Is the clinically positive axilla in breast cancer really a contraindication to sentinel lymph node biopsy?

ORIGINAL SCIENTIFIC ARTICLES

Is the Clinically Positive Axilla in Breast Cancer Really a Contraindication to Sentinel Lymph Node Biopsy? Michelle C Specht, MD, Jane V Fey, MPH, Patrick I Borgen, MD, FACS, Hiram S Cody III, MD, FACS Clinically positive axillary nodes are widely considered a contraindication to sentinel lymph node (SLN) biopsy in breast cancer, yet no data support this mandate. In fact, data from the era of axillary lymph node dissection (ALND) suggest that clinical examination of the axilla is falsely positive in as many as 30% of cases. Here we report the results of SLN biopsy in a selected group of breast cancer patients with palpable axillary nodes classified as either moderately or highly suspicious for metastasis. STUDY DESIGN: Among 2,027 consecutive SLN biopsy procedures performed by two experienced surgeons, clinically suspicious axillary nodes were identified in 106, and categorized as group 1 (asymmetric enlargement of the ipsilateral axillary nodes moderately suspicious for metastasis, n ⫽ 62) and group 2 (clinically positive axillary nodes highly suspicious for metastasis, n ⫽ 44). RESULTS: Clinical examination of the axilla was inaccurate in 41% of patients (43 of 106) overall, and was falsely positive in 53% of patients (33 of 62) with moderately suspicious nodes and 23% of patients (10 of 44) with highly suspicious nodes. False-positive results were less frequent with larger tumor size (p ⫽ 0.002) and higher histologic grade (p ⫽ 0.002), but were not associated with age, body mass index, or a previous surgical biopsy. CONCLUSIONS: Clinical axillary examination in breast cancer is subject to false-positive results, and is by itself insufficient justification for axillary lymph node dissection. If other means of preoperative assessment such as palpation- or image-guided fine needle aspiration are negative or indeterminate, then SLN biopsy deserves wider consideration as an alternative to routine axillary lymph node dissection in the clinically node-positive setting. (J Am Coll Surg 2005;200:10–14. © 2005 by the American College of Surgeons) BACKGROUND:

Although lymph node metastasis remains overwhelmingly the most important prognostic factor in operable breast cancer, and clinically positive nodes are a particularly poor prognostic sign, clinical examination of the axilla remains unreliable, and this unreliability is the basis of surgical lymph node staging.1,2 The greatest emphasis to date has been on false-negative results (about one-third of clinically node-negative examinations are falsely negative), but false-positive results are equally frequent (about one-third of clinically node-positive exam-

Presented as a poster at the Society of Surgical Oncology Meeting, New York, NY, March 2004.

inations are falsely positive) and deserve greater attention in an era when breast cancer surgery is governed by the principle of conservatism. Sentinel lymph node (SLN) biopsy has become the new standard of care for axillary node staging in breast cancer3 and appears suitable for virtually all patients with nonmetastatic disease,4 yet many authorities still consider palpable axillary nodes to be a contraindication.5,6 Because the diagnosis of clinically positive nodes is falsely positive in onethird of cases, we hypothesize that SLN biopsy deserves a wider role in this setting. Here we report the results of SLN biopsy in a selected group of breast cancer patients with a clinical axillary examination either moderately or highly suspicious for nodal metastasis.

Received June 2, 2004; Revised September 8, 2004; Accepted September 16, 2004. From The Breast Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY. Correspondence address: Hiram S Cody III, MD, FACS, Memorial Sloan Kettering Cancer Center, 1275 York Ave, New York, NY 10021.

METHODS Between September 1, 1996 and August 1, 2003, we performed SLN biopsy in 5,818 breast cancer patients at Memorial Sloan Kettering Cancer Center. All records

No competing interests declared.

© 2005 by the American College of Surgeons Published by Elsevier Inc.

10

ISSN 1072-7515/05/$30.00 doi:10.1016/j.jamcollsurg.2004.09.010

Vol. 200, No. 1, January 2005

Specht et al

Sentinel Lymph Node Biopsy and Palpable Axilla

11

Abbreviations and Acronyms

ALND FNA PPV SLN US

⫽ ⫽ ⫽ ⫽ ⫽

axillary lymph node dissection fine needle aspiration positive predictive value sentinel lymph node ultrasound

were maintained in a prospective database. After excluding patients with previous breast cancer, earlier axillary surgery, breast irradiation, and neoadjuvant chemotherapy, 5,082 patients were left, of whom 2,027 had operations by two experienced surgeons (PB and HSC) and are the focus of this study. Clinical axillary node status was documented in a standardized fashion at the initial consultation, and 106 patients (5%) were identified as having palpable clinically suspicious axillary nodes. In 62 (group 1), the nodes were believed to be moderately suspicious and were described as being firm, shotty, and more prominent than those on the contralateral side. In 44 (group 2), the nodes were thought to be either highly suspicious or unequivocally positive. Our technique of SLN biopsy has been described previously in detail.7 In brief, we used a combined dyeisotope mapping technique (intradermally injected unfiltered technetium 99m sulfur colloid and intraparenchymally injected isosulfan blue dye), removing all blue, hot, or palpably suspicious nodes. Intraoperative frozen sectioning was performed on all SLNs. If positive, an immediate ALND was done; if negative, the remainder of the node was examined by taking two adjacent sections (one hematoxylin and eosin-stained and one anticytokeratin-stained) from each of two levels 50␮ apart. Non-SLN from ALND specimens were examined routinely by single-section hematoxylin and eosin. Positive predictive value (PPV) for clinical axillary examination was defined as the proportion of all positive examinations that were truly positive: (true positive)/ (true positive ⫹ false positive). The univariate signifi-

Figure 1. Proportion of true-positive and false-positive results by level of clinical suspicion. Gray bars, false positives; black bars, true positives.

cance of differences was assessed by Fisher’s exact test or the chi-square test for binary or categoric covariates, or by the Wilcoxon rank correlation for ordered covariates. RESULTS SLNs were identified in all 106 patients with palpable axillary nodes. Clinical characteristics are identified in Table 1; as expected, Group 2 patients had larger tumors and more frequent lymphovascular invasion, both of which correlate with more frequent SLN positivity. An ALND was performed in 59% of patients (62), and in this subset, the accuracy of SLN biopsy was 100% (there were no false-negative results). The PPV for clinical examination of the axilla was 59% overall, 47% in the moderately suspicious group 1 patients, and 77% in the highly suspicious group 2 patients (Table 1). The proportion of false-positive results was higher in group 1 than in group 2 patients (Fig. 1). False-positive results, as one would expect, were associated with more favorable tumor characteristics (ie, a greater likelihood of being node negative) and true- positive results with less favorable tumor characteristics (ie, a greater likelihood of being node positive, Table 2,

Table 1. Patient and Tumor Characteristics Characteristics

Age, y, median (range) Median tumor size, cm LVI present, n SLN positive, n ALND done, n

Total (n ⴝ 106)

Group 1: moderate suspicion (n ⴝ 62)

Group 2: high suspicion (n ⴝ 44)

51 (26–88) 2.0 (n ⫽ 98) 38/106 (36%) 63/106 (59%) 62/106 (59%)

48 (32–88) 1.6 (n ⫽ 56) 20/62 (32%) 29/62 (48%) 31/62 (50%)

54 (26–85) 2.2 (n ⫽ 42) 18/44 (41%) 34/44 (77%) 31/44 (70%)

ALND, axillary lymph node dissection (defined as the removal of ⱖ10 nodes in toto); LVI, lymphovascular invasion; SLN, sentinel lymph node.

12

Specht et al

Sentinel Lymph Node Biopsy and Palpable Axilla

Table 2. Comparison of Patients with a False-Positive Versus True-Positive Clinical Examination

Characteristics

Age, y, median Previous surgical biopsy, n Body mass index, median Tumor size, cm High grade, histologic, n

Clinical axillary examination False True positive positive (n ⴝ 43) (n ⴝ 63)

p Value

49 52 23/43 (54%) 24/63 (38%) 24 24 1.6 2.6 16/37 (43%) 44/57 (77%)

0.982 0.163 0.743 0.002 0.002

p ⬍ 0.01). But age, body mass index, and previous surgical biopsy were not predictive of a false-positive examination. DISCUSSION By current treatment guidelines,5,6 clinically positive axillary nodes are a contraindication to SLN biopsy. But clinical examination of the axilla is imperfect and the results of five series1,2,8,9 from the era of ALND (spanning 50 years) indicate a PPV of 64% to 82%, a negative predictive value of 50% to 63%, and an overall accuracy of 63% to 68% (Table 3). In fact, patients with clinically suspicious axillary nodes comprise a wide spectrum of pathologic findings. Normal lymph nodes vary widely in size, consistency, and fat content. Lymphadenopathy is an element of many nonmalignant diseases. Even at operation, reactive adenopathy may be grossly indistinguishable from gross metastasis and frankly malignant nodes may appear completely normal. This inaccuracy suggests a wider role for SLN biopsy in selected patients with clinically suspicious nodes. Many breast cancer patients with clinically suspicious axillary nodes do not require SLN biopsy at all and can proceed directly to ALND because recent advances in breast imaging allow preoperative diagnosis for a considerable majority. For patients with locally advanced cancers, Herrada and colleagues10 found that axillary ultrasound (US) was superior to either physical examination or mammography in identifying nodal metastasis, and Oruwari and associates11 showed that US-guided fine needle aspiration (FNA) was 100% accurate in identifying nodal metastasis. Krishnamurthy and coworkers12 reported an overall accuracy for preoperative US-guided FNA of 79%, and noted that most of the discrepancies be-

J Am Coll Surg

tween the results of US-guided FNA and final pathology were related to incomplete US imaging of the axilla, to metastatic deposits smaller than 0.5 mm, and to neoadjuvant chemotherapy. Other studies of US-guided FNA showed similar results,13-17 and it is clear that preoperative axillary staging by US-guided FNA will play a growing role in the future. Preliminary studies of axillary CT,18 MRI,19 and PET scanning20 are promising, but at present appear less useful than US in this context. SLN biopsy worked well in our patients with clinically positive axillary nodes; we identified the SLN in 100% of patients, with no false-negative results among those who had a completion ALND. Early in our experience21 we defined the SLN as “blue” or “hot,” and using this definition, we observed a falsenegative rate of 14%.22 In reviewing our false-negative patients, we observed that about two-thirds of them were correctly staged at operation by identification of palpably suspicious nodes that were neither blue nor hot, and that by expanding the definition of an SLN to include palpability, our false-negative rate fell to 4%.22 Of note, for one patient in this series, the positive SLN was grossly suspicious but was neither blue nor hot. Although gross tumor involvement of nodes might interfere with the uptake of both dye and isotope, our results suggest that by expanding the definition of the SLN to include palpability, SLN biopsy is equally accurate for patients with clinically positive and clinically negative axillae. In our population of patients with clinically suspicious axillae having SLN biopsy, we observed a PPV of 59%, results quite comparable with those of the previous studies from the era of ALND. These results support a role for SLN biopsy in this setting, but some caveats apply to our analysis. The study population Table 3. Studies of Clinical Axillary Examination (Relative to Final Pathology) Lead author, y

Cutler1 (Memorial, 1968) Cutler1 (End Results, 1970) Fisher2 (NSABP B-04, 1981) de Freitas8 (1991) Vaidya9 (1996)

n

PPV, %

NPV, %

Accuracy, %

1,747 1,210 641 115 200

64 65 73 82 76

63 58 61 50 58

63 65 66 68 66

NPV, negative predictive value (true negative/[true negative ⫹ false negative]); PPV, positive predictive value (true positive/[true positive ⫹ false positive]).

Vol. 200, No. 1, January 2005

Specht et al

represents a subset of our entire experience with SLN biopsy (we chose to analyze two surgeons’ results based on their consistent documentation of both clinical findings and degree of preoperative suspicion), and may not be representative of our entire experience. In addition, the 106 study patients do not represent all patients with a clinically positive axilla, some of whom proceeded directly to ALND based on either a positive FNA result or simply on a high level of clinical suspicion. Had these patients (most of them node positive) been included in our study population, the PPV of clinical examination might have appeared higher. We were unable to clearly define a group of patient or tumor characteristics that were predictive of a falsepositive examination. Of note, previous surgical biopsy appeared to be somewhat more frequent among patients with a false-positive (versus a false-negative) clinical axillary examination (Table 2). Reactive adenopathy after a surgical biopsy can be striking and might be grossly indistinguishable from metastatic disease; this is a setting in which preoperative USguided FNA may be particularly useful, and in which patients with a benign result should proceed to SLN biopsy rather than an immediate ALND based solely on clinical examination. But as we have previously argued,7 if the surgeon encounters grossly suspicious axillary nodes during SLN biopsy, it is absolutely reasonable to default to ALND on the basis of clinical suspicion and despite a negative frozen section (or touch prep) on the SLN. In conclusion, clinical axillary examination is falsely positive in a considerable proportion of patients with either moderately or highly suspicious findings. For those patients in whom a preoperative palpation- or US-guided FNA is nondiagnostic, SLN biopsy is preferable to ALND, as it is for breast cancer patients in general. Author Contributions

Study conception and design: Specht, Cody Acquisition of data: Specht Analysis and interpretation of data: Specht, Fey, Cody Drafting of manuscript: Specht, Cody Critical revision: Borgen, Cody Statistical expertise: Fey Supervision: Cody

Sentinel Lymph Node Biopsy and Palpable Axilla

13

REFERENCES 1. Cutler SJ, Axtell LM, Schottenfeld D, et al. Clinical assessment of lymph nodes in carcinoma of the breast. Surg Gynecol Obstet 1970;131:41–52. 2. Fisher B, Wolmark N, Banes M. The accuracy of clinical nodal staging and of limited axillary dissection as a determinant of histologic nodal status in carcinoma of the breast. Gynecol Obstet 1981;152:765–772. 3. Cody HS. Clinical aspects of sentinel node biopsy. Breast Cancer Res 2001;3:104–108. 4. Cody HS III. Sentinel lymph node biopsy for breast cancer: does anybody not need one? Ann Surg Oncol 2003;10:1131– 1132. 5. NCCN. The complete library of practice guidelines in oncology: breast cancer, version 1. Available at http://www.nccn.org Accessed October 5, 2004. 6. American Society of Breast Surgeons. Consensus statement on guidelines for performance of sentinel lymphadenectomy for breast cancer, 2003. Available at http://breastsurgeons. org/officialstmts/sentinel.shtml Accessed October 5, 2004. 7. Cody HS, Borgen PI. State-of-the-art approaches to sentinel node biopsy for breast cancer: study design, patient selection, technique, and quality control at Memorial Sloan-Kettering Cancer Center. Surg Oncol 1999;8:85–91. 8. de Freitas R Jr, Costa MV, Schneider SV, et al. Accuracy of ultrasound and clinical examination in the diagnosis of axillary lymph node metastases in breast cancer. Eur J Surg Oncol 1991; 17:240–244. 9. Vaidya JS, Vyas JJ, Thakur MH, et al. Role of ultrasonography to detect axillary node involvement in operable breast cancer. Eur J Surg Oncol 1996;22:140–143. 10. Herrada J, Iyer RB, Atkinson EN, et al. Relative value of physical examination, mammography, and breast sonography in evaluating the size of the primary tumor and regional lymph node metastases in women receiving neoadjuvant chemotherapy for locally advanced breast carcinoma. Clin Cancer Res 1997;3:1565–1569. 11. Oruwari JU, Chung MA, Koelliker S, et al. Axillary staging using ultrasound-guided fine needle aspiration biopsy in locally advanced breast cancer. Am J Surg 2002;184:307–309. 12. Krishnamurthy S, Sneige N, Bedi DG, et al. Role of ultrasoundguided fine-needle aspiration of indeterminate and suspicious axillary lymph nodes in the initial staging of breast carcinoma. Cancer 2002;95:982–988. 13. Bonnema J, van Geel AN, van Ooijen B, et al. Ultrasoundguided aspiration biopsy for detection of nonpalpable axillary node metastases in breast cancer patients: new diagnostic method. World J Surg 1997;21:270–274. 14. de Kanter AY, van Eijck CH, van Geel AN, et al. Multicentre study of ultrasonographically guided axillary node biopsy in patients with breast cancer. Br J Surg 1999;86:1459–1462. 15. Feu J, Tresserra F, Fabregas R, et al. Metastatic breast carcinoma in axillary lymph nodes: in vitro US detection. Radiology 1997; 205:831–835. 16. Verbanck J, Vandewiele I, De Winter H, et al. Value of axillary ultrasonography and sonographically guided puncture of axillary nodes: a prospective study in 144 consecutive patients. J Clin Ultrasound 1997;25:53–56. 17. Kuenen-Boumeester V, Menke-Pluymers M, de Kanter AY, et al. Ultrasound-guided fine needle aspiration cytology of axillary

14

Specht et al

Sentinel Lymph Node Biopsy and Palpable Axilla

lymph nodes in breast cancer patients. A preoperative staging procedure. Eur J Cancer 2003;39:170–174. 18. Yuen S, Sawai K, Ushijima Y, et al. Evaluation of axillary status in breast cancer. CT-based determination of sentinel lymph node size. Acta Radiol 2002;43:579–586. 19. Yamagami T, Yuen S, Sawai K, et al. MR imaging-guided axillary node biopsy for breast cancer: initial findings. Eur Radiol 2004; 14:151–156. 20. Wahl RL, Siegel BA, Coleman RE, et al. Prospective multicenter

J Am Coll Surg

study of axillary nodal staging by positron emission tomography in breast cancer: a report of the staging breast cancer with PET Study Group. J Clin Oncol 2004;22:277–285. 21. Hill ADK, Tran KN, Akhurst T, et al. Lessons learned from 500 cases of lymphatic mapping for breast cancer. Ann Surg 1999; 229:528–535. 22. Martin RCG, Derossis A, Fey J, et al. Intradermal isotope injection is superior to intramammary in sentinel node biopsy for breast cancer. Surgery 2001;130:432–438.

ANNOUNCEMENT Beginning September 2004 JACS editorial office will test an all electronic editorial system.

JANUARY 2005 All manuscripts will be submitted on the electronic system ONLY. WATCH FOR GUIDELINES All new electronic editorial office for JACS manuscripts.

JACS ONLINE CME-1 PROGRAM CME-1 COURSE EVERY MONTH 2 CREDITS AVAILABLE EACH MONTH 48 CREDITS AVAILABLE AT ALL TIMES www.jacsme.facs.org