on surgical management of primary breast cancer. Ann Surg Oncol 2011;18:738–744. Pitman MB, Yaeger KA, Brugge WR, Mino-Kenudson M. Prospective analysis of atypical epithelial cells as a high-risk cytologic feature for malignancy in pancreatic cysts. Cancer Cytopathol 2013;121:29–36. Sacre RA. Clinical evaluation of axillar lymph nodes compared to surgical and pathological findings. Eur J Surg Oncol 1986;12:169–173. Shin K, Caudle AS, Kuerer HM, Santiago L, Candelaria RP, Dogan B, Leung J, Krishnamurthy S, Yang W. Radiologic mapping for targeted axillary dissection: Needle biopsy to excision. AJR Am J Roentgenol 2016;207:1372–1379. Tucker NS, Cyr AE, Ademuyiwa FO, Tabchy A, George K, Sharma PK, Jin LX, Sanati S, Aft R, Gao F, Margenthaler JA, Gillanders WE. Axillary ultrasound accurately excludes clinically significant lymph node disease in patients with early stage breast cancer. Ann Surg 2016;264:1098–1102. Veronesi U, Paganelli G, Galimberti V, Viale G, Zurrida S, Bedoni M, Costa A, de Cicco C, Geraghty JG, Luini A, Sacchini V, Veronesi P. Sentinel-node biopsy to avoid axillary dissection in breast cancer with clinically negative lymph-nodes. Lancet 1997;349:1864–1867. Widra EA, Dunton CJ, McHugh M, Palazzo JP. Endometrial hyperplasia and the risk of carcinoma. Int J Gynecol Cancer 1995;5:233–235. Wilke LG, McCall LM, Posther KE, Whitworth PW, Reintgen DS, Leitch AM, Gabram SG, Lucci A, Cox CE, Hunt KK, Herndon JE 2nd, Giuliano AE. Surgical complications associated with sentinel lymph node biopsy: Results from a prospective international cooperative group trial. Ann Surg Oncol 2006;13:491–500. You S, Kang DK, Jung YS, An YS, Jeon GS, Kim TH. Evaluation of lymph node status after neoadjuvant chemotherapy in breast cancer patients: Comparison of diagnostic performance of ultrasound, MRI and 18F-FDG PET/CT. Br J Radiol 2015;88:20150143.
http://dx.doi.org/10.1016/j.ultrasmedbio.2017.07.011
d
Original Contribution ULTRASOUND-GUIDED FINE-NEEDLE ASPIRATION OF NON-PALPABLE AND SUSPICIOUS AXILLARY LYMPH NODES WITH SUBSEQUENT REMOVAL AFTER TATTOOING: FALSE-NEGATIVE RESULTS AND CONCORDANCE WITH SENTINEL LYMPH NODES WON HWA KIM,* HYE JUNG KIM,* JIN HYANG JUNG,y HO YONG PARK,y JEEYEON LEE,y WAN WOOK KIM,y JI YOUNG PARK,z HYEJIN CHEON,* SO MI LEE,* SEUNG HYUN CHO,* KYUNG MIN SHIN,* and GAB CHUL KIM* * Department of Radiology, Kyungpook National University Medical Center, Daegu, Korea; y Department of Surgery, Kyungpook National University Medical Center, Daegu, Korea; and z Department of Pathology, Kyungpook National University Medical Center, Daegu, Korea (Received 10 May 2017; revised 21 June 2017; in final form 14 July 2017)
Abstract—Ultrasonography-guided fine-needle aspiration (US-guided FNA) for axillary lymph nodes (ALNs) is currently used with various techniques for the initial staging of breast cancer and tagging of ALNs. With the implementation of the tattooing of biopsied ALNs, the rate of false-negative results of US-guided FNA for non-palpable and suspicious ALNs and concordance with sentinel lymph nodes were determined by node-to node analyses. A total of 61 patients with breast cancer had negative results for metastasis on US-guided FNA of their non-palpable and suspicious ALNs. The biopsied ALNs were tattooed with an injection of 1–3 mL Charcotrace (Phebra, Lane Cove West, Australia) ink and removed during sentinel lymph node biopsy or axillary dissection. We determined the rate of false-negative results and concordance with the sentinel lymph nodes by a retrospective review of surgical and pathologic findings. The association of false-negative results with clinical and imaging factors was evaluated using logistic regression. Of the 61 ALNs with negative results for USguided FNA, 13 (21%) had metastases on final pathology. In 56 of 61 ALNs (92%), tattooed ALNs corresponded to the sentinel lymph nodes. Among the 5 patients (8%) without correspondence, 1 patient (2%) had 2 metastatic ALNs of 1 tattooed node and 1 sentinel lymph node. In multivariate analysis, atypical cells on FNA results (odds ratio 5 20.7, p 5 0.040) was independently associated with false-negative FNA results. False-negative ALNs after US-guided FNA occur at a rate of 21% and most of the tattooed ALNs showed concordance with sentinel lymph nodes. (E-mail: [email protected]) Ó 2017 World Federation for Ultrasound in Medicine & Biology. Key Words: Fine-needle aspiration, Axillary lymph node, False-negative results, Breast carcinoma, Tattooing.
(Lucci et al. 2007; Veronesi et al. 1997). For patients expected to have a high risk of advanced nodal disease, a two-stage surgical procedure of SLNB with subsequent ALND is associated with an increased cost of lymphoscintigraphy and surgery time. SLNB is also an invasive procedure with several complications, which have been documented through large prospective trials (Lucci et al. 2007; Wilke et al. 2006). Thus, the preliminary staging of the axilla before surgery to help decide whether to perform SLNB has been of interest to date (Hyun et al. 2016; Tucker et al. 2016). For preliminary staging of the axilla, physical examination alone is neither sensitive nor specific, because metastatic ALNs are often non-palpable, and palpable ALNs can be reactive (de Freitas et al. 1991; Pamilo
INTRODUCTION Surgical staging of the axilla is crucial in patients with breast cancer, because the presence of metastatic axillary lymph nodes (ALNs) is a key prognostic and decisive factor to tailor the treatment (Krag et al. 1998). Sentinel lymph node biopsy (SLNB) has become a standard method of axillary staging in patients with clinically node-negative breast cancer, because SLNB achieves similar clinical outcomes with a lower morbidity compared with axillary lymph node dissection (ALND)
Address correspondence to: Hye Jung Kim, Department of Radiology, Kyungpook National University Medical Center, 807 Hoguk-ro, Buk-gu, Daegu 41404, Republic of Korea. E-mail: [email protected] 1
2
Ultrasound in Medicine and Biology
et al. 1989; Sacre 1986). Ultrasonography (US) of the axilla has been increasingly performed to overcome the limitation of physical examination. In a systematic review, US for the diagnosis of non-palpable axillary metastases is moderately sensitive (49%–87%) and fairly specific (56%–97%) (Alvarez et al. 2006). Nonetheless, because the signs of US for metastatic ALNs overlap with those of benign reactive change, US-guided fineneedle aspiration (FNA) is often added (Cho et al. 2009; Ciatto et al. 2007). However, although the addition of US-guided FNA increases the specificity, the sensitivity might be lowered by 10%–30% as a result of the false-negatives from the US-guided FNA (Baruah et al. 2010; Park et al. 2011). Furthermore, SLNB is not a perfect method to evaluate the status of the axilla, and the rates of false-negative SLNB have varied at a range of 0%–19% in previous studies (Kelley et al. 2004; Martin et al. 2005; Veronesi et al. 1997). Therefore, to compensate for the false-negatives from US-guided FNA and SLNB, suspicious ALNs with negative results for metastasis on US-guided FNA were localized through pre-operative tattooing and consecutively removed along with SLNB. This procedure allowed node-to-node analysis in which the suspicious ALNs on US were exactly matched to the ALNs with final pathologic results, in contrast to previous studies that have demonstrated the diagnostic performance of US or US-guided FNA based on a patient-based analysis. In addition, the concordance of suspicious ALNs on US to the sentinel nodes has rarely been evaluated. In this study, we performed node-to-node analyses in the two aspects as follows: (i) correlation of the histologic results on US-guided FNA to the final pathologic results for a suspicious node identified at US, and (ii) correlation of the biopsied node with the sentinel node. Therefore, the purpose of the present study was to determine the rate of false-negative results of USguided FNA for suspicious ALNs and to evaluate the concordance of suspicious ALNs with sentinel lymph nodes by node-to-node analyses. MATERIALS AND METHODS Patients The institutional review board of Kyungpook National University Medical Center approved this retrospective review of data, and the requirement of an informed consent was waived. A total of 67 patients with breast cancer had negative results for metastasis on US-guided FNA for their non-palpable and suspicious ALNs June 2013–October 2016. Nodes were defined as suspicious ALNs when 1 of the following characteristics was observed, on the basis of previous studies: cortical thickness .2 mm, eccentric cortical thickening, loss of fatty
Volume -, Number -, 2017
hilum and round shape (Cho et al. 2009; Park et al. 2013; You et al. 2015). Of the 67 patients, we excluded those with excisional breast biopsy (n 5 1) before USguided FNA and those who underwent neoadjuvant chemotherapy (n 5 5). A total of 61 patients with 61 ALNs were included in this study. US-guided FNA and tattooing FNA and tattooing were performed by 1 of 3 attending radiologists who have 5, 10 and 18 y of experience in breast imaging and intervention using real-time visualization with US. For FNA, a 23-gauge needle was inserted into the cortex of the ALN using a to-and-fro method with manual aspiration. Tattooing of the aspirated ALNs was performed through the injection of 1–3 mL of Charcotrace black ink (Phebra, Lane Cove West, Australia) into the cortex of ALNs and the adjacent soft tissue after local anesthesia (Fig. 1). Tattooing was generally performed on the same day or 1 d before surgery by the same radiologist who performed the FNA. The US-guided tattooing took approximately 5 min per patient. The radiologist marked the location of the ALNs on the skin with an oil-based pen to guide the surgical incision. All patients underwent FNA of the most suspicious-appearing ALN with subsequent tattooing. SLNB and intra-operative evaluation To identify sentinel lymph nodes, a dual method of radioisotopes and blue dye was used (n 5 56). However, in 5 patients, only a radioisotope (n 5 4) or blue dye (n 5 1) was used because of the unavailability of the agent or lymphoscintigraphy. The radioisotope (technetium-99 m phytate) or blue dye was injected before or at the time of surgery. ALNs identified by a gamma probe or ALNs containing blue dye were regarded as sentinel nodes and removed. All tattooed ALNs were identified and subsequently removed through inspection of the axilla during the SLNB. We reviewed the surgical records for the correspondence between the tattooed ALNs and the sentinel lymph nodes. All sentinel lymph nodes and tattooed ALNs were submitted for intra-operative frozen sections. If 3 or more ALNs were positive for metastasis on the frozen section, then axillary lymph node dissection was performed immediately after SLNB. Three attending breast surgeons and trainees performed all the operative procedures. Pathologic evaluation Acquired cells obtained from FNA were stained according to the classic Papanicolaou method. The cytologic results were classified into inadequate sampling, benign, atypical, suspicious for malignancy or definitely malignant (Goyal et al. 2013). A negative result for
False-negative results of fine-needle aspiration for axillary lymph nodes d W. H. KIM et al.
3
Fig. 1. A 45-y-old woman diagnosed with left breast cancer underwent tattooing for her non-palpable and suspicious ALN with the result of atypical cells on FNA. This ALN was removed during the axillary surgery and confirmed as metastasis at final pathology. Ultrasonographic image shows an ALN (arrow) before (a) and after (b) injection of charcoal-based ink. The echogenic ink material is identified in and around (arrowheads) the cortex of the ALN. ALN 5 axillary lymph node; FNA 5 fine-needle aspiration.
metastasis was defined as benign or atypical, and a positive result for metastasis was suspicious for malignancy or definitely malignant. For intra-operative frozen sections, the ALN was bisected and a single 5-mm-thick section was stained with hematoxylin and eosin (H&E). The ALNs after the frozen section were fixed in formalin, embedded in paraffin and sectioned for routine H&E staining. Each ALN was finally classified as negative or positive for metastasis and the numbers of ALNs sampled and with metastasis were recorded. ALNs with micrometastasis were classified as positive for metastasis, and ALNs with isolated tumor cells were classified as negative for metastasis. All cytologic and pathologic evaluation was performed by a pathologist with 18 y of experience in breast pathology. Data collection and statistical analysis The clinical data collected included age at cancer diagnosis, clinical manifestation (symptomatic vs. asymptomatic) and menopausal status. The following histopathological information was included in the study: FNA results; histologic tumor characteristics; tumor size; tumor grade; estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor (HER2) status; and lymphovascular invasion. A false-negative result was defined as a pathologically proven metastatic ALN in which US-guided FNA yielded negative results including benign or atypical. The factors associated with false-negative ALNs were evaluated using the c2 test and a t-test as appropriate. The odds ratio (OR) and 95% confidence interval (CI) for false-negative ALNs were calculated with a univariate logistic regression analysis, and variables with a value of p , 0.2 were selected for the final multivariate model. All statistical analyses were performed with statistical
software (SPSS software, v. 24.0; Chicago, IL, USA), and p , 0.05 was considered to indicate a statistically significant difference. RESULTS The mean age of the patients was 49 y (range, 31– 71 y). The median tumor size of 56 invasive cancers was 21 mm (range, 1–55 mm), and the median size of 5 in situ cancers was 34 mm (range, 10–60 mm). On cytologic report of FNA, no patients had inadequate sampling, 6 patients had atypical cells and the remaining 55 patients had benign results. The median time interval between FNA and surgery was 15 d (ranges, 3–40 d). The median number of retrieved sentinel lymph nodes was 2 (range, 1–6). No adverse reactions were associated with the tattooing procedure. Intra-operatively, all the tattooed ALNs were identified easily with visual inspection, and they were differentiated from the ALNs containing blue dye by verification of the black ink in the perinodal soft tissue. Regarding the node-to-node analysis, among the 61 suspicious ALNs with negative results on US-guided FNA, 12 ALNs (20%; 95% CI, 10%–34%) had metastases in the frozen sections. One ALN that was negative for metastasis on the frozen section had micrometastasis on the final pathology. Ultimately, 13 of 61 ALNs had false-negative results (21%; 95% CI, 11%–36%). Two ALNs that were tattooed nodes had isolated tumor cells. With respect to the patient-based analysis, 14 of 61 patients had false-negative results (23%; 95% CI, 13%– 39%). One patient (2%; 95% CI, 0%–9%) in whom the tattooed ALN was negative for metastasis had 2 metastatic ALNs of 1 sentinel and 1 non-sentinel lymph node. Among the 14 patients, 12 patients (20%; 95% CI, 6%– 21%) had a single (n 5 8) or 2 (n 5 4) metastatic ALNs (pN1), and the remaining 2 patients (3%; 95%
4
Ultrasound in Medicine and Biology
CI, 0.2%–7%) had 12 metastatic ALNs each (pN3). Of note, patients with advanced nodal disease (pN3) had atypical cells on their FNA results. In 56 of 61 patients (92%; 95% CI, 69%–119%), there was a correspondence between the tattooed ALNs and sentinel lymph nodes. Among the 5 patients (8%; 95% CI, 3%–19%) in whom the correspondence was not made, 1 patient (2%; 95% CI, 0%–9%) had 2 metastatic ALNs of 1 tattooed node and 1 sentinel lymph node. In another patient (2%; 95% CI, 0%–9%), the tattooed ALN was negative for metastasis; however, 2 metastatic ALNs of sentinel and non-sentinel lymph nodes were finally confirmed. The remaining 3 patients had no metastatic ALNs in either tattooed or sentinel lymph nodes. The imaging characteristics of the included ALNs are shown in Table 1. No significant differences were found in the imaging characteristics of ALNs. We observed a higher proportion of the presence of eccentric cortical thickening in false-negative ALNs than in true-negative ALNs, but this failed to reach statistical significance (p 5 0.192). In terms of patients and tumor characteristics (Table 2), we found significantly higher proportions of false-negative ALNs in patients who were asymptomatic (p 5 0.032) and had atypical cells on FNA results (p 5 0.005), in comparisons with those who were symptomatic and had no atypical cells on FNA results. In multivariate analysis, atypical cells on FNA results (OR 5 20.7 [95% CI, 1.1–373.7], p 5 0.040) was independently associated with false-negative FNA results. DISCUSSION The sensitivity or specificity of US-guided FNA for preliminary staging of ALNs has been evaluated in multiple studies. However, those studies could not precisely Table 1. Imaging characteristics of true-negative and false-negative ALNs on US-guided FNA* Imaging characteristics Cortical thickness #2 mm .2 mm Loss of fatty hilum Absence Presence Eccentric cortical thickening Absence Presence Shape Oval (L/S ratio $1.5) Round (L/S ratio ,1.5)
True-negative False-negative ALNs (n 5 48) ALNs (n 5 13) p value 0.343 9 (18.8) 39 (81.2)
1 (7.7) 12 (92.3)
44 (91.7) 4 (8.3)
12 (92.3) 1 (7.7)
35 (72.9) 13 (27.1)
7 (53.8) 6 (46.2)
40 (83.3) 8 (16.7)
11 (84.6) 2 (15.4)
0.941 0.192 0.913
ALNs 5 axillary lymph nodes; L/S 5 ratio of long- to short-axis diameter; US-guided FNA 5 ultrasonography-guided fine-needle aspiration. * Data are numbers of ALNs, with percentages in parentheses.
Volume -, Number -, 2017
Table 2. Patients and primary tumor characteristics of true-negative and false-negative axillary ALNs on US-guided FNA* Clinical and tumor characteristics Age (y) Clinical manifestation Symptomatic Asymptomatic Menopausal status Premenopausal Postmenopausal Atypical cell on FNA result Absence Presence Histologic tumor characteristics Ductal Lobular Othersy Tumor size (mm) Tumor gradez Low Intermediate High Hormone receptor status Negative Positive HER2 status Negative Positive Lymphovascular invasion Absence Presence
True-negative False-negative ALNs (n 5 48) ALNs (n 5 13) p value 48.1
53.0
37 (77.1) 11 (22.9)
6 (46.2) 7 (53.8)
30 (62.5) 18 (37.5)
5 (38.5) 8 (61.5)
46 (95.8) 2 (4.2)
9 (69.2) 4 (30.8)
0.360 0.032 0.123 0.005 0.447
38 (79.2) 5 (10.4) 5 (10.4) 22.1 (12.4)
11 (84.6) 0 (0.0) 2 (15.4) 23.0 (8.1)
6 (14.3) 10 (23.8) 26 (61.9)
1 (7.7) 6 (46.2) 6 (46.2)
25 (52.1) 23 (47.9)
4 (30.8) 9 (69.2)
35 (72.9) 13 (27.1)
9 (69.2) 4 (30.8)
30 (69.8) 13 (30.2)
6 (46.2) 7 (53.8)
0.795 0.291 0.176 0.794 0.123
ALNs 5 axillary lymph nodes; US-guided FNA 5 ultrasonographyguided fine needle aspiration. * Data are numbers of patients, with percentages in parentheses. HER2 5 human epidermal growth factor receptor 2. y Others include mucinous (n 5 4), metaplastic (n 5 1), and encapsulated papillary carcinoma (n 5 1). z Limited to patients (n 5 55) with available tumor grade results.
determine false-negative results because node-to-node analysis was unavailable. Furthermore, it remains to be answered whether SLNB is sufficient without the excision of biopsied but suspicious ALNs with negative results on FNA, because the sensitivity of US might be lowered by adding FNA. In our study, even with a negative FNA result, the probability of metastasis (1-negative predictive value [NPV]) was still 21% (13/61) by nodeto-node analysis and 23% (14/61) by patient-based analysis. This rate is comparable or slightly lower than the range of rate (28%–66%) reported by previous studies involving NPV (Alkuwari and Auger 2008; Ciatto et al. 2007; Kuenen-Boumeester et al. 2003; Park et al. 2013). The wide range of the NPV is most likely dependent on the following: palpable versus nonpalpable ALNs, morphologic criteria for determination of suspicious ALNs, radiologists versus clinicians performing FNA, characteristics of primary tumor and method of guidance (US guidance vs. with palpation only). This, together with our results, suggests that
False-negative results of fine-needle aspiration for axillary lymph nodes d W. H. KIM et al.
surgical staging of the axilla by SLNB or excision of biopsied ALNs remains essential. One of the major strengths of the present study is that the correlation between suspicious ALNs on US and sentinel lymph nodes was provided in the context of the feasibility of targeted ALN excision after tattooing. In our study, all the tattooed ALNs were visualized in the operation field and 92% (56/61) of the tattooed ALNs corresponded to sentinel lymph nodes. Furthermore, 1 tattooed ALN, which was not a sentinel lymph node, showed metastasis. To date, the result of pre-operative tattooing of ALNs has been shown in only 1 study with a limited population, and that study revealed a 96% correspondence rate between the tattooed ALNs and sentinel lymph nodes (Choy et al. 2015). Targeted ALN excision using other techniques, such as use of a metallic marker and radioactive seeds, has been described recently (Caudle et al. 2016; Shin et al. 2016). In comparison with these techniques, the tattoo technique has the advantages of a lower cost and higher convenience without the use of additional radioactive materials or a localization procedure to ascertain the location of the marker. The reason that most of the suspicious ALNs on US were the sentinel lymph nodes with a higher concordance rate is unclear. Sentinel lymph nodes are hypothetical lymph nodes or groups of nodes draining a cancer. Metastatic cells traveling in the afferent lymphatic channels are generally embedded in the marginal sinus at the cortex of a lymph node; thus, US classification of ALNs is based on their cortical thickness (Bedi et al. 2008). Hence, it appears to be a natural assumption that the most suspicious ALNs having the most thickened cortices are thus sentinel lymph nodes in many cases, even with benign ALNs. In the present study, we found that the atypical cells on FNA were associated with false-negative FNA results. Park et al. (2013) described that positive hormone receptor status and lymphovascular invasion (LVI) were associated with false-negative ALNs in US and US-guided FNA in their univariate analysis. In their multivariate analysis, high T-stage and LVI were associated with false-negative ALNs. In our study, positive hormone receptor status and LVI were also more frequently found in false-negative ALNs than in true-negative ALNs; however, those did not reach a statistical significance, probably because our selected population had negative FNA results whereas Park et al. (2013) included patients without FNA results. Furthermore, our findings indicated that the FNA result of atypical cells is a strong predictor of falsenegative results; atypical cells on FNA had an OR of 20.7 for metastasis in the multivariate analysis. Among the 6 ALNs with atypical cells on FNA, 4 ALNs (67%)
5
had metastasis in the final pathology, including 2 metastatic ALNs (33%) accompanied by advanced nodal disease. These findings are in line with results of previous studies showing that cytologic atypia is a high-risk feature for malignancy in various organs (Kelman et al. 2001; Olson et al. 2011; Pitman et al. 2013; Widra et al. 1995). Doubts may exist about the classification of the ALNs with atypical cells (n 5 6) on FNA as negative results for metastasis. In our institution, patients who have ALNs with atypical cells undergo SLNB with the tattooing technique because we believe a chance exists that they will have final negative pathologic results; in these cases, ALND is unnecessary. Indeed, about one third of the patients who had ALNs with atypical cells have negative pathologic results after surgery. Even if the negative results of ALNs were confined to only the benign results on US-guided FNA, the rate of falsenegative ALNs (20%) was similar to the results (21%) shown in our current analysis. Although we observed a considerable number of false-negative results for US-guided FNA, US-guided FNA is useful to guide axillary surgery or to tailor systemic treatment. US-guided FNA allows us to find a subgroup eligible for one-stage axillary surgery. If a patient has a positive FNA result for metastasis, direct axillary dissection can be performed without SLNB. In addition, if a patient were proven to have a metastatic ALN at initial staging, the patient would have the opportunity to have neoadjuvant chemotherapy before curative surgery. Our study has several limitations. First, the determination of suspicious ALNs on US during the preliminary staging of the axilla was subjective, potential variability among radiologists can occur, and inter-reader agreement was not calculated. US-guided FNA is an operatordependent procedure for obtaining the desired tissue. Second, because we focused on a node-to-node analysis of the suspicious ALNs with negative FNA results, the overall sensitivity and specificity of US-guided FNA could not be determined. The node-to-node analysis was unfeasible for patients who had positive results for metastasis on US-guided FNA because they usually undergo neoadjuvant chemotherapy or ALND without the tattooing procedure. Third, although the inadequate sampling of cytologic reports of US-guided FNA causes higher false-negative rates, the ALNs included in the present study did not have the results of inadequate sampling. In our institution, the rate of inadequate sampling is low, at approximately 2.6%. This might be in part a result of the following at our institution: relatively slim patients without abundant axillary fat, radiologists experienced in breast imaging and an FNA protocol with obligatory acquisition of at least two samples for each ALN. Finally, this study is limited to a single center in Korea. The characteristics of Asian patients can be
6
Ultrasound in Medicine and Biology
different from those from other countries (e.g., younger age, lower body mass index and different genetics) (Leong et al. 2010). In conclusion, the present study confirms that the false-negative ALNs after US-guided FNA occur at a rate of 21% and are associated with atypical cells on FNA results. Non-palpable and suspicious ALNs with negative FNA results can be excised after tattooing with an injection of the charcoal-based ink, and most of them corresponded to the sentinel lymph nodes. A technique of tattooing is a useful tool for the tagging of ALNs and thereby may potentially enable targeted ALN excision. Further investigations in a large number of patients are warranted. Acknowledgments—This work was supported by Biomedical Research Institute grant, Kyungpook National University Hospital (2017).
REFERENCES Alkuwari E, Auger M. Accuracy of fine-needle aspiration cytology of axillary lymph nodes in breast cancer patients: A study of 115 cases with cytologic-histologic correlation. Cancer 2008;114:89–93. Alvarez S, Anorbe E, Alcorta P, Lopez F, Alonso I, Cortes J. Role of sonography in the diagnosis of axillary lymph node metastases in breast cancer: A systematic review. AJR Am J Roentgenol 2006; 186:1342–1348. 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;97:680–683. Bedi DG, Krishnamurthy R, Krishnamurthy S, Edeiken BS, Le-Petross H, Fornage BD, Bassett RL Jr, Hunt KK. Cortical morphologic features of axillary lymph nodes as a predictor of metastasis in breast cancer: In vitro sonographic study. AJR Am J Roentgenol 2008;191:646–652. Caudle AS, Yang WT, Krishnamurthy S, Mittendorf EA, Black DM, Gilcrease MZ, Bedrosian I, Hobbs BP, DeSnyder SM, Hwang RF, Adrada BE, Shaitelman SF, Chavez-MacGregor M, Smith BD, Candelaria RP, Babiera GV, Dogan BE, Santiago L, Hunt KK, Kuerer HM. Improved axillary evaluation following neoadjuvant therapy for patients with node-positive breast cancer using selective evaluation of clipped nodes: Implementation of targeted axillary dissection. J Clin Oncol 2016;34:1072–1078. Cho N, Moon WK, Han W, Park IA, Cho J, Noh DY. Preoperative sonographic classification of axillary lymph nodes in patients with breast cancer: Node-to-node correlation with surgical histology and sentinel node biopsy results. AJR Am J Roentgenol 2009;193:1731–1737. Choy N, Lipson J, Porter C, Ozawa M, Kieryn A, Pal S, Kao J, Trinh L, Wheeler A, Ikeda D, Jensen K, Allison K, Wapnir I. Initial results with preoperative tattooing of biopsied axillary lymph nodes and correlation to sentinel lymph nodes in breast cancer patients. Ann Surg Oncol 2015;22:377–382. Ciatto S, Brancato B, Risso G, Ambrogetti D, Bulgaresi P, Maddau C, Turco P, Houssami N. 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. de Freitas R Jr, Costa MV, Schneider SV, Nicolau MA, Marussi E. Accuracy of ultrasound and clinical examination in the diagnosis of axillary lymph node metastases in breast cancer. Eur J Surg Oncol 1991;17:240–244. Goyal P, Sehgal S, Ghosh S, Aggarwal D, Shukla P, Kumar A, Gupta R, Singh S. Histopathological correlation of atypical (c3) and suspicious (c4) categories in fine needle aspiration cytology of the breast. Int J Breast Cancer 2013;2013:965498. Hyun SJ, Kim EK, Moon HJ, Yoon JH, Kim MJ. Preoperative axillary lymph node evaluation in breast cancer patients by breast magnetic resonance imaging (MRI): Can breast MRI exclude advanced nodal disease? Eur Radiol 2016;26:3865–3873.
Volume -, Number -, 2017 Kelley MC, Hansen N, McMasters KM. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Am J Surg 2004;188:49–61. Kelman AS, Rathan A, Leibowitz J, Burstein DE, Haber RS. Thyroid cytology and the risk of malignancy in thyroid nodules: Importance of nuclear atypia in indeterminate specimens. Thyroid 2001;11:271–277. Krag D, Weaver D, Ashikaga T, Moffat F, Klimberg VS, Shriver C, Feldman S, Kusminsky R, Gadd M, Kuhn J, Harlow S, Beitsch P. The sentinel node in breast cancer—A multicenter validation study. N Engl J Med 1998;339:941–946. Kuenen-Boumeester V, Menke-Pluymers M, de Kanter AY, Obdeijn IM, Urich D, Van Der Kwast TH. Ultrasound-guided fine needle aspiration cytology of axillary lymph nodes in breast cancer patients. A preoperative staging procedure. Eur J Cancer 2003;39:170–174. Leong SP, Shen ZZ, Liu TJ, Agarwal G, Tajima T, Paik NS, Sandelin K, Derossis A, Cody H, Foulkes WD. Is breast cancer the same disease in Asian and Western countries? World J Surg 2010;34:2308–2324. Lucci A, McCall LM, Beitsch PD, Whitworth PW, Reintgen DS, Blumencranz PW, Leitch AM, Saha S, Hunt KK, Giuliano AE, American College of Surgeons Oncology Group. Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J Clin Oncol 2007;25:3657–3663. Martin RC 2nd, Chagpar A, Scoggins CR, Edwards MJ, Hagendoorn L, Stromberg AJ, McMasters KM, University of Louisville Breast Cancer Sentinel Lymph Node Study. Clinicopathologic factors associated with false-negative sentinel lymph-node biopsy in breast cancer. Ann Surg 2005;241:1005–1012. discussion 1012–1015. Olson MT, Clark DP, Erozan YS, Ali SZ. Spectrum of risk of malignancy in subcategories of ‘atypia of undetermined significance.’. Acta Cytlo 2011;55:518–525. 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;8:115–120. Park SH, Kim EK, Park BW, Kim SI, Moon HJ, Kim MJ. False negative results in axillary lymph nodes by ultrasonography and ultrasonography-guided fine-needle aspiration in patients with invasive ductal carcinoma. Ultrashall Med 2013;34:559–567. Park SH, Kim MJ, Park BW, Moon HJ, Kwak JY, Kim EK. Impact of preoperative ultrasonography and fine-needle aspiration of axillary lymph nodes on surgical management of primary breast cancer. Ann Surg Oncol 2011;18:738–744. Pitman MB, Yaeger KA, Brugge WR, Mino-Kenudson M. Prospective analysis of atypical epithelial cells as a high-risk cytologic feature for malignancy in pancreatic cysts. Cancer Cytopathol 2013;121:29–36. Sacre RA. Clinical evaluation of axillar lymph nodes compared to surgical and pathological findings. Eur J Surg Oncol 1986;12:169–173. Shin K, Caudle AS, Kuerer HM, Santiago L, Candelaria RP, Dogan B, Leung J, Krishnamurthy S, Yang W. Radiologic mapping for targeted axillary dissection: Needle biopsy to excision. AJR Am J Roentgenol 2016;207:1372–1379. Tucker NS, Cyr AE, Ademuyiwa FO, Tabchy A, George K, Sharma PK, Jin LX, Sanati S, Aft R, Gao F, Margenthaler JA, Gillanders WE. Axillary ultrasound accurately excludes clinically significant lymph node disease in patients with early stage breast cancer. Ann Surg 2016;264:1098–1102. Veronesi U, Paganelli G, Galimberti V, Viale G, Zurrida S, Bedoni M, Costa A, de Cicco C, Geraghty JG, Luini A, Sacchini V, Veronesi P. Sentinel-node biopsy to avoid axillary dissection in breast cancer with clinically negative lymph-nodes. Lancet 1997;349:1864–1867. Widra EA, Dunton CJ, McHugh M, Palazzo JP. Endometrial hyperplasia and the risk of carcinoma. Int J Gynecol Cancer 1995;5:233–235. Wilke LG, McCall LM, Posther KE, Whitworth PW, Reintgen DS, Leitch AM, Gabram SG, Lucci A, Cox CE, Hunt KK, Herndon JE 2nd, Giuliano AE. Surgical complications associated with sentinel lymph node biopsy: Results from a prospective international cooperative group trial. Ann Surg Oncol 2006;13:491–500. You S, Kang DK, Jung YS, An YS, Jeon GS, Kim TH. Evaluation of lymph node status after neoadjuvant chemotherapy in breast cancer patients: Comparison of diagnostic performance of ultrasound, MRI and 18F-FDG PET/CT. Br J Radiol 2015;88:20150143.