The status of the axillary lymph nodes is the most important prognostic factor in breast cancer, making accurate axillary staging information essential
Royal Surrey County
There are almost 430,000 women diagnosed with breast cancer in Europe every year, with 45,000 of these women diagnosed in the UK. The annual incidence of breast cancer is rising and this is thought to be related to an increased detection rate due to screening as well as lifestyle factors.[1,2] Since the early 1980s, breast cancer mortality has been falling due to a combination of screening, increased specialisation of care and improvements in treatment options.
The UK five-year survival rate for breast cancer is over 80%4 so it is crucial that our surgical treatment offers the highest chance of cure with the minimal risk of lasting complications for the patient. The surgical management of breast cancer has come a long way from a patient having an operation to have a lump removed and not knowing whether or not she will wake up to find she has had a Halstead radical mastectomy. Modern techniques are minimally invasive and offer excellent cosmetic results without compromising oncological safety.
The importance of the axilla in breast cancer
It has long been known that the status of the axillary lymph nodes is the most important prognostic factor in breast cancer. This is a factor taken into account in the widely used Nottingham Prognostic Index,[5,6] which is a validated predictive tool for breast cancer survival. The presence of metastatic disease in the axilla determines the adjuvant therapy offered to a patient and the likelihood of disease elsewhere. It is therefore important that we obtain accurate axillary staging information about all our patients with invasive breast cancer.
The aims of surgery in the axilla are both staging and local control of disease.[7,8] Some patients are proven to have metastatic cancer in the axilla preoperatively and local control could be gained with radiotherapy in this situation with no adverse impact on survival.[l9,10] However, the number of positive nodes is an important prognostic indicator which helps to guide adjuvant therapy and this information to date can only be gleaned from axillary dissection. The current BASO and NHSBSP guidelines recommend surgical staging of the axilla in all patients with invasive cancer.
With the advent of screening programmes, the proportion of the women we detect with breast cancer with node-negative disease has increased. Up to 80% of patients in the NHS Breast Screening Programme are node negative. [12 ]Traditionally, all these women would have had axillary clearance. The main complications of axillary clearance are lymphoedema, shoulder dysfunction and severe neuralgia. Numbness and seromas are common but mild complications.
Despite this, the gold standard for assessment of axillary nodes remains surgery with preoperative assessment with ultrasound and fine needle aspiration cytology having a sensitivity of 65%.
Development of the sentinel node technique
The low rate of lymph node positivity in screen detected cancer was noticed early on in the history of the screening programme by many surgeons. It was recognised that in patients with a high chance of a clear axilla and a long life expectancy everything possible should be done to minimise the complications associated with axillary clearance. This led to the development of node sampling. A minimum of four nodes were taken from level I in the axilla and by 2000 this technique was offered by up to a third of UK surgeons. The concept of sentinel node biopsy had been around for some time at this stage and had been used in penile carcinoma and melanoma.[15,16] The definition of the sentinel node, according to Morton, is ‘Any node receiving direct lymphatic drainage from the primary tumour’. The idea behind sentinel lymph node biopsy is that lymph node flow is ordered and the tumour cells flow through a sentinel node or nodes before spreading higher in the axilla. Analysis of the sentinel node allows us to predict the presence of cancer elsewhere in the axilla.
The first paper that recognised that this technique may be applied to breast cancer patients was published in 1993. McMasters performed an analysis of 1,385 breast cancer patients who had sentinel node biopsy followed by axillary clearance and proved that sentinel node was very accurate with a sensitivity and specificity of 94% and 100%, respectively.
The NSABP-32 trial showed that this technique was technically feasible on a large number of patients and the UK ALMANAC trial proved that sentinel node biopsy achieved the aim of accurately staging the axilla while reducing morbidity and improving quality of life in a growing number of lymph node negative patients.[19,20] It became apparent that widespread adoption of the sentinel node technique could save 20,000 women in the UK every year from having a full axillary clearance.
Sentinel node biopsy in practice
The NSABP B-32 trial included a training phase for all the surgeons who entered patients into the trial, which varied according to the prior experience of the surgeon. It has also been clearly demonstrated in the UK that there is a learning curve associated with the procedure, and so a rigorous training programme has been developed to ensure that all surgeons reach a minimum standard before using this technique. The UK New Start Program is the largest structured validated surgical training programme. The programme starts with a theory day to which the entire team is invited. Accredited New Start trainers then visit the surgeon in their own hospital to proctor them on an initial five cases and then surgeons must submit data on 25 further patients who have both sentinel node
and axillary clearance. Validation standards were a localisation rate of >90% and false-negative rate of <10%.
From the validation studies of the sentinel node technique, we know that the procedure is associated with a false-negative rate. Axillary recurrence is rare in patients who have had full axillary dissection[23,24] but recent studies with long-term follow-up data of patients who have had sentinel node biopsy show that axillary recurrence rate is lower than expected.
All patients who have cancer should now have a preoperative ultrasound of the axilla. If the node appears suspicious, then a fine needle aspirate is taken. If this proves nodal involvement the patient proceeds to axillary clearance in the usual way. All patients who are node negative on preoperative staging should be offered a sentinel node biopsy.
A combination of blue dye and isotope is used to reduce the risk of failing to identify the sentinel node and the consensus is that this is the preferred technique. The isotope, Tc99mcolloid is injected intradermally and the blue dye, Patent Bleu V, is injected subdermally in the periareolar area in the tumour quadrant. Allergic reactions to the blue dye have been documented but in an analysis of the ALMANAC data and New Start experience the incidence of adverse reactions to the blue dye was low (0.9%) which were mostly mild.
As the technique has developed it has been shown that sentinel node biopsy can be used safely in pregnancy and accurately in patients who have primary chemotherapy.[28,29]
Dealing with the positive sentinel node
The increased focus on macroscopic examination of a lymph node, made possible by the sentinel node technique and the ability to see individual cells with immunohistochemistry, has increased our understanding of tumour deposits in the node.
Lymph node metastases are now grouped according to size. Macrometastases are tumour deposits which measure more than 2mm. Micrometastases are <0.2mm in size and isolated tumour cells are individual cells seen in the node.
The clinical significance of isolated tumour cells within a sentinel lymph node remains to be seen. Current guidelines are that patients with macro or micro metastases should undergo further axillary surgery but that patients with isolated tumour cells should be treated as node negative. The treatment of the patient with a positive sentinel node is the subject of a number of ongoing trials.[30,31] Nomograms have been developed to predict the positivity of non-sentinel nodes after a sentinel node biopsy and these can be discussed with the patient before opting for a second operation.[32,33]
Routine intraoperative analysis
In most centres, patients return to clinic for their pathology results at least a week after sentinel node biopsy. It became apparent that there were a number of patients who were staged negative preoperatively who had a positive sentinel node biopsy. These patients are currently brought back to theatre for a second operation. Operating in the axilla within a month of the first operation can prove difficult due to scarring and the anatomy may be difficult to define. Although there is no data to support this, current feeling is that in these patients’ axillary morbidity may actually be increased.
This has led to many centres exploring options to analyse the sentinel node while the patient is anaesthetised during the initial operation. Patients who have tumour in the sentinel node can then proceed to completion axillary clearance during the same anaesthetic. Although many of these techniques increase the operating time for individual patients, there is evidence to support the fact that overall theatre time and bed days are saved due to a decreased second operation rate for these patients. The aim is to find a technique that is highly sensitive and specific, reproducible, easy to use and of benefit to the patient.
Many centres in Europe use the intraoperative frozen section of the sentinel node. This technique has a reported accuracy of 90.7% and requires a pathologist for diagnosis. Imprint cytology is a simpler technique to prepare but this can be difficult to interpret and has a reported sensitivity of 57-69%. These techniques may not detect micrometastases in the sentinel node, which do appear to have an effect on prognosis. Rapid molecular-based techniques use PCR for mRNA amplification and much work has been done on finding the perfect marker combination that allows us to detect the presence of tumour in the lymph node accurately. The initial problem appeared to be that the entire node needed to be
submitted for analysis but it has since become apparent that this allows more accurate assessment of the node.
The ‘OSNA’ – One Stop Nucleic Acid Amplification – enables rapid analysis of the node for Cytokeratin 19 – a gene protein present in metastatic breast cancer. It has the advantage of being relatively quick to use, allowing intraoperative analysis, and it requires a technician rather than a pathologist to operate it. A similar system – Veridex – measures Cytokeratin 19 and mammaglobin. Both systems have been trialled and proven to be sensitive and specific for nodal metastases.[36, 37, 38]
Patients find intraoperative analysis of the sentinel node acceptable, despite uncertainty about whether or not they are going to wake up having had an axillary clearance.
Axillary nodal status remains the most important prognostic feature in breast cancer patients. The most accurate method to date of staging the axilla remains surgery. With an increasing number of patients presenting with node-negative disease and having a good outlook, it is important that we achieve this with minimal complications from surgery. Sentinel node biopsy has proven to be an accurate method of staging the axilla with reduced morbidity compared to axillary clearance. The roll out of sentinel node biopsy in the UK with the New Start programme has introduced an element of quality assurance to this new technique. The removal of a smaller number of nodes per patient has allowed us to focus more on the analysis of the node and to gain a greater understanding of the implications of a positive node. The emphasis of nodal analysis may shift from histological to molecular and to intraoperative analysis, thus improving the accuracy of axillary staging and the patients’ surgical experience.
2. J. Ferlay et al. Ann Oncol 2007;18(3):581-92.
3. Office for National Statistics Mortality Statistics: Cause. England and Wales 2007 London TSO 2009.
4. Office for National Statistics, Long-term Breast cancer Survival, England and Wales, up to 2003.
5. Galea MH et al. Breast Cancer Res Treat 1992;22(3):207-19.
6. Blamey RW et al. Eur J Cancer 2007;43(10):1548-55.
7. Veronesi U et al. Eur J Surg Oncol 1990;16(2):127-33.
8. Graverson HP et al. Eur J Surg Oncol 1998;14(5):407-12.
9. Louis-Sylvestre C et al. J Clin Oncol 2004;22(1):97-101.
10. Fisher et al. N Engl J Med 1985;312:674-81.
11. Quality Assurance Guidelines for Surgeons in Breast Cancer Screening, NHSBSP Publication No 20, 4th Edition, March 2009.
12. ABS at BASO Breast Screening Audit 2007-2008.
13. Alkuwari E et al. Cancer 2008;114(2):89-93.
14. Mansel RE. Breast Cancer Res 2000;2(S2):A8.
15. Cabanas RM. Cancer 1977;39(2):456-66.
16. Morton DL et al. Arch Surg 1992;127(4):392-9.
17. Krag DN. Surg Oncol 1993;2(6):335-9.
18. McMasters et al. N Engl J Med 1998;339(14):990-5.
19. Krag DN et al. Lancet Oncol 2007;8(10):881-8.
20. Mansel RE et al. J Natl Cancer Inst 2006;98(9):599-609.
21. Krag DN et al. J Natl Cancer Inst 2009;101(19):1356-62.
22. Clarke D et al. Ann Surg Oncol 2004;11(3):211S-15S.
23. Rosen PP et al. Ann Surg 1983;197(3):276-83.
24. Van Lancker et al. Am J Clin Oncol 1995;18(3):267-72.
25. Veronesi U et al. Eur J Cancer 2009;45(8):1381-8.
26. Schwartz GF et al. Hum Pathol 2002;33:579-89.
27. Barthelmes L et al. Eur J Surg Oncol 2009 Nov 23. [Epub ahead of print].
28. Spanheimer PM et al. Ann Surg Oncol 2009;16(5):1143-7.
29. Xing Y et al. Br J Surg 2006;93(5):539-46.
30. EORTC AMAROS trial 10981/22023.
31. Olson JA et al. J Clin Oncol 2008;26(21):3530-5.
32. Van Zee KJ al. Ann Surg Oncol 2003;10(10):1140-51.
33. Gur AS et al. J Am Coll Surg 2009;208(2):229-35.
34. van de Vrande S et al. Eur J Surg Oncol 2009;35(3):276-80.
35. Tew K et al. Br J Surg 2005;92(9):1068-80.
36. Tamaki Y et al. Clin Cancer Res 2009;15(8):2879-84.
37. Mansel RE et al. Breast Cancer Res Treat . 2009;115(3):595-600.
38. Visser M et al. Int J Cancer 2008;122(11):2562-7.
39. Chicken DW et al. Int J Surg 2007;5(2):76-80