Ann Thorac Surg 2005;80:1881-1886
© 2005 The Society of Thoracic Surgeons
Original article: General thoracic
A Strategy for Determining Which Thoracic Esophageal Cancer Patients Should Undergo Cervical Lymph Node Dissection
Shiro Nagatani, MD,
Yutaka Shimada, MD, PhD
*
,
Masato Kondo, MD,
Junichi Kaganoi, MD, PhD,
Masato Maeda, MD, PhD,
Go Watanabe, MD, PhD,
Masayuki Imamura, MD, PhD
Department of Surgery and Surgical Basic Science, Graduate School of Medicine, Kyoto University, Kyoto, Japan
Accepted for publication April 26, 2005.
* Address correspondence to Dr Shimada, Department of Surgery and Surgical Basic Science, Graduate School of Medicine, Kyoto University, Kawaracho 54, Shogoin Sakyo-ku, Kyoto 606-8507, Japan (Email: shimada{at}kuhp.kyoto-u.ac.jp).
 |
Abstract
|
|---|
BACKGROUND: There is controversy about performing cervical lymph node dissections in all middle and lower thoracic esophageal squamous cell carcinoma patients. The purpose of this study was to evaluate whether intraoperative examination of thoracic paratracheal lymph node by real-time reverse transcription-polymerase chain reaction was worthwhile for selecting patients for cervical lymph node dissection.
METHODS: Under informed consent, 30 middle and lower thoracic esophageal squamous cell carcinoma patients were examined for thoracic paratracheal lymph node metastasis intraoperatively by hematoxylin-eosin staining and real-time reverse transcription-polymerase chain reaction for messenger RNA encoding squamous cell carcinoma antigen. When thoracic paratracheal lymph node metastasis was found, cervical lymph node dissection was performed. If the patients had no thoracic paratracheal lymph node metastasis, a randomized study for selection of cervical lymph node dissection was performed.
RESULTS: Eleven of 30 patients with middle or lower third thoracic esophageal squamous cell carcinoma had thoracic paratracheal lymph node metastasis. Five of these 11 patients had cervical lymph node metastasis. Nineteen patients who had no metastasis in the thoracic paratracheal lymph nodes were enrolled in a randomized study. Eight of the 19 patients received cervical lymph node dissection and they were found not to have cervical lymph node metastasis. The other 11 patients did not receive cervical lymph node dissection, and there was no cervical lymph node recurrence.
CONCLUSIONS: The intraoperative diagnosis of metastasis in the thoracic paratracheal lymph node may be used as an indicator for cervical lymph node dissection in middle and lower thoracic esophageal cancer patients.
|
Introduction
|
|---|
The use of cervical lymph node (LN) dissection in thoracic esophageal squamous cell carcinoma (ESCC) has been controversial. There is still some debate as to whether performing cervical LN dissection on ESCC patients is necessary because of the lack of agreement of the benefits of such procedures on survival or quality of life. For example, cervical LN dissection may reduce cervical recurrence, although an increase in the incidence of postoperative complications, such as recurrent nerve palsy or difficulty in swallowing, may occur [1–3]. Therefore, the establishment of indications for cervical LN dissection is considered a priority.
Our retrospective report showed that surgery for thoracic ESCC, both with and without cervical LN dissection, based on the results from selective intraoperative pathologic diagnosis assessment (hematoxylin-eosin [HE] staining) for thoracic paratracheal LN was not fully acceptable owing to the lack of micrometastasis detection [4].
A previous report indicated that there was a significant relationship between cervical LN metastasis, including micrometastasis, and thoracic paratracheal LN metastasis in middle and lower thoracic ESCC by immunohistochemistry, and suggested that there was a high rate of cervical LN metastasis in the upper third of thoracic ESCC patients [5]. Multivariate analysis indicated that only thoracic paratracheal LN status was an independent predictive factor for cervical LN metastasis.
Recently, molecular biological methods using polymerase chain reaction (PCR) analysis were developed and used widely to detect micrometastasis in patients with a variety of malignant tumors. Another report suggested that a real-time reverse transcription–PCR (RT-PCR) assay might be useful for intraoperative diagnosis of subsequent cervical LN dissections in esophageal cancer [6]. We also reported that reverse transcription–nested PCR for messenger RNA (mRNA) encoding squamous cell carcinoma antigen mRNA was useful for detecting LN micrometastasis [7] and circulating cancer cells [8]. However, the reverse transcription–nested PCR method in previous reports [7, 8] could not be used for the intraoperative diagnosis of LN metastasis owing to the long assay time required. Thus, we established a method for intraoperative detection of thoracic paratracheal LN metastasis using real-time RT-PCR targeting ESCC with a Light Cycler (Roche Molecular Biochemicals, Mannheim, Germany) [9]. The Light Cycler technique can also quantify cancer-specific mRNA with real-time monitoring of PCR products.
The purpose of this prospective study was to evaluate whether intraoperative examination of the thoracic paratracheal LN by real-time RT-PCR is worthwhile for selecting patients for cervical LN dissection from middle and lower thoracic esophageal cancer patients.
|
Patients and Methods
|
|---|
Patients and Operations
We first obtained written informed consent from patients for surgery and for selection for cervical LN dissection. (The approval number of the Kyoto University Institutional Review Board was 347) The thoracic paratracheal LN of 40 thoracic ESCC patients were examined intraoperatively to detect metastasis, including micrometastasis, and 30 of the 40 were middle and lower thoracic ESCC patients. The patients were curatively operated on between August 2001 and July 2003 in our department. The standard surgical method used has been described previously [10]. In brief, esophagectomy with systematic LN dissection was performed using a right thoracotomy, and reconstruction was carried out using an esophagogastrostomy with a gastric tube through the retrosternal or intrathoracic route. To analyze lymphatic mapping in detail, LNs were categorized according to the mapping scheme of the American Thoracic Society, modified by Casson and colleagues [11]. The number of nodal station of the thoracic paratracheal lymph node, including recurrent nerve chain LN, was 2 and 4 in the classification of lymph node classification number of the North American multi-institutional clinical trial of the surgical treatment of esophageal cancer.
Eligibility Inclusion and Exclusion Criteria for the Randomization
Our criteria for the randomized study were as follows: (1) aged between 20 and 75 years, with an Eastern Cooperative Oncology Group performance status (ECOG-PS) of 0 through 1; (2) tumor located in the middle or lower third of the thoracic esophagus; (3) no adjacent organ involvement (less than T4); (4) no distant organ metastasis; (5) no obvious cervical LN metastasis (M0); (6) no micrometastasis in the thoracic paratracheal LN; (7) no LN metastasis within the circumference of the celiac axis (M0); (8) no recurrent nerve palsy; and (9) no in situ tumors.
In addition, patients were excluded for (1) a history of radiation therapy or chemotherapy, (2) the presence of serious cardiovascular, pulmonary, renal, or hepatic diseases, (3) the coexistence of an active malignant neoplasm, and (4) any conditions that the physician considered would preclude the trial.
Eligibility Criteria for Prospective Cervical Lymph Node Dissection
The criteria for the prospective cervical LN dissection were as follows: (1) patients who had micrometastasis in the thoracic paratracheal LN by RT-PCR, (2) patients with upper thoracic esophageal tumors, and (3) patients who were suspected of having cervical lymph node metastasis by preoperative examination such as computed tomography or ultrasonography.
Detection of Thoracic Paratracheal Lymph Node Metastasis
Protocol for evaluating thoracic paratracheal LN metastasis, as previously reported, was that all LNs more than 2 mm were divided into two parts by a sharp razor, and each part was checked for metastasis. One part was subjected to intraoperative frozen sectional HE staining, and the other part was subjected to real-time RT-PCR for squamous cell carcinoma antigen mRNA with a Light Cycler during the operation. The Light Cycler is an innovative thermal cycler platform with a fluorescence detection system, and this system enables rapid PCR while simultaneously quantifying and analyzing the results by monitoring fluorescence during amplification. The procedure for real-time RT-PCR was previously described [9]. In brief, real-time RT-PCR was performed in Light Cycler capillaries using a commercially available master mix containing Taq DNA polymerase and SYBR-Green I deoxyribonucleoside triphosphates (Light Cycler DNA master SYBR-Green I; Roche Molecular Biochemicals). After the addition of primers (final concentration, 0.25 pM), MgCl2 (4 mM), and template DNA to the master mix, 45 cycles of denaturation (94°C for 1 second), annealing (58°C for 10 seconds), and extension (72°C for 10 seconds) were performed. After the completion of PCR amplification, a melting curve analysis was performed. First, we dissected thoracic paratracheal LNs by a right thoracotomy and sent them to the laboratory. Then, during the LN dissection of the abdominal area and the construction of a gastric tube, the evaluation of RT-PCR was reported from the laboratory to the operation room. Eligibility was finally decided according to the operative findings of the laparotomy, and eligible patients were assigned either to receive or not receive cervical LN dissection by telephone from the operating theater to our central office.
Postoperative Follow-Up
The duration of follow-up time for the patients in this prospective study was, on average, 23.4 months (range, 10 to 33). Postoperative follow-up was done by computed tomography and ultrasonography every 6 months.
Study Design and Statistical Analysis
The primary endpoint of the randomized study was disease-free survival. The secondary endpoint was cervical LN recurrence. This study was designed to include 230 randomly assigned patients with a one-sided alpha of 0.05 and beta of 0.2. A proportional analysis between the sample groups was performed with Fisher's exact probability test in 2 x 2 contingency tables and with the 
2 test in 2 x n contingency tables. Statistical calculations were performed with Stat View for Windows software, version 5.0 (SAS Institute, Cary, North Carolina). The Kaplan-Meier survival model was used to estimate overall survival. The log-rank test was used to determine statistical differences between groups.
|
Results
|
|---|
Patient Characteristics
The clinicopathologic findings (5th Union Internationale Contre le Cancer [UICC] TNM classification, 1997) of the 30 patients are summarized in Table 1. The patients consisted of 27 men (90%) and 3 women (10%). All patients had squamous cell carcinoma of the esophagus. There were 9 T1 patients, 8 T2 patients, and 13 T3 patients according to the TNM classification. There were 9 N0 patients and 21 N1 patients, and no patients had distant organ metastasis (M0); however, 5 patients had distant LN metastasis (M1, LN; Table 1).
View this table:
[in this window]
[in a new window]
|
Table 1. Clinicopathologic Findings of the Eligible Patients With Middle and Lower Thoracic Esophageal Squamous Cell Carcinoma
|
|
Prospective Study Process
The prospective process is shown in Figure 1. Eleven patients (group A in Fig 1) underwent cervical LN dissection because they had metastasis of the thoracic paratracheal LN according to the intraoperative diagnosis. Nineteen patients who had no thoracic paratracheal LN metastasis were randomized (groups B and C in Fig 1), and 8 patients received cervical LN dissection (group B). The clinicopathologic findings of the 19 patients of this randomized study (groups B and C in Fig 1) are summarized in Table 2. There was no significant difference between the two groups. The numbers of LNs examined by intraoperative PCR techniques were 4.67 on average in the 30 patients, ranging from 2 to 6 (SD 1.07).
View larger version (27K):
[in this window]
[in a new window]
|
Fig 1. The prospective study process. Group A patients underwent cervical lymph node (LN) dissection because they were diagnosed as having upper mediastinal LN metastasis by intraoperative diagnosis. Group B patients received cervical LN dissection with randomized selection. Group C patients did not receive cervical LN dissection with randomized selection. (ESCC = esophageal squamous cell carcinoma.)
|
|
Comparison Between Intraoperative Detection and Metastatic Status in Cervical Lymph Nodes in Eligible Cases
Intraoperative detection was performed in 30 patients, including 4 HE positive or PCR positive, 2 HE positive or PCR negative, 5 HE negative or PCR positive, and 19 HE negative or PCR negative (Table 3). Five of 11 patients (45.5%) who underwent cervical LN dissection because they had metastasis of upper mediastinal LN on intraoperative diagnosis were found to have cervical LN metastasis (group A in Table 3, Fig 2). Regarding the randomized study, there was no cervical LN metastasis by real-time RT-PCR or conventional HE staining in group B (Table 3, Fig 2); and there was no cervical LN recurrence in group C, in which cervical LN dissection was not performed (Table 3, Fig 2).
View larger version (21K):
[in this window]
[in a new window]
|
Fig 2. Results of intraoperative detection of thoracic paratracheal lymph node (LN) metastasis and metastatic status in cervical lymph nodes. Cervical LN metastasis was detected by conventional hematoxylin and eosin staining and real-time reverse transcription-polymerase chain reaction. Cervical LN recurrence was followed up by commonly used diagnostic techniques, such as computed tomography and ultrasonography.
|
|
Accuracy of Predicting Cervical Metastasis Using Intraoperative Diagnosis of the Thoracic Paratracheal Lymph Node
To calculate sensitivity, specificity and predictive values of the intraoperative diagnosis of upper mediastinal LN, analyses included all patients except those who did not undergo cervical LN dissection. As shown in Table 4, the sensitivity of predicting cervical metastasis was 100%, specificity was 61.5%, the positive predictive value was 54.5%, and the negative predictive value was 100%.
View this table:
[in this window]
[in a new window]
|
Table 4. Accuracy of Intraoperative Pathological Diagnosis of the Paratracheal Lymph Node (LN) in the Prediction of Cervical Metastasis
|
|
Surgical Duration and Time Requirement for Real-Time RT-PCR
Surgical duration among the 30 cases was 589 minutes on average, and the mean duration of real-time RT-PCR using the Light Cycler was 179.4 minutes on average, ranging from 162 to 210 minutes (SD 16.2).
Morbidity, Mortality, and Prognosis
Concerning the postoperative complications, there was no significant difference in vocal cord paralysis or postoperative pulmonary complications between groups B and C. And there was no cervical LN recurrence or operative deaths among the 30 patients. Furthermore, there was no statistical difference in the survival according to doing the neck dissection or not in 30 patients (log-rank test, p = 0.116; Fig 3).
View larger version (15K):
[in this window]
[in a new window]
|
Fig 3. Kaplan-Meier curves showing overall survival rate according to doing the neck dissection or not in 30 patients (log-rank test, p = 0.116).
|
|
|
Comment
|
|---|
Cervical LN dissection in ESCC has been controversial. Some authors showed that 35% of ESCC had metastasis and stated that the cervical field of dissection was as likely as the mediastinum to be a site of nodal disease [12]. On the other hand, Orringer and coworkers [13] claimed that cervical LN metastasis of the ESCC should be considered a systematic disease, and even if one third of all patients with ESCC had cervical LN metastasis, routine cervical LN dissection should not be performed. However, there remains some debate regarding the performance of cervical LN dissection on ESCC patients.
Our results showed that 45.5% of patients who were detected as having thoracic paratracheal LN metastasis intraoperatively, with middle and lower third thoracic ESCC, had cervical LN metastasis. These results suggest that these patients should receive cervical LN dissection. In our prospective randomized study, 19 patients who had no thoracic paratracheal LN metastasis were found to have no cervical LN metastasis or recurrence. This finding indicated that patients who do not have metastasis in the thoracic paratracheal LN may not require cervical LN dissection for middle and lower third thoracic ESCC.
Regarding procedural disadvantages, some authors have showed that cervical LN dissections impaired swallowing and increased the risk of recurrent nerve palsy [13, 14]. In our randomized study, there was no difference between two-field LN dissection (with cervical dissection) and three-field LN dissection (without cervical dissection) in terms of the occurrence of postoperative pulmonary complications or vocal code paralysis.
The real-time RT-PCR format can also quantify cancer-specific mRNA with real-time monitoring of PCR products during the operation, and the results are more reproducible and reliable compared with conventional intraoperative diagnoses such as HE staining [6, 9]. Although, the time requirement of real-time RT-PCR was 179.4 minutes on average, this intraoperative diagnosis method was useful for esophageal surgery because we obtained the results of RT-PCR during an abdominal procedure. However, 3 hours is too long for shorter surgery such as gastric or colon resection.
With regard to esophageal adenocarcinoma, Altorki and associates [15] reported that cervical LN metastasis occurs with a similar frequency in adenocarcinoma and squamous cell carcinoma. Thus, our results suggest that the real-time RT-PCR detection method may be useful for esophageal adenocarcinoma, although carcinoembryonic antigen mRNA may suitable for adenocarcinoma cases.
Our results also showed that some patients had metastasis in the thoracic paratracheal LN that was not detected by the real-time RT-PCR reaction. The reason for real-time RT-PCR negativity in the case of positive HE staining could be that the portion of the LN subjected to the RT-PCR assay contained no cancer cells. We assume that if the whole LN were analyzed for metastasis by real-time RT-PCR, the detection rate of LN metastasis might increase.
In this prospective study, there was no statistical difference in the survival according to doing the neck dissection or not in 30 patients. The duration of follow-up for the patients was short (average, 23.8 months; range, 10 to 33) and the number of enrolled patients was small. Thus, we were unable to obtain a definite conclusion from our study, and research must continue. A larger number of patients are required to prove the validity of this novel approach.
In conclusion, we considered which middle and lower thoracic ESCC patients should be selected for cervical LN dissection. Our study indicated that an intraoperative diagnosis of metastasis in the thoracic paratracheal LN may be used as an indicator for cervical LN dissection.
|
Acknowledgments
|
|---|
This work was supported in part by a Grant-in-Aid (no. 14370385) from the Japanese Ministry of Education, Culture, Sports, Science and Technology.
|
References
|
|---|
- Law S, Wong J. Two-field dissection is enough for esophageal cancer Dis Esophagus 2001;14:98-103.[Medline]
- Dresner SM, Griffin SM. Pattern of recurrence following radical oesophagectomy with two-field lymphadenectomy Br J Surg 2000;87:1426-1433.[Medline]
- Nishimaki T, Suzuki T, Suzuki S, Kuwabara S, Hatakeyama K. Outcomes of extended radical esophagectomy for thoracic esophageal cancer J Am Coll Surg 1998;186:306-312.[Medline]
- Shimada Y, Sato F, Maeda M, et al. Validity of intraoperative pathological diagnosis of paratracheal lymph node as a strategy for selection of patients for cervical lymph node dissection during esophagectomy Dis Esophagus 2003;16:246-251.[Medline]
- Sato F, Shimada Y, Li Z, et al. Paratracheal lymph node metastasis is associated with cervical lymph node metastasis in patients with thoracic esophageal squamous cell carcinoma Ann Surg Oncol 2002;9:65-70.[Abstract/Free Full Text]
- Yoshioka S, Fujiwara Y, Sugita Y, et al. Real-time rapid reverse transcriptase-polymerase chain reaction for intraoperative diagnosis of lymph node micrometastasisclinical application for cervical lymph node dissection in esophageal cancers. Surgery 2002;132:34-40.[Medline]
- Kano M, Shimada Y, Kaganoi J, et al. Detection of lymph node metastasis of oesophageal cancer by RT-nested PCR for SCC antigen gene mRNA Br J Cancer 2000;82:429-435.[Medline]
- Kaganoi J, Shimada Y, Kano M, Okumura T, Watanabe G, Imamura M, et al. Detection of circulating oesophageal squamous cancer cells in peripheral blood and its impact on prognosis Br J Surg 2004;91:1055-1060.[Medline]
- Nagatani S, Shimada Y, Li Z, et al. Validation of intraoperative detection of paratracheal lymph node metastasis using real-time RT-PCR targeting esophageal squamous cell carcinoma Jp J Clin Oncol 2003;33:549-555.
- Imamura M, Ohishi K, Tobe T. Retrosternal esophagogastrostomy with the EEA stapler Surg Gynecol Obstet 1987;164:368-371.[Medline]
- Casson AG, Rusch VW, Ginsberg RJ, Zankowicz N, Finley RJ. Lymph node mapping of esophageal cancer Ann Thorac Surg 1994;58:1569-1570.[Medline]
- Altorki N, Skinner D. Should en bloc esophagectomy be the standard of care for esophageal carcinoma? Ann Surg 2001;234:581-587.[Medline]
- Orringer MB. Occult cervical nodal metastasis in esophageal cancerpreliminary results of three-field lymphadenectomy. J Thorac Cardiovasc Surg 1997;113:538-539.[Free Full Text]
- Tachimori Y, Kato H, Watanabe H. Surgery for thoracic esophageal carcinoma with clinically positive cervical nodes J Thorac Cardiovasc Surg 1998;116:954-959.[Abstract/Free Full Text]
- Altorki N, Kent M, Ferrara C, Port J. Three-field lymph node dissection for squamous cell and adenocarcinoma of the esophagus Ann Surg 2002;236:177-183.[Medline]
Top
Abstract
Introduction
