Lymphatic Mapping in the Treatment of Breast Cancer

Cox and colleagues from the H. Lee Moffitt Cancer Center summarize their experience with lymphatic mapping and sentinel lymph node biopsy for breast cancer and discuss this experience in the context of the existing literature on the subject. Since 1994, the authors have treated over 466 patients with early-stage breast cancer using this new technology. In summarizing their results, they conclude that lymphatic mapping shows "great promise as the next major advance in the treatment of breast cancer." However, they emphasize that the technique is still investigational.

We agree that this technology has the potential to replace standard lymph node dissection for breast cancer, in the same way that it has offered a rational alternative to elective lymph node dissection in the treatment of patients with melanoma. However, prior to adopting sentinel lymph node biopsy as the standard of care, several issues need to be addressed. These include: (1) variations in methodology, which appear to impact the rate of identification of sentinel lymph nodes; (2) the definition of a sentinel lymph node; (3) the appropriate use of breast lymphoscintigraphy; (4) the extent of pathologic evaluation of the sentinel lymph node needed; and (5) how to incorporate sentinel lymph node biopsy into the management of patients with carcinoma of the breast in the 1990s and beyond.

Variations in Methodology

Lymphatic mapping was first described by Morton and colleagues in 1992; they used an intradermal injection of isosulfan blue dye to evaluate the nodal basin for blue-stained nodes in melanoma patients.[1] Subsequently, Giuliano and associates described a similar technique in breast cancer patients. Using peritumoral injections of blue dye alone, they achieved an identification rate of about 65%.[2] In a subsequent report, again using blue dye alone, their identification rate improved to 93.5%.[3]

Krag et al were one of the first groups to describe the use of technetium-labeled sulfur colloid in breast lymphatic mapping; they reported an identification rate of 81.8% using radioisotope alone.[4] Albertini and colleagues from the Moffitt Cancer Center achieved a 91.9% identification rate using a combination of blue dye and technetium-labeled sulfur colloid.[5] Cox et al from the same institution cite an identification rate of 94.4% in 466 patients. Like the experience of Morton et al with melanoma patients and Giuliano et al with breast cancer patients, the current report demonstrates that the identification rate of a sentinel lymph node significantly improves as the experience (number of patients mapped) of the operator increases.[1,2]

Whether blue dye alone is adequate or whether a combination of blue dye and radioisotope is needed has not been clearly demonstrated. Giuliano et al reported a 93.5% identification rate using blue dye alone, and yet the current Moffitt series states that only 59.8% of the sentinel lymph nodes were blue. This discrepancy may be due to one or more of the following technical issues: (1) the time elapsed between blue dye injection and axillary exploration; (2) the volume and site of blue dye injection; (3) whether the breast tissue is excised before or after axillary exploration for the sentinel lymph node; and (4) whether the blue dye is injected into the breast before or after the segmental mastectomy is performed. A lower sentinel node identification rate using blue dye alone may be improved by the addition of a radiolabeled colloid, as it did in the current Moffitt report.

Definition of the Sentinel Node

The definition of a sentinel node has to be clarified, depending on the technique used. In their article, Cox et al state that a sentinel node is "blue," "hot," or "blue and hot." The definition of a "blue node" seems clear, but the definition of a "hot node" is not.

The investigators presume that a node is considered a sentinel node if the in vivo radioactive count is at least three times that of the background count, or if the ex vivo radioactive count of the sentinel node is 10 times greater than the count in a neighboring nonsentinel node. Since one cannot assess the ex vivo count ratio until these nodes are removed, the initial assessment of the sentinel node is based largely on in vivo count ratios. As the authors clearly state, very little of the radiocolloid (1% to 5%) actually travels to the sentinel node within 6 hours. This makes the background counts in the breast and axilla extremely high, causing "shine-through."

Ideally, if both radiocolloid and blue dye are employed, the operator can use a handheld gamma probe to transcutaneously localize the node, and co-localization of the blue dye in this same sentinel node would offer a visual clue that the correct or "sentinel" node was being excised. This would limit the amount of dissection required, making it more feasible to complete the procedure under local anesthesia. Importantly, the operator must clearly understand the definition of a sentinel lymph node in the context of different localization techniques.

Even though different centers employ varying methodologies for identifying sentinel lymph nodes, the end results may be the same. A curious, relatively small study from the Netherlands used blue dye injected intradermally in the skin above the tumor mass and sulfur colloid injected peritumorally into the breast parenchyma around the index lesion.[6] Interestingly, there was co-localization of the blue dye and the sulfur colloid into the same lymph node.

The important issue is not necessarily the exact method used to identify the sentinel lymph node, as long as it is reliable and reproducible. Rather, the key is whether we have quality control that mandates high identification rates and low false-negative rates.

What Is an Acceptable False-Negative Rate?

Once the definition of a sentinel lymph node has been established, and the identification rate of the sentinel node is sufficiently high (> 90%), it should follow that the false-negative rate will be low. The definition of an acceptably low false-negative rate has not established, however, and may vary with the experience of the operator and the technique employed.

The false-negative rate quoted by Cox and coauthors of 0.0017%, obtained by combining the experience of Albertini et al, Giuliano et al, and Cox et al, is misleading and probably incorrect. Not all of the 573 patients described underwent a complete lymph node dissection, and not all of them had positive lymph nodes. The false-negative rate should be calculated by using the number of patients who had positive axillary nodes without a positive sentinel node as the numerator and by using all of the patients who had at least one positive axillary node as the denominator. Based on this assessment, a false-negative rate ranging from 0.2% to 5% is much more realistic.

Within this context, we must decide what an acceptable false-negative rate should be. A rate of 5% may be perfectly acceptable in patients who will receive systemic chemotherapy regardless of the results of the sentinel node biopsy. However, this rate may be unacceptable in patients with T1a or T1b tumors, in whom the decision of whether or not to receive chemotherapy is based on the sentinel node biopsy results. Although it is assumed that a false-negative event is due to the identification of the wrong node, it may, in fact, result from inadequate histologic evaluation.

Breast Lymphoscintigraphy and Pathologic Evaluation

Two additional factors that need to be addressed in the multidisciplinary effort of breast lymphatic mapping are breast lymphoscintigraphy and pathologic evaluation of the sentinel node. The use of lymphoscintigraphy in melanoma has helped discern the appropriate nodal basin in areas of ambiguous nodal drainage. A similar procedure would be extremely useful in breast cancer patients, especially those with central or inner quadrant tumors. The technique of breast lymphoscintigraphy has not been clearly described in the literature, however, and few centers can accurately perform this procedure in order to specifically delineate the location of the axillary sentinel lymph node and to exclude patients from axillary surgery who have drainage only to the internal mammary nodes.

The extent of pathologic evaluation of the sentinel node also remains to be defined. Giuliano and colleagues demonstrated that more patients were proven to harbor micrometastases when the sentinel node was examined with serial sectioning and immunohistochemistry for cytokeratin.[7] Whether both serial sectioning and immunohistochemistry are needed is unclear.

Certainly, performing more studies on each node will raise the overall costs, which will negate some of the potential cost-benefits of performing a more selective surgical procedure. However, if patients are staged more accurately (truly node-negative or truly node-positive) by providing the pathologist with 1 or 2 sentinel nodes for study rather than 15 to 20 axillary nodes, we may be able to more carefully direct our adjuvant therapies. With more accurate staging, we may be able to spare truly node-negative patients the costs and morbidities associated with cytotoxic chemotherapy.

Incorporating Lymphatic Mapping Into Current Practice

The real challenge lies in our ability to place lymphatic mapping and sentinel lymph node biopsy in practice in the context of contemporary breast cancer management issues. Systemic therapy is now recommended not only for patients with positive lymph nodes but also for those who are node-negative with tumors > 1 cm in size. Indeed, chemotherapy has moved into the forefront of treatment and is now being used to "downstage" tumors preoperatively, even in patients with relatively early-stage disease (T2 N0).

Our current practice at the M. D. Anderson Cancer Center is to perform lymphatic mapping and sentinel lymph node biopsy in all patients who are clinically node-negative, including those treated with chemotherapy preoperatively. Breast lymphoscintigraphy is employed for all patients, regardless of tumor location.

On the day of surgery, technetium-labeled sulfur colloid is injected into and around the primary tumor, directly in the case of palpable lesions or under ultrasound or mammographic guidance in the case of nonpalpable lesions. After 2 to 6 hours, patients are taken to the operating room, where they receive an intraoperative peritumoral injection of 5 mL of iso-sulfan blue dye. A hand-held gamma probe is used to transcutaneously localize the sentinel node, and a mark is placed over the skin. A small skin incision is then made in the axilla, and initial visual inspection is performed in order to identify a blue-stained node. In most cases, this corresponds to the node that is also hot, as determined with the gamma probe.

The sentinel node or nodes are excised, and the axilla is checked again for a drop in radioactivity. If the radioactivity has not decreased, further inspection with the gamma probe is undertaken to identify additional sentinel nodes. The primary tumor is then excised to reduce background activity, and the axilla is checked once again for a decrease in radioactivity.

The sentinel node is sectioned, and a portion of the node is frozen for polymerase chain reaction (PCR) assays, while the remainder is prepared for serial sectioning and immunohistochemistry. Current protocols are underway to assess newer PCR techniques and additional markers that can be examined using immunohistochemistry.

Need for Randomized Trials

With all of this in mind, it is clear that the safety and efficacy of sentinel lymph node biopsy as a replacement for axillary dissection in breast cancer patients can be assessed only in the context of prospective, randomized trials. Two groups that are currently designing studies of this nature are the American College of Surgeons and the National Surgical Adjuvant Breast and Bowel Project (NSABP). We eagerly await the results of these trials and, in the interim, continue our attempts to better define the methodology for all of the disciplines involved in this effort. Our hope is to minimize morbidity and unnecessary surgery without compromising the primary goals of accurate staging, locoregional control, and potential cure of breast cancer.

References:

1. Morton DL, Wen DR, Wong JH, et al: Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 127:392-399, 1992.

2. Giuliano AE, Kirgan DM, Guenther JM, et al: Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 3:391-401, 1994.

3. Giuliano AE, Jones RC, Brennan M, et al: Sentinel lymphadenectomy in breast cancer. J Clin Oncol 15:2345-2350, 1997.

4. Krag DN, Weaver DL, Alex JC, et al: Surgical resection and radiolocalization of the sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 2:335-339, 1993.

5. Albertini JJ, Lyman GH, Cox C, et al: Lymphatic mapping and sentinel node biopsy in the patient with breast cancer. JAMA 276:1818-1822, 1996.

6. Borgstein PJ, Meijer P, Pijpers R: Intradermal blue dye to identify sentinel lymph node in breast cancer. Lancet 349:1668-1669, 1997.

7. Giuliano AE, Dale PS, Turner RR, et al: Improved axillary staging of breast cancer with sentinel lymphadenectomy. Ann Surg 222:394-401, 1995.