Abstract
Objective:
Clips are often placed to mark axillary nodes with biopsy-confirmed metastases in breast cancer patients. Evaluation of the clipped node after chemotherapy can identify patients who have eradication of nodal disease. The goal of this study was to determine if preoperative fine needle aspiration of clipped nodes after neoadjuvant chemotherapy could predict the presence of residual disease.
Methods:
This prospective registry study enrolled 50 patients with a clip placed to mark nodes with biopsy-confirmed metastasis who completed neoadjuvant chemotherapy. Participants had a fine needle aspiration performed of the clipped node before seed-localized lymph node excision. Fine needle aspiration pathology was compared to surgical pathology.
Results:
Of the 36 (72%) with residual disease on surgical pathology, 3 (8%) had a nondiagnostic aspirate, carcinoma was seen in 14 (39%), and 19 (53%) had a false negative result. The sensitivity for fine needle aspiration was 42.4%, specificity 100%, and negative predictive value of 40.6%. In univariate analysis, the odds of a true positive result increased significantly with the mean initial size of the clipped node (OR 4.3, p=0.004) and the size of the metastatic focus after neoadjuvant chemotherapy (OR 1.3, p=0.006) while normalization of nodes after chemotherapy (OR 0.1) or lack of response on ultrasound (OR=0.11) were associated with a false negative result (p=0.01).
Conclusion:
Fine needle aspiration of marked nodes after chemotherapy has a high false negative rate. This highlights the need for surgical staging of the axilla after neoadjuvant chemotherapy to assess response.
Keywords: breast neoplasms, lymph nodes, chemotherapy, surgery, fine needle biopsy
Precis:
Clips can be placed to mark nodes with biopsy-confirmed metastases. The goal of this study was to determine if needle biopsy of the clipped node after chemotherapy can predict nodal response.
The status of axillary lymph nodes is an important prognostic feature in breast cancer patients. Because of the impact of nodal disease on management, ultrasound is often used to evaluate newly diagnosed patients. Suspicious lymph nodes identified by ultrasound can be biopsied to assess for metastases. Fine needle aspiration (FNA) enables immediate on-site assessment and real-time staging which expedites treatment planning.1–3 This approach has a superior sensitivity compared to ultrasound alone or physical examination.2,4 Advocates of nodal ultrasound with FNA point to potential cost savings because unnecessary surgical staging procedures, specifically sentinel lymph node dissection (SLND), can be omitted if metastases are identified preoperatively.5,6
Patients with pathologically-confirmed nodal metastasis often receive neoadjuvant chemotherapy (NAC) which can result in conversion to pathologic node negative status in 21–75% of patients depending on tumor subtype.7,8 Historically, patients with proven nodal metastasis have required completion axillary lymph node dissection (ALND) after chemotherapy regardless of the response. However, recent trials have opened consideration for minimally invasive surgical techniques. The ACOSOG Z1071 trial was designed to determine if SLND could accurately assess nodal response after NAC thus identifying patients that could potentially avoid ALND.9 While the reported false negative rate (FNR) of 12.6% exceeded the prespecified success threshold of 10%, an interesting observation was made in subsequent analyses. In 170 patients a clip was placed to mark the node with biopsy-confirmed metastases prior to chemotherapy. When this clipped node was retrieved as a SLN (N=107), the FNR of SLND was lowered to 6.8%.10 The utility of evaluating the clipped node specifically was confirmed in a subsequent single-institution registry trial that reported a FNR of 4.2% for evaluation of the clipped node after NAC.11 Assessment of the node known to initially contain metastasis is a logical step in axillary staging and has been incorporated into NCCN guidelines.12 There are now reports of surgical techniques such as targeted axillary dissection, which ensures removal of the clipped node along with SLND, showing FNRs as low as 2% demonstrating that evaluation of the clipped node is critical to axillary assessment after NAC.11,13
While the use of SLND and techniques ensuring removal of the clipped node are gaining acceptance,14 all of these approaches require pre-operative procedures such as injections or placement of seeds as well as the need for surgical staging. The identification of residual nodal metastases still requires ALND unless the patient is enrolled on a clinical trial that dictates a different therapy. Preoperative identification of residual disease would allow for improved surgical planning for completion ALND when indicated without the unnecessary procedures and expense required for SLND or marking and removal of clipped nodes. Because the node with confirmed metastasis can be marked and followed through chemotherapy, there is potential use for FNA, which is proven to be sensitive and specific in the pre-chemotherapy setting, after completion of NAC to assess response. However, there is currently no data concerning the accuracy of FNA of axillary nodes in the post-NAC setting. The goal of this study was to determine if FNA of the clipped lymph node reliably identifies residual nodal disease after NAC.
Methods
Study Design
This prospective study was conducted between 2014 and 2017 with approval from the Institutional Review Board. To be eligible for enrollment, patients had to have biopsy-proven axillary node metastasis with a clip placed in the node at the time of biopsy, complete NAC, and be scheduled for seed-localized excision of the clipped node at the time of surgery. Eligible patients were enrolled after signing informed consent.
Study Protocol
Per our institutional standard, all patients had routine imaging with mammogram and ultrasound of the breast and regional nodal basins at the time of diagnosis. Nodes with an abnormal appearance on ultrasound either because of size or morphology2 were biopsied using a 21 gauge needle, stained by Papanicolaou and Diff-Quick methods with on-site evaluation by a dedicated breast cytopathologist.1 The diagnosis of carcinoma was based on the presence of features such as enlarged nucleus, nuclear membrane irregularities, prominent nucleolus, scant or moderate amounts of delicate or vacuolated cytoplasm, and mitotic figures in tumor cells. If more than one abnormal node was seen, the most suspicious node was biopsied. If nodal metastasis was confirmed pathologically, a clip was placed in the biopsied node. All enrolled patients received anthracycline- and/or taxane-based chemotherapy regimens. Patients with HER2-positive tumors had additional HER2-targeted therapy.
After the completion of chemotherapy, the surgical approach was determined by the treating breast surgical oncologist. In order to localize the clip-containing lymph nodes intra-operatively, patients had ultrasound-guided I125 radioactive seed placement in the clipped lymph node up to 5 days before surgery. Trial participants underwent FNA with a 21-gauge needle under ultrasound guidance during this encounter before the seed was placed. A sample was considered “nondiagnostic” if no cellular tissue (either lymphoid or carcinoma) was seen in the aspirate.
During surgery, the clip-containing nodes were removed separately and sent for intra-operative radiograph. All enrolled patients had a seed and clip identified in the specimen. The nodes were serially sectioned and processed using routine hematoxylin & eosin staining with the addition of pancytokeratin staining at the discretion of the pathologist. Any residual metastatic focus, including isolated tumor cells, was considered as node positive.
Data Analysis
Data was analyzed using R version 3.3.1.15 The association between categorical variables and clinical outcomes was assessed using Fisher’s exact test. Continuous variables were assessed using unequal variance t-test. Among patients with residual disease, logistic regression was used to evaluate clinical and imaging characteristics for association with the occurrence of a true positive biopsy result. Firth penalized logistic models were fitted for covariates with zero count cells. All statistical tests used a significance level of 0.05.
Results
Fifty patients were enrolled. Clinicopathologic features of the participants are summarized in Table 1. Twenty-one (42%) patients had one abnormal node on the initial pretreatment ultrasound, 9 (18%) had 2 abnormal nodes, 8 (16%) had 3 nodes, and 12 (24%) had ≥ 4 suspicious nodes. The last US performed during or after NAC showed no nodal response or progression in 10% (n=5), improvement but not normalization in 33% (n=16), normalization in 57% (n=28), and was not performed in one case. The last recorded size of the clipped node ranged from 0.4–4 cm with a mean of 1.3 cm.
Table 1 –
Characteristics of All Subjects
| All Patients | Path Node Positive | Path Node Negative | |
|---|---|---|---|
| N=50 | N=36 | N=14 | |
| Gender | |||
| Female | 49 (98%) | 35 (97%) | 14 (100%) |
| Male | 1 (2%) | 1 (3%) | |
| Mean Clinical Tumor Size | 3.9 cm (1–16.9) | 4.0 cm (1–16.9) | 3.6 cm (1.2–7.2) |
| Clinical T Category | |||
| T1 | 9 (18%) | 8 (22%) | 1 (7%) |
| T2 | 29 (58%) | 19 (53%) | 10 (71%) |
| T3 | 11 (22%) | 8 (22%) | 3 (21%) |
| T4 | 1 (2%) | 1 (3%) | 0 |
| Tumor Histology | |||
| Ductal | 47 (94%) | 33 (92%) | 14 (100%) |
| Lobular | 2 (4%) | 2 (6%) | |
| Other | 1 (2%) | 1 (3%) | |
| Tumor Subtype | |||
| HR+/HER2− | 32 (64%) | 26 (72%) | 6 (43%) |
| HR+/HER2+ | 10 (20%) | 6 (17%) | 4 (29%) |
| HR-/HER2+ | 4 (8%) | 1 (3%) | 3 (21%) |
| HR-/HER2− | 4 (8%) | 3 (8%) | 1 (7%) |
| Number of Abnormal Nodes on initial Ultrasound | |||
| 1 node | 21 (42%) | 12 (33%) | 9 (64%) |
| 2 nodes | 9 (18%) | 8 (22%) | 1 (7%) |
| 3 nodes | 8 (16%) | 5 (14%) | 3 (21%) |
| ≥4 nodes | 12 (24%) | 11 (31%) | 1 (7%) |
| Mean Size of Clipped Node Before NAC | 1.9 cm (0.5 – 5.2) | 2.0 cm (0.5–5.2) | 1.7 cm (1.2–2.8) |
| Presence of extra-axillary nodal disease | 5 (10%) | 4 (11%) | 1 (7%) |
| Ultrasound-assessed response to NAC | |||
| No response/progression | 5 (10%) | 5 (14%) | 0 |
| Partial Response | 16 (33%) | 14 (40%) | 2 (14%) |
| Normalization | 28 (57%) | 16 (46%) | 12 (86%) |
| Not performed | 1 | 1 | |
| Last recorded size of clipped node (Mean) | 1.3 cm (Range 0.4–4) | 1.3 cm (0.4–4) | 1.2 cm (0.7–2.3) |
| Breast Surgery | |||
| BCT | 24 (48%) | 16 (44%) | 8 (57%) |
| Mastectomy | 26 (52%) | 20 (56%) | 6 (43%) |
| Number of Pathologic Positive Nodes | |||
| 0 | 14 (28%) | 0 | 14 (100%) |
| 1 | 7 (14%) | 7 (19%) | |
| 2–3 | 15 (30%) | 14 (39%) | |
| ≥4 | 14 (28%) | 15 (42%) | |
| Size of clipped node metastasis (mean) | 8.3 mm | ||
BCT – breast conserving therapy; HR – hormone receptor; NAC – neoadjuvant chemotherapy
Of the 50 patients, 14 (28%) converted to pathologic node negative status. The proportion achieving nodal pCR varied by tumor subtype from 19% for HR+/HER2- tumors to 75% for HR-/HER2- tumors. The needle aspirate was non-diagnostic in 4 patients (1 pathologic node negative and 3 node positive). Of the 36 (72%) with residual disease on surgical pathology, 3 (8%) had a nondiagnostic aspirate, carcinoma was seen in 14 (39%), and 19 (53%) had a false negative result. Overall, 8% (4/50) had a nondiagnostic result, 54% (27/50) had an accurate results (either true negative or true positive), and 38% (19/50) had an inaccurate result. (Summarized in Figure 1) This results in sensitivity of 42.4% (95% CI 25.5–60.8), specificity of 100% (95% CI 75.3–100), and negative predictive value of 40.6% (95% CI 23.7–59.4).
Figure 1 1.
Study Schema
Ultrasound Characteristics after NAC
When looking at ultrasound characterization of nodal response, all patients (5/5) with no response or progression had residual nodal metastasis on surgical pathology. Four of these 5 (80%) had a false negative FNA result. In the 16 patients with a partial response on ultrasound, 14 (88%) had residual nodal disease. Four of these (28.6%) had a false negative FNA before surgery. Finally, 28 patients had normalization of the nodes on ultrasound even though 16 (57%) had nodal disease on surgical pathology. Of these 16, 2 had nondiagnostic FNA results while 11 of the remaining 14 (78.6%) had a false negative FNA biopsy (figure 2). The remaining patient who did not have an ultrasound performed after chemotherapy had residual nodal disease and a nondiagnostic FNA.
Figure 2 –
Correlation of ultrasound characteristics with Pathologic results
Ability of FNA to Detect Residual Disease
We next performed analysis of patients with residual nodal disease who had a diagnostic FNA biopsy (n=33) in order to determine differences between those that had a true positive biopsy (n=14) and those that had an inaccurate false negative result (n=19). The false negative rate for FNA biopsy in this group was 57.6% (95% CI 39.2–74.5). When comparing patients with a true positive result to those with a false negative result, the groups has similar clinical T categories, tumor histology, number of suspicious nodes on pre-NAC ultrasound, size of the clipped node on post-NAC ultrasound, and number of positive nodes on surgical pathology. However, patients with an accurate FNA had larger size of the clipped node before initiating chemotherapy (mean 2.5 vs 1.6 cm, p=0.015), and a larger focus of disease in the clipped node after NAC (1.2 vs. 0.6 cm , p=0.006). While it did not reach statistical significance, patients with a false negative result were more likely to have normalization of the nodes on US (58%, 11/19) than those with a true positive result (21.4%, 3/14) (p=0.22). These results are summarized in Table 2. On univariate analysis, the odds of a true positive biopsy result increased significantly with the size of the clipped node before NAC (OR 4.3, p=0.004) and the size of the metastatic focus after NAC (OR 1.3, p=0.003). Post-chemotherapy ultrasound evaluations showing normalization (OR 0.1) or progression (OR=0.1) were associated with a false negative FNA biopsy result (p=0.013). Sample size limits multivariate analysis (table 3).
Table 2 –
Characteristics of patients with residual disease and diagnostic needle biopsy*
| FNA Accurate | FNA Inaccurate | P value | |
|---|---|---|---|
| (True Positive) | (False Negative) | ||
| N=14 | N=19 | ||
| Mean Clinical Tumor Size | 4.4 (1–16.9) | 3.8 (1.3–11) | P=0.6 |
| Clinical T Category | |||
| T1 | 2 (14.3%) | 4 (21.1%) | 0.94 |
| T2 | 9 (64.3%) | 10 (52.6%) | |
| T3 | 3 (21.4%) | 4 (21.1%) | |
| T4 | 0 | 1 (5.3%) | |
| Tumor Histology | 0.32 | ||
| Ductal | 13 (92.9%) | 17 (89.5%) | |
| Lobular | 0 | 2 (10.5%) | |
| Other | 1 (7.1%) | 0 | |
| Tumor Subtype | 0.25 | ||
| HR+/HER2− | 9 (64.3%) | 15 (78.9%) | |
| HR+/HER2+ | 4 (28.6%) | 1 (5.3%) | |
| HR-/HER2+ | 0 | 1 (5.3%) | |
| HR-/HER2− | 1 (7.1%) | 2 (10.5%) | |
| Number of Abnormal Nodes on | |||
| initial Ultrasound | |||
| 1 node | 3 (21.4%) | 8 (42%) | 0.4 |
| 2 nodes | 3 (21.4%) | 5 (26.3%) | |
| 3 nodes | 2 (14.3%) | 3 (15.8%) | |
| ≥4 nodes | 6 (42.9%) | 3 (15.8%) | |
| Mean Size of Clipped Node | 2.5 (1.2–5.2) | 1.6 (0.5–3.1) | 0.015 |
| Presence of extra-axillary nodal disease | 2 (14.3%) | 2 (10.5%) | 1 |
| Ultrasound-assessed response to NAC | |||
| No response/progression | 1 (7.1%) | 4 (21.1%) | 0.22 |
| Partial Response | 10 (71.4%) | 4 (21.1%) | |
| Normalization | 3 (21.4%) | 11 (57.9%) | |
| Not performed | 1 | ||
| Last recorded size of clipped node (Mean) | 1.5 (0.4–4) | 1.3 (0.6–3.8) | 0.45 |
| Number of Pathologic Positive | |||
| Nodes | 3 (21.4%) | 3 (15.8%) | 0.81 |
| 1 | 5 (35.7%) | 9 (37.4%) | |
| 2–3 | 6 (42.9%) | 7 (36.8%) | |
| ≥4 | |||
| Size of clipped node metastasis | 1.2 cm | 0.6 cm | 0.006 |
HR – hormone receptor; NAC – neoadjuvant chemotherapy
Three patients had nondiagnostic FNA results
Table 3 –
Univariate Analysis to identify features associated with true positive results among patients with residual nodal disease and a diagnostic needle biopsy
| Univariate Analysis | ||
|---|---|---|
| Odds Ratio | P Value | |
| Variable | (95% CI) | |
| Clinical Tumor Size | 1.1 | 0.5 |
| (0.9–1.4) | ||
| Clinical T Category | ||
| T1 (ref) | -- | 0.9 |
| T2 | 1.6 (0.3–11.2) | |
| T3 | 1.4 (0.2–12.5) | |
| T4 | 0.6 (0–16.8) | |
| Tumor Histology | ||
| Ductal (ref) | -- | 0.4 |
| Lobular | 0.3 (0–3.5) | |
| Other | 3.9 (0.2–586.3) | |
| Tumor Subtype | ||
| HR-/HER2− (ref) | -- | 0.4 |
| HR-/HER2+ | 0.6 (0–21.3) | |
| HR+/HER2− | 1.0 (0.1–12.4) | |
| HR+/HER2+ | 5 (0.4–109.9) | |
| Number of Abnormal Nodes on initial Ultrasound | ||
| 1 node (ref) | -- | 0.3 |
| 2 nodes | 1.6 (0.2–12.1) | |
| 3 nodes | 1.8 (0.2–17.4) | |
| ≥4 nodes | 5.3 (0.9–42.7) | |
| Mean Size of Clipped Node | 4.3 (1.5–19.6) | 0.004 |
| Presence of Extra-Axillary Nodal Disease | 1.4 (0.2–13.2) | 0.7 |
| Ultrasound Response to NAC | ||
| Partial Response (ref) | -- | 0.01 |
| Progression | 0.1 (0–0.9) | |
| Normalization | 0.1 (0.02–0.6) | |
| Last Recorded Size of Clipped Node | 1.5 (0.6–4.3) | 0.4 |
| Number of Pathologic Positive Nodes | ||
| 1 (ref) | -- | 0.8 |
| 2–3 | 0.6 (0.1–4.0) | |
| ≥4 | 0.9 (0.1–6.3) | |
| Size of Clipped Node Metastasis | 1.3 (1.1–1.6) | 0.003 |
Discussion
This trial evaluated the use of FNA of lymph nodes with biopsy-confirmed metastasis to assess response to NAC. We found that FNA has a FNR of 57.6% and a negative predictive value of 40.6%, making it unreliable for nodal assessment. We can conclude from this data that a negative FNA after chemotherapy is non-informative. This highlights the need for surgical staging of the axilla after chemotherapy.
The use of needle biopsy to assess for nodal disease has allowed for improved multidisciplinary management of breast cancer patients since nodal disease often determines the need for chemotherapy, extent of surgery, and the possibility of adjuvant radiation. This has been studied extensively in the diagnostic setting before initiation of therapy. The use of physical examination alone is unreliable with reported FNRs of 30–50%4,16 and ultrasound has emerged as a preferred technique for nodal assessment before therapy. One benefit of ultrasound is that abnormal nodes can be assessed in the same setting with needle biopsy to allow for pathologic confirmation of disease. In a study from our institution, axillary ultrasound with fine needle aspiration of abnormal nodes was found to have a sensitivity of 93% in patients with metastatic deposits measuring >5 mm compared to 44% when the largest focus was <5 mm.1 Another study of FNA showed the overall sensitivity was 65% with 82% of the false negative cases occurring when the metastatic focus was < 1 cm .3
It has been hypothesized that ultrasound could be used to assess nodal response since some nodes will resume a normal appearance after chemotherapy. Central review of post-chemotherapy axillary ultrasound images was performed in patients enrolled on the ACOSOG Z1071 trial. In patients with normal-appearing nodes after chemotherapy, 57% (243/430) were found to have residual nodal disease. 9 This is similar to our finding that 57% of patients with normalization on ultrasound had disease on surgical pathology. In contrast, 28% (51/181) of patients in the ACOSOG Z1071 who had persistently abnormal appearing nodes on ultrasound, were found to be node negative at the time of surgery.17 Thus relying on ultrasound would have committed a significant proportion of women to potentially unnecessary axillary surgery. The SENTINA (SENTinel NeoAdjuvant) trial had similar findings. In their trial, physical examination alone had a sensitivity of 8.3%, specificity of 94.8%, and a negative predictive value of 46.6%. The addition of ultrasound assessment improved the sensitivity to 24.4% and negative predictive value slightly to 50.3%, but the specificity was lowered to 91.4%. In patients with normalized nodes on ultrasound, 49.7% (294/592) were pathologically node positive.18 These trials support our finding that ultrasound findings are not adequate for assessment of nodal response.
One significant finding from the ACOSOG Z1071 trial was that there may be utility to marking nodes with confirmed metastasis with clips before chemotherapy. In that trial, 170 patients had a clip placed to mark the node biopsy proven disease. When the clipped node was retrieved as a SLN, the FNR was 6.8% compared to 19% if the clipped node was not retrieved as a SLN.10 In a study from our institution which enrolled patients who had biopsy-confirmed nodal metastasis with a clip placed to mark the biopsied node, specific evaluation of that node after chemotherapy resulted in a FNR of 4.2%.11 If specific evaluation of the clipped node was combined with evaluation of the sentinel nodes, this FNR fell to 1.4%. In 23% of cases, the clipped node was not retrieved as a sentinel node, highlighting the need for specific marking and retrieval of this node in axillary evaluation.11 Building on this concept, we developed a surgical technique called targeted axillary dissection which involves selectively removing the clipped node in addition to removal of the sentinel nodes.13 To accomplish this, the clipped node is marked with a seed under ultrasound guidance up to 5 days before surgery. The surgeon can then use a Neoprobe to identify the node as well as the sentinel nodes for removal. In the first series published with this procedure including 85 patients, the FNR was 2%.11 Other groups been evaluating this technique as well including Diego et al. which reported on 29 patients with successful localization and removal of the clipped node in addition to SLND. In the 11 patients with residual nodal disease, all had disease seen in the clipped node.19 Investigators from the Mayo clinic have also published a series of 56 patients with clipped nodes removed with a variety of techniques.20 A group from the Netherlands have used MARI ( Marking Axillary nodes with Radioactive Iodine seeds) which involves placing the seed at the time of initial biopsy. They report a FNR of 7% in their first 100 patients.21 To date, they have not published on this technique in addition to SLND however trials to assess this are ongoing.22
This study used FNA instead of the larger sampling possible with a core needle biopsy. In the studies of pre-therapy nodal assessment, core needle biopsy had similar sensitivity and specificity to FNA. 23,24 While accuracy may improve with a larger volume of tissue, we felt that the core needle approach is problematic in this setting. First, nodes that have treatment effect after NAC tend to contract and become fibrotic and dense. This makes core needle biopsy technically more difficult. Secondly, all of these patients were having I125 seeds placed in the nodes in preparation for surgery in the following days. The increased tissue trauma associated with core needle assessment would have created hematomas that could potentially complicate accurate placement of the I125 seeds and may have resulted in removal of the clip making it impossible to confirm that the node removed at surgery was the one originally marked with a clip. Lastly, in planning the trial we felt that FNA would be optimal from a workflow perspective if this had been a positive trial. FNA allows for immediate cytological assessment, so the decision to place or not place an I125 seed could be made in the same setting. Core biopsy would have required biopsy done in one setting, a wait time of 24–48 hours for pathologic assessment, and then another procedure if the decision to place an I125 seed was made.
Pathologic evaluation of nodes after chemotherapy can be challenging. This study used conventional cytopathological criteria for classifying specimens as described in the Methods. Cytomorphological changes in the tumor cells that are commonly noted following therapy such as pleomorphism, prominent nucleolus and vacuolated cytoplasm were not felt to impact identification of residual metastasis. We hypothesize that that therapy-related fibrosis seen in nodes resulted in decreased cellular contributing to the high FNR.
While reporting negative results, this trial is relevant for several reasons. First, this is one of the first protocols with the ability to correlate the radiologic appearance and surgical pathology of the specific node known to harbor disease at the time of diagnosis. Most other reports use the pathologic assessment of all axillary nodes removed at surgery as comparable standard. This also allowed for specific evaluation of the accuracy of needle biopsy in a specific lymph node. Secondly, as the field moves towards minimizing breast and axillary surgery in patients who undergo NAC, it is important to understand the limitations of preoperative nodal assessment after NAC; this is the first trial to assess the use of FNA of nodes after NAC.25 There are additional limitations to this study including the small sample size and fact that it is a single institution study. Selection bias is also possible as potential participants were identified by the treating surgeons. Lastly, while each FNA was performed under ultrasound guidance with visualization of the clipped node, there is a small possibility that another non-clipped node could have been sampled leading to false negative results.
In conclusion, the goal of this study was to determine if FNA of the clipped node after NAC could reliably identify patients with persistent nodal disease. We show that FNA is not accurate in this setting and would misclassify >50% of patients with residual nodal disease. This highlights the continued need for surgical evaluation of the axilla, either with SLND, targeted axillary dissection or complete axillary node dissection after chemotherapy.
Acknowledgments
Funding: NIH Cancer Center Support Grant from NIH (CA16672), NIH P30 grant (CA016672) (BPH), Mike Hogg foundation (ASC), MD Anderson Clinical Innovator Award (ASC),PH and Fay Etta Robinson Distinguished Professorship in Cancer Research (HMK).
Footnotes
There are no conflicts of interest or financial disclosures.
References
- 1.Krishnamurthy S, Sneige N, Bedi D, et al. Role of ultrasound-guided fine-needle aspiration of indeterminate and suspicious axillary lymph nodes in the initial staging of breast carcinoma. Cancer 2002;95(5):982–988. [DOI] [PubMed] [Google Scholar]
- 2.Bedi D, Krishnamurthy R, Krishnamurthy S, et al. 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(3):646–652. [DOI] [PubMed] [Google Scholar]
- 3.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(2):89–93. [DOI] [PubMed] [Google Scholar]
- 4.Sacre R Clinical evaluation of axillary lymph nodes compared to surgical and pathological findings. Eur J Surg Oncol 1986;12(2):169–173. [PubMed] [Google Scholar]
- 5.Boughey J, Moriarty J, Degnim A, Gregg M, Egginton J, Long K. Cost modeling of preoperative axillary ultrasound and fine-needle aspiration to guide surgery for invasive breast cancer. Ann Surg Oncol 2010;17(4):953–958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Turaga K, Chau A, Eatrides J, et al. Selective application of routine preoperative axillary ultrasonography reduces costs for invasive breast cancers. Oncologist 2011;16(7):942–946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Boughey J, Ballman K, McCall A, et al. Tumor Biology and Response to Chemotherapy Impact Breast Cancer-specific Survival in Node-positive Breast Cancer Patients Treated With Neoadjuvant Chemotherapy: Long-term Follow-up From ACOSOG Z1071 (Alliance). Ann Surg 2017;266(4):667–676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Dominici L, Negron Gonzalez V, Buzdar A, et al. Cytologically proven axillary lymph node metastases are eradicated in patients receiving preoperative chemotherapy with concurrent trastuzumab for HER2-positive breast cancer. Cancer 2010;116(12):2884–2889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Boughey J, Suman V, Mittendorf E, et al. Sentinel lymph node surgery after neoadjuvant chemotherapy in patients with node-positive breast cancer: the ACOSOG Z1071 (Alliance) clinical trial. JAMA 2013;310(14):1455–1461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Boughey J, Ballman K, Le-Petross H, et al. Identification and Resection of Clipped Node Decreases the False-negative Rate of Sentinel Lymph Node Surgery in Patients Presenting With Node-positive Breast Cancer (T0-T4, N1-N2) Who Receive Neoadjuvant Chemotherapy: Results From ACOSOG Z1071 (Alliance). Ann Surg 2016;263(4):802–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Caudle A, Yang W, Krishnamurthy S, et al. 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(10):1072–1078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology: Breast, Version 2.2017, BINV-11. 2017; 2.2017:www.nccn.org. Accessed Sept 1, 2017.
- 13.Caudle A, Yang W, Mittendorf E, et al. Selective Surgical Localization of Axillary Lymph Nodes Containing Metastases in Patients With Breast Cancer: A Prospective Feasibility Trial. JAMA Surg 2014;34(10):1072–1078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Caudle A, Bedrosian I, Milton D, et al. Use of Sentinel Lymph Node Dissection After Neoadjuvant Chemotherapy in Patients with Node-Positive Breast Cancer at Diagnosis: Practice Patterns of American Society of Breast Surgeons Members. Ann Surg Oncol 2017;24(10):2925–2934. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Team RC. R: A language and environment for statistical computing R Foundation for Statistical Computing; 2017; https://www.R-project.org. [Google Scholar]
- 16.de Freitas R, Costa M, Schneider S, Nicolau M, 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(3):240–244. [PubMed] [Google Scholar]
- 17.Boughey J, Ballman K, Hunt K, et al. Axillary Ultrasound After Neoadjuvant Chemotherapy and Its Impact on Sentinel Lymph Node Surgery: Results From the American College of Surgeons Oncology Group Z1071 Trial (Alliance). J Clin Oncol 2015;33(30):3368–3393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schwentner L, Helms G, Neklijudova V, et al. Using ultrasound and palpation for predicting axillary lymph node status following neoadjuvant chemotherapy - Results from the multi-center SENTINA trial. Breast 2017;31:202–207. [DOI] [PubMed] [Google Scholar]
- 19.Diego E, McAuliffe P, Soran A, et al. Axillary Staging After Neoadjuvant Chemotherapy for Breast Cancer: A Pilot Study Combining Sentinel Lymph Node Biopsy with Radioactive Seed Localization of Pre-treatment Positive Axillary Lymph Nodes. Ann Surg Oncol 2016;23(5):1549–1553. [DOI] [PubMed] [Google Scholar]
- 20.Nguyen T, Heiken T, Glazebrook K, Boughey J. Localizing the Clipped Node in Patients with Node-Positive Breast Cancer Treated with Neoadjuvant Chemotherapy: Early Learning Experience and Challenges. Ann Surg Oncol 2017;24(10):3011–3016. [DOI] [PubMed] [Google Scholar]
- 21.Donker M, Straver ME, Wesseling J, et al. Marking Axillary Lymph Nodes With Radioactive Iodine Seeds for Axillary Staging After Neoadjuvant Systemic Treatment in Breast Cancer Patients: The MARI Procedure. Ann Surg 2015;261(2):378–382. [DOI] [PubMed] [Google Scholar]
- 22.van Nijnatten T, Simons J, Smidt M, et al. A Novel Less-invasive Approach for Axillary Staging After Neoadjuvant Chemotherapy in Patients With Axillary Node-positive Breast Cancer by Combining Radioactive Iodine Seed Localization in the Axilla With the Sentinel Node Procedure (RISAS): A Dutch Prospective Multicenter Validation Study. Clin Breast Cancer 2017;17(5):399–402. [DOI] [PubMed] [Google Scholar]
- 23.Abe H, Schmidt R, Kulkarni K, Sennett C, Mueller J, Newstead G. Axillary lymph nodes suspicious for breast cancer metastasis: sampling with US-guided 14-gauge core-needle biopsy--clinical experience in 100 patients. Radiology 2009;250(1):41–49. [DOI] [PubMed] [Google Scholar]
- 24.Rao R, Lilley L, Andrews V, Radford L, Ulissey M. Axillary staging by percutaneous biopsy: sensitivity of fine-needle aspiration versus core needle biopsy. Ann Surg Oncol 2009;16(5):1170–1175. [DOI] [PubMed] [Google Scholar]
- 25.van la Parra R, Kuerer H. Selective elimination of breast cancer surgery in exceptional responders: historical perspective and current trials. Breast Cancer Res 2016;18(1):28. [DOI] [PMC free article] [PubMed] [Google Scholar]