Original Article
Sentinel Lymph Node Biopsy Assessment Using Intraoperative Imprint Cytology in Breast Cancer Patients: Results of a Validation Study Suryanarayana Deo, Atul Samaiya, Paresh Jain,1 Sonal Asthana, Mona Anand,1 Nootan K. Shukla and Rajive Kumar,1 Departments of Surgical Oncology and 1Laboratory Oncology, Institute Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi, India.
OBJECTIVE: Sentinel lymph node biopsy (SLNB) in breast cancer patients is emerging as a promising minimallyinvasive tool. There has been an exponential increase in the literature related to sentinel lymph nodes (SLN) in breast cancer patients, mainly from Western centres. This study was carried out to address issues relevant to breast cancer patients in developing countries, including the method of SLN detection, the role of imprint cytology in the assessment of SLN, and the role of SLNB in locally advanced breast cancer (LABC). METHODS: This study included 76 women with breast cancer. The blue-dye method was used to identify the sentinel node. Touch imprint smears were prepared from the sectioned node, stained using the Jenner-Geimsa technique, and examined for tumour deposits. RESULTS: Sentinel nodes were identified in 69 of 76 patients. The sensitivity, specificity and accuracy of SLNB in predicting axillary node status were 84.2%, 100% and 91.3%, respectively. The sensitivity, specificity and accuracy of intraoperative imprint cytology were 96.9%, 100% and 98.6%, respectively. CONCLUSIONS: These results prove that high levels of SLN detection can be achieved using the blue-dye method alone. Its role in LABC patients needs further evaluation. In view of promising results, imprint cytology should be used more frequently as an alternative to frozen section for the assessment of sentinel nodes. [Asian J Surg 2004;27(4):294–8]
Introduction Over the past few decades, surgery for breast carcinoma has evolved from radical mastectomy to breast conservation. At present, axillary lymph node dissection (ALND) remains the standard of care for surgical management of all patients with invasive breast carcinoma, as it provides precise staging and excellent local control.1,2 However, routine ALND exposes a large number of patients, particularly those with nodenegative breast cancer, to unnecessary perioperative risk and long-term morbidity in the form of neuropathies, lymphoedema and shoulder syndrome.3 Contrary to the Halste-
dian concept of breast cancer, the Fisherian hypothesis postulates that breast cancer is a systemic disease and also questions the validity of routine axillary dissection in the quest for breast cancer cure.4 The concept of sentinel lymph node biopsy (SLNB) emerged in the early 1970s to obviate the need for routine lymphadenectomy and to identify the subset of patients at risk of regional and systemic spread. The initial work on sentinel nodes was performed in malignant melanoma patients;5 with the encouraging results, SLNB was tried in other solid tumours,6 including breast cancer.7–9 There has been an exponential increase in the literature related to SLNB in breast cancer, and
Address correspondence and reprint requests to Dr. Suryanarayana Deo, Department of Surgical Oncology, Institute Rotary Cancer Hospital, All India Institute of Medical Sciences, New Delhi–110029, India. E-mail:
[email protected] • Date of acceptance: 25 September, 2003 © 2004 Elsevier. All rights reserved.
294
ASIAN JOURNAL OF SURGERY VOL 27 • NO 4 • OCTOBER 2004
070/2001
■ IMPRINT CYTOLOGY OF SENTINEL LYMPH NODES ■
most Western centres have started advanced work to evaluate the therapeutic role of SLNB in early breast cancer patients.7–9 More recently, there have also been reports of SLNB in locally advanced breast cancer (LABC).10,11 There is, however, a paucity of literature on SLNB from developing countries. There is a need to address certain issues relevant to breast cancer patients in developing countries, including standardization and validation of the SLNB technique, the role of SLNB in LABC, and finding cheaper and reliable alternatives to frozen-section and gamma-probe methods. This study was undertaken to address these issues.
Patients and methods This prospective validation study, undertaken in a tertiary care cancer centre (Institute Rotary Cancer Hospital, All India Institute of Medical Sciences), included 76 women with operable breast carcinoma (Stage I, II or III) between January 2000 and December 2002. Six patients with LABC also received neoadjuvant chemotherapy. Clinical examination and mammogram were performed for all patients to assess the primary tumour and axillary node status. Informed consent was obtained from all patients. Breast conservation surgery was performed in 22 and modified radical mastectomy in 54 patients. Level I, II or III ALND was performed in all patients after SLNB evaluation.
SLNB technique A standard blue-dye method was used as described by Giuliano et al.12 Isosulfan blue dye (4 mL of 1%; Sigma-Aldrich, Milwaukee, WI, USA) was injected peritumourally at the 12, 3, 6 and 9 o’clock positions. Centripetal massage was applied around the injection site for 2 to 3 minutes to facilitate the diffusion of dye into the lymphatics. The axilla was explored through an oblique 5 cm incision behind the anterior axillary fold 5 minutes after the injection of dye. Meticulous dissection was performed in the subcutaneous plane to identify the blue lymph vessel. Once identified, the blue lymph vessel was traced to the blue node distally and primary tumour proximally. The first draining blue node was labelled the sentinel node and removed for analysis. Further exploration was performed for any second-echelon nodes before proceeding to definitive surgery, including complete axillary dissection irrespective of the outcome of SLNB.
Intraoperative assessment of SLNB Intraoperative imprint cytology (IIC) assessment of the SLN
ASIAN JOURNAL OF SURGERY VOL 27 • NO 4 • OCTOBER 2004
070/2001
was performed instead of standard frozen sections. After obtaining imprint smears, the SLN was submitted separately for histopathological examination. The final histopathology report was taken as a gold standard for analysing the accuracy of IIC results. The sentinel node was sectioned at two levels. Touch imprint smears were taken from all four cut surfaces. The smears were air-dried for 5–6 minutes and then stained using a rapid Jenner-Geimsa technique: the slide was stained with Jenner stain for 2 minutes and counterstained with Geimsa for 7 minutes. Two oncopathologists evaluated the slides independently. All sections were scanned for metastatic tumour deposits and were reported as positive or negative for malignancy (Figure). The identification rate of SLN, sensitivity, specificity and accuracy of SLNB and IIC, with final histopathology as gold standard, were calculated in a standard fashion. In addition, the negative predictive value of SLNB was calculated.
Results Sixty-one patients had stage I or II breast cancer and 15 had stage III breast cancer according to pTNM. The mean tumour size was 3 cm (1–7 cm). The mean age of patients was 50.6 years (range, 30–68 years). The tumour was located in the upper outer quadrant in 48 patients, lower outer quadrant in eight, upper inner quadrant in eight and lower inner quadrant in three. The tumour was central in six patients and multicentric in three.
SLNB The SLN (blue node) was identified in 69 of 76 patients, with an overall identification rate of 90.8%. With increasing ex-
Figure. Touch imprint of an axillary sentinel lymph node showing a cluster of large tumour cells against a background of lymphocytes (original magnification ×1,000).
295
■ DEO AND OTHERS ■
perience, the rate of identification increased to 96% in the last 50 cases. The median time taken for SLN identification was 12 minutes (range, 8–20 minutes). A total of 99 SLNs were identified in the 69 patients. The mean number of SLNs per patient was 1.43 (range, 1–4). In 62 patients, the first draining node was found at level I, while six patients had an SLN at level II. One patient had a positive intramammary SLN away from the axillary basin. Sentinel node identification could correctly predict the nodal status of 63 of the 69 patients evaluated, with an accuracy of 91.3%. The overall sensitivity, specificity and negative predictive value of SLNB in predicting axillary node status were 84.2%, 100% and 83.4%, respectively (Table). The SLN was the only node involved in 12 of the 32 cases with positive axillary nodes. Among the 15 LABC patients, the SLN was identified in 11 (73.3%), and it was truly representative of remaining axillary node status in 10 patients (66.7%). The average number of nodes harvested from each patient at ALND was 17 (range, 7–33), and the mean number of positive nodes was three (range, 1–28) in 32 patients with nodal positivity. Extranodal spread was seen in only one patient.
Intraoperative imprint cytology The SLNs from 69 patients were analysed using the imprint cytology technique. In the final histopathology assessment, SLNs showed metastases in 32 of 69 patients and IIC correctly identified 31 of these 32 positive patients. There was only one false-negative report on IIC and all 37 patients with a negative SLN on final histopathology were accurately diagnosed using IIC. The average reporting time for IIC was 15 minutes. The overall sensitivity, specificity and accuracy of IIC were 96.9%, 100% and 98.6%, respectively. The overall accuracy of SLNB in combination with IIC in predicting axillary node status was 89.9%.
Discussion Traditionally, surgery plays a major role in the management of breast cancer, and axillary dissection is considered an integral part of surgical management. Recently, however, its curative role has been questioned, and less invasive methods such as SLNB are being used to stage the disease accurately. The basis for this paradigm shift in the management of the axilla was the questionable therapeutic role of axillary dissection in nodenegative patients and its attendant morbidity in 30–40% of patients.8,13,14 With the standardization of technique and encouraging results obtained in the field of skin and penile cancer,5,15 SLNB
296
Table. Results of sentinel lymph node biopsy (SLNB) and axillary lymph node status on histopathological examination Axillary lymph node status
SLNB Positive Negative
Positive
Negative
32 16
10 31
has been studied extensively in early breast cancer patients in the West. The overall trend seems to be shifting towards conservative axillary dissection, as happened with breast conservation surgery in the 1980s. However, the spectrum of disease is different in developing countries and, in the absence of proper guidelines, SLNB can be misused, leading to inappropriate and unscientific application of the technique. The issues that need to be studied in developing countries are the technique and validation of the SLNB procedure, the role of SLNB in LABC patients, and finding reliable and less expensive alternatives to frozen-section and gamma-probe methods. Two techniques of SLNB are described in the literature, the blue-dye method and the radiocolloid method (gamma probe). In 1993, Krag et al reported an 82% success rate in SLN identification in 22 patients using 99mTc sulfur colloid.7 Giuliano et al first described intraoperative SLN detection using blue dye in 174 patients, with a 66% SLN identification rate and 12% false-negative rate.13 A combination of blue dye and radiocolloid scan may yield better results. Albertini et al used a combination of the two techniques and reported a 92% success rate in 62 breast cancer patients.16 Many authors have reported increased accuracy using the double technique (dye and radiocolloid).16,17 However, in a randomized trial, Morrow et al showed no difference in the rate of SLN identification when using blue dye alone or in combination with the radiocolloid technique.18 SLN identification has an initial learning curve.12,13,18 In the present series, during the latter half of the study, the SLN identification rate reached 96%. The overall SLN identification rate of 90.8% using blue dye alone is comparable to those with the combination method in the literature. The learning curve of SLN using blue dye is small and surgeons doing routine breast surgery can easily master the dye method by performing 20–25 cases. The overall sensitivity of the SLN in predicting axillary node status in the current study was relatively low (84.2%), although specificity was 100% and accuracy was 91.3%. The negative predictive value of SLNB for axillary lymph node status was 83.8%. The reported rates of sensitivity and specificity
ASIAN JOURNAL OF SURGERY VOL 27 • NO 4 • OCTOBER 2004
070/20
■ IMPRINT CYTOLOGY OF SENTINEL LYMPH NODES ■
for SLNB range from 66% to 100% in the literature.9,13 The reasons for low sensitivity include previous excisional biopsy, multicentric tumour, heavy nodal disease and large tumour size.19,20 The low sensitivity in the current study could be explained by the patient profiles. In two of six patients with Tx lesions, one of three patients with multicentric tumours, and all three patients with heavy nodal disease, the SLN was falsely negative and could not predict axillary status accurately. The second issue to be addressed is the role of SLNB in LABC. The inherent problems of LABC are large tumour size, prior chemotherapy and heavy nodal disease leading to altered lymphatic pathways. However, recent studies indicate that the SLN can be identified accurately despite the large tumour size and prior chemotherapy.10,11,21,22 In the present study, the SLN was identified in 11 of 15 LABC patients (73.3%), and the false-negative rate was 9.1%. Stearns et al report an SLN identification rate of 85% with a false-negative rate of 14% in 34 patients with LABC after neoadjuvant chemotherapy.11 Similarly, Cohen et al reported an 82% identification rate and a 12% false-negative rate in 38 patients after neoadjuvant chemotherapy.22 In the current study, the number of patients in this subgroup was too small to allow a definitive conclusion. The role of SLNB in LABC patients, especially after neoadjuvant chemotherapy, remains investigational. The third issue that was evaluated in the present study was intraoperative assessment of SLNB. Traditionally, frozensection analysis is considered the method of choice for intraoperative assessment.9,23,24 However, these facilities are not widely available in many centres in developing countries, and the accuracy of frozen section varies from 78% to 95%, with a false-negative rate of 6–24%.9,23,24 In addition, the frozensection method is associated with 25–50% tissue loss and freezing artefacts.23,24 IIC is emerging as an alternative to frozen section, with promising results. Rubio et al first reported experience with IIC in SLN assessment in breast cancer patients with 88.2% sensitivity and 100% specificity.25 Motomura et al reported sensitivity, specificity and accuracy of 96.7%, 91.9% and 92.8%, respectively, using IIC for SLN assessment in 101 breast cancer patients.26 Recently, however, Shiver et al and Lee at al reported relatively low sensitivities (56% and 65%) using IIC for SLN assessment.27,28 The specificity of IIC was 100% in both studies. In comparison, the sensitivity and specificity of IIC in the present study were 96.9% and 100%, resulting in a high accuracy of 98.6%. IIC was preferred over frozen section because it is less expensive, technically easy and does not require additional infrastructure.27
ASIAN JOURNAL OF SURGERY VOL 27 • NO 4 • OCTOBER 2004
070/2001
Conclusions These results show that a high SLNB detection rate can be achieved using the blue-dye method alone. The role of SLNB in LABC patients, especially after neoadjuvant chemotherapy, remains investigational. The overall results of IIC are very promising, and we feel that it should be used as an alternative to frozen section in view of its simplicity, low cost and high accuracy.
References 1. Senofsky GM, Moffat FL Jr, Davis K, et al. Total axillary lymphadenectomy in management of breast cancer. Arch Surg 1997;126: 1336–42. 2. Noguchi M, Miwa K, Michigishi T, et al. The role of axillary lymph node dissection in breast cancer management. Breast Cancer 1997;4: 143–53. 3. Ivens D, Hoe L, Podd TJ, et al. Assessment of morbidity from complete axillary dissection. Br J Surg 1992;66:136–8. 4. Fisher B. The surgical dilemma in the primary therapy of invasive breast cancer: a critical appraisal. Curr Probl Surg 1970;Oct:1–53. 5. Leong SP. The role of sentinel lymph node in malignant melanoma. Surg Clin North Am 2000;80:1741–57. 6. Lim RB, Wong JH. Sentinel lymphadenectomy in gynecological malignancies and solid tumors other than melanoma and breast cancer. Surg Clin North Am 2000;80:1787–98. 7. Krag DN, Weaver DL, Alex JC, et al. Surgical resection and radio localization of sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993;2:335–40. 8. Giuliano AE. Sentinel lymphadenectomy in primary breast carcinoma: an alternative to routine axillary dissection. J Surg Oncol 1996;62: 75–7. 9. Veronesi U, Paganelli G, Galimberti V, et al. Sentinel node biopsy to avoid axillary dissection in breast cancer with clinically node negative lymph nodes. Lancet 1997;349:1864–9. 10. Breslin TM, Cohen L, Sahin A, et al. Sentinel lymph node biopsy is accurate after neo-adjuvant chemotherapy for breast cancer. J Clin Oncol 2000;18:3480–6. 11. Stearns V, Ewing CA, Slack R, et al. Sentinel lymphadenectomy after neo-adjuvant chemotherapy for breast cancer may reliably represent the axilla except for inflammatory breast cancer. Ann Surg Oncol 2002;9:235–42. 12. Giuliano AE, Jones RC, Brenna M, et al. Sentinel lymphadenectomy in breast cancer. J Clin Oncol 1997;15:2345–50. 13. Giuliano AE, Kirgan DM, Guentnor JM, et al. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391–401. 14. Noguchi M. Sentinel lymph node biopsy as alternative to routine axillary lymph node dissection in breast cancer patients. J Surg Oncol 2001;76:144–56. 15. Cabanas RM. An approach for the treatment of penile carcinoma. Cancer 1977;39:456–66. 16. Albertini JJ, Lyman GH, Cox C, et al. Lymphatic mapping and sentinel lymphadenectomy in patients with breast cancer. JAMA 1996;
297
■ DEO AND OTHERS ■
276:1818–22. 17. Borgstein P, Pijpers R, Comans EF, et al. Sentinel lymph node in breast cancer; guidelines and pitfalls of lymphoscintigraphy and gamma probe detection. J Am Coll Surg 1998;186:275–83. 18. Morrow M, Rademaker AW, Bethke KP, et al. Learning sentinel lymph node biopsy: results of a prospective randomized trial of two techniques. Surgery 1999;126:714–22. 19. Veronesi U, Paganelli G, Viale G, et al. Sentinel lymph node biopsy and axillary dissection in breast carcinoma: results in a large series. J Natl Cancer Inst 1999;91:368–73. 20. Noguchi M, Bando E, Tsugawa K, et al. Staging efficacy of breast cancer and sentinel lymphadenectomy. Breast Cancer Res Treat 1999; 57:221–9. 21. Offodile R, Hoh C, Barsky SH, et al. Minimal invasive breast carcinoma staging using lymphatic mapping with radio colloid Dextran. Cancer 1998;82:1704–8. 22. Cohen LF, Breslin TM, Kuerer HM, et al. Identification and evaluation of axillary sentinel lymph node in patients with breast carci-
298
noma treated with neo-adjuvant chemotherapy. Am J Surg Pathol 2000;24:1266–72. 23. Von Diest PJ, Torrenga H, Borgstein PJ, et al. Reliability of intraoperative frozen section and imprint cytology investigation of sentinel lymph node in breast cancer. Histopathology 1999;35:14–8. 24. Pfeifer JD. Sentinel lymph node biopsy. Am J Clin Pathol 1999;112: 599–602. 25. Rubio IT, Korourian S, Cowan C, et al. Sentinel lymph node biopsy for staging of breast cancer. Am J Surg 1998;176:532–7. 26. Motomura K, Inaji H, Komoike Y, et al. Intra-operative sentinel lymph node examination by imprint cytology and frozen section during breast surgery. Br J Surg 2000;87:597–607. 27. Shiver SA, Creager AJ, Geisinger K, et al. Intraoperative analysis of sentinel lymph nodes by imprint cytology for cancer of the breast. Am J Surg 2002;184:424–7. 28. Lee A, Krishnamurthy S, Sahin A, et al. Intraoperative touch by imprint of sentinel lymph nodes in breast cancer patients. Cancer 2002;96;225–31.
ASIAN JOURNAL OF SURGERY VOL 27 • NO 4 • OCTOBER 2004
070/20
