14 Rapid Immunohistochemistry Enhances the Intraoperative Diagnosis of Sentinel Lymph Node Metastases in Invasive Lobular Breast Carcinoma Junnu P. Leikola, M.D. 1 Terttu S. Toivonen, M.D. 2 Leena A. Krogerus, M.D., Ph.D. 2 Karl A. J. von Smitten, M.D., Ph.D. 1 Marjut H. K. Leidenius, M.D., Ph.D. 1 1 Breast Surgery Unit, Maria Hospital, Helsinki, Finland. 2 Department of Pathology, Helsinki University Central Hospital, Helsinki, Finland. BACKGROUND. The sensitivity of the intraoperative diagnosis of sentinel lymph node (SLN) micrometastases and the metastases of invasive lobular carcinoma (ILC) is low. The goal of the current study was to assess whether the use of intraoperative, rapid immunohistochistochemistry (IHC) enhances the intraoperative detection of micrometastases and metastases of ILC. METHODS. The sensitivity of the intraoperative diagnosis of SLN metastasis was evaluated in 438 patients when using rapid IHC with a cytokeratin biomarker. The results were compared with those obtained for 557 patients without rapid IHC but with conventional staining. RESULTS. For patients with ILC, the sensitivity of the intraoperative diagnosis was 87% (45 of 52) in the IHC group and 66% (39 of 59) in the non-ihc group (P 0.02). The sensitivity of the intraoperative diagnosis was similar for patients with other types of invasive cancer regardless of the use of rapid IHC. However, rapid IHC enhanced marginally the intraoperative diagnosis of the smallest micrometastases, isolated tumor cells (P 0.06). CONCLUSIONS. Rapid IHC with cytokeratin labeling enhanced the intraoperative diagnosis of SLN metastases in patients with ILC. It may also improve the intraoperative diagnosis of micrometastases. Cancer 2005;104:14 9. 2005 American Cancer Society. Address for reprints: Junnu P. Leikola, M.D., Munkkiniemen puistotie 8 B, FIN- 00330 Helsinki, Finland; Fax: (011) 358-9-47163490; E-mail: junnu.leikola@hus.fi Received November 29, 2004; revision received February 10, 2005; accepted February 28, 2005. KEYWORDS: sentinel lymph node biopsy, breast carcinoma, immunohistochemistry, lobular carcinoma, lymphatic metastasis, micrometastasis. Intraoperative diagnosis of sentinel lymph node (SLN) metastases enables axillary staging and treatment during the same surgical procedure in the majority of patients undergoing SLN biopsy, thus reducing the number of surgical procedures as well as hospital costs. 1,2 A nearly 100% sensitivity in intraoperative diagnosis has been reached using an exhaustive intraoperative frozen section (FS) method. 3 This includes serial sectioning of the whole SLN for FS, requiring vast resources from the pathology laboratories beyond the capability and financial constraints of most institutions. The sensitivity of intraoperative immunohistochemistry (IHC) to determine the diagnosis has been essentially lower when using other less laborious FS methods. The sensitivity of intraoperative diagnosis has been especially low in connection with the metastases and micrometastases of invasive lobular carcinoma (ILC). 1,4,5 To date, studies have attempted to solve the problem of low diagnostic sensitivity by implementing intraoperative (IHC). 6 9 These studies are based on a small study population 2005 American Cancer Society DOI 10.1002/cncr.21133 Published online 31 May 2005 in Wiley InterScience (www.interscience.wiley.com).
Intraoperative IHC and SLN Biopsy/Leikola et al. 15 and have been inconclusive in resolving the added value of IHC in histologic subgroups of breast carcinoma. The aim of the current study was to assess whether the use of intraoparative IHC with a cytokeratin biomarker enhances the intraoperative detection of micrometastases and metastases of ILC in a large study population. MATERIALS AND METHODS The current study was based on prospectively collected data. It was carried out between February 2001 and September 2004 at the Breast Surgery Unit of Maria Hospital, Helsinki University Hospital (Helsinki, Finland). Two hundred seventy patients were excluded from the study because they received surgery between January 17, 2003 and September 1, 2003, a time frame during which another rapid IHC technique was evaluated. The remaining 995 patients who underwent a successful SLN biopsy were included in the study. The SLN biopsy specimens of the first 557 patients were evaluated without the use of intraoperative IHC (the non-ihc group). The remaining 438 patients in the rapid IHC group were incorporated into the intraoperative diagnosis. The SLN metastases were assessed by the same experienced breast carcinoma pathologists in both patient groups. The project plan was approved by the ethical committee of Helsinki City University Hospitals. Written informed consent was obtained from each patient. Patient characteristics are presented in Table 1. Surgery For both groups of patients, preoperative lymphatic mapping, a hand-held gamma detector, and blue dye were used to identify the SLNs in the axilla. Lymphoscintigraphy was performed the day before surgery a median of 4 hours after a single intratumoral injection of 100 MBq of 99m Tc-labeled human albumin colloid Nanocoll (Nycomed Amersham Sorin s.r.l., Saluggia, Italy), with particle size 80 nm in a volume of 0.2 ml. At least 5 minutes before incision, 1 ml of patent blue dye was injected intratumorally (Bleu Patenté V; Laboratoire Geuerbet, Aulnay-sous-Bois, France). All focally radioactive and/or blue lymph nodes in the axilla were harvested. Level I II axillary clearance (AC) was performed during the primary surgery for patients with SLN metastases identified in the FS. Patients with false-negative findings in the FS diagnosis underwent Level I II AC as a second surgical procedure. TABLE 1 Patient and Tumor Characteristics of 995 Patients with Breast Carcinoma with Successful Sentinel Lymph Node Biopsy Characteristics Non- IHC group no. (%) (n 557) IHC group no. (%) (n 438) Median age in yrs (range) 57 (25 91) 58 (24 95) Histologic tumor size in mm (range) 15 (2 70) 15 (1 110) Tumor histology Invasive ductal 307 (55) 216 (49) Invasive lobular 134 (24) 111 (25) Other invasives 97 (18) 82 (19) Ductal carcinoma in situ 19 (3) 29 (7) T-classification T0 19 (3) 29 (7) T1a 11 (2) 19 (4) T1b 109 (20) 81 (19) T1c 277 (50) 184 (42) T2 140 (25) 119 (28) Tumor grade I 191 (34) 109 (25) II 235 (42) 183 (42) III 108 (19) 107 (24) Not applicable 4 10 Tumor location Central 49 (9) 42 (10) Upper medial 116 (21) 91 (21) Lower medial 49 (9) 35 (8) Upper lateral 269 (48) 216 (49) Lower lateral 74 (13) 54 (12) IHC: immunohistochemistry. Histologic Methods The SLN specimens were labeled in the operation room to indicate the site of origin and sent to the pathology laboratory as separate specimens. The pathologist removed all extracapsular fat and measured the SLN specimens. The SLN specimens were sliced into 1 1.5-mm thick sections perpendicular to their long axis, as described in our previous study. 10 The slices were arranged flat on prefrozen Tissue-Tek OCT compound (Sakura Fine Technical Co. Ltd., Tokyo, Japan). Touch preparations (imprints) from the surface were first made by pressing the glass slide gently against the still soft surface of the slices, already frozen in place from the bottom. The imprints were then stained with toluidine blue. As the pathologist evaluated the imprints, FS were cut from two levels and stained with toluidine blue. During intraoperative IHC, the adjacent anteceding section of the second cutting level was removed for rapid cytokeratin immunostaining (Cyto-nel Ultrarapid IHC, Immuno Diagnostic Oy, Hämeenlinna, Finland) according to the manufacturer s instructions. As soon as the malignant cells were observed, the pathologist reported to the surgeon in the operation
16 CANCER July 1, 2005 / Volume 104 / Number 1 theatre. The remaining tissue specimen from the FS was thawed, set in cassettes flat between sponges, fixed in buffered formalin, and embedded in paraffin. Specimens for hematoxylin and eosin (H & E) staining were sectioned from two levels. If no cancer cells were detected in the permanent H & E-stained sections, an immunostain for cytokeratin was performed. The first specimens were stained with Cam 5.2 (Becton Dickinson Immunocytometry Systems, San Jose, CA) and later, with Cam 5.2 (which is no longer available), with AE1/AE3- keratin stain (Zymed Laboratories, South San Francisco, CA) as recommended by the manufacturer. After the intraoperative evaluations were performed, all of the remaining SLN tissue specimen was fixed directly in phosphate-buffered 10% formalin. After fixation, the SLN tissue specimens were sliced and embedded wholly in paraffin. H&E sections were made from two levels of each lymph node. In addition, cytokeratin immunostaining was done from one level from the formalin-fixed tissue specimen from patients with a negative FS finding. Metastases 2mmbut 0.2 mm were considered micrometastases. Those not 0.2 mm were called isolated tumor cells (ITC). 11 If the micrometastasis was detectable by IHC only, the prerequisite for diagnosis was that immunostained cells were morphologically cancer cells. Lymph nodes in the AC specimens were embedded wholly in paraffin. H&E sections were prepared from two levels, 200 m apart. Statistical Analysis The Fisher exact test was used to compare the proportional data. The medians were compared using the Mann Whitney U test. RESULTS A median of 2 (range, 1 14) SLNs were harvested in the non-ihc group and a median 2 (range, 1 11) SLNs were harvested in the IHC group. In patients with AC, a median of 17 (3 43) and 19 (2 39) axillary SLNs and non-slns, respectively, were evaluated. SLN metastases were found in 219 of 557 patients (39%) in the non-ihc group and in 175 of 438 patients (40%) in the IHC group. The SLNs were the only metastatic lymph nodes in 150 of 219 patients (69%) in the non-ihc group and in 129 of 175 patients (74%) in the IHC group (P 0.27). The median diamaters of the SLN metastases were 3 mm (range, 0.05 30 mm) and 3 mm (range, 0.05 35 mm) in the non-ihc and IHC groups, respectively. The only tumor-positive SLN findings supported a tendency towards an increased frequency of detection in the IHC group (82 of 175 patients [47%]) compared TABLE 2 Sentinel Lymph Node Metastases and the Use of Intraoperative Immunohistochemistry Characteristics Non-IHC group no. (%) IHC group no. (%) No. of SLN metastases according to metastasis size (mm) 0.2 5/219 (2) 27/175 (15) 0.2 1.0 52/219 (24) 42/175 (24) 1.1 2.0 27/219 (12) 13/175 (8) Micrometastases total 84/219 (38) 82/175 (47) (P 0.206) Macrometastases ( 2.0 mm) 135/219 (62) 93/175 (53) (P 0.101) Micrometastases and ITC as only tumor positive finding 84/219 (38) 82/175 (47) (P 0.206) Micrometastases detected only in FS slides 12/84 (14) 8/82 (10) (P 0.822) IHC: immunohistochemistry; SLN: sentinel lymph node; ITC: isolated tumor cells; FS: frozen section. FIGURE 1. A 0.5-mm sentinel lymph node micrometastasis of invasive ductal carcinoma stained with rapid immunohistochemistry. with the non-ihc group (84 of 219 patients [38%]), but the difference was not statistically significant. The micrometastasis (or ITC) was found only in the intraoperative examination in 12 of 84 patients (14%) in the non-ihc group and in 8 of 82 patients (10%) in IHC group (Table 2). An important finding in the current study was that the ILC sensitivity of the intraoperative diagnosis was 87% (45 of 52 patients) in the IHC group and 66% (39 of 59 patients) in the non-ihc group (P 0.02). The sensitivity of the intraoperative diagnosis was similar in patients with invasive ductal carcinoma (IDC) and other types of invasive cancer in both patient groups (Figs. 1, 2). ITC, the smallest ( 0.2 mm) lymph node metastases, were revealed intraoperatively only in the IHC group (Fig. 3). The sensitivity of the intraoperative diagnosis was similar in the non-ihc and the IHC
Intraoperative IHC and SLN Biopsy/Leikola et al. 17 FIGURE 2. The same micrometastasis stained with toluidine blue. FIGURE 3. Isolated tumor cells, a metastatic finding of invasive lobular carcinoma, was stained by rapid immunohistochemistry. groups, regardless of histologic tumor stage or grade (Table 3). DISCUSSION The major finding in the current study was that the use of rapid IHC cytokeratin biomarker analysis significantly reduced the false-negative rate (FNR) in the intraoperative diagnosis of ILC metastases. With the incorporation of the cytokeratin biomarker, the intraoperative diagnosis of ILC equaled that obtained by IDC. This is not surprising, because ILC as a histologic subtype of breast carcinoma remains a challenge for histopathologic evaluation. 4,12 It metastasizes to lymph nodes in a scattered fashion, lacking severe cytologic atypia. Metastatic cells tend to fill the sinuses, mimicking sinus histiocytosis. 12 Adding IHC to the routine paraffin H&Ediagnostic methods has been shown to provide crucial advantage in revealing ILC metastasis from such morphologic camouflage. IHC reduces the FNR by 10 36% and is, therefore, recommended for routine use especially in patients with ILC. 12 14 ILC is the second most common type of breast carcinoma with increasing incidence rates, especially among women 50 years. 15,16 This has recently been attributed to hormone replacement therapy in this age group. 17 Increasing the sensitivity of diagnosis of ILC intraoperatively is particularly important in this disease because of the difficult clinical and mammographic detection of ILC. 15 Overall, the intraoperative sensitivity (82 92%) for detecting metastasis in IDC has not been a clinicopathologic diagnostic problem. 1 Therefore, no improvement in the intraoperative diagnosis was realized in the most common histopathologic type (IDC) of breast carcinoma when biomarker analysis was incorporated into conventional immunohistochemistry. Thus, no significant differences were realized in the intraoperative diagnosis of micrometastases with or without rapid IHC. However, in the case of detection of micrometastasis in IDC, there was a small benefit of rapid IHC for detecting micrometastases that otherwise would have gone undetected. However, the magnitude of the impact of IHC for upstaging effects was not realized because the proportions of micrometastasis noted only in the FS slides were similar whether IHC was incorporated as an additional biomarker. The use of rapid IHC in IDC did not enhance the overall sensitivity of the intraoperative diagnosis. This is in agreement with previous reports. 5,18 Previous reports did not subclassify the analysis based on tumor histology. Neither did they report the sensitivity of the intraoperative diagnosis separately for metastases of ILC. As regards to micrometastases, the addition of intraoperative IHC to such tedious cryosectioning techniques seems unlikely to improve the already high sensitivity. Because of the distinguishing pathologic features of this neoplasm, the number of tissue sections evaluated is of primary importance when compared with the addition of biomarkers in IHC for finding the small metastatic sites. 5 However, the value of rapid IHC may be contributory in centers that incorporate fewer sections in their intraoperative diagnostic procedures and that routinely experience a lower sensitivity for the intraoperative detection of the disease. Preoperative ultrasonography of the axilla is an efficient method of detecting lymph node metastases, thus reducing the need for SLN procedures in patients with clinically negative lymph nodes. 19 With the in-
18 CANCER July 1, 2005 / Volume 104 / Number 1 TABLE 3 Sensitivity of Intraoperative Diagnosis in Relation to the Size of SLN Metastasis and the Tumor Characteristics Characteristics Non-IHC group (%) IHC group (%) P value Overall intraoperative 170/219 (78) 139/175 (79) 0.712 Sensitivity histologic group DCIS 0/0 (0) 3/3 (100) IDC 106/127 (83) 67/85 (79) 0.470 ILC 39/59 (66) 45/52 (87) 0.015 Other invasive 25/33 (76) 24/35 (69) 0.594 Size of SLN metastasis 0.2 0/5 (0) 13/27 (48) 0.064 0.2 1.0 27/52 (52) 25/42 (60) 0.534 1.1 2.0 16/27 (59) 10/13 (77) 0.316 2.0 127/135 (94) 91/93 (98) 0.206 T-classification T0 0/0 (0) 3/3 (100) T1a 1/1 (100) 4/5 (80) 1.000 T1b 19/29 (66) 14/20 (70) 1.000 T1c 78/109 (71) 45/62 (73) 0.861 T2 4 73/82 (89) 71/83 (86) 0.478 Tumor grade I 49/67 (73) 20/31 (65) 0.536 II 82/106 (77) 69/86 (80) 0.694 III 38/45 (84) 43/50 (86) 0.832 NA 4 6 IHC: immunohistochemistry; SLN: sentinel lymph node; DCIS: ductal carcinoma in situ; IDC: invasive ductal carcinoma; ILC: invasive lobular carcinoma; NA: not applicable. creasing awareness of the benefit of preoperative ulrasonography, our unit has implemented this examination in the diagnostics of an increasing number of patients initially scheduled to receive an SLN biopsy. Because the patients in the IHC group were diagnosed and treated chronologically later than patients in the non-ihc group, the patients in the IHC group received preoperative ultrasonography more frequently for axillary screening than patients in the non-ihc group (data not shown). Therefore, the proportion of patients with SLN micrometastases or ITC only may have been accentuated in the IHC group, because the metastases revealed by axillary ultrasonography are most probably large tumors. The intraoperative examination of SLN specimens is challenging and tedious work. Confirming no evidence of metastasis may be especially timeconsuming, because the pathologist is forced to evaluate multiple cytologic and morphologic criteria in a cancer simulating other cell types that mimic this pathologic group (e.g., histiocytes). Part of this difficulty is attributed to the finding that blue dye, as well as the colloid medium for the radioactive tracer, is ingested by the histiocytes, making the endothelial cells swell and resulting in a pathologic phenotype that is suspicious for malignancy. Furthermore, follicular centers in the lymph nodes may be cut in a plane that simulates a focus of poorly differentiated carcinoma. Differential brown staining of IHC greatly simplifies the pathologic decision because it is easier and quicker when cancer cells are stained brown by a cytokeratin marker to enhance the differential diagnosis. Even though unspecific staining in dendritic cells 20 is common, a differential diagnosis between benign and malignant immunostained cells is not usually a confounding histopathologic dilemma. In addition, the small scattered metastatic lobular carcinoma cells are difficult or even impossible to tell apart from lymphatic cells without IHC. Preparation of the rapid IHC-stained slides takes 15 minutes, and is performed simultaneously with the other preparations that enable an intraoperative diagnosis. Therefore, rapid IHC did not cause any delay in reporting the results to the surgeon. In conclusion, rapid IHC enhances the intraoperative diagnosis of SLN metastases in patients with ILC. It also may improve the intraoperative diagnosis of micrometastases. REFERENCES 1. Leidenius MH, Krogerus LA, Toivonen TS, Von Smitten KJ. The feasibility of intraoperative diagnosis of sentinel lymph node metastases in breast cancer. J Surg Oncol. 2003;84:68 73. 2. Ronka R, Smitten K, Sintonen H, et al. The impact of sentinel node biopsy and axillary staging strategy on hospital costs. Ann Oncol. 2004;15:88 94. 3. Veronesi U, Paganelli G, Viale G, et al. Sentinel lymph node biopsy and axillary dissection in breast cancer: results in a large series. J Natl Cancer Inst. 1999;91:368 373. 4. Holck S, Galatius H, Engel U, Wagner F, Hoffmann J. Falsenegative frozen section of sentinel lymph node biopsy for breast cancer. Breast. 2004;13:42 48. 5. Viale G, Bosari S, Mazzarol G, et al. Intraoperative examination of axillary sentinel lymph nodes in breast carcinoma patients. Cancer. 1999;85:2433 2438. 6. D Errico A, Grassigli A, Gruppioni E, et al. Thorough intraoperative analysis of breast sentinel lymph node biopsies: histologic and immunohistochemical findings. Surgery. 2004;135:248 254; discussion 255 257, 357. 7. Karsten U, Stosiek P. Fast and sensitive immunodetection of carcinoma cells in sentinel nodes. Virchows Arch. 2002;440: 325 329. 8. Nahrig JM, Richter T, Kuhn W, et al. Intraoperative examination of sentinel lymph nodes by ultrarapid immunohistochemistry. Breast J. 2003;9:277 281. 9. Beach RA, Lawson D, Waldrop SM, Cohen C. Rapid immunohistochemistry for cytokeratin in the intraoperative evaluation of sentinel lymph nodes for metastatic breast carcinoma. Appl Immunohistochem Mol Morphol. 2003; 11:45 50.
Intraoperative IHC and SLN Biopsy/Leikola et al. 19 10. Krogerus LA, Leidenius MH, Toivonen TS, von Smitten KJ. Towards reasonable workload in diagnosis of sentinel lymph nodes: comparison of two frozen section methods. Histopathology. 2004;44:29 34. 11. Sobin LH, Wittekind C, editors. TNM classification of malignant tumours. 6th ed. New York: John Wiley & Sons, 2002. 12. Bussolati G, Gugliotta P, Morra I, Pietribiasi F, Berardengo E. The immunohistochemical detection of lymph node metastases from infiltrating lobular carcinoma of the breast. Br J Cancer. 1986;54:631 636. 13. Turner RR, Giuliano AE. Intraoperative pathologic examination of the sentinel lymph node. Ann Surg Oncol. 1998;5: 670 672. 14. Cote RJ, Peterson HF, Chaiwun B, et al. Role of immunohistochemical detection of lymph-node metastases in management of breast cancer. International breast cancer study group. Lancet. 1999;354:896 900. 15. Li CI, Anderson BO, Daling JR, Moe RE. Trends in incidence rates of invasive lobular and ductal breast carcinoma. JAMA. 2003;289:1421 1424. 16. Li CI, Anderson BO, Porter P, Holt SK, Daling JR, Moe RE. Changing incidence rate of invasive lobular breast carcinoma among older women. Cancer. 2000;88:2561 2569. 17. Li CI, Weiss NS, Stanford JL, Daling JR. Hormone replacement therapy in relation to risk of lobular and ductal breast carcinoma in middle-aged women. Cancer. 2000;88:2570 2577. 18. Zurrida S, Mazzarol G, Galimberti V, et al. The problem of the accuracy of intraoperative examination of axillary sentinel nodes in breast cancer. Ann Surg Oncol. 2001;8:817 820. 19. Deurloo EE, Tanis PJ, Gilhuijs KG, et al. Reduction in the number of sentinel lymph node procedures by preoperative ultrasonography of the axilla in breast cancer. Eur J Cancer. 2003;39:1068 1073. 20. Xu X, Roberts SA, Pasha TL, Zhang PJ. Undesirable cytokeratin immunoreactivity of native nonepithelial cells in sentinel lymph nodes from patients with breast carcinoma. Arch Pathol Lab Med. 2000;124:1310 1313.