The Sentinel Lymph Node in Non-Small-Cell Carcinoma of the Lung
Michael J. Liptay
The concept of a sentinel node associated with primary solid tumors was initially put forth by Cabanas in reference to penile carcinoma and refined to its current use by Morton and colleagues beginning with melanoma. The description of the first nodal target of lymph from a primary tumor site—and therefore the node most likely to receive metastatic cells—defined the sentinel node by its function rather than a fixed anatomic location.
Sentinel node identification has become the standard of care in melanoma and breast cancer surgical care. The idea of a first nodal station draining a tumor theoretically would allow the assessment of that node(s) to represent the state of the remaining regional nodes. The sentinel node is used to limit potentially morbid lymph node dissections. Another benefit of the technique is directing applications of more focused pathologic or molecular staging techniques (serial sections, immunohistochemistry, real-time polymerase chain reaction, etc.). This ability to direct a more focused search for metastatic disease in the sentinel node rather than all of the nodes removed is a primary benefit of the technique in lung cancer.
With current indications for adjuvant chemotherapy in resected lung cancer largely based on the status of the locoregional lymph nodes, the accurate identification of positive nodes has gained therapeutic importance.
Sentinel Node Theory and Application
First site of nodal drainage—representative of other nodes’ status
Limit extent of node dissection
Focus pathologic or molecular staging techniques
Lymphatic Anatomy of the Lung and Pleura
The lymph drainage of the lung consists of intrapulmonary and extrapulmonary lymphatic systems. This complex network of lymphatic vessels generally flows out of the lobes to the hilar and mediastinal nodes into the thoracic duct. Anatomic studies have demonstrated that the subpleural lymphatic plexus can drain directly into the mediastinal nodes without involving bronchopulmonary nodes with a frequency up to 25%.15 This phenomenon has been described clinically in lung cancer as “skip metastases.” This pattern of nodal spread has been attributed to the subpleural plexus, which is most frequently involved with visceral tumor involvement (T2).
Sentinel node mapping would most logically benefit those with clinical early stage tumors. Tumor size, differentiation, and cell type have been shown to correlate with lymph node metastases. Squamous cell tumors <2 cm in size had a 6% nodal metastasis rate, while adenocarcinomas under <2 cm have up to a 19% rate. Tumors >3 cm have up to a 32% nodal metastasis rate.3
With our inability to accurately predict nodal drainage patterns of specific tumors, the rationale for sentinel node mapping in lung cancer surgery became apparent.
Intraoperative Tracer Injection
The application of the sentinel node technique to lung cancer began with Little et al. in 1999.10 The use of Isosulfan Blue dye resulted in a slightly less than 50% identification rate. The primary drawback of the blue dye was the frequent black anthracosis encountered in thoracic nodes, making dye detection difficult.
Our group first reported the use of intraoperative radioactive tracer in 2000.8,9 The technique used technetium-99 (99mTc) sulfur colloid filtered thru a 20-μm filter. Prior to the multicenter phase II CALGB trial, intraoperative sentinel lymph node mapping was performed on 165 consecutive patients presenting as candidates for anatomic resection of a suspected primary lung cancer.7 Of these, 148 consecutive patients had completely resected non-small-cell lung cancer (NSCLC) and were included in this study group.
Successful migration of the radioisotope through lymphatics was seen in 120 of 148 patients (81%). A sentinel node was identified in 104 of 120 patients (87%) with successful migration of
radioisotope, or 70%, in 104 of all 148 attempted mapping procedures. Our initial experience included all patients undergoing resection for suspected lung cancers, regardless of the presence of hilar or mediastinal adenopathy or large necrotic tumors. In 28 of the 148 patients (19%), we failed to demonstrate migration of the radioisotope through the lymphatics. Hilar and/or mediastinal adenopathy was present in eight patients, whereas nine patients had tumors >5 cm. Two patients underwent preoperative chemoradiation, and no explanation was found for the technical failure in 11 patients.
radioisotope, or 70%, in 104 of all 148 attempted mapping procedures. Our initial experience included all patients undergoing resection for suspected lung cancers, regardless of the presence of hilar or mediastinal adenopathy or large necrotic tumors. In 28 of the 148 patients (19%), we failed to demonstrate migration of the radioisotope through the lymphatics. Hilar and/or mediastinal adenopathy was present in eight patients, whereas nine patients had tumors >5 cm. Two patients underwent preoperative chemoradiation, and no explanation was found for the technical failure in 11 patients.
In our series the sentinel node was positive for metastatic disease in 33 of 104 patients (32%); the sentinel node was the only metastatic node in 12 of 33 patients (36%). We detected micrometastatic disease in the sentinel node with immunohistochemistry or serial sectioning in 8 of 33 patients (24%). Thus, in our first experience with the sentinel node procedure, lung cancers were upstaged in 8 of 148 cases (5.5%).
Mediastinal lymph node involvement without concurrent spread to the intraparenchymal and hilar nodal basins has been termed “skip metastases.” The incidence of this phenomenon in patients with positive N2 mediastinal nodes is between 20% and 30% in most series.13 In our study, 25 of 104 sentinel nodes (24%) were mediastinal.
Since our initial reports, other groups had reported sentinel node identification rates of 74% with 99mTc alone16 and 81% using a combination of 99mTc and blue dye.17
Multicenter Phase II Trial: Cancer and Leukemia Group B 140203
The Cancer and Leukemia Group B designed a phase II trial to test a standardized method of this technique in a multi-institutional setting. CALGB 140203 opened for accrual in September 2004. The technique was based on our intraoperative mapping with 99mTc sulfur colloid. The inclusion criterion was suspected clinical stage I non-small-cell lung cancer.
Intraoperatively, tumors were injected with 99mTc sulfur colloid (0.25 mCi). The tumor and lymph nodes were measured in vivo with a handheld Geiger counter and resection of the tumor and nodes was carried out. Sentinel nodes, all other nodes, and the tumor were analyzed with standard histologic assessment. Negative sentinel nodes were also evaluated with immunohistochemistry.
In this phase II trial, eight surgeons participated (1–13 patients enrolled per surgeon), and 46 out of the planned 150 patients were enrolled before the trial was terminated early based on disappointing accuracy, slower than expected accrual, and reduced funding from the National Cancer Institute. Of the 46 patients on study, 43 had cancer and an attempted complete resection and 39 patients underwent sentinel node mapping. One or more sentinel nodes were identified in 24 of the 39 patients (61.5%). The sentinel node(s) were found to be accurate (no other nodes were positive for cancer if the sentinel node was negative) in 20 of 24 patients (83.3%). In the overall group, the sentinel node mapping procedure was found to be accurate in 20 of 39 patients (51.2%).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
