| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of General and Surgical Oncology (LF, MF, SR, SB, CEV), Radiology (GA), and Medicine (PP), MG Vannini Hospital, Rome, Italy; and Department of Pathology (MA, LC, FRP), San Giovanni-Addolorata Hospital, Rome, Italy.
Correspondence: Address correspondence and reprint requests to: Lucio Fortunato, MD, Ospedale MG Vannini, Via Acqua Bullicante, 4, 00177 Rome, Italy; Fax: 39-06-24-29-1326; E-mail: lfortunato{at}tiscali.it
| ABSTRACT |
|---|
|
|
|---|
Methods: In a group of 402 consecutive patients with primary breast cancer who underwent sentinel lymph node biopsy, an intraoperative examination (IE) was obtained in 236 cases either by frozen section (FS; n = 68) or by touch preparation cytology (TP; n = 168).
Results: IE had an accuracy of 89% (209 of 236), but it identified only 52 of 77 positive cases (sensitivity, 68%). There were 25 false-negative cases (13.7%), of which 7 were macrometastases and 18 by micrometastases (P < .001). Six macrometastases were missed by TP and one by FS (P = .9). There were two false-positive cases (3.7%). Overall, 48 (20%) of 236 patients avoided a delayed return to the operating room for a completion lymphadenectomy because of IE findings. This occurred in 10% of patients with tumors <1 cm in diameter, in 20% of those with tumors between 1 and 2 cm, and in 34% of those with tumors >2 cm in diameter (P = .05). The cost savings for the Italian Health System amounted to 198,040 (US$223,794) in these patients.
Conclusions: IE has acceptable sensitivity for lymph node macrometastases, but it is a weak tool for diagnosing micrometastases. FS and TP are roughly equivalent. IE allows management changes, because approximately 20% of all patients are expected to undergo synchronous axillary dissection, and it is particularly helpful in T2 patients. This may allow substantial cost savings for the health-care system.
Key Words: Breast cancer Sentinel lymph node Frozen section Cytology Immunohistochemistry
| INTRODUCTION |
|---|
|
|
|---|
To date, the lymph node status remains the most important prognostic factor for women with breast cancer. However, it is now emphatically recognized in most centers that the removal of a large number of negative lymph nodes in women with breast cancer is useless and potentially harmful. Furthermore, standard axillary staging is not always accurate, and retrospective analysis has demonstrated that positive cases can be missed.4
SLN biopsy, on the other hand, is easy, reproducible, accurate and carries minimal morbidity. In experienced hands, the SLN is found in >95% of cases, and it has an accuracy of 97% to 98%.5 It is suggested that particularly women with small tumors should not undergo axillary dissection to avoid its potential consequences.
Intraoperative examination (IE) of the SLN is desirable because it may identify positive cases and may avoid the unpleasant and sometimes stressful return of the patient to the operating room. It also expedites the surgical process and, therefore, may allow a more timely beginning of adjuvant therapies.
Because no guidelines have been published regarding IE of the SLNs, this is still a matter of debate and, sometimes, of controversy. We present our experience with IE of SLNs for breast cancer at the cancer unit of a community hospital in Rome, Italy.
| METHODS |
|---|
|
|
|---|
Patients were usually mapped the night before surgery with a median dose of 0.6 mCi of 99mTc-filtered nanocolloid (Nanocoll; Nycomed Amersham-Sorin, Saluggia-VC, Italy) injected intradermally at the site of the primary tumor or around the areola, and/or with 2 to 3 mL of isosulfan blue at the time of surgery. After primary tumor removal, the SLN was identified with a handheld probe (Scintiprobe MR100; Pol.Hi.Tech, Carsoli, AQ, Italy) and removed. A median of two SLNs were found per patient. In these patients, the SLN was identified in 97% of cases and showed a correlation with the final lymph node status in 97% of cases.
In 236 patients (59%), an IE of the SLN was performed either with frozen section (FS; n = 68; from January 1999 to April 2000) or with touch preparation cytology (TP; n = 168; from April 2000 to June 2003). This change was determined by the desire to preserve tissue for subsequent step analysis of the SLN and eventual identification of micrometastases.
TP was performed after the SLN was cleared of the surrounding fat. The SLN was halved, and an imprint was obtained on one slide. The slide was then fixed with Cytospray (Streck Laboratories, Omaha, NE), air-dried, and stained with toluidine blue, an easy and fast method with which the pathologists involved in the study have a long lasting experience and confidence. IE usually required no more than 5 to 10 minutes for each case. Usually, one pathologist (M.A.) read the slides and made the intraoperative diagnosis.
Afterward, SLNs were paraffin-embedded and step-sectioned at 75-µm intervals. At least six additional couples of sections were obtained from each SLN, other than the conventional first one, and each section was stained with both hematoxylin and eosin and immunohistochemistry (IHC) with a cytokeratin panel (MNF-116 monoclonal antibodies).
Patients were staged according to the 2002 edition of the International Union Against Cancer staging manual.6 Micrometastases were defined as those between 0.2 and 2 mm in diameter. Isolated tumor cells (ITCs) were clusters <.2 mm or single or aggregated cells, usually diagnosed by IHC. The presence of only ITCs staged the patient N0 (itc+); therefore, these patients were excluded from the sensitivity and accuracy computations.
Statistical analysis was performed with a statistical package (True Epistat; Epistat Services, Richardson, TX). Fishers exact or
2 tests were used to compare categorical variables. Significance was defined as P < .05 (two tailed).
| RESULTS |
|---|
|
|
|---|
|
IE was accurate in 209 (89%) of 236 cases but identified only 52 of 77 positive cases (sensitivity, 68%; Fig. 1). Of these 77 positive cases, 52 were represented by macrometastases and 25 by micrometastases.
|
|
|
Among the 52 patients with macrometastases, FS missed 1 of 11 and TP missed 6 of 41 cases (P = .9). The median diameter of the metastatic deposit in the lymph node in these seven cases was 3.5 mm (range, 35 mm).
Overall, 48 (20%) of 236 patients avoided a delayed return to the operating room for a deferred axillary node dissection because of the information gained by IE. They represent the majority of node-positive patients in our group (48 of 77; 62%). Four additional patients with positive IE results of the SLN were not included in this group because a backup lymphadenectomy would have been performed regardless of the intraoperative findings (patients preference or macroscopic lymphadenopathy). Such a change in surgical management occurred in 6 (9%) of 64 patients with tumors <1 cm in diameter, in 24 (20%) of 119 patients with tumors between 1 and 2 cm, and in 18 (34%) of 53 patients with tumors >2 cm in diameter (P = .05; Fig. 2).
|
| DISCUSSION |
|---|
|
|
|---|
However, IE has several drawbacks, including a lack of sensitivity, a potential loss of frozen lymph node material for step analysis and IHC, and a false sense of security given to the patient in anticipation of the definitive result. Furthermore, it requires the pathologist to give a diagnostic commitment on the basis of objectively scanty material.
Several methods of IE have been described. These include FS and TP,7 scrape cytology,8 IHC stain,9,10 and an exhaustive method of serial intraoperative sectioning described by Veronesi et al.11 The latter is appealing because the entire lymph node is step-analyzed during the operation. However, it is very difficult to implement because it requires human and financial resources not easily available in most centers.
The sensitivity of IE in the literature is variable8,1229 (44%94%; Table 4). Only a few series, including this one, have studied a cohort of more than 200 patients. One reason for this wide range in variability is probably related to the extent of the search for metastatic tumors in the SLN, i.e., the number of serial sections and the IHC stain, and the proportion of SLN macrometastases in that group of patients. Therefore, these values may be seriously biased. It has also been reported that the sensitivity of IE may be higher for larger tumors because the proportion of macrometastases in these patients increases. In T1a patients, the actual benefit from IE due to the avoidance of a second operation was as little as 4% in one report.14 In the present experience, there was a correlation between tumor diameter and positive IE. Although this confirms the usefulness of IE for patients with T2 tumors, the clinical benefit of IE in smaller tumors is questionable and needs to be better elucidated.
|
False-positive cases with cytology are a rare possibility, as demonstrated in our series, and pitfalls of IE are described.30 False positives are probably due to clusters of histiocytes or to virus-related changes in lymphoid cells in the SLN. These cells may be erroneously recognized as atypical in the imprints and therefore considered metastatic. Conversely, false-negative cytology results in cases of macrometastases are probably due to sclerosis of the SLN, with more difficult detachment of the neoplastic cells on the slide. We are presently studying the usefulness of a combination of TP and scrape cytology with the hypothesis that both examinations together may increase the sensitivity of the test and that some lymph node metastases are more easily identified by one examination than the other.
The worth of intraoperative identification of SLN metastases relies on the belief that further dissection is warranted in most positive cases. Nevertheless, all cases may not be equal, and prospective randomized trials are under way to clarify this.3133
At the present time, the presence of ITCs should not be considered by the clinician for further treatment planning, and this information should be reserved for experimental clinical studies. However, the increased incidence of ITCs in the SLN in patients with lobular cancer is of interest, because it may express a particular behavior of less cohesive tumor cells. A different pattern of geographical distribution of lymph node metastases has been also reported: they are usually found within the lymph nodes in ductal cancer, and in the subcapsular or medullary sinuses in lobular carcinoma.34,35
Micrometastases in the SLN are most often an isolated event, and the presence of further metastases in other non-SLNs varies between 5% and 36%.3641 Their prognostic significance is not yet clearly defined, although several retrospective studies suggest that they may play a role in identifying patients at higher risk for relapse.4244 The need for further axillary dissection in these cases is debated, and observation or sampling of only a few additional nodes may be found to be alternative and equally effective methods.
A mathematical model to predict the risk of further lymph node involvement has been recently developed,45 and it may be clinically useful, particularly in these cases, in treatment planning and in discussing with the patients the risks and benefits of further surgical intervention. Therefore, the worth of IE consists mainly in the diagnosis of macrometastases in the SLN, the only ones that carry a significant risk for other lymph node involvement. This risk can be stratified according to several factors, including the diameter of the primary tumor, and it is roughly 50% in most experiences.3,12,4650 From this point of view, IE seems to be a reliable test, because it is quite sensitive and accurate in diagnosing macrometastases.
We report that, in our experience, IE resulted in a significant cost savings for the health-care system by avoiding a second operation and, therefore, additional diagnosis-related group charges. We did not perform a formal cost-effectiveness analysis, but this is an important issue that needs to be better studied. Of course, international differences may apply in terms of the costs of single procedures, and our experience may not be replicated in other health-care systems.
In summary, IE identifies most lymph node macrometastases but misses almost all micrometastases. FS and TP are roughly equivalent. The advantage of the latter is mainly in the preservation of tissue for subsequent step analysis and IHC. A change in management occurs in approximately 20% of cases, thus allowing these patients to avoid a second operation for completion lymphadenectomy. Management changes are more frequent in patients with larger tumors, whereas the clinical benefit of IE in patients with tumors <1 cm in diameter needs to be better elucidated. In our experience, IE allowed significant cost savings for the health-care system. Standardization of the IE technique is needed.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Received for publication December 2, 2003. Accepted for publication July 27, 2004.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Shimazu, Y. Tamaki, T. Taguchi, F. Tsukamoto, T. Kasugai, and S. Noguchi Intraoperative Frozen Section Analysis of Sentinel Lymph Node in Breast Cancer Patients Treated with Neoadjuvant Chemotherapy Ann. Surg. Oncol., June 1, 2008; 15(6): 1717 - 1722. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Arora, D. Martins, T. L. Huston, P. Christos, S. Hoda, M. P. Osborne, A. J. Swistel, E. Tousimis, P. I. Pressman, and R. M. Simmons Sentinel Node Positivity Rates With and Without Frozen Section for Breast Cancer Ann. Surg. Oncol., January 1, 2008; 15(1): 256 - 261. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. D. Nathanson, M. Burke, R. Slater, and A. Kapke Preoperative Identification of the Sentinel Lymph Node in Breast Cancer Ann. Surg. Oncol., November 1, 2007; 14(11): 3102 - 3110. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Teixeira Soares, U. Frederigue-Junior, and L. A. de Luca Anatomopathological Analysis of Sentinel and Nonsentinel Lymph Nodes in Breast Cancer: Hematoxylin-Eosin Versus Immunohistochemistry International Journal of Surgical Pathology, October 1, 2007; 15(4): 358 - 368. [Abstract] [PDF] |
||||
![]() |
M. Tsujimoto, K. Nakabayashi, K. Yoshidome, T. Kaneko, T. Iwase, F. Akiyama, Y. Kato, H. Tsuda, S. Ueda, K. Sato, et al. One-step Nucleic Acid Amplification for Intraoperative Detection of Lymph Node Metastasis in Breast Cancer Patients Clin. Cancer Res., August 15, 2007; 13(16): 4807 - 4816. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |