Endobronchial Ultrasonography Added to Endoscopic Ultrasonography Improves Staging in Esophageal Cancer

Transcription

1 Endobronchial Ultrasonography Added to Endoscopic Ultrasonography Improves Staging in Esophageal Cancer Moishe Liberman, MD, PhD, Nawar Hanna, MD, Andre Duranceau, MD, Vicky Thiffault, RN, and Pasquale Ferraro, MD Department of Surgery, Division of Thoracic Surgery, Centre Hospitalier de l Universite de Montreal Endoscopic Tracheobronchial and Oesophageal Center, University of Montreal, Montreal, Quebec, Canada Background. The gold standard for staging the local extension (T stage) and lymph node (LN) status (N stage) of esophageal cancer is endoscopic ultrasonography (EUS). When biopsy of the peritumoral LNs is performed using EUS, there is a risk of specimen contamination secondary to piercing the primary tumor; this shortcoming can be circumvented with endobronchial ultrasonography (EBUS). Moreover, EBUS allows for biopsy of LN stations not accessible with EUS. Methods. The study consisted of a prospective clinical trial. Fifty-two consecutive patients with potentially resectable esophageal cancer referred for endoscopic staging were prospectively enrolled. Radial and convex EUS followed by convex EBUS were performed during a single staging procedure. The LNs not accessible by EUS were biopsied using EBUS. Results of the EBUS procedure were compared to those of EUS in terms of the addition of staging information, upstaging, and confirmation of stage. Results. The combined EBUS-EUS procedure was performed in 42 patients. Ten patients were excluded. In all, 54 LNs were biopsied under EUS guidance and 48 LNs were biopsied under EBUS guidance. The EUS results were positive for metastatic esophageal cancer in 29 LNs (54%), and EBUS was positive in 10 LNs (21%). The addition of EBUS to EUS in the staging of esophageal cancer led to nodal and patient upstaging in 5 patients (12%) and confirmedtheeusstagewith additional negative or positive LN sampling in 29 patients (69%). Positive EBUS that led to upstaging (5 patients) changed the treatment plan from potentially resectable to palliative. There was no morbidity related to EBUS. Conclusions. A combined EBUS-EUS staging procedure improves precision in staging, leads to upstaging, and can change the treatment plan in patients with esophageal cancer. (Ann Thorac Surg 2013;96:232 8) Ó 2013 by The Society of Thoracic Surgeons The gold standard for evaluating the local extension of esophageal cancer is endoscopic ultrasonography (EUS). The tumoral extension into the esophageal wall (T stage) and lymph node (LN) status (N stage) are best assessed using EUS. Suspicious lymph nodes in the mediastinum and in lymph node basins drained by the tumor can be biopsied through the esophagus with fineneedle aspiration (FNA) under real-time EUS guidance to obtain pathologic confirmation [1 3]. Computed tomography (CT), positron emission tomography (PET) and PET-CT have low sensitivity and specificity for lymph node staging and are primarily used for the assessment of metastatic disease [4 6]. Accepted for publication March 7, Presented at the Forty-ninth Annual Meeting of The Society of Thoracic Surgeons, Los Angeles, CA, Jan 26 30, Address correspondence to Dr Liberman, CETOC, Division of Thoracic Surgery, Centre Hospitalier de l Universite de Montreal, 1560 rue Sherbrooke E, 8e CD Pavillon Lachapelle, Bureau D-8051, Montreal, Quebec H2L 4M1, Canada; moishe.liberman@umontreal.ca. Lymph node status is essential for preoperative esophageal cancer staging and aids the clinician in making decisions between surgery alone, neoadjuvant chemoradiotherapy followed by surgery, surgery followed by chemoradiation, or inoperability [7]. The most important decision-making information required when treating a patient with esophageal cancer is procuring an accurate preoperative clinical stage [8]. When biopsy of the peritumoral LNs is performed using EUS, there is a risk of specimen contamination secondary to piercing the primary tumor; this shortcoming can be circumvented with endobronchial ultrasound (EBUS). EBUS also allows for biopsy of LN stations not accessible with EUS. To date, no study has explored the utility of EBUS for LN biopsy in esophageal cancer patients. Patients and Methods The study consisted of a prospective clinical trial (clinicaltrials.gov NCT ). Consecutive patients with potentially resectable esophageal cancer (based on CT and PET-CT), referred for echoendoscopic staging at the Centre Hospitalier de l Universite de Montreal (CHUM) Ó 2013 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc

2 Ann Thorac Surg LIBERMAN ET AL 2013;96:232 8 EBUS ADDED TO EUS IN ESOPHAGEAL CANCER STAGING 233 Endoscopic Tracheobronchial and Oesophageal Center (C.E.T.O.C.), Division of Thoracic Surgery, University of Montreal, were prospectively enrolled. Potential resectability was determined by one of five general thoracic surgeons and a multidisciplinary foregut cancer tumor board. The study was approved by the Institutional Review Board at the Centre de Recherche du CHUM. Inclusion criteria included biopsy-proven esophageal cancer, endoluminal esophageal mass without previous biopsy, medical suitability for endoscopic procedure, and ability to consent. Exclusion criteria included patient receiving oral anticoagulant drugs including warfarin or clopidogrel with inability to stop medication for 5 days before procedure, anatomy precluding EBUS, and endobronchial tumor invasion. Aspirin was not a contraindication to inclusion. The EUS followed by EBUS was performed during a single staging procedure. The order of endoscopic procedures for all patients was as follows: (1) flexible esophagogastroduodenoscopy with endoluminal biopsy (in patients without prior tissue diagnosis); (2) radial EUS for T staging; (3) linear EUS for lymph node biopsy and N staging; (4) flexible bronchoscopy for evaluation of the left mainstem bronchus and trachea for tumor invasion; and (5) linear EBUS for transtracheal and transbronchial lymph node biopsy. Lymph nodes not accessible by EUS owing to tumor interposition were biopsied using EBUS. For patients in whom the tumoral stenosis was too severe to allow passage of the radial and linear EUS scopes, the EBUS scope was used in the esophagus to provide complete tumoral T staging as well as to assess the liver, adrenal glands, celiac axis. and gastrohepatic ligament lymph nodes [9]. All FNA specimens were evaluated by dedicated cytopathologists in the Department of Pathology at the Centre Hospitalier de l Universite de Montreal. Results were evaluated for (1) positivity; (2) negativity (no tumor cells, presence of normal lymphocytes); and (3) inadequacy (no tumor cells, no lymphocytes). A minimum of two passes were performed into each lymph node. All procedures were performed under local anesthesia with sedation or under general anesthesia. All procedures were performed under the supervision of a single thoracic surgeon (M.L.); however, the majority of procedures were performed by residents, fellows, and visiting endoscopists. All patients were followed up at 3 to 4 weeks after their procedure to evaluate procedure-related morbidity. Data were prospectively recorded during the staging procedure. Lymph node staging designation was based on the International Association for the Study of Lung Cancer mediastinal lymph node staging map [10]. Esophageal cancer staging was based on the American Joint Committee on Cancer (AJCC) staging manual, sixth edition [11]. Results of the EBUS procedure were compared with those of EUS in terms of the addition of staging information, upstaging and confirmation of stage. Univariate and multivariate analysis were performed using SPSS, version 16 (SPSS, Chicago, IL). Outcomes of interest included (1) ability of EBUS to sample lymph nodes not able to be reached by EUS secondary to the requirement to traverse tumor during EUS-FNA; (2) ability of EBUS to sample lymph nodes not attainable by EUS secondary to anatomic location in proximity to the airway and lack of proximity to the esophagus/stomach; and (3) change in treatment plan based on the addition of EBUS to EUS staging of esophageal cancer. Results Between January 2010 and September 2011 (21 months), 52 patients were enrolled into the trial. Ten patients were excluded (1, no tumor; 1, gastric cancer; 8, no EBUS). The combined EBUS-EUS procedure was performed in 42 consecutive patients. There were 8 patients enrolled in the trial who had esophageal cancer but did not undergo EBUS, owing to the endoscopists judgment at the time of echoendoscopic staging that the addition of EBUS to EUS in the clinical setting was unnecessary or superfluous. There were 11 female and 31 male patients. Mean age was 64 years (SD 10.6). Tumoral histopathology was adenocarcinoma in 27 patients (64%) and squamous cell carcinoma in 15 patients (36%). In all, 54 LNs were biopsied under EUS guidance and 48 LNs were biopsied under EBUS guidance. Lymph node sampling descriptive map is shown in Table 1. The EUS results were positive for metastatic esophageal cancer in 29 LNs (54%), and EBUS was positive in 10 LNs (21%). The addition of EBUS to EUS in the staging of esophageal cancer led to nodal and patient upstaging in 5 patients (12%; Table 2) and confirmed the EUS stage with additional negative or positive LN sampling in 29 patients (69%; Table 3). Positive EBUS leading to upstaging (5 patients) changed the treatment plan from potentially resectable to palliative. That change was due to finding malignant lymph nodes behind esophageal tumors or high nodular Barrett s in 2 patients and Table 1. Endobronchial Ultrasonography/Endoscopic Ultrasonography Lymph Node Sampling Descriptive Map Lymph Node Station EUS EBUS 2R 2 1 2L 3 4R 2 6 4L R 1 8L 10 9R 1 9L 1 10R 3 Celiac axis 6 Gastrohepatic ligament 7 Left adrenal gland 1 Total EBUS ¼ endobronchial ultrasonography; EUS ¼ endoscopic ultrasonography.

3 234 LIBERMAN ET AL Ann Thorac Surg EBUS ADDED TO EUS IN ESOPHAGEAL CANCER STAGING 2013;96:232 8 Table 2. Endobronchial Ultrasonography Upstaging in Patients Undergoing Combined EBUS/EUS Procedure Tumor Location (From Incisors) EUS LN Biopsy Site Positive EUS LN EUS Stage EBUS LN Biopsy Site Positive EBUS LN EBUS Stage cm GH ligament 1 IVA #7, 10R 2 IVB cm None 0 IIA #7 1 IVB cm 2R 1 III #7 1 IVB cm None 0 IIA 4R, #7 2 IVB cm Celiac axis, GH ligament 2 IVA #7 1 IVB EBUS ¼ endobronchial ultrasonography; EUS ¼ endoscopic ultrasonography; GH ¼ gastrohepatic; LN ¼ lymph node. lymph node sites not accessible by EUS because of EUS limitations or patient anatomy in 3 patients (lymph node sites: 10R, 4R, and #7). EBUS added to EUS led to upstaging in 5 patients, confirmed stage with additional lymph node biopsies not attainable using EUS alone in 29 patients, and did not add any additional staging information in 8 patients. The EBUS confirmed the clinical suspicion of lymph node positivity on PET and CT in 3 of the 5 patients who were upstaged with the addition of EBUS to EUS. In 2 of the EBUS upstaged patients, lymph nodes that changed the stage were either equivocal or nonpathologic by CT or PET criteria for malignancy. In cases where the tumoral stenosis was too severe to allow passage of the radial and linear EUS scopes, the EBUS scope was used in the esophagus to provide complete tumoral T staging as well as assess the liver, adrenal glands, celiac axis, and gastrohepatic ligament lymph nodes. EBUS utilized in the esophagus for T and N staging was required in 5 cases (11.9%) and was Table 3. Endobronchial Ultrasonography Confirmation of Stage in Patients Undergoing Combined EBUS-EUS Procedure Tumor Location a EUS LN Biopsied No. of EUS LN Biopsied Positive EUS LN EUS Stage EBUS LN Biopsied No. of EBUS LN Biopsied Positive EBUS LN EBUSþEUS Stage Lower Celiac 1 0 III Lung lesion, #7 2 0 III Lower 2L, 4L 2 2 IVB 2R 1 0 IVB Middle 0 0 IIA #7 1 0 IIA Lower #5 1 0 IIA #7 1 0 IIA Lower 8L 1 0 IIA #7 1 0 IIA Lower 8R, 4R 3 0 IIA #7 1 0 IIA Middle GHL 1 1 IVA #7 1 0 IVA Lower 0 0 IIA #7 1 0 IIA Lower GHL, #5 2 2 IVB #7 1 0 IVB Lower #5, 8L 2 1 III #7 1 0 III Lower 0 IIA #7 1 0 IIA Upper 8L, 2L 2 2 IVB 4R 1 1 IVB Lower #5, #7 2 2 IVB #7 1 0 IVB Lower #7, 4L 2 2 IVB 4R, #7 2 2 IVB Lower 0 IIA 10R 1 0 IIA Middle 0 IIA 4R 1 0 IIA Lower Celiac 1 1 IVA #7 1 0 IVA Middle 0 0 IIA #7 1 0 IIA Lower 8L 1 0 I #7, 4R 2 0 I Middle 9R 1 1 III #7 1 0 III Lower 0 IIA #7 1 0 IIA Middle Celiac 1 0 IIA #7 1 0 IIA Lower Celiac 1 1 IVA #7 1 0 IVA Middle Celiac, #5 2 0 III #7 1 0 III Lower #5, #7 2 0 IIA #7 1 0 IIA Lower 8L 1 1 III #7 1 0 III Lower GHL, 8L 2 0 I #7 1 0 I Lower #5, 8L 2 0 IIA #7 1 0 IIA Lower #5, 4L 2 0 IIA #7 1 0 IIA a Tumor location: upper <24 cm, middle 24 cm to 32 cm, lower >32 cm. EBUS ¼ endobronchial ultrasonography; EUS ¼ endoscopic ultrasonography; GHL ¼ gastrohepatic ligament; LN ¼ lymph node.

4 Ann Thorac Surg LIBERMAN ET AL 2013;96:232 8 EBUS ADDED TO EUS IN ESOPHAGEAL CANCER STAGING 235 successful at passing through all stenotic tumors in this study. No tumor required dilation before EBUS in the esophagus. Therefore, EBUS allowed for complete T assessment in 5 patients who would have otherwise not have been able to have their tumor completely T staged. There was no morbidity related to EBUS, either in the perioperative period or at 30-day follow-up. All procedures were performed under local anesthesia with moderate sedation or under general anesthesia. Procedural times ranged between 14 minutes and 113 minutes; median procedural time was 35 minutes. EBUS added an average of 9.2 minutes (SD 10.0) to the total endoscopy procedure time. Median length of stay after the procedure was 0 days (SD 1.8). Reasons for hospitalization after endoscopy included concomitant esophageal stenting (7 patients), hospitalization before the procedure for general status (1 patient), and pneumonia (1 patient). Comment Progression of disease from primary site to lymph nodes follows lymphatic drainage. Any lymph node surrounding the esophagus can be affected; that is especially true in esophageal cancer because the lymphatic drainage of the esophagus is not necessarily segmental and predictable, but more longitudinal with skip metastasis [12]. More often than not lymph nodes surrounding the primary tumor are affected by metastatic spread and biopsy is necessary to obtain a tissue diagnosis. Even though several studies have demonstrated the precision of EUS plus FNA for N stage (with surgical control of the pathological specimen), when biopsy of the peritumoral lymph nodes is performed, there is always a risk of contamination of the sample secondary to piercing the primary tumor. Therefore, when the lymph nodes which require sampling require biopsy through the tumor, no FNA can be performed. That makes accurate pathological N staging impossible. Some investigators have suggested that the risk of false positive biopsy is minimized if one avoids both placing the needle through the primary tumor and avoiding the removal of the stylet from the aspiration needle until the lymph node has been penetrated [3]. These researchers also mention that the cytopathologist should also be able to identify lymphoid tissue in the sample [1]. Nevertheless, the risk of false positivity is not abolished when biopsy of these nodes is performed after passing through the tumor, and this has a great impact on the choice of therapy. The only way to be absolutely sure of the lack of contamination of the lymph node sample is to find a method of biopsy of the lymph node without traversing the primary tumor. Lymph nodes hidden by the tumor on EUS that were able to be biopsied by EBUS were positive for malignant cells in 2 of 5 patients who were upstaged in this study. There were 7 patients who had confirmation of stage by EBUS after peritumoral lymph nodes seen on EUS were biopsied using EBUS without requiring transtumoral lymph node biopsy. Endoscopic ultrasonography techniques are becoming more and more popular [13 16]. The safety of these technologies has been demonstrated on numerous occasions. These minimally invasive transluminal techniques do not require general anesthesia and can be performed rapidly, accurately, and at low cost [17 19]. Endobronchial ultrasonography has been used for many years by pulmonologists and thoracic surgeons primarily for the management of lung cancer [13 16]. Its efficacy and safety has been demonstrated [17 19]. The combined EBUS- EUS procedure has the ability to access all mediastinal lymph node stations during a single endoscopic staging procedure [20]. EBUS is not currently utilized by most centers in the staging of esophageal cancer. To date, no study has explored the possibility of the use of EBUS neither for peritumoral lymph node biopsy in esophageal cancer patients nor for any suspicious mediastinal lymph nodes not attainable owing to anatomic constraints of the stomach and esophagus during EUS. At Notre Dame Hospital, Centre Hospitalier de L Universite de Montreal, all patients diagnosed with esophageal cancer are scheduled for a combined EUS-EBUS procedure to examine all the lymph nodes near or far from the primary tumor amenable to EBUS-guided transbronchial biopsy. However, the decision to perform EUS alone or EUS and EBUS is left up to the echoendoscopist s clinical judgment at the time of the procedure. That is important because it allows physicians to bypass the primary tumor and pass through the normal (noncancerous) bronchial wall, eliminating any chance of tumoral contamination. In addition, this also provides a much more complete mediastinal staging as it allows biopsy of a population of lymph nodes inaccessible to EUS biopsy. Adding EBUS to EUS for staging esophageal cancer in this study added the most valuable information for patients with upper and mid esophageal tumors owing to the inability to assess local lymph node invasion in the peritracheal area and subcarinal region. However, in certain cases of distal (gastroesophageal junction) tumors, EBUS did add valuable additional information secondary to anteriorly positioned subcarinal lymph nodes not attainable by EUS or right lower paratracheal (station 4R) lymph nodes, which are often difficult to visualize echographically with EUS because of air interposition from the trachea. Twelve percent of patients were upstaged with the addition of EBUS to EUS in this study, and that has an important impact on patient prognostication and treatment planning. Changes in stage also impact on cost of care as upstaging in these patients changed the treatment plan from a definitive multimodality treatment strategy including surgical resection to definitive chemoradiation therapy. EBUS used in the esophagus to bypass stenotic tumors and provide a complete EUS T staging of the tumor as well as assessment of upper abdominal organs and lymph nodes has been described by Buxaum and Eloubeidi [9]. This technique has been used for assessment of tight esophageal strictures and tumors in our center since In this study, the technique allowed for complete EUS (using an EBUS scope) in 5 patients. Without EBUS used in the esophagus, 12% of patients in this

5 236 LIBERMAN ET AL Ann Thorac Surg EBUS ADDED TO EUS IN ESOPHAGEAL CANCER STAGING 2013;96:232 8 study would have had incomplete EUS staging owing to the inability to pass through the tumor. The addition of EBUS to EUS in this study added approximately 9 minutes to the standard esophagoscopy and EUS procedure. There was no additional morbidity with the addition of EBUS to EUS. The cost of procedure was certainly increased, as EBUS requires additional endoscopy, disposable equipment, cytology preparation, and interpretation. Costs were not assessed in the current trial. We do not advocate that EBUS be routinely added to EUS for all patients; however, it has a definite benefit when used selectively, especially in upper and middle third esophageal tumors. Furthermore, if it is possible to prevent surgical resection in patients who would have otherwise been understaged using EUS alone and not benefited from surgical resection, cost savings would certainly result. In the current study, the AJCC sixth edition of the esophageal staging system, as opposed to the current seventh edition, was used to describe the preoperative stage of patients based on both EUS and EBUS. This was purposeful as it is very difficult to preoperatively stage patients with esophageal cancer using the seventh edition of the AJCC esophageal cancer staging system. We, similarly to all thoracic surgeons in the Western World, utilize the new staging system in the postresectional setting because it is superior to the sixth edition staging system in terms of prognostication. However, the N status in the seventh edition is based on the number and not location of positive lymph nodes. It is impossible to use EUS, EBUS, CT, or PET to count positive lymph nodes. Lymph node stations often contain multiple nodes, and these nodes often meet CT, PET, or EUS criteria for malignancy yet upon biopsy are nonpathologic. We therefore find it difficult to precisely N stage patients in the preoperative setting (with EUS with or without EBUS) using the new staging system. In conclusion, a combined EBUS-EUS staging procedure improves precision in staging, leads to upstaging, and can change the treatment plan in patients with esophageal cancer. EBUS added to EUS should be used selectively for patients for whom complete T and N staging is not possible using EUS alone. Funding for this study was provided by grants from the Canadian Foundation for Innovation, the Marcel and Rolande Gosselin Chair in Thoracic Surgical Oncology, and the Thoracic Surgery Research Foundation of Montreal. References 1. Saltzman JR. Endoscopic and other staging techniques. Semin Thorac Surg Cardiovasc Surg 2003;15: Lightdale CJ, Kulkarni KG. Role of endoscopic ultrasonography in the staging and follow-up of esophageal cancer. J Clin Oncol 2005;23: Puli SR, Reddy JBK, Bechtold ML, Antillon D, Ibdah JA, Antillon MR. Staging accuracy of esophageal cancer by endoscopic ultrasound: a meta-analysis and systematic review. World J Gastroenterol 2008;14: Erasmus JJ, Munden RF. The role of integrated computed tomography positron-emission tomography in esophageal cancer: staging and assessment of therapeutic response. Semin Radiat Oncol 2006;17: Monden RF, Macapinlac HA, Erasmus JJ. Esophageal cancer: the role of integrated CT-PET in initial staging and response assessment after preoperative therapy. J Thorac Imaging 2006;21: Wong WL, Chambers RJ. Role of PET/PET CT in the staging and restaging of thoracic oesophageal cancer and gastrooesophageal cancer: a literature review. Abdom Imaging 2008;33: Vazquez-Sequeiros E, Wiersema MJ, Clain JE, et al. Impact of lymph node staging on therapy of esophageal carcinoma. Gastroenterology 2003;125: Townsend CM, Beauchamp RD, Evers BM, Mattox KL, eds. Sabiston textbook of surgery: the biological basis of modern surgical practice. 18th ed. Philadelphia: ElsevierSaunders; Buxaum JL, Eloubeidi M. Transgastric endoscopic ultrasound (EUS) guided fine needle aspiration (FNA) in patients with esophageal narrowing using the ultrasonic bronchovideoscope. Dis Esophagus 2011;24: Rusch VW, Asamura H, Watanabe Hea. The IASLC lung cancer staging project: a proposal for a new international lymph node map in the forthcoming seventh edition of the TNM classification for lung cancer. J Thorac Oncol 2009;4: American Joint Committee on Cancer. AJCC cancer staging manual. 6th ed. Philadelphia: Lippincott-Raven; Attila T, Faigel DO. Role of endoscopic ultrasound in superficial esophageal cancer. Dis Esophagus 2009;22: Annema JT, Rabe KF. State of the art lecture: EUS and EBUS in pulmonary medicine. Endoscopy 2006;(Suppl 1): Becker H, Herth F. Is endobronchial ultrasound indispensable in clinical practice? Pro: endobronchial ultrasound. J Bronchol 2002;9: Janes SM, Spiro SG. Esophageal endoscopic ultrasound/ endobronchial ultrasound-guided fine needle aspiration: a new dawn for the respiratory physician? Am J Respir Crit Care Med 2007;175: Herth FJ, Rabe KF, Gasparini S, Annema JT. Transbronchial and transoesophageal (ultrasound-guided) needle aspirations for the analysis of mediastinal lesions. Eur Respir J 2006;28: Aabakken L, Silvestri GA, Hawes R, Reed CE, Marsi V, Hoffman B. Cost-efficacy of endoscopic ultrasonography with fine-needle aspiration vs. mediastinotomy in patients with lung cancer and suspected mediastinal adenopathy. Endoscopy 1999;31: Manaker S, Ernst A, Marcus L. Affording endobronchial ultrasound. Chest 2008;133: Ernst A, Gangadharan SP. A good case for a declining role for mediastinoscopy just got better. Am J Respir Crit Care Med 2008;177: Liberman M, Duranceau A, Grunenwald E, Thiffault V, Khereba M, Ferraro P. Initial experience with a new technique using EUS access for biopsy of para-aortic (station #6) mediastinal lymph nodes without traversing the aorta. J Thorac Cardiovasc Surg 2012;144: DISCUSSION DR K. ROBERT SHEN (Rochester, MN): Moishe, all 5 of the patients who were upstaged by the addition of endobronchial ultrasonography (EBUS) to the endoscopic ultrasonography (EUS) had a subcarinal lymph node that was positive, and

6 Ann Thorac Surg LIBERMAN ET AL 2013;96:232 8 EBUS ADDED TO EUS IN ESOPHAGEAL CANCER STAGING 237 I would think anatomically that would be a station as being accessible with EUS. It sounded like at least for 1 of them there was peritumoral involvement so you didn t want to cross the tumor. What was the situation for the other 4? DR LIBERMAN: That is a good question. Not all of them were upstaged based on the station #7; I think 3 of the 5. One of them was tumor; one was a nodular Barrett s extending very high up in the esophagus and we were worried about getting a false positivity not knowing whether there was carcinoma in that specimen; and the other one was a very anterior #7, which we biopsied the #7 on EUS and it was negative, but then we did that anterior #7 by EBUS, that was the one that was positron emission tomography (PET) positive and it was positive on histologic analysis. So definitely, the hilar nodes are more important lymph nodes, nodes that are in locations that are more difficult to get, and, of course, EUS or EBUS is probably much more important in the proximal tumors than in the ones that are very distal or in the cardia extending to the esophagus. DR STEVEN DEMEESTER (Los Angeles, CA): Very nice study, intriguing use of the technology. I wondered if you could break down whether it was as valuable for adenocarcinomas, which are typically located distally and are less likely to have involved paratracheal and subcarinal nodes, as it was for squamous cancers? And then just a comment. It s useful to use the new staging system where lymph nodes along the gastrohepatic ligament or even celiac axis are not stage IV disease. DR LIBERMAN: Thank you very much for those questions and comments. Yes, I definitely agree that it s definitely a useful staging system and we just tried to keep it as tight as we could with the staging system that we chose because, again, it s very hard for us as echoendoscopists, or people doing EUS staging for esophageal cancer, to use purely the new staging system even though we appreciate it is a better staging system in the postoperative setting. In terms of adeno versus squamous, again it s a small study and the power really isn t there to differentiate. But definitely proximal tumors are much more important, or I think EBUS is much more important, especially for N1 staging in the proximal esophagus because of the risk of false positivity when passing through tumor to biopsy peri-tumoral N1 nodes, if you believe that it is important, which we do. Adeno versus squamous, we don t have enough numbers to really tell you, but I think, yes; however, I cannot definitively say that based on our numbers. DR DANIEL BOFFA (New Haven, CT): I just want to clarify, at your institution subcarinal lymph nodes would constitute somebody becoming unresectable? DR LIBERMAN: No, a subcarinal lymph node would not. However, using the staging system, someone in the old staging system who has a gastroesophageal (GE) junction tumor, celiac axis lymph nodes are considered IVa and any other lymph node out of that area would be IVb. The treatment is left up to the treating surgeon. And we often will include those patients in our neoadjuvant chemoradiation. DR BOFFA: So how many patients actually had their treatment changed? That s what I m a little confused about. DR LIBERMAN: In this study, again, there were only 5 patients who changed their stage based on the study. And I can t tell you the exact number, but I think 3 of those patients were made, especially the 1 with the hilar node, the #7, and the very high tumor, and 1 of the other patients did go on to nonoperative treatment, the other 2 were operated on. But it definitely changed our radiation field, and we often outline the radiation doses with our radiation oncologists based on what lymph nodes are positive on EBUS-EUS. DR BOFFA: And did you look at hilar nodes on every person you did the EBUS or only if imaging suggested there was something going on? DR LIBERMAN: In general we only looked at them if there was a suggestion on imaging. However, when we re staging lung cancer we re much more systematic. It s quite rare, you know very well, to have hilar lymph node positivity in a patient with esophageal cancer without other pleurimetastatic lymph nodes. DR GAIL DARLING (Toronto, Ontario, Canada): Moishe, nice work as always. My question relates, again, to the change in treatment that you re going to offer these patients. And how often did your EBUS or EUS staging change the radiation field from the PET scan, for example? DR LIBERMAN: That s a good question. So again, only 3 of those patients had a change in their treatment. Those 3 patients were made nonoperable. So they got palliative chemoradiation. One of those patients just actually got an esophageal stent and nothing else. And so for the 2 other patients, it did change how we planned our radiation. Whether we would have planned the same way on PET scan and CT alone, it s very possible that we would have done the same radiation planning. But again, I think by doing more accurate staging and being precise and knowing that that node is positive, you may better dose your radiation. And if you know that node is negative as opposed to positive, you may give the patient a lesser radiation dose especially when the tumor is very, very low. DR DARLING: And knowing that most T3 tumors are going to have positive nodes and a large percentage of T2 tumors will have positive nodes, how often would your biopsies have changed, your treatment algorithm for example if you just went on the T stage you would have treated them with induction therapy. So how often was your invasive staging changing that treatment algorithm? DR LIBERMAN: Very rare. I think that we all agree that anyone with a T3 and above should get neoadjuvant, whatever. My feeling is they should get neoadjuvant chemoradiation. And most of those patients are either N0 or N1 in the majority of cases. So it doesn t change our treatment plan that much in that way. However, I find that EUS or EUS plus EBUS is more to exclude patients from curative intent treatment. And looking at the celiac axis, gastrohepatic ligament, even a GE junction tumor with celiac axis, multiple bulky metastatic nodes at the celiac axis, which are proven positive on EUS, in general we don t operate on those patients in our center. So I think in the opposite to your question, I think it changes more when you upstage them, but confirmation of stage is not as important. We mentioned the confirmational stage, but it s really not that important for us. DR DARLING: My last question is, when you do a radial EUS and you see these round nodes as opposed to flat nodes they are almost always positive I don t ever recall finding negative ones. I think it is different than lung cancer. When you do an EBUS in lung cancer or you do the CT scan, big nodes may be negative, right? But in esophageal cancer, if you see those nodes, like they re almost always positive. I don t recall ever finding

7 238 LIBERMAN ET AL Ann Thorac Surg EBUS ADDED TO EUS IN ESOPHAGEAL CANCER STAGING 2013;96:232 8 negative ones. And certainly if you saw bulky celiac lymph nodes, you re not going to believe it if it came back negative. DR LIBERMAN: True. It s the opposite. It s for the patients with the small nodes that end up being positive, especially when they re not peritumoral. Because I don t want to operate on a patient and find out 6 months later on a GE junction tumor that there is a positive 2R that was non-bulky that I saw on EBUS and I biopsied, and it would have been positive at that time. We have seen that multiple times in our center. There is a reason that esophageal cancer has a very, very poor prognosis and some of that is due to the disease and some of that is because we are very poor stagers and we re deciding to operate sometimes, not always, on the wrong patients. I think the more we can be precise with our staging, the more patients we can operate upon who actually derive a benefit from our operation and our chemoradiation, the better we serve these patients. And so I think that the better we can stage these patients, the better treatment plan we can have for them. DR SCOTT SWANSON (Boston, MA): Quick question. A great talk. Have you done any EBUS or EUS after neoadjuvant to look at nodes; and if you have, is it more difficult or are there tricks to that? DR LIBERMAN: Thank you, great question. And it s a topic of much controversy in the EUS and EBUS literature. We have done it on occasion, but we don t like to do it. For the T staging it s extremely unreliable even with a radial EUS scope and even with a very experienced operator. There are so much sclerotic changes, there is so much liquid, that T staging is extremely inaccurate. For N staging, I would say it s pretty accurate because even though, yes; big nodes, rounds nodes, dark nodes are more likely to be malignant, we don t base any decision on what nodes look like and we get tissue on every node. So if you re worried in a patient who had neoadjuvant chemoradiation, in my practice we always do a CT-PET scan before going on to operation to look for distant metastatic disease that may have popped up during the chemoradiation. If we see something like an adrenal gland or a node way outside the field, let s say it s a GE junction tumor and there is a 2L node that I know I m not going to be able to get, we ll go back and do a post-neoadjuvant EUS. And if it s positive, it s positive, because cancer cells are cancer. DR SWANSON: My question was really, is it hard to needle aspirate a node that s been radiated in your experience? DR LIBERMAN: No, not for us, no. DR SHEN: Moishe, does the location of the tumor, if it s ange junction versus middle or upper third, make any difference in your mind in terms of how useful you think that the technique is, or in those cases you particularly find it to be in addition to just doing one versus the other? DR LIBERMAN: I think that s a great question. And definitely the proximal and mid tumors there is a lot more benefit in doing your EBUS with your EUS. We use it much more selectively in the GE junction or cardial tumors invading the esophagus, but we re hyperactive stagers and we like to get all the nodes we can. But definitely the bang for your buck is much higher in the proximal and middle third tumors.