Implementation of molecular-based intraoperative sentinel node analysis: practical, logistical and controversial issues.
Sentinel node biopsy is the established standard for staging the axilla in clinically node-negative breast cancer . It can be carried out in all patients who have undergone preoperative assessment showing no evidence of axillary disease on imaging, or needle biopsy
The ability to analyse the sentinel node status during the operation allows the patient to have a therapeutic axillary clearance at the same time if malignancy is detected within the node. This potentially avoids a delayed second hospital admission, anaesthetic and operation. Between 25% and 30% of patients undergoing sentinel node biopsy have a positive biopsy [2,3].
Intraoperative techniques for assessing sentinel nodes date back over a decade. Today's established techniques include cytological, histological and, most recently, molecular-based analyses. A review of the different techniques has reported differences in sensitivity and specificity for frozen-section and imprint cytology [4,5], with each having specific advantages and disadvantages (Figure 1). Molecular techniques have the potential to achieve greater sensitivity, in part due to the reduction in sampling error associated with a less thorough examination of a node .
We have been carrying out sentinel node biopsy in our unit since 1999 (ALMANAC Trial) and since 2004 as standard treatment. We introduced the GeneSearch [TM] Breast Lymph Node Assay (BLNA) (Veridex, Warren, New Jersey, USA), the first commercially available qRT-PCR (quantitative reverse transcriptase- polymerase chain reaction) assay for intraoperative assessment of sentinel lymph node material, into our breast unit in November 2007. The kit uses two gene markers, mammoglobin (MGB) 1, a breast-specific marker, and cytokeratin (CK) 19, an epithelial marker. This FDA-approved/CE-marked test detects mRNA from the marker genes that are expressed in breast and epithelial cells but not in lymphoid cells. The other commercially available test. One Step Nucleic Acid Amplification (OSNA) (Sysmex, Kobe, Japan) works in a similar fashion but only looks at one gene marker: CK19.
Before any new intervention or procedure is carried out in a trust, internal governance approval has to be sought and a business case made that evaluates the pros and cons. Costs have to be taken into account for equipment, additional personnel required and theatre utility, and these balanced against the additional costs associated with a second hospital admission for a delayed axillary clearance. On a broader scale, the additional clinical benefits being provided to a local population should be considered. The current tariff structure of remuneration makes intraoperative testing non-profitable from an individual trust viewpoint, but it becomes cost neutral, or offers minor savings to the health economy, by reducing the additional costs associated with a second hospital admission .
Validation, training and learning curve
Our biomedical scientists had to undergo formal training to carry out the technique. This training was provided by Johnson & Johnson, the parent company of Veridex, as a 3-day course in Strasbourg. Newer members of the team can now be trained in house by those carrying out the procedure regularly. There was a short learning curve, with 98% concordance being reached after 50 cases, and we carried out 125 cases in our unit before we went live, performing immediate axillary clearance when the intraoperative testing was positive. This enabled us to fully validate our intraoperative test results with conventional histological analysis and look at any safety aspects. Since introduction of the procedure in the unit, the time taken from removing the sentinel node to having a definitive positive or negative result has been reduced from 60 minutes to 30 minutes. Positive results can be obtained even earlier if there is early detection of enough positive material.
The hardware costs for intraoperative molecular-based testing vary from approximately 30,000 [pounds sterling] to 60,000 [pounds sterling], depending on which method of testing is adopted. The equipment includes basic standard laboratory consumables, reagents, homogeniser, weighing scales, printer, freezer and a cell cycler. Although not essential, the equipment and laboratory will ideally be set up in close vicinity to the operating theatre. This facilitates cross-specialty learning from the theatre staff, surgeons and biomedical scientists and, in our unit, has definitely helped in implementing the new technique.
Patient consent is obtained preoperatively from all patients undergoing sentinel node biopsy so that material can be used and stored as required in the future. The sentinel nodes are identified using dual localisation with a blue dye and radioactive marker. The nodes are then harvested, and each node removed can start being processed without any need to wait for all sentinel nodes to be retrieved. The fat surrounding the node is removed by either the surgeon or biomedical scientist. The nodes are sectioned at 2-mm intervals, and alternate slices analysed using the BLNA. Remaining slices of node are sent for histology, including CK19 immunohistochemistry at multiple levels. For molecular analysis, the samples are weighed, homogenised and RNA eluted. The PCR mix is then prepared with the primers and probes and finally the RNA is added before being placed in the cell cycler. The product of the RT-PCR reaction is quantified by fluorescence and recorded in graph and numerical form on a computer screen. Positive and negative controls are used in all PCR runs to ensure quality control issues are met anc to assess for any contamination. Final results of each run for each node are classified as positive (micrometastases/ macrometastases), negative--or invalid, if the controls fail.
Patients who undergo immediate axillary clearance will have an axillary drain placed at the time of surgery. When the patient recovers from their anaesthetic, they are immediately aware that they have had more extensive surgery than they had been expecting, with the implication that their cancer is no longer confined to the breast. Although it has not yet been formally evaluated, this can potentially lead to increased anxiety--both from a prognostic viewpoint and because of the increased likelihood of additional adjuvant therapies in the form of chemotherapy or radiotherapy. Patients should therefore be given enough information prior to the surgery to understand the implications of intraoperative testing. We have designed and distributed written patient information leaflets to all patients undergoing intraoperative testing as an additional resource. Patients are also made fully aware of the possibility of discordance between the intraoperative test result and final histology, and the implications of this.
Discordant patient results
A prospective database has to be set up and maintained in order to enable the analysis of discordant results. Our previously published results have shown overall concordance of 95% between histological analysis, including micrometastases, and molecular analysis (sensitivity 96%, specificity 95%) [7). If there is discordance between the intraoperative test and conventional histology, this is likely to be due to sampling error with each evaluating alternate slices of the node. As the node is sectioned in 2-mm slices, it is likely that any discordant result is based on tissue less than 2 mm thick and therefore a positive result on molecular analysis with no histological evidence of metastases could be reasonably treated as micrometastatic disease. Where there is discordance, there is no way of knowing whether the test has failed (although the positive and negative controls confirm the internal validity of each test run) and has produced a false-negative or a false-positive result, or whether the finding is due to sampling error. We have so far resisted testing the entire node using the molecular-based assay but this may be something that will require further exploration in the future.
Predicting and stratifying which patients are more likely to have positive sentinel nodes, based on preoperative tumour grade, size and presence of lymphovascular invasion, can be helpful when planning a list. A list can potentially over-run if two or three patients undergoing sentinel node biopsy in a full day's operating session end up having an immediate axillary clearance--or under-run if all sentinel nodes are negative. Flexible working patterns of theatre staff can be useful if there are regular overruns. Sentinel nodes are harvested prior to any surgery on the breast which allows evaluation of the sentinel nodes in time for the surgeon to carry out appropriate surgery to the breast. We have found that this does not increase the theatre time if the sentinel node is negative, and that 40 minutes' additional theatre time is required per axillary clearance . Patients having sentinel node biopsy without axillary clearance are often fit to be discharged on the day of their surgery. Those having additional immediate axillary clearance are likely to remain in hospital for an extra day or two.
Issues around micrometastases
There is outcome controversy as to the best practice when dealing with micrometastases in the sentinel node and there are ongoing prospective studies evaluating their prognostic significance. However, studies have shown an incidence of 10-15% of finding further involvement of non-sentinel nodes when axillary clearance is subsequently carried out . The current American Society of Clinical Oncology (ASCO) guidelines state that all patients who have sentinel node micrometastases should undergo axillary clearance. The software initially provided with the Veridex kit was unable to distinguish whether a positive result was indicative of micrometastases or macro metastases. Modification to the software and interpretation of the analyses now allows this distinction.
In December 2009, Johnson & Johnson indicated that it would stop the manufacturing of the BLNA kits and would be withdrawing its product support in August 2010. The reason for this was not publicly stated but was most likely the result of poor uptake in the USA. This led to our unit looking at alternative methods for carrying out the intraoperative analysis. In collaboration with the pathology department in a separate NHS trust, a new assay, Metasin, was developed and validated. Metasin was based on the technique of qRT-PCR along similar lines to Veridex . This was validated using BLNA kits in parallel, as controls. The data we have analysed so far has indicated that the assay is as robust and safe as Veridex, and significantly cheaper.
Intraoperative analysis of sentinel nodes can be carried out by a number of different techniques. This allows patients to undergo a single operation to treat and stage the breast cancer. There is widespread variation in practice in the UK, with many units not actually carrying out any form of intraoperative testing, but there is a trend towards using molecular-based assays as the technique of choice, with increasing evidence of its greater sensitivity. It is likely to become the gold standard for analysis over conventional histological and cytological techniques. Automation and molecular whole-node analysis may reduce the intense workload of the histopathologist but this may have to be tempered with the loss of some prognostic information and also the potential loss of long-term archiving of paraffin blocks. However the choice of best technique of intraoperative nodal assessment for an individual unit is likely to be determined by cost, expertise and local availability of service.
The authors would like to thank: I Cree, Consultant Pathologist, M Jeffrey, Consultant Pathologist, A McDowell, Biomedical Scientist, and G Gabriel, Biomedical Scientist, of Queen Alexandra Hospital, Portsmouth, who have all been instrumental in the facilitation of intraoperative sentinel node testing in our unit.
[1.] National Institute for Health and Clinical Excellence. Guideline 80: Early and locally advanced breast cancer. Diagnosis and treatment. National Institute for Health and Clinical Excellence, 2009. Available at: www.nice.org.uk/nicemedia/pdf/ CG80NICEGuideline.pdf (last accessed November 2010).
[2.] Turner R, Ollila D, Krasne D and Giuliano A. Histopathologic validation of the sentinel lymph node hypothesis for breast carcinoma. Ann Surg. 1997, 226, 271-278.
[3.] Mansel R, Fallowfield, L, Kissin M eta/. Randomised multicentre trial of sentinel node biopsy versus standard axillary treatment in operable breast cancer: the ALMANAC Trial. I Natl Cancer Inst, 2006, 98, 599-609.
[4.] Tew K, Irwig L Matthews Aef al. Meta-analysis of sentinel node imprint cytology in breast cancer. Br J Surg, 2005, 92, 10681080
[5.] Aihara T, Munakata S, Morino H and Takatsuka Y. Comparison of frozen section and touch imprint cytology for evaluation of sentinel lymph node metastasis in breast cancer. Ann Surg Oncol, 2004, 11, 747- 750.
[6.] Layfield D, Agrawal A, Roche H and Cutress R. Intraoperative assessment of sentinel lymph nodes in breast cancer. Br J Surg, 2010. September [epub ahead of print], doi: 10.1002/bjs.7229.
[7.] Cutress R, McDowell A, Gabriel F ef al. Observational and cost analysis of the implementation of breast cancer sentinel node intra-operative molecular diagnosis. J Clin Pathol, 2010, 6, 522529
[8.] Cserni G, Gregori D, Merletti F et al. Meta-analysis of non-sentinel node metastases associated with micrometastatic sentinel nodes in breast cancer. Br J Surg, 2004, 91, 1245-1252.
[9.] Al-Ramadhani S, Balaraman P, Mascall A et al. Validation of metasin: a novel real-time PCR assay for metastatic carcinoma. Poster presentation (P10) at Pathological Society of Great Britain and Ireland, Winter meeting, 2010. Available at: www.pathsoc org/files/meetings/winter2010/05.01.106552ProgMAINv10(web). pdf (last accessed October 2010).
Avi Agrawal, Martin Wise and Constantinos Yiangou
Department of Breast Surgery, Portsmouth Hospitals NHS Trust, Hampshire, UK
Correspondence to: Avi Agrawal
Breast Unit, Queen Alexandra Hospital
Southwick Hill Road, Cosham, Portsmouth
Hampshire P06 3LY, UK
Figure 1: Advantages and disadvantages of various techniques of intraoperative sentinel lymph node assessment Technique Advantages Disadvantages Frozen-section Very established technique Expensive -76% sensitivity Equipment widely available Loss of tissue -99% specificity Labour-intensive Operator-dependent Histopathologist and technician required Imprint/scrape Low preparation time Cytologist required cytology -63% sensitivity Low cost Higher false-negative -99% specificity No loss of tissue results in micro metastatic and lobular carcinoma Molecular- Reduced sample error High initial capital based assay outlay -96% sensitivity Trained technician Requires specific only required training/learning curve -95% specificity Greater volume analysis Unable to assess of lymph node extranodal/ extracapsular spread No intra-observer No archival paraffin variation section material available
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Feature Article|
|Author:||Agrawal, Avi; Wise, Martin; Yiangou, Constantinos|
|Publication:||Advances in Breast Cancer|
|Article Type:||Disease/Disorder overview|
|Date:||Dec 1, 2010|
|Previous Article:||Preoperative staging of the axilla.|
|Next Article:||Sentinel lymph node biopsy in patients presenting with breast cancer: a review of special cases.|