- Email: [email protected]
Contrast-enhanced ultrasound -guided axillary lymph node core biopsy: Diagnostic accuracy in preoperative staging of invasive breast cancer
European Journal of Radiology 84 (2015) 2130–2136
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Contrast-enhanced ultrasound -guided axillary lymph node core biopsy: Diagnostic accuracy in preoperative staging of invasive breast cancer Rautiainen Suvi a,f,∗ , Sudah Mazen a , Joukainen Sarianna b , Sironen Reijo c,e,f , Vanninen Ritva a,d,f , Sutela Anna a a
Department of Clinical Radiology, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland Department of Plastic Surgery, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland c Department of Clinical Pathology, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland d Unit of Radiology, Biocenter Kuopio, University of Eastern Finland, P.O.B. 1627, FI-70211 Kuopio, Finland e Unit of Pathology and Forensic Medicine at Institute of Clinical Medicine, Biocenter Kuopio, University of Eastern Finland, P.O.B. 1627, FI-70211 Kuopio, Finland f Cancer Center of Eastern Finland, University of Eastern Finland, P.O.B. 1627, FI-70211 Kuopio, Finland b
a r t i c l e
i n f o
Article history: Received 1 June 2015 Received in revised form 9 July 2015 Accepted 12 August 2015 Keywords: Breast neoplasms Axilla Biopsy Ultrasonography Contrast media
a b s t r a c t Objectives: To evaluate accuracy of contrast enhanced ultrasound (CEUS)-sentinel procedure followed by core biopsy (CB) and marking in patients with breast cancer. To compare the axillary metastatic tumour burden in patients with positive vs. negative CB results. Methods: Two radiologists in our tertiary care hospital performed axillary CEUS sentinel procedures on consecutive US node negative breast cancer patients. The first enhancing lymph node (LN) was core biopsied and marked with a breast coil. The results were compared to final histopathology. We analysed the diagnostic performance of CEUS CB and its ability to detect patients with higher axillary burden (>2 metastasis). Results: During the study period between January 2013 and December 2014, altogether 54 patients (mean age 60.4 years) were included in the statistical analysis. The sensitivity for CEUS CB was 66.7%, specificity 100%, PPV 100%, NPV 93.8% and overall accuracy 94.4%. The method correctly recognised all the axillae with higher tumour burdens (sensitivity 100%, N = 3) and 59.3% of coils marking the LNs were discovered. Conclusion: CEUS -guided axillary CB proved to be feasible and accurate procedure with moderate sensitivity and it clearly identified the higher axillary tumour burden. The coil marking of LNs as used cannot be recommended. In clinical routine, CEUS procedure might be recommended in selective patient populations. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction During the past few decades, staging of the axilla in patients with breast cancer has shifted from routine axillary lymph node dissec-
Abbreviations: ALND, axillary lymph node dissection; CB, core biopsy; CEUS, contrast-enhanced ultrasound; FNA, fine-needle aspiration; LN, lymph node; SLN, sentinel lymph node; SLNB, sentinel lymph node biopsy; UNB, ultrasound-guided needle biopsy; US, ultrasound. ∗ Corresponding author at: Department of Clinical Pathology, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland. Fax: +358 17 173 341. E-mail addresses: suvi.rautiainen@kuh.fi (S. Rautiainen), mazen.sudah@kuh.fi (M. Sudah), sarianna.joukainen@kuh.fi (S. Joukainen), reijo.sironen@kuh.fi (R. Sironen), ritva.vanninen@kuh.fi (R. Vanninen), anna.sutela@kuh.fi (A. Sutela). http://dx.doi.org/10.1016/j.ejrad.2015.08.006 0720-048X/© 2015 Elsevier Ireland Ltd. All rights reserved.
tion (ALND) towards sentinel lymph node biopsy (SLNB), which has become the standard of care. SLNB has achieved high sensitivity (97–98%), with similar survival outcomes but with less morbidity compared to ALND [1,2]. Moreover, the need for routine ALND after positive SLNB findings was recently challenged by randomized trials (ACOSOG and AMAROS) in patients with T1-2 tumours and clinically node negative axilla [3,4]. The concept of low (<2 positive LNs) vs. high axillary tumour burden (>2 positive LNs) has recently been introduced also in radiological studies, which aim to determine the role of preoperative investigations in the evolving axillary staging [5–7]. Although the search for a non-invasive approach to preoperatively stage the axilla has attracted much interest, the diagnostic figures have not achieved the results of SLNB [6,8,9]. The sensitivity and specificity of axillary ultrasound (US) has been reported
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
between 61% - 62% and 82% - 86%, respectively [6,10]. US-guided needle biopsy (UNB) has improved the sensitivity (80%) with perfect specificity (100%) [6,9,10] and finds approximately half of axillary metastasis preoperatively [11]. The US-positive axillae is more frequently associated with higher axillary tumour burden (>2 positive LNs) [12] and the metastases found by US tend to be larger than average [13,14]. However, the majority of patients with small breast tumours have a low axillary tumour burden [15,16] and more accurate preoperative staging would be desired. The false negative findings in axillary UNB can arise from failure to sample the sentinel lymph node (SLN) (45%), or the metastatic deposit within the LN (55%) [13]. The latter can be reduced by using core biopsy (CB) instead of fine needle aspiration (FNA) [17,18]. Moreover, the ability to correctly localize the SLN during biopsy would improve the detection rate of metastatic LNs. The use of contrast enhanced US (CEUS) to localize the SLN has shown it to be both safe and feasible [19–21]. In comparison to traditional isotope SLNB and Patent Blue dye, the sensitivity of CEUS to detect SLN correctly was 89% [20]. In clinically node negative patients, the sensitivity of CEUS -guided biopsy (with either CB or FNA) was 61–65% [21,22]. However, the literature on CEUS -guided SLN biopsies is still scanty and details such as localization methods prior to subsequent surgical excision remains to be determined [20,23]. The goals of the present study were: (1) to assess the accuracy of CEUS-sentinel procedure followed by CB in patients with no ultrasonographic signs indicative of axillary metastasis and (2) to evaluate the metastatic tumour burden in patients with positive vs. negative CB results and (3) to investigate the accuracy of marking the CEUS SLN with a breast coil. 2. Materials and methods This prospective study was approved by the National Medicines Agency (EudraCT 2012-005349-20) and by the hospital ethics committee according to the recommendation from the national ethical board for medical research. All consecutive newly diagnosed patients at least 18 years of age and with invasive breast cancer or BI-RADS 5 breast lesions were invited to participate in the study. The exclusion criteria were as follows: patients who could not provide written informed consent; clinically suspicious finding of axillary metastasis; history of previous ipsilateral breast cancer, axillary surgery or radiotherapy; no axillary surgery planned or performed; contraindication for the use of US contrast agent. In addition, exclusion criteria for axillae in the grey scale US study included morphological abnormality of the LNs were any of the following findings: any cortical thickening ≥5.0 mm or dislocated or absent fatty hilum. No limits were set for minimum thickness of the axillary LN, breast tumour size or tumour multifocality. During the study period between January 2013 and December 2014, the patients in this study were referred from a screening centre, two district hospitals and private institutions. All patients were evaluated by a breast radiologist before surgery and breast lesions were classified according to the ACR BI-RADS® lexicon (American College of Radiology, Breast Imaging Reporting and Data System) [24,25]. The findings were evaluated at a multidisciplinary breast meeting. 2.1. US—contrast enhancement studies and biopsies The CEUS procedures and consecutive biopsies were performed by two breast radiologists with 5 and 21 years of experience in breast imaging and interventions (SR, MS). Before the start of the study, both radiologists had limited experience in performing axillary CEUS procedures. The US examinations were undertaken
2131
with an Esaote MyLabClassC (Esaote S.p.A., Genova, Italy) with a 7–13 MHz linear transducer and the contrast studies using the contrast-specific software program and linear probe. During the grey scale axillary US assessment, special attention was paid to the typical location of the SLN at the caudal part of level I. If no exclusion criteria were present, the possible SLN with the thickest cortex at that location was documented. If during CEUS procedure this LN was found not the enhancing sentinel node, then the enhanced LN was biopsied and CEUS was deemed to have changed the biopsy plan. The morphology of the CEUS LN was noted as was the number of other enhancing LNs. All US and CEUS findings were recorded to a structured study form. 2.1.1. CEUS and biopsy procedures The CEUS procedure was performed according to Sever et al. [20]. In short; the contrast agent powder, SonoVue® (Bracco Imaging, Milan, Italy) was mixed with 2.0 ml of sterile saline. The contrast was then injected intradermally into the upper outer quadrant close to the areola with a 26G needle using 1.0 ml syringe. The amount of a single injection was approximately 0.4 ml and the injection was repeated 1–2 times if needed. Breast and axillae were scanned with US and the contrast injection area was massaged for 5–15 s in cases where no enhancing lymph vessels or axillary LNs could be visualized. The rate of lymph vessel enhancement was categorised as (1) spontaneous, (2) delayed, or (3) delayed after massage. The core biopsies were obtained from the first enhancing LN aiming at the cortex and subcapsular area using a 16G core-needle with 20-mm notch exposure (Temno Evolution, CareFusion Corporation, McGaw Park, IL, USA) or with an automated CNB gun with a 22-mm throw (Bard Magnum; Bard Biopsy Systems, Tempe, AZ, USA). A total of two samples that visually included solid non-fatty tissue were obtained. More biopsies were collected at the discretion of the radiologist if the first biopsies were deemed as being nonrepresentative. In the case where two or more LNs were enhancing, then only the first LN to enhance was biopsied (Fig. 1). The successfully biopsied and enhanced LNs were marked with a breast coil (MReye Breast Localization coil, COOK Medical, Bloomington, IN, USA) adjacent to the biopsy area, for possible further recognition. The total number of enhancing LNs and biopsies were recorded, as well as any possible complications. The procedure was deemed as unsuccessful if no axillary LN enhancement could be observed. 2.1.2. Patient management and histopathological evaluation Core biopsy specimens were placed into 10% formalin and embedded in paraffin after fixation. The samples were cut into 5m slices at 4 different levels then stained with haematoxylin and eosin (HE). CB samples were evaluated by 2 pathologists, first at diagnosis, then at the multidisciplinary meeting. Patients underwent SLNB in case axillary CB showed micrometastasis and were directed to ALND in the cases where the preoperative CB revealed metastases. The CEUS-positive LNs were recognized from the evacuation specimens either by the surgeon during the operation or from specimen X-rays and were marked separately for the pathologist. If the biopsy result was negative, then SLNB was performed according to a standardized 2-day protocol including injections of technetium-labelled nanocolloid (Nanocoll, Technetium Tc-99M Albumin Aggregated Kit, GE Healthcare, Milan, Italy) and Patent Blue dye preoperatively as well as the intraoperative frozen sections. During surgery, the performing breast surgeon recognized the CEUS LN with the coil and removed this and any radioactive LN’s and any other clinically suspicious LNs. After formalin fixation, all of the obtained LNs were paraffin embedded, serially sectioned into 2-mm slices and stained with HE. For the detection of micrometastases, an immunohistochem-
2132
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
Fig. 1. US images of a 40-year-old breast cancer patient with a multifocal breast tumour BI-RADS 5. (a) An enhancing axillary LN with (b) visual afferent lymph vessel (arrow) was identified in CEUS. (c) This was the first and only visual enhancing LN which in the grey scale US had a cortical thickness of 2.7 mm. (d) Metastasis was detected by core biopsy (arrow heads) and confirmed at final histopathology, which showed 3/15 metastatic LNs after axillary dissection. Table 1 Patient demographics and tumour characteristics (N = 54). Characteristic
N
(%)
Tumour pathological T classification T1 T2 T3 T4
35 15 4 0
64.8 27.8 7.4 0
Tumour N classification N0 N1 N1 (mi) N2 N3
42 7 3 1 1
77.8 13.0 5.6 1.9 1.9
Stage 1 2 3 4
32 20 2 0
59.3 37.0 3.7 0
Tumour histology Ductal Lobular Mucinous Other
42 9 1 2
77.8 16.7 1.9 3.7
Grade of invasive carcinoma 1 2 3 Multifocality
9 34 11 13
16.7 63.0 20.4 24.1
ical staining was then performed against low molecular weight cytokeratin (Techmate 500 Plus, DAKO, Denmark) using a labelled polymer-based method (DAKO kit K4068, low molecular cytokeratin, Cam 5.2, Becton Dickinson, NJ, USA). If the frozen section during surgery was positive for either macro- or micrometastasis, the patient underwent ALND. Patients with negative frozen sections but with micrometastasis at final histology were treated according to the recommendation of the multidisciplinary group experts, either with completion ALND or follow-up. LNs were deemed positive for malignancy if they contained macrometastases (>2 mm). Both micrometastases (>0.2–2.0 mm) and isolated tumour cells (ITC, <0.2 mm) on histology were classified as negative. In the further subgroup analysis, the axillary positive patients were divided according to the number of LNs with macrometastases; high axillary burden with >2 positive LNs and low burden with 1–2 positive LNs. 2.1.3. Statistical analysis Statistical analysis was performed with SPSS version 19 (IBM Corporation, Somers, NY, USA). SLNB-negative as well as micrometastases and isolated tumour cells (ITC) were classified as non-metastatic LNs. Diagnostic performance (sensitivity, specificity, positive predictive value, negative predictive value and overall accuracy) was calculated for CEUS CB. Positive and negative predictive values were calculated with Bayes’ formula [26] and 95% confidence intervals were calculated with Wilson’s nocontinuity correction formula [27]. In the subgroup analysis, the statistical comparisons were performed using McNemar’s test or
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
2133
Fig. 2. A unifocal 15 mm breast tumour (**) of a 49-year-old woman (a). In the CEUS study (b) the contrast-filled lymph vessel (arrow) is amputated at the site of the tumour (**) with lymphovascular invasion in the final histopathology. In addition, the traditional isotope sentinel procedure failed and the ALND was performed with 2/17 metastatic LNs (largest macrometastasis 4 mm).
Fisher’s exact test. p < 0.05 was considered to indicate statistical significance. 3. Results During the study period between January 2013 and December 2014, a total of 60 patients fulfilled the inclusion criteria and were
enrolled. The success rate of the CEUS procedure was 95.0% (57/60). In two cases, the lymph vessels were detected but no enhancing LNs could be observed. In the third patient, a visual amputation of the enhancing lymph vessel by breast tumour infiltration was detected (Fig. 2), also with no enhancing LNs. In two cases, it was not possible to perform the CB due to technical or safety reasons. Another patient with DCIS at CB and final histology was also excluded.
Fig. 3. Flow chart of patients.
2134
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
Table 2 (a–b) The contrast study details of the patients (N = 54) with enhancing axillary lymph nodes (LN). (a) Characteristic CEUS LN as the lowest axillary LN Enhancement type Spontaneous Delayed Delayed after massage Coils localized by palpation/X-ray (b) Characteristic Injection volume (ml) Number of injections per axilla CEUS LN’s features Cortical thickness (mm) Length (mm) Width (mm)
N
(%)
41
75.9
41 8 5 32
83.3 14.8 9.3 59.3
Mean
Range
0.4 2.2
0.1–0.5 1–3
2.4 12.8 5.8
0.2–5.0 5.8–25.0 2.5–10.0
Table 3 Information of the axillary positive patients’ (N = 12) lymph node morphology in preoperative ultrasound and the core biopsy results in comparison to the histopathology of the axillary metastases and breast tumour. No
LN cortex (mm)
CB true positive
1 2 3 4 5 6 7 8 9 10 11 12
3.0 3.7 5.0 3.4 1.9 2.2 2.7 3.4 2.1 2.4 2.3 3.9
Yes Yes Yes Yes No Yes No Yes No No Yes No
Metastatic LN 3 2 1 3 2 1 1 6 1 1 mi 1 mi 1 mi
Size of metastatic deposit (mm)
Tumour final size (mm)
Tumour histology
Grade
Multifocality
14.0 11.0 11.0 6.0 4.0 3.0 2.1 10.0 10.0 2.0 1.0 0.2
36.0 70.0 35.0 36.0 17.0 15.0 10.0 25.0 22.0 11.0 18.0 29.0
ILC ILC IDC IDC IDC IDC IDC IDC IDC ILC IDC IDC
3 2 3 3 2 2 1 3 3 2 2 2
Yes Yes No Yes No No No Yes No Yes Yes No
mi = micrometastasis. ILC = invasive lobular carcinoma. IDC = invasive ductal carcinoma. Table 4 The diagnostic performance of the contrast enhanced ultrasound—guided axillary lymph node core biopsy and the comparison to the literature.
Present study CEUS + CB CEUS + CB Cox et al. [22] Cox et al. [22] CEUS + FNA or CB Sever et al. [20]CEUS + FNA or CB
N
Sensitivity % with 95% CI Specificity % with 95% CI PPV % with 95% CI
NPV % with 95% CI
Overall accuracy % with 95% CI
54 235 295 126
66.7 [6/9]35.4–88.0 65.0 59.0–71.0 61.0 56.0–66.0 65.0
93.8 [45/48]83.2–97.9 95.0 81.0 92.0
94.4 [51/54]84.8–98.1 n n n
100 [45/45]92.1–100 100 100 100
100 [6/6]61.0–100 100 100 100
Diagnostic performance figures calculated against the macrometastases in SNLB or ALND. Positive and negative predictive values were calculated by using the Bayes formula with an estimate of prevalence of 16.7% for macrometastasis [26]. The 95% confidence intervals were calculated by using the Wilson no-continuity correction formula [27]. Data in brackets [] are nominators and denominators.
Hence, a total of 54 patients were included in the statistical analysis (Fig. 3). The mean age of the patients was 60.4 years (range 34–88 years). The tumour charasteristics are shown in Table 1. Marking of the CEUS LNs with breast coil was successful in 59.3% of cases (32/54). The detailed information on the contrast studies are presented in Table 2a and b. The CEUS procedure changed the biopsy target LN in 24.0% (13/54) of the cases compared to grey scale US. Three of these were metastatic LNs (in two cases, this was the only metastastic LN). In one case (1.9%, 1/54), the core biopsy specimen was considered as inadequate (cortical thickness 1.1 mm), and the axilla was determined as being benign in SLNB. Axillary macrometastases were found in the final histopathology in 16.7% (9/54) of the patients with invasive carcinoma. In multifocal tumours (N = 13, 24.1%), axillary macrometastases were found in 4 (30.8%) cases and in unifocal tumours (N = 41, 75.9%) in 2 (4.9%). A total of 6 axillae had either micrometastases (N = 3)
or ITC (N = 3). CEUS CB was false negative in three (33.3%) of the macrometastases having a maximum size of 2.1–10.0 mm. The detailed information on the axillary positive patients is presented in Table 3. CEUS -guided CB had a sensitivity of 66.7% and a specificity of 100% to detect metastases. The detailed diagnostic performance of CEUS-CB is shown in Table 4 with a comparison to the literature. A total of 3 of the 9 patients with axillary macrometastases had a high axillary tumour burden (>2 metastatic LNs) while 6 patients had a low axillary tumour burden (1–2 metastatic LNs). The CEUS CB was true positive in all of the patients with a high axillary tumour burden (3/3) and in half of the patients with a low tumour burden (3/6) (p = 0.081). When the high vs. low axillary tumour burden groups were compared, no statistically significant differences in US features could be detected i.e. mean number of LNs with >2 mm cortex was 1.3 (high) vs. 0.7 (low) (p = 0.149), cortical thickness maximum 3.3 mm
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
(high) vs. 2.9 mm (low) mm (p = 0.268) and CEUS LN’s mean cortical thickness 3.0 mm (high) vs. 2.8 mm (low) (p = 0.548).
4. Discussion Previously reports from a single institution have described the technical feasibility of axillary CEUS biopsy procedure, which detects additional metastases in nodes interpreted as normal in grey scale US [21,22]. Our comparable results confirm that axillary CEUS procedure can be obtained to clinical use without hands-on training and is a feasible, consistent procedure. The low sensitivities in previous studies (61.0–65.0%) and the present (66.7%), can be attributed mainly to the strict inclusion criteria of US node negativity and the low prevalence of axillary metastases. It is well known that a LN’s smooth cortex is less predictive for metastasis, while other morphological features such as the absence of a central fatty hilum and multi-lobulated cortical thickening, are highly predictive for metastasis [12–14]. The LNs with the latter features were excluded from this study. Nevertheless, CEUS CB was able to correctly detect the majority of axillary metastasis. Unfortunately, a few patients with macrometastasis were not identified. The CEUS LN may not be the metastatic node or the biopsy may hit an area free of metastasis within the correct SLN especially in case of very small macrometastasis. During the SLNB procedure, often more than one SLN is removed, leading to higher sensitivity. In addition, CB is not technically feasible in all patients due to inadequate samples, even if more accurate than FNA [17,18]. In our study, there were fewer inadequate samples than previously reported for CEUS CB, reflecting our longstanding experience with axillary CB. The CEUS -guided biopsy can accidentally detect small tumour deposits (micrometastases or ITC), which might lead to ALND and thus to overtreatment. This risk should be acknowledged and the patients with these findings should be subjected to SLNB. Therefore, although CEUS is useful in its current form, it cannot replace SLNB in routine clinical practice. Our study confirms that CEUS CB procedure is safe. The precise intradermal contrast injection is essential but, fortunately, the injections can be safely repeated with no adverse effects. In the majority of patients, the contrast filled the lymph vessels immediately and could be followed directly to the axillary LNs, which makes it attractive when searching for the SLN. The fact that CEUS changed the biopsy plan compared to grey scale US in a considerable number of patients might indicate that CEUS could increase the sensitivity of axillary needle biopsies. Recent results emerging from randomized controlled trials (ACOSOG, AMAROS) [3,4] have led to a less aggressive axillary staging in patients fulfilling the studies’ inclusion criteria in some countries. In our study, CEUS CB accurately excluded the high axillary burden in otherwise US node negative axillae. Nonetheless the number of events is too small to draw firm conclusions on this issue, which is a limitation of this study. The absence of LN’s pathological features correlated well with the axillary tumour burden, as only 5% (3/54) had macrometastasis in >2 LNs. On the other hand, the majority of the patients had no axillary metastasis and thus underwent additional intervention, questioning the utility of CEUS CB as routine clinical procedure. However, CEUS CB may prove useful for selective patient groups i.e. when ACOSOG or AMAROS criteria are not fulfilled or when the presence of metastasis would change the treatment plans, or upon the recommendations of multidisciplinary team. The localization and marking the CEUS LN is another topic of interest. A relatively large and palpable coil was chosen in this study, which was believed to be well localized. This was not the case since only 59% of coils could be localized, corresponding to the results of a recent study [23]. The abundant axillary fat around the
2135
small LNs combined with the different position of patients during CEUS and SLNB procedure might well explain the failure. Preoperative axillary US is a topic of considerable interest and changes in practices are inevitable. At the moment, the lack of clarity and guidelines is confusing. Two ongoing randomized trials (INSEMA, SOUND) will hopefully shed more light on the future role of axillary US and US-guided biopsies [28,29]. Positive outcomes might increase the future role of axillary US and consequently CEUS procedures. In conclusion, CEUS -guided axillary CB in newly diagnosed breast cancer is a feasible and accurate procedure with moderate sensitivity in node negative patients. It detected all patients with high axillary tumour burden. The coil marking beside the CEUS SLN cannot be recommended. In routine clinical practice, the CEUS CB procedure might be recommended in selective patient populations. Acknowledgements This study was supported in part by Kuopio University Hospital (grant number 5063525), Kuopio University Foundation, Sirkka and Mauri Wiljasalo Foundation, Instrumentarium Science Foundation and Cancer Center of Eastern Finland provided by the University of Eastern Finland. We express our special appreciation for to nurse Tiina Föhr for her help in the recruitment of patients to the study and Tuomas Selander M.Sc. for the statistical advice. References [1] D.N. Krag, S.J. Anderson, T.B. Julian, et al., Sentinel-lymph-node resection compared with conventional axillary-lymph-node dissection in clinically node-negative patients with breast cancer: overall survival findings from the NSABP B-32 randomised phase 3 trial, Lancet Oncol. 11 (10) (2010) 927–933. [2] R.E. Mansel, L. Fallowfield, M. Kissin, et al., Randomized multicenter trial of sentinel node biopsy versus standard axillary treatment in operable breast cancer: the ALMANAC trial, J. Natl. Cancer Inst. 98 (9) (2006) 599–609. [3] A.E. Giuliano, K.K. Hunt, K.V. Ballman, et al., Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: a randomized clinical trial, JAMA 305 (6) (2011) 569–575. [4] M. Donker, G. van Tienhoven, M.E. Straver, et al., Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981–22023 AMAROS): a randomised, multicentre, open-label, phase 3 non-inferiority trial, Lancet Oncol. 15 (12) (2014) 1303–1310. [5] T.J. Hieken, B.C. Trull, J.C. Boughey, et al., Preoperative axillary imaging with percutaneous lymph node biopsy is valuable in the contemporary management of patients with breast cancer, Surgery 154 (4) (2013) 831–838, discussion 838–40. [6] N. Houssami, R.M. Turner, Staging the axilla in women with breast cancer: the utility of preoperative ultrasound-guided needle biopsy, Cancer Biol. Med. 11 (2) (2014) 69–77. [7] C. Reyna, J.V. Kiluk, A. Frelick, N. Khakpour, C. Laronga, M.C. Lee, Impact of axillary ultrasound (AUS) on axillary dissection in breast conserving surgery (BCS), J. Surg. Oncol. 111 (4) (2015) 813–818. [8] S.E. Harnan, K.L. Cooper, Y. Meng, et al., Magnetic resonance for assessment of axillary lymph node status in early breast cancer: a systematic review and meta-analysis, Eur. J. Surg. Oncol. 37 (11) (2011) 928–936. [9] N. Houssami, S. Ciatto, R.M. Turner, H.S. Cody 3rd, P. Macaskill, 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. 254 (2) (2011) 243–251. [10] S. Alvarez, E. Anorbe, P. Alcorta, F. Lopez, I. Alonso, J. Cortes, Role of sonography in the diagnosis of axillary lymph node metastases in breast cancer: a systematic review, AJR Am. J. Roentgenol. 186 (5) (2006) 1342–1348. [11] S.C. Diepstraten, A.R. Sever, C.F. Buckens, 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. 21 (1) (2014) 51–59. [12] N. Houssami, S.C. Diepstraten, H.S. Cody 3rd, R.M. Turner, A.R. Sever, Clinical utility of ultrasound-needle biopsy for preoperative staging of the axilla in invasive breast cancer, Anticancer Res. 34 (3) (2014) 1087–1097. [13] P.D. Britton, E. Provenzano, S. Barter, et al., Ultrasound guided percutaneous axillary lymph node core biopsy: how often is the sentinel lymph node being biopsied? Breast 18 (1) (2009) 13–16. [14] M.J. Garcia-Ortega, M.A. Benito, E.F. Vahamonde, P.R. Torres, A.B. Velasco, M.M. Paredes, Pretreatment axillary ultrasonography and core biopsy in patients with suspected breast cancer: diagnostic accuracy and impact on management, Eur. J. Radiol. 79 (1) (2011) 64–72.
2136
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
[15] R.M. Nadeem, The feasibility of a randomised controlled trial for the axillary management of a select group of invasive breast cancer patients: SLNB vs. no-SLNB, Breast Cancer 22 (4) (2015) 343–349. [16] J. Cools-Lartigue, A. Sinclair, N. Trabulsi, et al., Preoperative axillary ultrasound and fine-needle aspiration biopsy in the diagnosis of axillary metastases in patients with breast cancer: predictors of accuracy and future implications, Ann. Surg. Oncol. 20 (3) (2013) 819–827. [17] S. Rautiainen, A. Masarwah, M. Sudah, et al., Axillary lymph node biopsy in newly diagnosed invasive breast cancer: comparative accuracy of fine-needle aspiration biopsy versus core-needle biopsy, Radiology 269 (1) (2013) 54–60. [18] M.A. Ganott, M.L. Zuley, G.S. Abrams, et al., Ultrasound guided core biopsy versus fine needle aspiration for evaluation of axillary lymphadenopathy in patients with breast cancer, ISRN Oncol. 2014 (Feb 4) (2014) 703160. [19] A. Sever, S. Jones, K. Cox, J. Weeks, P. Mills, P. Jones, Preoperative localization of sentinel lymph nodes using intradermal microbubbles and contrast-enhanced ultrasonography in patients with breast cancer, Br. J. Surg. 96 (11) (2009) 1295–1299. [20] A.R. Sever, P. Mills, S.E. Jones, W. Mali, P.A. Jones, Sentinel node identification using microbubbles and contrast-enhanced ultrasonography, Clin. Radiol. 67 (7) (2012) 687–694. [21] A.R. Sever, P. Mills, J. Weeks, et al., Preoperative needle biopsy of sentinel lymph nodes using intradermal microbubbles and contrast-enhanced ultrasound in patients with breast cancer, AJR Am. J. Roentgenol. 199 (2) (2012) 465–470.
[22] K. Cox, A. Sever, S. Jones, et al., Validation of a technique using microbubbles and contrast enhanced ultrasound (CEUS) to biopsy sentinel lymph nodes (SLN) in pre-operative breast cancer patients with a normal grey-scale axillary ultrasound, Eur. J. Surg. Oncol. 39 (7) (2013) 760–765. [23] M.W. Barentsz, H.M. Verkooijen, R.M. Pijnappel, et al., Sentinel lymph node localization with contrast-enhanced ultrasound and an I-125 seed: an ideal prospective development study, Int. J. Surg. 14 (2015) 1–6. [24] C. D’Orsi, E. Mendelson, D. Ikeda, et al., Breast Imaging Reporting and Data System: ACR BI-RADS-Breast Imaging Atlas, American College of Radiology, Reston, VA, 2003. [25] C.L. Mercado, BI-RADS update, Radiol. Clin. North Am. 52 (3) (2014) 481–487. [26] R.F.S. Fletcher, Clinical Epidemiology: The Essentials, 4th ed., Lippincott Williams & Wilkins, Baltimore, MD, 2005, pp. 45–48. [27] R.G. Newcombe, Two-sided confidence intervals for the single proportion: comparison of seven methods, Stat. Med. 17 (8) (1998) 857–872. [28] O. Gentilini, U. Veronesi, Abandoning sentinel lymph node biopsy in early breast cancer? A new trial in progress at the European Institute of Oncology of Milan (SOUND: sentinel node vs observation after axillary UltraSouND), Breast 21 (5) (2012) 678–681. [29] T. Reimer, S. Hartmann, A. Stachs, B. Gerber, Local treatment of the axilla in early breast cancer: concepts from the national surgical adjuvant breast and bowel project B-04 to the planned intergroup sentinel mamma trial, Breast Care (Basel) 9 (2) (2014) 87–95.
Contents lists available at ScienceDirect
European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad
Contrast-enhanced ultrasound -guided axillary lymph node core biopsy: Diagnostic accuracy in preoperative staging of invasive breast cancer Rautiainen Suvi a,f,∗ , Sudah Mazen a , Joukainen Sarianna b , Sironen Reijo c,e,f , Vanninen Ritva a,d,f , Sutela Anna a a
Department of Clinical Radiology, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland Department of Plastic Surgery, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland c Department of Clinical Pathology, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland d Unit of Radiology, Biocenter Kuopio, University of Eastern Finland, P.O.B. 1627, FI-70211 Kuopio, Finland e Unit of Pathology and Forensic Medicine at Institute of Clinical Medicine, Biocenter Kuopio, University of Eastern Finland, P.O.B. 1627, FI-70211 Kuopio, Finland f Cancer Center of Eastern Finland, University of Eastern Finland, P.O.B. 1627, FI-70211 Kuopio, Finland b
a r t i c l e
i n f o
Article history: Received 1 June 2015 Received in revised form 9 July 2015 Accepted 12 August 2015 Keywords: Breast neoplasms Axilla Biopsy Ultrasonography Contrast media
a b s t r a c t Objectives: To evaluate accuracy of contrast enhanced ultrasound (CEUS)-sentinel procedure followed by core biopsy (CB) and marking in patients with breast cancer. To compare the axillary metastatic tumour burden in patients with positive vs. negative CB results. Methods: Two radiologists in our tertiary care hospital performed axillary CEUS sentinel procedures on consecutive US node negative breast cancer patients. The first enhancing lymph node (LN) was core biopsied and marked with a breast coil. The results were compared to final histopathology. We analysed the diagnostic performance of CEUS CB and its ability to detect patients with higher axillary burden (>2 metastasis). Results: During the study period between January 2013 and December 2014, altogether 54 patients (mean age 60.4 years) were included in the statistical analysis. The sensitivity for CEUS CB was 66.7%, specificity 100%, PPV 100%, NPV 93.8% and overall accuracy 94.4%. The method correctly recognised all the axillae with higher tumour burdens (sensitivity 100%, N = 3) and 59.3% of coils marking the LNs were discovered. Conclusion: CEUS -guided axillary CB proved to be feasible and accurate procedure with moderate sensitivity and it clearly identified the higher axillary tumour burden. The coil marking of LNs as used cannot be recommended. In clinical routine, CEUS procedure might be recommended in selective patient populations. © 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction During the past few decades, staging of the axilla in patients with breast cancer has shifted from routine axillary lymph node dissec-
Abbreviations: ALND, axillary lymph node dissection; CB, core biopsy; CEUS, contrast-enhanced ultrasound; FNA, fine-needle aspiration; LN, lymph node; SLN, sentinel lymph node; SLNB, sentinel lymph node biopsy; UNB, ultrasound-guided needle biopsy; US, ultrasound. ∗ Corresponding author at: Department of Clinical Pathology, Kuopio University Hospital, Kuopio, P.O. Box 100, Puijonlaaksontie 2, 70029 KYS, Finland. Fax: +358 17 173 341. E-mail addresses: suvi.rautiainen@kuh.fi (S. Rautiainen), mazen.sudah@kuh.fi (M. Sudah), sarianna.joukainen@kuh.fi (S. Joukainen), reijo.sironen@kuh.fi (R. Sironen), ritva.vanninen@kuh.fi (R. Vanninen), anna.sutela@kuh.fi (A. Sutela). http://dx.doi.org/10.1016/j.ejrad.2015.08.006 0720-048X/© 2015 Elsevier Ireland Ltd. All rights reserved.
tion (ALND) towards sentinel lymph node biopsy (SLNB), which has become the standard of care. SLNB has achieved high sensitivity (97–98%), with similar survival outcomes but with less morbidity compared to ALND [1,2]. Moreover, the need for routine ALND after positive SLNB findings was recently challenged by randomized trials (ACOSOG and AMAROS) in patients with T1-2 tumours and clinically node negative axilla [3,4]. The concept of low (<2 positive LNs) vs. high axillary tumour burden (>2 positive LNs) has recently been introduced also in radiological studies, which aim to determine the role of preoperative investigations in the evolving axillary staging [5–7]. Although the search for a non-invasive approach to preoperatively stage the axilla has attracted much interest, the diagnostic figures have not achieved the results of SLNB [6,8,9]. The sensitivity and specificity of axillary ultrasound (US) has been reported
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
between 61% - 62% and 82% - 86%, respectively [6,10]. US-guided needle biopsy (UNB) has improved the sensitivity (80%) with perfect specificity (100%) [6,9,10] and finds approximately half of axillary metastasis preoperatively [11]. The US-positive axillae is more frequently associated with higher axillary tumour burden (>2 positive LNs) [12] and the metastases found by US tend to be larger than average [13,14]. However, the majority of patients with small breast tumours have a low axillary tumour burden [15,16] and more accurate preoperative staging would be desired. The false negative findings in axillary UNB can arise from failure to sample the sentinel lymph node (SLN) (45%), or the metastatic deposit within the LN (55%) [13]. The latter can be reduced by using core biopsy (CB) instead of fine needle aspiration (FNA) [17,18]. Moreover, the ability to correctly localize the SLN during biopsy would improve the detection rate of metastatic LNs. The use of contrast enhanced US (CEUS) to localize the SLN has shown it to be both safe and feasible [19–21]. In comparison to traditional isotope SLNB and Patent Blue dye, the sensitivity of CEUS to detect SLN correctly was 89% [20]. In clinically node negative patients, the sensitivity of CEUS -guided biopsy (with either CB or FNA) was 61–65% [21,22]. However, the literature on CEUS -guided SLN biopsies is still scanty and details such as localization methods prior to subsequent surgical excision remains to be determined [20,23]. The goals of the present study were: (1) to assess the accuracy of CEUS-sentinel procedure followed by CB in patients with no ultrasonographic signs indicative of axillary metastasis and (2) to evaluate the metastatic tumour burden in patients with positive vs. negative CB results and (3) to investigate the accuracy of marking the CEUS SLN with a breast coil. 2. Materials and methods This prospective study was approved by the National Medicines Agency (EudraCT 2012-005349-20) and by the hospital ethics committee according to the recommendation from the national ethical board for medical research. All consecutive newly diagnosed patients at least 18 years of age and with invasive breast cancer or BI-RADS 5 breast lesions were invited to participate in the study. The exclusion criteria were as follows: patients who could not provide written informed consent; clinically suspicious finding of axillary metastasis; history of previous ipsilateral breast cancer, axillary surgery or radiotherapy; no axillary surgery planned or performed; contraindication for the use of US contrast agent. In addition, exclusion criteria for axillae in the grey scale US study included morphological abnormality of the LNs were any of the following findings: any cortical thickening ≥5.0 mm or dislocated or absent fatty hilum. No limits were set for minimum thickness of the axillary LN, breast tumour size or tumour multifocality. During the study period between January 2013 and December 2014, the patients in this study were referred from a screening centre, two district hospitals and private institutions. All patients were evaluated by a breast radiologist before surgery and breast lesions were classified according to the ACR BI-RADS® lexicon (American College of Radiology, Breast Imaging Reporting and Data System) [24,25]. The findings were evaluated at a multidisciplinary breast meeting. 2.1. US—contrast enhancement studies and biopsies The CEUS procedures and consecutive biopsies were performed by two breast radiologists with 5 and 21 years of experience in breast imaging and interventions (SR, MS). Before the start of the study, both radiologists had limited experience in performing axillary CEUS procedures. The US examinations were undertaken
2131
with an Esaote MyLabClassC (Esaote S.p.A., Genova, Italy) with a 7–13 MHz linear transducer and the contrast studies using the contrast-specific software program and linear probe. During the grey scale axillary US assessment, special attention was paid to the typical location of the SLN at the caudal part of level I. If no exclusion criteria were present, the possible SLN with the thickest cortex at that location was documented. If during CEUS procedure this LN was found not the enhancing sentinel node, then the enhanced LN was biopsied and CEUS was deemed to have changed the biopsy plan. The morphology of the CEUS LN was noted as was the number of other enhancing LNs. All US and CEUS findings were recorded to a structured study form. 2.1.1. CEUS and biopsy procedures The CEUS procedure was performed according to Sever et al. [20]. In short; the contrast agent powder, SonoVue® (Bracco Imaging, Milan, Italy) was mixed with 2.0 ml of sterile saline. The contrast was then injected intradermally into the upper outer quadrant close to the areola with a 26G needle using 1.0 ml syringe. The amount of a single injection was approximately 0.4 ml and the injection was repeated 1–2 times if needed. Breast and axillae were scanned with US and the contrast injection area was massaged for 5–15 s in cases where no enhancing lymph vessels or axillary LNs could be visualized. The rate of lymph vessel enhancement was categorised as (1) spontaneous, (2) delayed, or (3) delayed after massage. The core biopsies were obtained from the first enhancing LN aiming at the cortex and subcapsular area using a 16G core-needle with 20-mm notch exposure (Temno Evolution, CareFusion Corporation, McGaw Park, IL, USA) or with an automated CNB gun with a 22-mm throw (Bard Magnum; Bard Biopsy Systems, Tempe, AZ, USA). A total of two samples that visually included solid non-fatty tissue were obtained. More biopsies were collected at the discretion of the radiologist if the first biopsies were deemed as being nonrepresentative. In the case where two or more LNs were enhancing, then only the first LN to enhance was biopsied (Fig. 1). The successfully biopsied and enhanced LNs were marked with a breast coil (MReye Breast Localization coil, COOK Medical, Bloomington, IN, USA) adjacent to the biopsy area, for possible further recognition. The total number of enhancing LNs and biopsies were recorded, as well as any possible complications. The procedure was deemed as unsuccessful if no axillary LN enhancement could be observed. 2.1.2. Patient management and histopathological evaluation Core biopsy specimens were placed into 10% formalin and embedded in paraffin after fixation. The samples were cut into 5m slices at 4 different levels then stained with haematoxylin and eosin (HE). CB samples were evaluated by 2 pathologists, first at diagnosis, then at the multidisciplinary meeting. Patients underwent SLNB in case axillary CB showed micrometastasis and were directed to ALND in the cases where the preoperative CB revealed metastases. The CEUS-positive LNs were recognized from the evacuation specimens either by the surgeon during the operation or from specimen X-rays and were marked separately for the pathologist. If the biopsy result was negative, then SLNB was performed according to a standardized 2-day protocol including injections of technetium-labelled nanocolloid (Nanocoll, Technetium Tc-99M Albumin Aggregated Kit, GE Healthcare, Milan, Italy) and Patent Blue dye preoperatively as well as the intraoperative frozen sections. During surgery, the performing breast surgeon recognized the CEUS LN with the coil and removed this and any radioactive LN’s and any other clinically suspicious LNs. After formalin fixation, all of the obtained LNs were paraffin embedded, serially sectioned into 2-mm slices and stained with HE. For the detection of micrometastases, an immunohistochem-
2132
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
Fig. 1. US images of a 40-year-old breast cancer patient with a multifocal breast tumour BI-RADS 5. (a) An enhancing axillary LN with (b) visual afferent lymph vessel (arrow) was identified in CEUS. (c) This was the first and only visual enhancing LN which in the grey scale US had a cortical thickness of 2.7 mm. (d) Metastasis was detected by core biopsy (arrow heads) and confirmed at final histopathology, which showed 3/15 metastatic LNs after axillary dissection. Table 1 Patient demographics and tumour characteristics (N = 54). Characteristic
N
(%)
Tumour pathological T classification T1 T2 T3 T4
35 15 4 0
64.8 27.8 7.4 0
Tumour N classification N0 N1 N1 (mi) N2 N3
42 7 3 1 1
77.8 13.0 5.6 1.9 1.9
Stage 1 2 3 4
32 20 2 0
59.3 37.0 3.7 0
Tumour histology Ductal Lobular Mucinous Other
42 9 1 2
77.8 16.7 1.9 3.7
Grade of invasive carcinoma 1 2 3 Multifocality
9 34 11 13
16.7 63.0 20.4 24.1
ical staining was then performed against low molecular weight cytokeratin (Techmate 500 Plus, DAKO, Denmark) using a labelled polymer-based method (DAKO kit K4068, low molecular cytokeratin, Cam 5.2, Becton Dickinson, NJ, USA). If the frozen section during surgery was positive for either macro- or micrometastasis, the patient underwent ALND. Patients with negative frozen sections but with micrometastasis at final histology were treated according to the recommendation of the multidisciplinary group experts, either with completion ALND or follow-up. LNs were deemed positive for malignancy if they contained macrometastases (>2 mm). Both micrometastases (>0.2–2.0 mm) and isolated tumour cells (ITC, <0.2 mm) on histology were classified as negative. In the further subgroup analysis, the axillary positive patients were divided according to the number of LNs with macrometastases; high axillary burden with >2 positive LNs and low burden with 1–2 positive LNs. 2.1.3. Statistical analysis Statistical analysis was performed with SPSS version 19 (IBM Corporation, Somers, NY, USA). SLNB-negative as well as micrometastases and isolated tumour cells (ITC) were classified as non-metastatic LNs. Diagnostic performance (sensitivity, specificity, positive predictive value, negative predictive value and overall accuracy) was calculated for CEUS CB. Positive and negative predictive values were calculated with Bayes’ formula [26] and 95% confidence intervals were calculated with Wilson’s nocontinuity correction formula [27]. In the subgroup analysis, the statistical comparisons were performed using McNemar’s test or
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
2133
Fig. 2. A unifocal 15 mm breast tumour (**) of a 49-year-old woman (a). In the CEUS study (b) the contrast-filled lymph vessel (arrow) is amputated at the site of the tumour (**) with lymphovascular invasion in the final histopathology. In addition, the traditional isotope sentinel procedure failed and the ALND was performed with 2/17 metastatic LNs (largest macrometastasis 4 mm).
Fisher’s exact test. p < 0.05 was considered to indicate statistical significance. 3. Results During the study period between January 2013 and December 2014, a total of 60 patients fulfilled the inclusion criteria and were
enrolled. The success rate of the CEUS procedure was 95.0% (57/60). In two cases, the lymph vessels were detected but no enhancing LNs could be observed. In the third patient, a visual amputation of the enhancing lymph vessel by breast tumour infiltration was detected (Fig. 2), also with no enhancing LNs. In two cases, it was not possible to perform the CB due to technical or safety reasons. Another patient with DCIS at CB and final histology was also excluded.
Fig. 3. Flow chart of patients.
2134
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
Table 2 (a–b) The contrast study details of the patients (N = 54) with enhancing axillary lymph nodes (LN). (a) Characteristic CEUS LN as the lowest axillary LN Enhancement type Spontaneous Delayed Delayed after massage Coils localized by palpation/X-ray (b) Characteristic Injection volume (ml) Number of injections per axilla CEUS LN’s features Cortical thickness (mm) Length (mm) Width (mm)
N
(%)
41
75.9
41 8 5 32
83.3 14.8 9.3 59.3
Mean
Range
0.4 2.2
0.1–0.5 1–3
2.4 12.8 5.8
0.2–5.0 5.8–25.0 2.5–10.0
Table 3 Information of the axillary positive patients’ (N = 12) lymph node morphology in preoperative ultrasound and the core biopsy results in comparison to the histopathology of the axillary metastases and breast tumour. No
LN cortex (mm)
CB true positive
1 2 3 4 5 6 7 8 9 10 11 12
3.0 3.7 5.0 3.4 1.9 2.2 2.7 3.4 2.1 2.4 2.3 3.9
Yes Yes Yes Yes No Yes No Yes No No Yes No
Metastatic LN 3 2 1 3 2 1 1 6 1 1 mi 1 mi 1 mi
Size of metastatic deposit (mm)
Tumour final size (mm)
Tumour histology
Grade
Multifocality
14.0 11.0 11.0 6.0 4.0 3.0 2.1 10.0 10.0 2.0 1.0 0.2
36.0 70.0 35.0 36.0 17.0 15.0 10.0 25.0 22.0 11.0 18.0 29.0
ILC ILC IDC IDC IDC IDC IDC IDC IDC ILC IDC IDC
3 2 3 3 2 2 1 3 3 2 2 2
Yes Yes No Yes No No No Yes No Yes Yes No
mi = micrometastasis. ILC = invasive lobular carcinoma. IDC = invasive ductal carcinoma. Table 4 The diagnostic performance of the contrast enhanced ultrasound—guided axillary lymph node core biopsy and the comparison to the literature.
Present study CEUS + CB CEUS + CB Cox et al. [22] Cox et al. [22] CEUS + FNA or CB Sever et al. [20]CEUS + FNA or CB
N
Sensitivity % with 95% CI Specificity % with 95% CI PPV % with 95% CI
NPV % with 95% CI
Overall accuracy % with 95% CI
54 235 295 126
66.7 [6/9]35.4–88.0 65.0 59.0–71.0 61.0 56.0–66.0 65.0
93.8 [45/48]83.2–97.9 95.0 81.0 92.0
94.4 [51/54]84.8–98.1 n n n
100 [45/45]92.1–100 100 100 100
100 [6/6]61.0–100 100 100 100
Diagnostic performance figures calculated against the macrometastases in SNLB or ALND. Positive and negative predictive values were calculated by using the Bayes formula with an estimate of prevalence of 16.7% for macrometastasis [26]. The 95% confidence intervals were calculated by using the Wilson no-continuity correction formula [27]. Data in brackets [] are nominators and denominators.
Hence, a total of 54 patients were included in the statistical analysis (Fig. 3). The mean age of the patients was 60.4 years (range 34–88 years). The tumour charasteristics are shown in Table 1. Marking of the CEUS LNs with breast coil was successful in 59.3% of cases (32/54). The detailed information on the contrast studies are presented in Table 2a and b. The CEUS procedure changed the biopsy target LN in 24.0% (13/54) of the cases compared to grey scale US. Three of these were metastatic LNs (in two cases, this was the only metastastic LN). In one case (1.9%, 1/54), the core biopsy specimen was considered as inadequate (cortical thickness 1.1 mm), and the axilla was determined as being benign in SLNB. Axillary macrometastases were found in the final histopathology in 16.7% (9/54) of the patients with invasive carcinoma. In multifocal tumours (N = 13, 24.1%), axillary macrometastases were found in 4 (30.8%) cases and in unifocal tumours (N = 41, 75.9%) in 2 (4.9%). A total of 6 axillae had either micrometastases (N = 3)
or ITC (N = 3). CEUS CB was false negative in three (33.3%) of the macrometastases having a maximum size of 2.1–10.0 mm. The detailed information on the axillary positive patients is presented in Table 3. CEUS -guided CB had a sensitivity of 66.7% and a specificity of 100% to detect metastases. The detailed diagnostic performance of CEUS-CB is shown in Table 4 with a comparison to the literature. A total of 3 of the 9 patients with axillary macrometastases had a high axillary tumour burden (>2 metastatic LNs) while 6 patients had a low axillary tumour burden (1–2 metastatic LNs). The CEUS CB was true positive in all of the patients with a high axillary tumour burden (3/3) and in half of the patients with a low tumour burden (3/6) (p = 0.081). When the high vs. low axillary tumour burden groups were compared, no statistically significant differences in US features could be detected i.e. mean number of LNs with >2 mm cortex was 1.3 (high) vs. 0.7 (low) (p = 0.149), cortical thickness maximum 3.3 mm
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
(high) vs. 2.9 mm (low) mm (p = 0.268) and CEUS LN’s mean cortical thickness 3.0 mm (high) vs. 2.8 mm (low) (p = 0.548).
4. Discussion Previously reports from a single institution have described the technical feasibility of axillary CEUS biopsy procedure, which detects additional metastases in nodes interpreted as normal in grey scale US [21,22]. Our comparable results confirm that axillary CEUS procedure can be obtained to clinical use without hands-on training and is a feasible, consistent procedure. The low sensitivities in previous studies (61.0–65.0%) and the present (66.7%), can be attributed mainly to the strict inclusion criteria of US node negativity and the low prevalence of axillary metastases. It is well known that a LN’s smooth cortex is less predictive for metastasis, while other morphological features such as the absence of a central fatty hilum and multi-lobulated cortical thickening, are highly predictive for metastasis [12–14]. The LNs with the latter features were excluded from this study. Nevertheless, CEUS CB was able to correctly detect the majority of axillary metastasis. Unfortunately, a few patients with macrometastasis were not identified. The CEUS LN may not be the metastatic node or the biopsy may hit an area free of metastasis within the correct SLN especially in case of very small macrometastasis. During the SLNB procedure, often more than one SLN is removed, leading to higher sensitivity. In addition, CB is not technically feasible in all patients due to inadequate samples, even if more accurate than FNA [17,18]. In our study, there were fewer inadequate samples than previously reported for CEUS CB, reflecting our longstanding experience with axillary CB. The CEUS -guided biopsy can accidentally detect small tumour deposits (micrometastases or ITC), which might lead to ALND and thus to overtreatment. This risk should be acknowledged and the patients with these findings should be subjected to SLNB. Therefore, although CEUS is useful in its current form, it cannot replace SLNB in routine clinical practice. Our study confirms that CEUS CB procedure is safe. The precise intradermal contrast injection is essential but, fortunately, the injections can be safely repeated with no adverse effects. In the majority of patients, the contrast filled the lymph vessels immediately and could be followed directly to the axillary LNs, which makes it attractive when searching for the SLN. The fact that CEUS changed the biopsy plan compared to grey scale US in a considerable number of patients might indicate that CEUS could increase the sensitivity of axillary needle biopsies. Recent results emerging from randomized controlled trials (ACOSOG, AMAROS) [3,4] have led to a less aggressive axillary staging in patients fulfilling the studies’ inclusion criteria in some countries. In our study, CEUS CB accurately excluded the high axillary burden in otherwise US node negative axillae. Nonetheless the number of events is too small to draw firm conclusions on this issue, which is a limitation of this study. The absence of LN’s pathological features correlated well with the axillary tumour burden, as only 5% (3/54) had macrometastasis in >2 LNs. On the other hand, the majority of the patients had no axillary metastasis and thus underwent additional intervention, questioning the utility of CEUS CB as routine clinical procedure. However, CEUS CB may prove useful for selective patient groups i.e. when ACOSOG or AMAROS criteria are not fulfilled or when the presence of metastasis would change the treatment plans, or upon the recommendations of multidisciplinary team. The localization and marking the CEUS LN is another topic of interest. A relatively large and palpable coil was chosen in this study, which was believed to be well localized. This was not the case since only 59% of coils could be localized, corresponding to the results of a recent study [23]. The abundant axillary fat around the
2135
small LNs combined with the different position of patients during CEUS and SLNB procedure might well explain the failure. Preoperative axillary US is a topic of considerable interest and changes in practices are inevitable. At the moment, the lack of clarity and guidelines is confusing. Two ongoing randomized trials (INSEMA, SOUND) will hopefully shed more light on the future role of axillary US and US-guided biopsies [28,29]. Positive outcomes might increase the future role of axillary US and consequently CEUS procedures. In conclusion, CEUS -guided axillary CB in newly diagnosed breast cancer is a feasible and accurate procedure with moderate sensitivity in node negative patients. It detected all patients with high axillary tumour burden. The coil marking beside the CEUS SLN cannot be recommended. In routine clinical practice, the CEUS CB procedure might be recommended in selective patient populations. Acknowledgements This study was supported in part by Kuopio University Hospital (grant number 5063525), Kuopio University Foundation, Sirkka and Mauri Wiljasalo Foundation, Instrumentarium Science Foundation and Cancer Center of Eastern Finland provided by the University of Eastern Finland. We express our special appreciation for to nurse Tiina Föhr for her help in the recruitment of patients to the study and Tuomas Selander M.Sc. for the statistical advice. References [1] D.N. Krag, S.J. Anderson, T.B. Julian, et al., Sentinel-lymph-node resection compared with conventional axillary-lymph-node dissection in clinically node-negative patients with breast cancer: overall survival findings from the NSABP B-32 randomised phase 3 trial, Lancet Oncol. 11 (10) (2010) 927–933. [2] R.E. Mansel, L. Fallowfield, M. Kissin, et al., Randomized multicenter trial of sentinel node biopsy versus standard axillary treatment in operable breast cancer: the ALMANAC trial, J. Natl. Cancer Inst. 98 (9) (2006) 599–609. [3] A.E. Giuliano, K.K. Hunt, K.V. Ballman, et al., Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: a randomized clinical trial, JAMA 305 (6) (2011) 569–575. [4] M. Donker, G. van Tienhoven, M.E. Straver, et al., Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981–22023 AMAROS): a randomised, multicentre, open-label, phase 3 non-inferiority trial, Lancet Oncol. 15 (12) (2014) 1303–1310. [5] T.J. Hieken, B.C. Trull, J.C. Boughey, et al., Preoperative axillary imaging with percutaneous lymph node biopsy is valuable in the contemporary management of patients with breast cancer, Surgery 154 (4) (2013) 831–838, discussion 838–40. [6] N. Houssami, R.M. Turner, Staging the axilla in women with breast cancer: the utility of preoperative ultrasound-guided needle biopsy, Cancer Biol. Med. 11 (2) (2014) 69–77. [7] C. Reyna, J.V. Kiluk, A. Frelick, N. Khakpour, C. Laronga, M.C. Lee, Impact of axillary ultrasound (AUS) on axillary dissection in breast conserving surgery (BCS), J. Surg. Oncol. 111 (4) (2015) 813–818. [8] S.E. Harnan, K.L. Cooper, Y. Meng, et al., Magnetic resonance for assessment of axillary lymph node status in early breast cancer: a systematic review and meta-analysis, Eur. J. Surg. Oncol. 37 (11) (2011) 928–936. [9] N. Houssami, S. Ciatto, R.M. Turner, H.S. Cody 3rd, P. Macaskill, 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. 254 (2) (2011) 243–251. [10] S. Alvarez, E. Anorbe, P. Alcorta, F. Lopez, I. Alonso, J. Cortes, Role of sonography in the diagnosis of axillary lymph node metastases in breast cancer: a systematic review, AJR Am. J. Roentgenol. 186 (5) (2006) 1342–1348. [11] S.C. Diepstraten, A.R. Sever, C.F. Buckens, 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. 21 (1) (2014) 51–59. [12] N. Houssami, S.C. Diepstraten, H.S. Cody 3rd, R.M. Turner, A.R. Sever, Clinical utility of ultrasound-needle biopsy for preoperative staging of the axilla in invasive breast cancer, Anticancer Res. 34 (3) (2014) 1087–1097. [13] P.D. Britton, E. Provenzano, S. Barter, et al., Ultrasound guided percutaneous axillary lymph node core biopsy: how often is the sentinel lymph node being biopsied? Breast 18 (1) (2009) 13–16. [14] M.J. Garcia-Ortega, M.A. Benito, E.F. Vahamonde, P.R. Torres, A.B. Velasco, M.M. Paredes, Pretreatment axillary ultrasonography and core biopsy in patients with suspected breast cancer: diagnostic accuracy and impact on management, Eur. J. Radiol. 79 (1) (2011) 64–72.
2136
S. Rautiainen et al. / European Journal of Radiology 84 (2015) 2130–2136
[15] R.M. Nadeem, The feasibility of a randomised controlled trial for the axillary management of a select group of invasive breast cancer patients: SLNB vs. no-SLNB, Breast Cancer 22 (4) (2015) 343–349. [16] J. Cools-Lartigue, A. Sinclair, N. Trabulsi, et al., Preoperative axillary ultrasound and fine-needle aspiration biopsy in the diagnosis of axillary metastases in patients with breast cancer: predictors of accuracy and future implications, Ann. Surg. Oncol. 20 (3) (2013) 819–827. [17] S. Rautiainen, A. Masarwah, M. Sudah, et al., Axillary lymph node biopsy in newly diagnosed invasive breast cancer: comparative accuracy of fine-needle aspiration biopsy versus core-needle biopsy, Radiology 269 (1) (2013) 54–60. [18] M.A. Ganott, M.L. Zuley, G.S. Abrams, et al., Ultrasound guided core biopsy versus fine needle aspiration for evaluation of axillary lymphadenopathy in patients with breast cancer, ISRN Oncol. 2014 (Feb 4) (2014) 703160. [19] A. Sever, S. Jones, K. Cox, J. Weeks, P. Mills, P. Jones, Preoperative localization of sentinel lymph nodes using intradermal microbubbles and contrast-enhanced ultrasonography in patients with breast cancer, Br. J. Surg. 96 (11) (2009) 1295–1299. [20] A.R. Sever, P. Mills, S.E. Jones, W. Mali, P.A. Jones, Sentinel node identification using microbubbles and contrast-enhanced ultrasonography, Clin. Radiol. 67 (7) (2012) 687–694. [21] A.R. Sever, P. Mills, J. Weeks, et al., Preoperative needle biopsy of sentinel lymph nodes using intradermal microbubbles and contrast-enhanced ultrasound in patients with breast cancer, AJR Am. J. Roentgenol. 199 (2) (2012) 465–470.
[22] K. Cox, A. Sever, S. Jones, et al., Validation of a technique using microbubbles and contrast enhanced ultrasound (CEUS) to biopsy sentinel lymph nodes (SLN) in pre-operative breast cancer patients with a normal grey-scale axillary ultrasound, Eur. J. Surg. Oncol. 39 (7) (2013) 760–765. [23] M.W. Barentsz, H.M. Verkooijen, R.M. Pijnappel, et al., Sentinel lymph node localization with contrast-enhanced ultrasound and an I-125 seed: an ideal prospective development study, Int. J. Surg. 14 (2015) 1–6. [24] C. D’Orsi, E. Mendelson, D. Ikeda, et al., Breast Imaging Reporting and Data System: ACR BI-RADS-Breast Imaging Atlas, American College of Radiology, Reston, VA, 2003. [25] C.L. Mercado, BI-RADS update, Radiol. Clin. North Am. 52 (3) (2014) 481–487. [26] R.F.S. Fletcher, Clinical Epidemiology: The Essentials, 4th ed., Lippincott Williams & Wilkins, Baltimore, MD, 2005, pp. 45–48. [27] R.G. Newcombe, Two-sided confidence intervals for the single proportion: comparison of seven methods, Stat. Med. 17 (8) (1998) 857–872. [28] O. Gentilini, U. Veronesi, Abandoning sentinel lymph node biopsy in early breast cancer? A new trial in progress at the European Institute of Oncology of Milan (SOUND: sentinel node vs observation after axillary UltraSouND), Breast 21 (5) (2012) 678–681. [29] T. Reimer, S. Hartmann, A. Stachs, B. Gerber, Local treatment of the axilla in early breast cancer: concepts from the national surgical adjuvant breast and bowel project B-04 to the planned intergroup sentinel mamma trial, Breast Care (Basel) 9 (2) (2014) 87–95.