- Email: [email protected]
Sentinel node biopsy for staging lung cancer
Surg Clin N Am 82 (2002) 561–571
Sentinel node biopsy for staging lung cancer Alex G. Little, MD Department of Surgery, University of Nevada, School of Medicine, 2040 West Charleston Boulevard, Suite 601, Las Vegas, NV 89102, USA
Lung cancer is the leading cause of cancer related death in both men and women; it has surpassed breast cancer as the leading cause of cancer death in women. In 1995, 180,000 new cases were diagnosed in the United States and the incidence continues to increase [1]. While prevention is the Holy Grail, maximizing the therapeutic strategy for each patient is essential to improving outcomes. This requires the best possible information about the extent of disease as defined by the staging process. The prognosis of the patient with non-small cell lung cancer (NSCLC) is closely related to the stage of the disease as defined by the TNM classification system developed and endorsed by the American Joint Committee on Cancer (AJCC) [2]. The size, location, and invasiveness of the primary tumor (T); the extent and pattern of peribronchial, hilar, and mediastinal lymph node involvement (N); and the presence of metastasis (M) are the major determinants of staging. The American Thoracic Society/American Joint Committee on Cancer (ATS/AJCC) TNM descriptors for the current staging system of lung cancer are listed below (see box). Table 1 shows the TNM subsets that constitute each stage group. Accurate clinical and pathological staging are essential to define prognosis subgroups and to develop therapeutic strategies based on the presence or absence of metastatic disease and the extent of locoregional disease. The presence or absence of lymph node metastasis is the single most important predictor of recurrence of disease and survival in patients undergoing lung resection: the five-year survival rate is approximately 40% lower in patients who have mediastinal lymph node metastasis than in those who do not [3]. The staging process has three aspects: clinical, surgical, and pathological staging. The clinical stage is based on information gleaned from the full spectrum of non-invasive tests such as chest radiograph (CXR), chest E-mail address: [email protected] (A.G. Little). 0039-6109/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 0 3 9 - 6 1 0 9 ( 0 2 ) 0 0 0 1 6 - 6
562
A.G. Little / Surg Clin N Am 82 (2002) 561–571
Primary tumor (T) TX Primary tumor cannot be assessed, or tumor proven by the presence of malignant cells in sputum or bronchial washings but not visualized by imaging or bronchoscopy TO No evidence of primary tumor Tis Carcinoma in situ T1 Tumor £3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus (ie, not in the main bronchus) T2 Tumor with any of the following features of size or extent: >3 cm in greatest dimension Involves main bronchus, £2 cm distal to the carina Invades the visceral pleura Associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung. T3 Tumor of any size that directly invades any of the following: chest wall (including superior sulcus tumors), diaphragm, mediastinal pleura, parietal pericardium; or tumor in the main bronchus <2 cm distal to the carina, but without involvement of the carina; or associated atelectasis or obstruction of the entire lung T4 Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea, esophagus, vertebral body, carina; or tumor with a malignant pleural or pericardial effusion, or with satellite tumor nodules within the ipsilateral primary-tumor lobe of the lung. Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis to ipsilateral peribronchial or ipsilateral hilar lymph nodes, and intrapulmonary nodes involved by direct extension of the primary tumor N2 Metastasis to ipsilateral mediastinal or subcarinal lymph nodes N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral, or contralateral scalene, or supraclavicular lymph nodes Distant metastasis (M) MX Presence of distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis present
A.G. Little / Surg Clin N Am 82 (2002) 561–571
563
Table 1 Stage grouping - TNM subsets Stage
TNM subset
0
Carcinoma in situ
IA
T1N0M0
IB
T2N0M0
IIA
T1N1M0
IIB
T2N1M0 T3N0M0
IIIA
T3N1M0 T1N2M0 T2N2M0 T3N2M0
IIIB
T4N0M0 T4N1M0 T4N2M0 T1N3M0 T2N3M0 T3N3M0 T4N3M0
IV
Any T Any N M1
computerized axial tomography (CAT) scan, magnetic resonance imaging (MRI) scan, and positron emission tomography (PET) scan. Invasive procedures, such as mediastinoscopy, performed before definitive surgical intervention are also part of clinical staging. Surgical staging occurs during an operation and incorporates both gross observations by the surgeon, such as chest wall invasion by a primary tumor, and the results of frozen section analysis of biopsied or removed tissue, such as lymph nodes and bronchial margins. Finally, the pathologic stage is determined from histologic inspection of all tissue removed during the definitive operative procedure. Accurate clinical staging of mediastinal, hilar, and intrapulmonary lymph nodes is an important aspect of the process of determining the therapeutic strategy for an individual lung cancer patient, where the possibilities range from surgery alone to chemoradiation therapy or an integrated multimodality approach. Fig. 1 depicts the regional lymph node stations generally used for lung cancer staging [4]. Accurate surgical staging allows determination of the most appropriate extent of resection, wedge resection, lobectomy, or pneumectomy; the need for en bloc resection of adjacent chest wall; and the type and extent of mediastinal lymph node sampling or dissection. Pathologic staging establishes the definitive or ultimate stage and accuracy is paramount. Identification of residual disease by the finding of tumor at a bronchial margin is of course important, but the more common issue is accurate determination of whether or not mediastinal lymph nodes contain
564
A.G. Little / Surg Clin N Am 82 (2002) 561–571
Fig. 1. The joint ATS/AJCC system for regional lymph node classification for lung cancer staging. (From Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1719; with permission.)
metastatic disease. Because survival is significantly affected by the presence of nodal disease, this latter determination impacts both patient prognosis and decisions regarding the use of postoperative adjuvant therapy. Despite the importance of nodal staging as a predictor of prognosis and determinant of therapy in patients being operated on for lung cancer, consensus is lacking in the thoracic surgical community with respect to the preferred approach to the retrieval of lymph nodes for postoperative, pathological staging in patients undergoing lung resection. The options are (1) to
A.G. Little / Surg Clin N Am 82 (2002) 561–571
565
simply remove the nodes (hilar, segmented, and parenchymal) that are attached to and part of the lung specimen, (2) to biopsy selected mediastinal node groups, or (3) to dissect and remove the nodal groups en masse. When mediastinal lymph nodes are not routinely dissected or sampled, understaging occurs. Up to 20% of lung cancer patients with clinical and surgical Stage I or II disease have been demonstrated to have occult involvement of mediastinal lymph nodes (N2), identification of which results in upstaging to a Stage III classification [5]. Even when lymph nodes are removed and examined histologically, micrometastatic disease can be missed, further complicating the situation. Immunohistochemical staining of removed lymph nodes for a variety of tumor associated antigens is able to demonstrate the pressure of so-called micrometastatic disease in histologically tumor negative lymph nodes [6]. Because of the logistics of the process of obtaining this information, which involves sectioning, staining, and examining multiple lymph nodes, a way to simplify the process without sacrificing sensitivity would be helpful. Operative lymphatic mapping and selective lymphadenectomy History The technique of intraoperative lymphatic mapping and selective lymphadenectomy to identify the sentinel lymph node (SLN) was developed as a method of determining the regional lymph node status in patients with breast cancer or melanoma [7,8]. It was observed that pathological examination of the sentinel lymph node increased the accuracy of nodal staging [7,8]. This increased accuracy results from the observation that detailed pathologic examination of the sentinel lymph node, with multiple sections being inspected, results in increased detection of micrometastasis in comparison with routine examination of multiple lymph nodes. This technique, as applied to lung cancer, relies on the hypothesis that for a given segment or lobe of the lung, there is a dominant lymph node in the basin to which the lymphatic flow drains first; ie, the sentinel lymph node. Current status of mediastinal staging in lung cancer Intrapulmonary, peribronchial, and hilar lymph nodes are routinely removed in the process of performing a standard lobectomy or pneumectomy. Dissection of mediastinal lymph nodes, which requires an additional effort, is the challenge and is not routinely performed. Despite the fact that most or all patients have preoperative mediastinal staging with CAT, MRI, or PET scans, and some undergo mediastinoscopy, up to 20% of patients have occult disease found by histologic examination if mediastinal lymph nodes are removed by a formal mediastinal dissection [5]. Therefore, some form of routine surgical staging is necessary to ensure maximal staging accuracy, which in turn can be used to predict the patient’s prognosis and affect therapeutic
566
A.G. Little / Surg Clin N Am 82 (2002) 561–571
strategies. Either intraoperative mediastinal node dissection or node sampling (cherry picking) has been suggested as a means to provide more accurate staging than examination of lymph nodes attached to a lobe or even an entire lung removed in standard fashions. However, neither approach is universally accepted, perhaps because of the increase in operative time and potential morbidity associated with these activities. This has created interest in applying the lymphatic mapping technique, developed for and in patients with breast cancer or melanoma, to patients being operated on for lung cancer. University of Nevada sentinel lymph node experience Our group’s experience with the sentinel node technique was gained using the following technique [9]. A standard posterior lateral thoracotomy with or without a muscle-sparing technique is performed. In patients with tumors involving the chest wall, the incision is extended as necessary and a three-rib segment of chest wall is resected in continuity with the lung. Upon entering the chest, the lung is collapsed as contralateral ventilation is continued through a double lumen endotracheal tube. Visual and tactile inspection of the lung and mediastinum is performed to identify areas of abnormality. A total of 5 ml of 1% isosulfan blue dye (lymphazurin) is injected into the lung parenchyma in four quadrants around the primary tumor. Use of a relatively small needle, such as a 21 gauge, reduces dye spillage from the needle holes and staining of the surrounding tissues, which can increase the difficulty of identifying a sentinel node. The pleura is then circumferentially incised around the hilum, and routine lung resection is carried out. Ten minutes is allowed for lymphatic spread of the vital blue dye. Attention is then directed initially to the adjacent mediastinal node basins. The first lymph node to stain blue is considered the SLN and is removed and sent separately for permanent histological examination. A stained drainage lymphatic vessel feeding the SLN is also sought but, in our experience, rarely found. If needed, small incisions are created in the mediastinal pleural to allow identification of the blue-stained afferent lymphatic vessels leading to the SLN. Once the SLN is identified, and distinguished from adjacent anthracotic nodes, the entire node is removed by means of sharp dissection or electrocautery. Afferent and efferent lymphatic vessels proceeding from the SLN, some of which may be stained blue, are controlled with hemostatic clips for anatomic reference. Ultimately, lobectomy, bilobectomy, or pneumonectomy is performed. In our initial experience, at the completion of the lung resection, mediastinal node dissection was performed. During a right thoracotomy, levels 4, 7, 8, 9 and 10 (according to the unified lymph node mapping developed jointly by the ATS and AJCC [4]) were removed. Levels 4, 5, 6, 7, 8, and 9 were resected during a left thoracotomy. In our initial study, 36 patients with non-small cell lung cancer were analyzed [8]. Squamous cell carcinoma was present in 13 patients, adenocarcinoma in 19 patients, and other histologic types in 4 patients. All patients
A.G. Little / Surg Clin N Am 82 (2002) 561–571
567
were evaluated before the operation by computed tomographic scanning of the chest and the upper abdomen to evaluate mediastinal lymph nodes and to exclude liver and adrenal gland involvement. Seven of the 36 patients underwent cervical mediastinoscopy for pathological examination of paratracheal or subcarinal lymph nodes greater than 1 cm in short axis on the CT, and all analyzed lymph node biopsies from these 7 patients were negative for the presence of metastatic involvement. Twenty-nine patients were clinical Stage I, and 7 patients were clinical Stage II. Results Pathologic staging 1. Twenty-one patients (58.3%) had Stage I: T1N0M0 in 11 (30.6%) and T2N0M0 in 10 (27.8%). 2. Seven patients (19.4%) had Stage II: T1 or T2N1M0 in 6 (16.7%) and T3N0M0 in 1 (2.8%). 3. Eight patients (22.2%) had Stage IIIA: T1N2M0 in 4 (11.1%), T2N2M0 in 2 (5.5%) and T3N1M0 in 2 (5.5%). Sentinel node results The University of Nevada results from the initial experience with the sentinel lymph node technique in 36 consecutive patients with NSCLC were as follows (see box): Sentinel Lymph Node Identified 1. Total: 17 of 36 patients (47%) 2. Tumor Positive: 8 of 17 patients (47%) Path N1, 3 patients Path N2, 5 patients 3. Tumor Negative: 9 of 17 patients (53%) Path N0, 9 patients No Sentinel Lymph Node Identified 1. 2. 3. 4.
Total: 19 of 36 patients (53%) Path N0, 13 patients Path N1, 5 patients Path N2, 1 patient
Sentinel node positive A sentinel lymph node was found in 17 of the 36 patients (47%). In 9 of these 17 patients neither the sentinel node nor any other lymph node
568
A.G. Little / Surg Clin N Am 82 (2002) 561–571
contained metastatic disease, so nodal staging was N0. The sentinel nodes in the other 8 patients contained metastatic NSCLC: in 3 patients the sentinel node was either hilar or intrapulmonary and contained metastatic tumor (N1), but all mediastinal nodes were negative for metastatic involvement. In the remaining 5 patients the sentinel lymph node was in the mediastinum; paratracheal in 3 (station 4) and subcarinal in 2 (station 7). Pathologic examination of the SLN documented metastatic involvement (N2), despite both a negative preoperative CT scan and mediastinoscopy. In these 5 patients, the sentinel lymph node was the only site of metastasis; no other mediastinal lymph node stations contained metastatic tumor. Sentinel node negative In 19 patients (53%), no sentinel lymph node was identified, despite careful intra-operative mediastinal lymph node dissection. Among these patients, pathologic staging was N0 in 13, N1 in 5, and N2 in 1. Radioguided technique The only other reported experience with sentinel lymph node mapping in lung cancer patients is from Liptay and associates [10]. They also used a technique at the time of thoracotomy. They injected a total dosage of two mCi of a technetium–99-labeled colloid suspension at the outer margins of the tumor in four quadrants. Subsequently they traced and mapped the colloid migration with a handheld gamma probe counter. The migration was considered successful if a nodal station registered counts per second greater than three times background values. The migration time averaged 63 minutes in patients in whom sentinel nodes were identified. Out of 45 patients with lung cancer, 37 (82%) were said to have had a sentinel node identified. This is a considerably higher sensitivity rate of sentinel node identification than we were able to obtain at the University of Nevada. The actual identification, however, as the authors point out, was of ‘‘nodal stations’’ that consisted of several lymph nodes rather than a single sentinel node. While this result could be used to focus pathologic examination, it is still necessary for the pathologist to examine several lymph nodes to search for histological or micrometastatic disease.
Discussion The use of isosulfan blue dye for lymphatic mapping has been extensively studied in patients with melanoma and breast cancer and has become a standard staging procedure in breast cancer patients undergoing mastectomy [6,7]. In these groups of patients, examination of the sentinel node yields accurate pathologic staging information and makes possible rational and selectively aggressive surgical and medical approaches.
A.G. Little / Surg Clin N Am 82 (2002) 561–571
569
The initial premise of the sentinel node technique was that awareness of the presence or absence of metastatic tumor in the node would be of benefit by predicting the status of all other regional lymph nodes. If this were true, a negative intraoperative frozen section analysis of the identified node would obviate the need for further lymph node dissection and histological examination. A positive result would lead to a potentially therapeutic regional node dissection. This line of reasoning has held up to some extent in the surgical treatment of patients with breast carcinoma or melanoma. We were encouraged in our experience that the specificity of the sentinel node technique was excellent. All 9 patients with tumor negative sentinel lymph nodes were found to be free of disease on final pathologic examination of mediastinal lymph nodes. In addition, of the 19 patients without a sentinel lymph node, 18 were either N0 or N1. This potential intraoperative role for lymphatic mapping, however, seems unlikely to pertain to lung cancer patients for two reasons. First, the blue dye technique has a relatively low sensitivity for detection of a sentinel node. We were able to identify a sentinel node only in approximately half our patients. Although we hoped our continued experience would result in an improved sensitivity (the learning curve phenomenon), as occurred in the breast cancer and melanoma experiences, that has not happened—our yield remains at the 50% level. We feel this is due to several reasons. The lymphatic network draining the lung is extensive and provides multiple pathways for tumor dissemination. This frustrating anatomic reality has been demonstrated in studies that examine the routes of lung cancer drainage. One illustrative example is provided in Table 2, adapted from Asamura and his colleagues [11], which shows the diffuse pattern of mediastinal lymph node involvement in 88 patients with peripheral non-small cell carcinomas. In addition, blue staining is difficult to identify in anthracotic nodes surrounded by other blue/black nodes and collapsed lung. Second, the radiation-guided approach appears to detect a lymph node group rather than a single node, increasing the challenge of time-efficient frozen section analysis during surgery. Also, both the isosulfan blue dye and the radio guided techniques add at least a modest amount of time and tediousness to the lung resection procedure, decreasing their appeal to the thoracic surgeon. Table 2 Lymph node involvement in the mediastinum according to tumor location Mediastinal lymph node location Stations
1
2
3
4
5
6
RUL (122) 5 (4.1) 3 (2.5) 17 (13.9) 4 (3.3) RML (31) 2 (6.5) 1 (3.2) 5 (16.1) RLL (64) 3 (4.7) 3 (4.7) LUL (83) 3 (3.6) 2 (2.4) 6 (7.2) 5 (6.0) 6 (7.2) LLL (37) 1 (2.7) 1 (2.7) 1 (2.7)
7 2 5 8 4 3
8
9
N2/3
(1.6) 18 (14.8) (16.1) 7 (22.6) (12.5) 2 (3.1) 11 (17.2) (4.8) 15 (18.1) (8.1) 1 (2.7) 1 (2.7) 5 (13.5)
Adapted from Asamura H, Nakayama H, Kondo H, J Thorac Cardiovasc Surg 1996; 111; with permission.
570
A.G. Little / Surg Clin N Am 82 (2002) 561–571
Investigators have demonstrated that an upward staging occurs when this selective lymphatic mapping technique is employed in breast cancer patients or melanoma patients [6,7]. This is the most likely way that the sentinel node technique will play a useful role in the staging of lung cancer patients. It appears that allowing the pathologist to focus on a single lymph node results in finding micrometastases that otherwise would be missed by standard processing of multiple lymph nodes. With only a single node to examine, the use of specialized techniques for examination of the sentinel node, including immunohistochemical staining and polymerase chain reaction, may even further increase the ability to detect only a few metastatic cells. Early in our experience we identified the accumulation of blue dye in the central portion of lymph nodes in some of our patients. This was not detectable by surface inspection, and was only found when the nodes were bisected. This suggested initially that radionuclide-guided techniques might improve the sensitivity of this technique in lung cancer patients. However the experience of Liptay and associates is only moderately encouraging, because the sensitivity of approximately 80% that they found was still lower than hoped for, and dealing with a packet of identified nodes is not a major benefit. The approach we are now using is to dissect and remove all the ipsilateral mediastinal lymph node groups in toto. After the thoracotomy is closed, the node groups are separated into individual nodes, which are then bisected. This approach in our most recent 23 patients has resulted in the finding of one or at most two sentinel nodes in 22 of the patients. A multiple section analysis of the sentinel node, both histological and with immunohistological staining, is then compared to the results of routine histological examination of all other nodes. The completion of this ongoing study should help determine if this application of the principles of the technique of the sentinel node in non-small cell cancer patients is helpful in the process of pathological staging. The important understanding is that acquisition of sufficient tissue to determine the status of N1 and N2 lymph node basins is an essential responsibility of the thoracic surgeon performing resections for carcinoma of the lung. Although a therapeutic benefit may result from a complete mediastinal lymphadenectomy, the main benefit relates to more accurate staging, because increasingly effective multimodality therapies are available for patients with N2 disease. With the knowledge that N1 status does not reliably predict N2 node status, either routine mediastinal lymphadenectomy, lymph node sampling, or the sentinel lymph node technique should be routinely employed to accurately stage the disease and select the most appropriate therapy for each individual patient. References [1] Parker S, Tong T, Bolden S, et al. Cancer statistics. Cancer 1997;45:5. [2] Mountain C. Revisions in the international system for staging lung cancer. Chest 1997;111:1710–7.
A.G. Little / Surg Clin N Am 82 (2002) 561–571
571
[3] Ginsberg R, Kris M, Armstrong J. Cancer of the lung. In: DeVita V, Hellman S, Rosenberg S, editors. Cancer: principles and practice of oncology. 4th edition. Philadelphia: J.B. Lippincott; 1993. p. 673–758. [4] Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718–23. [5] Graham ANJ, Chan KJM, Pastorino V, Goldstraw P. Systematic nodal dissection in the intrathoracic staging of patients with non-small cell lung cancer. J Thorac Cardiovasc Surg 1999;117:246–51. [6] Dobashi K, Sugio K, Osabi T, et al. Murometastatic P 53-positive cells in the lymph nodes of non-small-cell lung cancer: prognostic significance. J Thorac Cardiovasc Surg 1997;114:339–46. [7] Giuliano A, Dale P, Turner R, et al. Improved axillary staging of breast cancer with sentinel lymphadenectomy. Ann Surg 1995;222:394–401. [8] Morton D, Wen D, Cochran A. Management of early-stage melanoma by intraoperative lymphatic mapping and selective lymphadenectomy or ‘‘watch and wait.’’ Surg Oncol Clin North Am 1992;1:247–1258. [9] Little A, de Hoyos A, Kirgan D, Arcomano T, et al. Intraoperative lymphatic mapping for non-small cell lung cancer: the sentinel node technique. J Thor Cardiovasc Surg 1999; 117:220–4. [10] Liptay MJ, Masters GA, Winchester DJ, et al. Intraoperative radioisotope sentinel lymph node mapping in non-small cell lung cancer. Ann Thorac Surg 2000;70:384–90. [11] Asamura H, Nakayama H, Kondo H, et al. Lymph node involvement, recurrence, and prognosis in resected small, peripheral, non-small-cell lung carcinomas. J Thorac Cardiovasc Surg 1996;111:1125–34.
Sentinel node biopsy for staging lung cancer Alex G. Little, MD Department of Surgery, University of Nevada, School of Medicine, 2040 West Charleston Boulevard, Suite 601, Las Vegas, NV 89102, USA
Lung cancer is the leading cause of cancer related death in both men and women; it has surpassed breast cancer as the leading cause of cancer death in women. In 1995, 180,000 new cases were diagnosed in the United States and the incidence continues to increase [1]. While prevention is the Holy Grail, maximizing the therapeutic strategy for each patient is essential to improving outcomes. This requires the best possible information about the extent of disease as defined by the staging process. The prognosis of the patient with non-small cell lung cancer (NSCLC) is closely related to the stage of the disease as defined by the TNM classification system developed and endorsed by the American Joint Committee on Cancer (AJCC) [2]. The size, location, and invasiveness of the primary tumor (T); the extent and pattern of peribronchial, hilar, and mediastinal lymph node involvement (N); and the presence of metastasis (M) are the major determinants of staging. The American Thoracic Society/American Joint Committee on Cancer (ATS/AJCC) TNM descriptors for the current staging system of lung cancer are listed below (see box). Table 1 shows the TNM subsets that constitute each stage group. Accurate clinical and pathological staging are essential to define prognosis subgroups and to develop therapeutic strategies based on the presence or absence of metastatic disease and the extent of locoregional disease. The presence or absence of lymph node metastasis is the single most important predictor of recurrence of disease and survival in patients undergoing lung resection: the five-year survival rate is approximately 40% lower in patients who have mediastinal lymph node metastasis than in those who do not [3]. The staging process has three aspects: clinical, surgical, and pathological staging. The clinical stage is based on information gleaned from the full spectrum of non-invasive tests such as chest radiograph (CXR), chest E-mail address: [email protected] (A.G. Little). 0039-6109/02/$ - see front matter 2002, Elsevier Science (USA). All rights reserved. PII: S 0 0 3 9 - 6 1 0 9 ( 0 2 ) 0 0 0 1 6 - 6
562
A.G. Little / Surg Clin N Am 82 (2002) 561–571
Primary tumor (T) TX Primary tumor cannot be assessed, or tumor proven by the presence of malignant cells in sputum or bronchial washings but not visualized by imaging or bronchoscopy TO No evidence of primary tumor Tis Carcinoma in situ T1 Tumor £3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus (ie, not in the main bronchus) T2 Tumor with any of the following features of size or extent: >3 cm in greatest dimension Involves main bronchus, £2 cm distal to the carina Invades the visceral pleura Associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung. T3 Tumor of any size that directly invades any of the following: chest wall (including superior sulcus tumors), diaphragm, mediastinal pleura, parietal pericardium; or tumor in the main bronchus <2 cm distal to the carina, but without involvement of the carina; or associated atelectasis or obstruction of the entire lung T4 Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea, esophagus, vertebral body, carina; or tumor with a malignant pleural or pericardial effusion, or with satellite tumor nodules within the ipsilateral primary-tumor lobe of the lung. Regional lymph nodes (N) NX Regional lymph nodes cannot be assessed N0 No regional lymph node metastasis N1 Metastasis to ipsilateral peribronchial or ipsilateral hilar lymph nodes, and intrapulmonary nodes involved by direct extension of the primary tumor N2 Metastasis to ipsilateral mediastinal or subcarinal lymph nodes N3 Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral, or contralateral scalene, or supraclavicular lymph nodes Distant metastasis (M) MX Presence of distant metastasis cannot be assessed M0 No distant metastasis M1 Distant metastasis present
A.G. Little / Surg Clin N Am 82 (2002) 561–571
563
Table 1 Stage grouping - TNM subsets Stage
TNM subset
0
Carcinoma in situ
IA
T1N0M0
IB
T2N0M0
IIA
T1N1M0
IIB
T2N1M0 T3N0M0
IIIA
T3N1M0 T1N2M0 T2N2M0 T3N2M0
IIIB
T4N0M0 T4N1M0 T4N2M0 T1N3M0 T2N3M0 T3N3M0 T4N3M0
IV
Any T Any N M1
computerized axial tomography (CAT) scan, magnetic resonance imaging (MRI) scan, and positron emission tomography (PET) scan. Invasive procedures, such as mediastinoscopy, performed before definitive surgical intervention are also part of clinical staging. Surgical staging occurs during an operation and incorporates both gross observations by the surgeon, such as chest wall invasion by a primary tumor, and the results of frozen section analysis of biopsied or removed tissue, such as lymph nodes and bronchial margins. Finally, the pathologic stage is determined from histologic inspection of all tissue removed during the definitive operative procedure. Accurate clinical staging of mediastinal, hilar, and intrapulmonary lymph nodes is an important aspect of the process of determining the therapeutic strategy for an individual lung cancer patient, where the possibilities range from surgery alone to chemoradiation therapy or an integrated multimodality approach. Fig. 1 depicts the regional lymph node stations generally used for lung cancer staging [4]. Accurate surgical staging allows determination of the most appropriate extent of resection, wedge resection, lobectomy, or pneumectomy; the need for en bloc resection of adjacent chest wall; and the type and extent of mediastinal lymph node sampling or dissection. Pathologic staging establishes the definitive or ultimate stage and accuracy is paramount. Identification of residual disease by the finding of tumor at a bronchial margin is of course important, but the more common issue is accurate determination of whether or not mediastinal lymph nodes contain
564
A.G. Little / Surg Clin N Am 82 (2002) 561–571
Fig. 1. The joint ATS/AJCC system for regional lymph node classification for lung cancer staging. (From Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1719; with permission.)
metastatic disease. Because survival is significantly affected by the presence of nodal disease, this latter determination impacts both patient prognosis and decisions regarding the use of postoperative adjuvant therapy. Despite the importance of nodal staging as a predictor of prognosis and determinant of therapy in patients being operated on for lung cancer, consensus is lacking in the thoracic surgical community with respect to the preferred approach to the retrieval of lymph nodes for postoperative, pathological staging in patients undergoing lung resection. The options are (1) to
A.G. Little / Surg Clin N Am 82 (2002) 561–571
565
simply remove the nodes (hilar, segmented, and parenchymal) that are attached to and part of the lung specimen, (2) to biopsy selected mediastinal node groups, or (3) to dissect and remove the nodal groups en masse. When mediastinal lymph nodes are not routinely dissected or sampled, understaging occurs. Up to 20% of lung cancer patients with clinical and surgical Stage I or II disease have been demonstrated to have occult involvement of mediastinal lymph nodes (N2), identification of which results in upstaging to a Stage III classification [5]. Even when lymph nodes are removed and examined histologically, micrometastatic disease can be missed, further complicating the situation. Immunohistochemical staining of removed lymph nodes for a variety of tumor associated antigens is able to demonstrate the pressure of so-called micrometastatic disease in histologically tumor negative lymph nodes [6]. Because of the logistics of the process of obtaining this information, which involves sectioning, staining, and examining multiple lymph nodes, a way to simplify the process without sacrificing sensitivity would be helpful. Operative lymphatic mapping and selective lymphadenectomy History The technique of intraoperative lymphatic mapping and selective lymphadenectomy to identify the sentinel lymph node (SLN) was developed as a method of determining the regional lymph node status in patients with breast cancer or melanoma [7,8]. It was observed that pathological examination of the sentinel lymph node increased the accuracy of nodal staging [7,8]. This increased accuracy results from the observation that detailed pathologic examination of the sentinel lymph node, with multiple sections being inspected, results in increased detection of micrometastasis in comparison with routine examination of multiple lymph nodes. This technique, as applied to lung cancer, relies on the hypothesis that for a given segment or lobe of the lung, there is a dominant lymph node in the basin to which the lymphatic flow drains first; ie, the sentinel lymph node. Current status of mediastinal staging in lung cancer Intrapulmonary, peribronchial, and hilar lymph nodes are routinely removed in the process of performing a standard lobectomy or pneumectomy. Dissection of mediastinal lymph nodes, which requires an additional effort, is the challenge and is not routinely performed. Despite the fact that most or all patients have preoperative mediastinal staging with CAT, MRI, or PET scans, and some undergo mediastinoscopy, up to 20% of patients have occult disease found by histologic examination if mediastinal lymph nodes are removed by a formal mediastinal dissection [5]. Therefore, some form of routine surgical staging is necessary to ensure maximal staging accuracy, which in turn can be used to predict the patient’s prognosis and affect therapeutic
566
A.G. Little / Surg Clin N Am 82 (2002) 561–571
strategies. Either intraoperative mediastinal node dissection or node sampling (cherry picking) has been suggested as a means to provide more accurate staging than examination of lymph nodes attached to a lobe or even an entire lung removed in standard fashions. However, neither approach is universally accepted, perhaps because of the increase in operative time and potential morbidity associated with these activities. This has created interest in applying the lymphatic mapping technique, developed for and in patients with breast cancer or melanoma, to patients being operated on for lung cancer. University of Nevada sentinel lymph node experience Our group’s experience with the sentinel node technique was gained using the following technique [9]. A standard posterior lateral thoracotomy with or without a muscle-sparing technique is performed. In patients with tumors involving the chest wall, the incision is extended as necessary and a three-rib segment of chest wall is resected in continuity with the lung. Upon entering the chest, the lung is collapsed as contralateral ventilation is continued through a double lumen endotracheal tube. Visual and tactile inspection of the lung and mediastinum is performed to identify areas of abnormality. A total of 5 ml of 1% isosulfan blue dye (lymphazurin) is injected into the lung parenchyma in four quadrants around the primary tumor. Use of a relatively small needle, such as a 21 gauge, reduces dye spillage from the needle holes and staining of the surrounding tissues, which can increase the difficulty of identifying a sentinel node. The pleura is then circumferentially incised around the hilum, and routine lung resection is carried out. Ten minutes is allowed for lymphatic spread of the vital blue dye. Attention is then directed initially to the adjacent mediastinal node basins. The first lymph node to stain blue is considered the SLN and is removed and sent separately for permanent histological examination. A stained drainage lymphatic vessel feeding the SLN is also sought but, in our experience, rarely found. If needed, small incisions are created in the mediastinal pleural to allow identification of the blue-stained afferent lymphatic vessels leading to the SLN. Once the SLN is identified, and distinguished from adjacent anthracotic nodes, the entire node is removed by means of sharp dissection or electrocautery. Afferent and efferent lymphatic vessels proceeding from the SLN, some of which may be stained blue, are controlled with hemostatic clips for anatomic reference. Ultimately, lobectomy, bilobectomy, or pneumonectomy is performed. In our initial experience, at the completion of the lung resection, mediastinal node dissection was performed. During a right thoracotomy, levels 4, 7, 8, 9 and 10 (according to the unified lymph node mapping developed jointly by the ATS and AJCC [4]) were removed. Levels 4, 5, 6, 7, 8, and 9 were resected during a left thoracotomy. In our initial study, 36 patients with non-small cell lung cancer were analyzed [8]. Squamous cell carcinoma was present in 13 patients, adenocarcinoma in 19 patients, and other histologic types in 4 patients. All patients
A.G. Little / Surg Clin N Am 82 (2002) 561–571
567
were evaluated before the operation by computed tomographic scanning of the chest and the upper abdomen to evaluate mediastinal lymph nodes and to exclude liver and adrenal gland involvement. Seven of the 36 patients underwent cervical mediastinoscopy for pathological examination of paratracheal or subcarinal lymph nodes greater than 1 cm in short axis on the CT, and all analyzed lymph node biopsies from these 7 patients were negative for the presence of metastatic involvement. Twenty-nine patients were clinical Stage I, and 7 patients were clinical Stage II. Results Pathologic staging 1. Twenty-one patients (58.3%) had Stage I: T1N0M0 in 11 (30.6%) and T2N0M0 in 10 (27.8%). 2. Seven patients (19.4%) had Stage II: T1 or T2N1M0 in 6 (16.7%) and T3N0M0 in 1 (2.8%). 3. Eight patients (22.2%) had Stage IIIA: T1N2M0 in 4 (11.1%), T2N2M0 in 2 (5.5%) and T3N1M0 in 2 (5.5%). Sentinel node results The University of Nevada results from the initial experience with the sentinel lymph node technique in 36 consecutive patients with NSCLC were as follows (see box): Sentinel Lymph Node Identified 1. Total: 17 of 36 patients (47%) 2. Tumor Positive: 8 of 17 patients (47%) Path N1, 3 patients Path N2, 5 patients 3. Tumor Negative: 9 of 17 patients (53%) Path N0, 9 patients No Sentinel Lymph Node Identified 1. 2. 3. 4.
Total: 19 of 36 patients (53%) Path N0, 13 patients Path N1, 5 patients Path N2, 1 patient
Sentinel node positive A sentinel lymph node was found in 17 of the 36 patients (47%). In 9 of these 17 patients neither the sentinel node nor any other lymph node
568
A.G. Little / Surg Clin N Am 82 (2002) 561–571
contained metastatic disease, so nodal staging was N0. The sentinel nodes in the other 8 patients contained metastatic NSCLC: in 3 patients the sentinel node was either hilar or intrapulmonary and contained metastatic tumor (N1), but all mediastinal nodes were negative for metastatic involvement. In the remaining 5 patients the sentinel lymph node was in the mediastinum; paratracheal in 3 (station 4) and subcarinal in 2 (station 7). Pathologic examination of the SLN documented metastatic involvement (N2), despite both a negative preoperative CT scan and mediastinoscopy. In these 5 patients, the sentinel lymph node was the only site of metastasis; no other mediastinal lymph node stations contained metastatic tumor. Sentinel node negative In 19 patients (53%), no sentinel lymph node was identified, despite careful intra-operative mediastinal lymph node dissection. Among these patients, pathologic staging was N0 in 13, N1 in 5, and N2 in 1. Radioguided technique The only other reported experience with sentinel lymph node mapping in lung cancer patients is from Liptay and associates [10]. They also used a technique at the time of thoracotomy. They injected a total dosage of two mCi of a technetium–99-labeled colloid suspension at the outer margins of the tumor in four quadrants. Subsequently they traced and mapped the colloid migration with a handheld gamma probe counter. The migration was considered successful if a nodal station registered counts per second greater than three times background values. The migration time averaged 63 minutes in patients in whom sentinel nodes were identified. Out of 45 patients with lung cancer, 37 (82%) were said to have had a sentinel node identified. This is a considerably higher sensitivity rate of sentinel node identification than we were able to obtain at the University of Nevada. The actual identification, however, as the authors point out, was of ‘‘nodal stations’’ that consisted of several lymph nodes rather than a single sentinel node. While this result could be used to focus pathologic examination, it is still necessary for the pathologist to examine several lymph nodes to search for histological or micrometastatic disease.
Discussion The use of isosulfan blue dye for lymphatic mapping has been extensively studied in patients with melanoma and breast cancer and has become a standard staging procedure in breast cancer patients undergoing mastectomy [6,7]. In these groups of patients, examination of the sentinel node yields accurate pathologic staging information and makes possible rational and selectively aggressive surgical and medical approaches.
A.G. Little / Surg Clin N Am 82 (2002) 561–571
569
The initial premise of the sentinel node technique was that awareness of the presence or absence of metastatic tumor in the node would be of benefit by predicting the status of all other regional lymph nodes. If this were true, a negative intraoperative frozen section analysis of the identified node would obviate the need for further lymph node dissection and histological examination. A positive result would lead to a potentially therapeutic regional node dissection. This line of reasoning has held up to some extent in the surgical treatment of patients with breast carcinoma or melanoma. We were encouraged in our experience that the specificity of the sentinel node technique was excellent. All 9 patients with tumor negative sentinel lymph nodes were found to be free of disease on final pathologic examination of mediastinal lymph nodes. In addition, of the 19 patients without a sentinel lymph node, 18 were either N0 or N1. This potential intraoperative role for lymphatic mapping, however, seems unlikely to pertain to lung cancer patients for two reasons. First, the blue dye technique has a relatively low sensitivity for detection of a sentinel node. We were able to identify a sentinel node only in approximately half our patients. Although we hoped our continued experience would result in an improved sensitivity (the learning curve phenomenon), as occurred in the breast cancer and melanoma experiences, that has not happened—our yield remains at the 50% level. We feel this is due to several reasons. The lymphatic network draining the lung is extensive and provides multiple pathways for tumor dissemination. This frustrating anatomic reality has been demonstrated in studies that examine the routes of lung cancer drainage. One illustrative example is provided in Table 2, adapted from Asamura and his colleagues [11], which shows the diffuse pattern of mediastinal lymph node involvement in 88 patients with peripheral non-small cell carcinomas. In addition, blue staining is difficult to identify in anthracotic nodes surrounded by other blue/black nodes and collapsed lung. Second, the radiation-guided approach appears to detect a lymph node group rather than a single node, increasing the challenge of time-efficient frozen section analysis during surgery. Also, both the isosulfan blue dye and the radio guided techniques add at least a modest amount of time and tediousness to the lung resection procedure, decreasing their appeal to the thoracic surgeon. Table 2 Lymph node involvement in the mediastinum according to tumor location Mediastinal lymph node location Stations
1
2
3
4
5
6
RUL (122) 5 (4.1) 3 (2.5) 17 (13.9) 4 (3.3) RML (31) 2 (6.5) 1 (3.2) 5 (16.1) RLL (64) 3 (4.7) 3 (4.7) LUL (83) 3 (3.6) 2 (2.4) 6 (7.2) 5 (6.0) 6 (7.2) LLL (37) 1 (2.7) 1 (2.7) 1 (2.7)
7 2 5 8 4 3
8
9
N2/3
(1.6) 18 (14.8) (16.1) 7 (22.6) (12.5) 2 (3.1) 11 (17.2) (4.8) 15 (18.1) (8.1) 1 (2.7) 1 (2.7) 5 (13.5)
Adapted from Asamura H, Nakayama H, Kondo H, J Thorac Cardiovasc Surg 1996; 111; with permission.
570
A.G. Little / Surg Clin N Am 82 (2002) 561–571
Investigators have demonstrated that an upward staging occurs when this selective lymphatic mapping technique is employed in breast cancer patients or melanoma patients [6,7]. This is the most likely way that the sentinel node technique will play a useful role in the staging of lung cancer patients. It appears that allowing the pathologist to focus on a single lymph node results in finding micrometastases that otherwise would be missed by standard processing of multiple lymph nodes. With only a single node to examine, the use of specialized techniques for examination of the sentinel node, including immunohistochemical staining and polymerase chain reaction, may even further increase the ability to detect only a few metastatic cells. Early in our experience we identified the accumulation of blue dye in the central portion of lymph nodes in some of our patients. This was not detectable by surface inspection, and was only found when the nodes were bisected. This suggested initially that radionuclide-guided techniques might improve the sensitivity of this technique in lung cancer patients. However the experience of Liptay and associates is only moderately encouraging, because the sensitivity of approximately 80% that they found was still lower than hoped for, and dealing with a packet of identified nodes is not a major benefit. The approach we are now using is to dissect and remove all the ipsilateral mediastinal lymph node groups in toto. After the thoracotomy is closed, the node groups are separated into individual nodes, which are then bisected. This approach in our most recent 23 patients has resulted in the finding of one or at most two sentinel nodes in 22 of the patients. A multiple section analysis of the sentinel node, both histological and with immunohistological staining, is then compared to the results of routine histological examination of all other nodes. The completion of this ongoing study should help determine if this application of the principles of the technique of the sentinel node in non-small cell cancer patients is helpful in the process of pathological staging. The important understanding is that acquisition of sufficient tissue to determine the status of N1 and N2 lymph node basins is an essential responsibility of the thoracic surgeon performing resections for carcinoma of the lung. Although a therapeutic benefit may result from a complete mediastinal lymphadenectomy, the main benefit relates to more accurate staging, because increasingly effective multimodality therapies are available for patients with N2 disease. With the knowledge that N1 status does not reliably predict N2 node status, either routine mediastinal lymphadenectomy, lymph node sampling, or the sentinel lymph node technique should be routinely employed to accurately stage the disease and select the most appropriate therapy for each individual patient. References [1] Parker S, Tong T, Bolden S, et al. Cancer statistics. Cancer 1997;45:5. [2] Mountain C. Revisions in the international system for staging lung cancer. Chest 1997;111:1710–7.
A.G. Little / Surg Clin N Am 82 (2002) 561–571
571
[3] Ginsberg R, Kris M, Armstrong J. Cancer of the lung. In: DeVita V, Hellman S, Rosenberg S, editors. Cancer: principles and practice of oncology. 4th edition. Philadelphia: J.B. Lippincott; 1993. p. 673–758. [4] Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718–23. [5] Graham ANJ, Chan KJM, Pastorino V, Goldstraw P. Systematic nodal dissection in the intrathoracic staging of patients with non-small cell lung cancer. J Thorac Cardiovasc Surg 1999;117:246–51. [6] Dobashi K, Sugio K, Osabi T, et al. Murometastatic P 53-positive cells in the lymph nodes of non-small-cell lung cancer: prognostic significance. J Thorac Cardiovasc Surg 1997;114:339–46. [7] Giuliano A, Dale P, Turner R, et al. Improved axillary staging of breast cancer with sentinel lymphadenectomy. Ann Surg 1995;222:394–401. [8] Morton D, Wen D, Cochran A. Management of early-stage melanoma by intraoperative lymphatic mapping and selective lymphadenectomy or ‘‘watch and wait.’’ Surg Oncol Clin North Am 1992;1:247–1258. [9] Little A, de Hoyos A, Kirgan D, Arcomano T, et al. Intraoperative lymphatic mapping for non-small cell lung cancer: the sentinel node technique. J Thor Cardiovasc Surg 1999; 117:220–4. [10] Liptay MJ, Masters GA, Winchester DJ, et al. Intraoperative radioisotope sentinel lymph node mapping in non-small cell lung cancer. Ann Thorac Surg 2000;70:384–90. [11] Asamura H, Nakayama H, Kondo H, et al. Lymph node involvement, recurrence, and prognosis in resected small, peripheral, non-small-cell lung carcinomas. J Thorac Cardiovasc Surg 1996;111:1125–34.