Lymph node invasion might have more prognostic impact than R status in advanced esophageal adenocarcinoma

The American Journal of Surgery (2013) 205, 711-717 Clinical Surgery Lymph node invasion might have more prognostic impact than R status in advanced esophageal adenocarcinoma Magali Cabau, M.D. a, Guillaume Luc, M.D. a, Eric Terrebonne, M.D. b, Geneviève Belleanne, M.D. c,véronique Vendrely, M.D. b, Antonio Sa Cunha, M.D. a, Denis Collet, M.D., Ph.D. a, * a Departments of Digestive Surgery, b Digestive Oncology, and c Pathology, University Hospital of Bordeaux, Avenue de Magellan, 33604 Pessac Cedex, France KEYWORDS: Esophagus; Cancer; Adenocarcinoma; Lymph node invasion; Radicality; Prognosis Abstract BACKGROUND: Advanced esophageal adenocarcinomas are associated with 5-year survival rates ranging from 14% to 35%. Nodal status and tumor clearance are the main prognostic factors. However, their respective prognostic values have not been compared to date. METHODS: Seventy consecutive patients with stage T3 adenocarcinomas of the esophagus or gastric cardia were retrospectively assessed. Neoadjuvant therapy was indicated in all cases. Prognostic values of R0 resection and nodal status were evaluated using univariate and multivariate analyses. RESULTS: Neoadjuvant therapy was achieved in 62 patients, 41 with radiochemotherapy and 21 with perioperative chemotherapy. Transthoracic esophagectomy and transhiatal esophagectomy were performed in 54 and 15 patients, respectively. Clavien-Dindo grade III or IV complications occurred in 16 patients (23%). Two patients died in the hospital (3%). In univariate and multivariate analyses, nodal status was the main independent factor predicting overall survival; tumor clearance (R0 or R1) had less prognostic impact and was not statistically significant. Furthermore, R1 resection was a prognostic indicator for metastatic recurrence. CONCLUSIONS: These results indicate that nodal status has more prognostic impact than R status in stage T3 adenocarcinomas of the esophagus or gastric cardia. Thus, local control in R1 patients by postoperative radiotherapy is not justified. Ó 2013 Elsevier Inc. All rights reserved. Esophageal carcinoma is the 6th most common cause of tumor-related death worldwide. The prognosis of this disease is poor, with a 5-year overall survival rate of 10%. 1 The 5-year survival rate increases to 25% for resectable tumors. 2 The incidence of adenocarcinomas of the esophagus has grown dramatically over the past 20 years. 3 6 In The authors declare no conflicts of interest. * Corresponding author. Tel.: 133-05-57-65-60-04; fax: 11-33-5-57- 65-60-27. E-mail address: denis.collet@chu-bordeaux.fr Manuscript received January 30, 2012; revised manuscript July 17, 2012 France, adenocarcinomas account for 30% of esophageal carcinomas. 7 The main risk factors are gastroesophageal reflux disease and Barrett s esophagus. In contrast, the incidence of squamous cell carcinoma (SCC) is decreasing because of reductions in the consumption of alcohol and tobacco. Adenocarcinoma and SCC have similar prognoses. However, SCC arises frequently in debilitated patients with associated benign or malignant pathologies induced by tobacco and/or alcohol. These comorbidities can affect short-term and long-term treatment outcomes for esophageal carcinoma. 0002-9610/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2012.08.009

712 The American Journal of Surgery, Vol 205, No 6, June 2013 Broadly, there are 2 types of adenocarcinomas, which are differentiated by the circumstances of diagnosis. Some tumors are detected at an early or very early stage during follow-up for Barrett s esophagus in patients who have no symptoms other than gastroesophageal reflux disease. For those patients with stage I tumors, the 5-year survival rate ranges from 70% to 85%. 8 10 Therefore, the majority of patients with this type of disease can be cured. In this patient population, clinicians are faced with the challenge of achieving a curative effect with minimal consequences. At present, these patients are treated with endoscopic or minimally invasive interventions. In contrast, most malignancies are still diagnosed as a result of dysphagia, which is present in 75% of patients, 4 or symptoms of local involvement. These are cases of locally advanced cancer. In these cases, the challenge is to determine the best therapeutic combination to achieve prolonged survival or even a curative effect while taking into account the patient s ability to undergo aggressive treatment. For nonmetastatic, locally advanced adenocarcinomas, the current standard of care is controversial, and the role of neoadjuvant chemoradiation is not well established. 11 13 However, neoadjuvant chemoradiotherapy is usually recommended 14 and can result in a complete pathologic response in 20% of patients. 3,15 A recent meta-analysis provides strong evidence for neoadjuvant chemotherapy or chemoradiotherapy for locally advanced esophageal cancer. 16 The prognostic roles of number and ratio of positive nodes in esophageal cancer have already been established by Mariette et al. 17 However, the importance of R status compared with nodal involvement in locally advanced adenocarcinoma has not been studied to date. The aim of this study was to compare the respective prognostic values of lymph node involvement and R status in patients who underwent surgery for advanced adenocarcinoma of the esophagus or gastric cardia. Methods From January 1, 2000, to September 30, 2010, 165 patients underwent surgery at our institution for esophageal malignancies, and 123 patients had adenocarcinomas of the cardia or esophagus. In this study, we focused on 70 consecutive patients with T3N0 or N1M0 tumors (the 6th and 7th Union for International Cancer Control/TNM classification) according to the pretherapeutic workup who eventually underwent surgery. For all patients, staging investigations included upper endoscopy with biopsies, computed tomography (CT), and endoscopic ultrasound whenever there was no esophageal stenosis. Combined positron emission tomography and CT was routinely available beginning in 2007, and it has been systematically performed since that time. For all patients, neoadjuvant treatment was planned and consisted of 5-fluorouracil and cisplatin with concomitant radiotherapy or chemotherapy alone. The chemotherapy regimen consisted of a 1-hour infusion of cisplatin 50 mg/m 2 on day 1, preceded and followed by hydration, and a 2-hour infusion of leucovorin 400 mg/m 2, followed by a 400 mg/m 2 bolus of 5-leucovorin, followed by a 46-hour continuous infusion of 2,400 mg/m 2 of 5- leucovorin. All patients were given prophylactic antiemetic treatment consisting of 5-hyroxytryptamine-3 antagonists and prednisolone. A central venous catheter was used for all patients. Treatment was repeated every 2 weeks if the neutrophil count was.1500/mm 3, the platelet count.10,000/mm 3, and serum creatinine,120 mmol/l. Patients received 3 preoperative cycles and 3 postoperative cycles. If radiochemotherapy was performed, radiation therapy consisted of 45 Gy, 1.8 Gy/fraction and per day, divided into 5 fractions per week. External-beam radiotherapy was performed with 18-MV photons from a linear accelerator. According to the International Commission on Radiation Units and Measurements 50 guidelines, target and avoidance volumes were delineated on planning CT. The treated volume included the primary tumor with a 5-cm longitudinal margin, involved lymph nodes, and celiac lymph nodes; when available, data from positron emission tomography were integrated into planning CT. Three cycles of concomitant chemotherapy were delivered using cisplatin and 5-fluorouracil during the radiation period of treatment. Surgery was performed 4 to 6 weeks after the completion of neoadjuvant therapy. Patients who could not receive combined therapy because of contraindications, either cardiovascular or respiratory, were directly referred to the surgeon. The surgical approach was governed by the general status of the patient. The standard operation was an Ivor- Lewis esophagectomy with an abdominal right thoracotomy associated with a standardized 2-field lymph node dissection. For patients with poor general status, a transhiatal esophagectomy with a more limited lymph node dissection, limited to the lower mediastinum, was preferred. 18 Whatever the surgical technique, patients received physiotherapy with immunonutrition supplementation for 8 days before surgery. The specimen was prepared by the surgeon at the end of the procedure as follows: nodes were separated from the esophagus, and the esophagus was longitudinally opened along the side opposite the lesion by the surgeon and then pinned and sent fresh to the pathologist. Gross dissection was performed after 24 to 48 hours of fixation in 10% neutral buffered formalin. After a succinct but precise gross description (localization and size of the lesion, distance from the edge of the tumor to the surgical margins), India ink was applied on the outer surface of the specimen to allow for precise evaluation of the lateral margin. Next, the tumor was carefully sampled (R3 cassettes), and all of the lymph nodes were submitted in their entirety. The histologic report was done using a standardized report on the basis of the World Health Organization and 7th TNM classifications. The operation was categorized as an R1 resection if the tumor margin was,1 mm. 19 No

M. Cabau et al. Impact of lymph node invasion vs radicality 713 retrospective review of the specimen was performed for the purpose of this study. Patients were followed for R1 year by a clinical evaluation. Thoracoabdominal CT and upper endoscopy were performed every 6 months for the 1st 2 years and then every year for the 3 following years. Follow-up was complete for every patient. Recurrence was defined as any changes on CT, and it was confirmed if possible by biopsy or positron emission tomography. Recurrent disease was defined as either locoregional recurrence, including mediastinal lymph nodes, or distant metastasis. Statistical analysis Data were prospectively collected on a specific database and were updated at each follow-up. For the purposes of this study, survival status was ascertained in December 2010. Survival was estimated using the actuarial method, including postoperative deaths. The log-rank test was used for comparison of survival curves. Prognostic factors were analyzed by Cox proportional-hazards regression analysis. The confidence interval was 95%. Statistical analysis was performed using SAS version 9.2 (SAS Institute Inc, Cary, NC). Results Demographics The ratio of men to women was.67, with a median age of 66.8 years (range, 41 82 years) (Table 1). Tumors were located in the middle 3rd of the esophagus in 3 cases (4.3%), and they were classified as Siewert type I tumors in 55 patients (78.5%) and Siewert type II tumors in 11 patients (15.7%). One patient had an extended tumor from the lower 3rd of the esophagus to the small curvature of the stomach. Table 1 Patient characteristics Variable Value Men/women 47/23 Age (y), median (range) 66.8 (41 82) Location Siewert type I 55 (78%) Siewert type II 11 (16%) Middle 3rd 3 (4%) Extended tumor 1 (1%) Neoadjuvant treatment Chemotherapy 21 (30%) Radiochemotherapy 41 (59%) None 8 (11%) Surgery Lewis-Santy 54 (77%) Transhiatal esophagectomy 15 (21.5%) Esophagogastrectomy 1 (1.5%) Neoadjuvant treatment and surgery were completed in 62 patients. A total of 41 patients received radiochemotherapy, and 21 patients received perioperative chemotherapy. Eight patients did not receive any preoperative treatment; therefore, results are expressed as intention to treat because they were judged unable to undergo combined therapy. Fifty-four patients (77%) underwent Ivor-Lewis procedures, and 15 (23%) underwent transhiatal esophagectomy. The patient with an extended tumor underwent esophagogastrectomy with an intrathoracic esophagojejunal anastomosis. Postoperative mortality and morbidity Major complications occurred in 16 patients (morbidity, 22%), which were grade III (6 patients) or IV (10 patients) according to the Clavien-Dindo classification. 20 Reoperation was required in 10 patients (3 fistulas, 3 chylothoraces, 3 pleural effusions, and 1 leakage of the jejunostomy). Two patients died in the hospital (mortality rate, 3%), 1 from necrosis of the esophageal substitute and 1 from pneumonia. Histopathology Pathologic results are reported in Table 2. Resection was macroscopically complete in all patients and microscopically complete in 57 patients. The median number of lymph nodes retrieved was 18 (range, 4 53). Nineteen patients had.4 positive nodes, and 18 patients had.20% positive nodes. Table 2 Pathology Variable Value Median number of retrieved nodes 18 Tumor clearance R0 85 (85%) R1 16 (16%) T stage pt0 13 (13%) pt1 11 (11%) pt2 11 (11%) pt3 52 (52%) N stage N0 47 (47%) N1 39 (39%) N2 12 (12%) N3 3 (3%).4 positive nodes 30 (30%) R20% positive nodes 30 (30%),20% of positive nodes 71 (71%) Vascular or neural invasion Not available 43 (43%) Absent 22 (22%) Nerve 13 (13%) Vascular 11 (11%) Vascular and nerve 12 (12%)

714 The American Journal of Surgery, Vol 205, No 6, June 2013 Vascular or nervous invasion was systematically evaluated beginning in 2007 and was available for only 50 patients (72%). Long-term results The overall survival rates at 1, 3, and 5 years were 77%, 48%, and 35%, respectively, with a median overall survival of 34 months (Fig. 1). Disease-free survival rates at 1, 3, and 5 years were 73%, 44%, and 34%, respectively, with a median of 29.70 months (95% confidence interval, 16.40 43.00 Months. Recurrence was observed in 24 patients; 13 patients had local recurrences, and 11 had distant metastases. The median time to recurrence was 9 months from operation; the median times to recurrence for subgroups were 9.2 months in patients with R1 resections, 8.7 months in patients with 20% positive nodes, and 7.6 months in patients with 4 positive nodes. Among 13 patients with R1 resections, tumor recurrence was locoregional in 2 and distant metastases in 5, whereas 6 patients had not experienced recurrences at the time of last data collection. Among the 8 patients who underwent only surgery, 1 died in the early postoperative course, 5 had recurrences that were locoregional in 2 and distant in 3, and 2 were free of disease at the time of data collection. Univariate and multivariate analyses Univariate analyses of factors influencing median overall survival in this study were calculated for overall and disease-free survival (Table 3). The univariate analyses identified common factors (eg, age and gender) and the following histologic parameters as risk factors for death: the number of lymph nodes positive for invasion and the ratio of 20% nodal invasion, as well as the degree of vascular and nervous invasion at the resection limits (Table 3). In multivariate analysis (Table 4), the only independent factors for overall survival were the ratio of positive nodes Table 3 Univariate analysis Prognostic variable n Relative risk (relative risk, 2.40; 95% confidence interval, 1.20 4.82; P 5.01) and number positive nodes.4 (relative risk, 2.88; 95% confidence interval, 1.38 6.01; P 5.005). In this model, neither R status (R0 vs R1) nor vascular or nervous invasion was a statistically significant prognostic indicator compared with nodal status. For disease-free survival, none of these factors reached significance. Overall survival rates categorized by positive lymph node ratio and R status are shown in Figs. 2 and 3. Comments 95% confidence interval Overall survival Age 70 1.01.98 to 1.05.53 Gender 70 2.34.56 to 9.77.24.4 positive nodes 70 2.95 1.42 to 6.13.004 R status (R0/R1) 70 1.81.81 to 4.02.14 Positive node 70 2.54 1.28 to 5.05.008 Vascular or nervous invasion Presence vs absence 50.59.26 to 1.35.21 Disease-free survival Age 70 1.04 26.93 to 2.11.29 Gender 70 2.99.67 to 29.43.08.4 positive nodes 70 1.94.61 to 6.15.20 R status (R0/R1) 70.70.20 to 2.44.79 Positive node 70 2.59.71 to 5.97.15 Vascular or nervous invasion Presence vs absence 70 1.58.60 to 4.16.24 Advanced esophageal adenocarcinoma still has a poor prognosis, with 5-year survival rates ranging from 14% to 40%. 8 10,21 24 This rate is often estimated to be about 20% P Figure 1 Overall actuarial survival in intention-to-treat analysis of patients with stage T3 adenocarcinomas treated with neoadjuvant chemotherapy or chemoradiotherapy followed by surgery.

M. Cabau et al. Impact of lymph node invasion vs radicality 715 Table 4 Multivariate analysis Prognostic variable n Relative risk 95% confidence interval P Overall survival.4 positive nodes 70 2.88 1.38 6.01.005 Positive node 70 2.40 1.20 4.82.01 R status (R0/R1) 70 1.70.76 3.80.19 Disease-free survival.4 positive nodes 70 1.20.18 7.64.84 Positive node 70 1.93.35 10.62.44 R status (R0/R1) 70 1.63.43 6.23.46 in most published series. In our study, we observed an overall survival rate of 35% in intention to treat, which is among the best published data. In most series, SCC and adenocarcinomas are studied together and have equivalent prognoses. It is unlikely that our results are due to the design of the study, which was specifically focused on adenocarcinomas. Our study focused on a group of patients with locally advanced adenocarcinoma of the esophagus or the esophagogastric junction. Pooling these 2 types of tumors is logical because they have the same prognosis and require the same treatment, except for Siewert type III tumors, 25 which are considered to be gastric cancers. Therefore, this series was very homogenous despite its small size. The use of neoadjuvant therapy is widely accepted, though there is a lack of clear data confirming its efficacy, and this approach is currently the standard of care in France. The specific role of radiotherapy is still controversial and tends to be restricted to stage T3 or T4 tumors 6 to increase the rate of R0 resections. As a result, the neoadjuvant therapy in this series consists of both chemotherapy and radiochemotherapy. It is a basic oncologic principle that positive margins are associated with a high risk for recurrence, and usually, R status is considered to be a major prognostic factor. This criterion is significant whenever statistical analysis includes advanced stage T3 tumors that involve R0 or R1 resection and limited stage T2 or T1 cancers, for which resection is always R0. Therefore, T stage and R status are not really independent. Our results show that lymph node involvement has more prognostic value than R status, and neither perineural nor perivascular involvement seems to have prognostic value, in contrast to pancreatic cancer. Lymph node involvement can be assessed by the absolute number of positive nodes or by the ratio of positive nodes divided by all of the examined nodes. Two recent studies used cutoff values of either 10% 10 or 20% 17 of positive nodes to determine the prognostic significance of nodal involvement, but these studies reached the same conclusion regarding the significance of the positive node ratio. The curative effect of lymphadenectomy has been debated for a long time and is still controversial. However, in a recent study, Rizk et al 26 showed that it can have a curative effect, except in patients with in situ tumors or patients with.7 positive nodes. Therefore, it is likely that extensive lymphadenectomy improves patients long-term survival and does not improve the prognosis only because of stage migration. 27 It is surprising to conclude that R status does not have prognostic significance. However, it has already been suggested that a positive margin does not affect the pattern of 1st recurrence in other digestive malignancies, such as pancreatic 28 or rectal 29 cancers. Histologic criteria for R1 resection are strict, even if it might be clouded by sampling error. Our pathologic method is quite standardized and Figure 2 Kaplan-Meier survival rates for patients with stage T3 esophageal adenocarcinoma who underwent neoadjuvant treatment followed by surgical esophagectomy. Patients with,20% positive nodes in the specimen are shown in blue, and patients with R20% positive nodes are shown in red.

716 The American Journal of Surgery, Vol 205, No 6, June 2013 Figure 3 Kaplan-Meier survival rates for patients with stage T3 esophageal adenocarcinoma who underwent neoadjuvant treatment followed by surgical Esophagectomy according R status. meets the quality criteria usually stated, aiming at reducing the variability of the results. The finding that R status does not have prognostic significance may be explained by the small sample size, which did not allow for significant statistics. However, in both the univariate and multivariate analyses, node status was highly significant. Instead, we feel that the strict criteria for R0 resection limits the significance of this factor. 19 Indeed, a margin of,1 mm does not always mean that malignant cells are left in place. In our study, of the 13 patients with R1 resections, 7 had recurrences; only 2 of these recurrences were locoregional. Most recurrences after R1 resection were distant from the tumor site. Thus, we believe that R1 resection is an indicator of both surgical quality and tumor aggressiveness, as suggested by other studies about rectal or pancreatic tumors. The practical consequence of these results are that locoregional control of a malignancy by postoperative radiotherapy is not mandatory in patients who did not have neoadjuvant radiotherapy and likely does not improve the chances of long-term survival. The retrospective nature of our study requires that readers be cautious when interpreting the results. However, all of the data were prospectively recorded in a specific database; in addition, all of the patients were consecutive, and there was no loss of patients during follow-up. Although it is not really a prospective study, we think that the methodology allows us to draw some conclusions that should be confirmed by future studies on larger number of patients. Indeed, it is difficult to draw strong conclusions on the basis of a group of 13 patients. However, to reach the size of 30 patients, which is generally considered the minimum required for statistical analysis, R150 patients would be necessary because of a rate of R1 resection of 20%, as usually shown. This means that our data are preliminary and should be confirmed by further study, which is currently ongoing in our department. In summary, our data suggest that R status has less prognostic impact than nodal stage in predicting overall survival in patients with advanced esophageal adenocarcinoma who have received neoadjuvant therapy. Our findings suggest that in cases involving R1 resections, postoperative radiotherapy is not necessary to improve local control of the disease. References 1. Faivre J, Forman D, Estève J, et al. Survival of patients with oesophageal and gastric cancers in Europe. Eur J Cancer 1998;34: 2167 75. 2. Mariette C, Piessen G, Triboulet J-P. Therapeutic strategies in oesophageal carcinoma: role of surgery and other modalities. Lancet Oncol 2007;8:545 53. 3. Shahbaz Sarwar CM, Luketich JD, Landreneau RJ, et al. Esophageal cancer: an update. Int J Surg 2010;8:417 22. 4. Daly JM, Fry WA, Little AG, et al. Esophageal cancer: results of an American College of Surgeons Patient Care Evaluation Study. J Am Coll Surg 2000;190:562 72. 5. Hasegawa S, Yoshikawa T. Adenocarcinoma of the esophagogastric junction: incidence, characteristics, and treatment strategies. Gastric Cancer 2010;13:63 73. 6. Tougeron D, Richer JP, Silvain C. Management of esophageal adenocarcinoma. J Visc Surg 2011;148:e161 70. 7. Lepage C, Bouvier AM, Manfredi S, et al. Trends in incidence and management of esophageal adenocarcinoma in a well-defined population. Gastroenterol Clin Biol 2005;29:1258 63. 8. Mirnezami R, Rohatgi A, Sutcliffe RP, et al. Multivariate analysis of clinicopathological factors influencing survival following esophagectomy for cancer. Int J Surg 2010;8:58 63. 9. Mariette C, Taillier G, Van Seuningen I, et al. Factors affecting postoperative course and survival after en bloc resection for esophageal carcinoma. Ann Thorac Surg 2004;78:1177 83. 10. Portale G, Hagen JA, Peters JH, et al. Modern 5-year survival of resectable esophageal adenocarcinoma: single institution experience with 263 patients. J Am Coll Surg 2006;202:588 96. 11. Wijnhoven BP, van Lanschot JJ, Tilanus HW, et al. Neoadjuvant chemoradiotherapy for esophageal cancer: a review of meta-analyses. World J Surg 2009;33:2606 14.

M. Cabau et al. Impact of lymph node invasion vs radicality 717 12. Jin HL, Zhu H, Ling TS, et al. Neoadjuvant chemoradiotherapy for resectable esophageal carcinoma: a meta-analysis. World J Gastroenterol 2009;15:5983 91. 13. Gebski V, Burmeister B, Smithers BM, et al. Survival benefits from neoadjuvant chemoradiotherapy or chemotherapy in oesophageal carcinoma: a meta-analysis. Lancet Oncol 2007;8:226 34. 14. Malthaner R, Wong RKS, Spithoff K. Preoperative or postoperative therapy for resectable oesophageal cancer: an updated practice guideline. Clin Oncol (R Coll Oncol) 2010;22:250 6. 15. Urschel JD, Vasan H. A meta-analysis of randomized controlled trials that compared neoadjuvant chemoradiation and surgery to surgery alone for resectable esophageal cancer. Am J Surg 2003;185: 538 43. 16. Sjoquist KM, Burmeister BH, Smithers BM, et al. Survival after neoadjuvant chemotherapy or chemoradiotherapy for resectable oesophageal carcinoma: an updated meta-analysis. Lancet Oncol 2011;12: 681 92. 17. Mariette C, Piessen G, Briez N, et al. The number of metastatic lymph nodes and the ratio between metastatic and examined lymph nodes are independent prognostic factors in esophageal cancer regardless of neoadjuvant chemoradiation or lymphadenectomy extent. Ann Surg 2008; 247:365 71. 18. Orringer MB, Sloan H. Esophagectomy without thoracotomy. J Thorac Cardiovasc Surg 1978;76:643 54. 19. Verhage RJ, Zandvoort HJ, ten Kate FJ, et al. How to define a positive circumferential resection margin in T3 adenocarcinoma of the esophagus. Am J Surg Pathol 2011;35:919 29. 20. Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009;250:187 96. 21. Swisher SG, Ajani JA, Komaki R, et al. Long-term outcome of phase II trial evaluating chemotherapy, chemoradiotherapy, and surgery for locoregionally advanced esophageal cancer. Int J Radiat Oncol Biol Phys 2003;57:120 7. 22. Meneu-Diaz JC, Blazquez LA, Vicente E, et al. The role of multimodality therapy for resectable esophageal cancer. Am J Surg 2000;179: 508 13. 23. Lee PC, Port JL, Paul S, et al. Predictors of long-term survival after resection of esophageal carcinoma with nonregional nodal metastases. Ann Thorac Surg 2009;88:186 93. 24. Kesler KA, Helft PR, Werner EA, et al. A retrospective analysis of locally advanced esophageal cancer patients treated with neoadjuvant chemoradiation therapy followed by surgery or surgery alone. Ann Thorac Surg 2005;79:1116 21. 25. Whitson BA, Groth SS, Li Z, et al. Survival of patients with distal esophageal and gastric cardia tumors: a population-based analysis of gastroesophageal junction carcinomas. J Thorac Cardiovasc Surg 2010;139:43 8. 26. Rizk NP, Ishwaran H, Rice TW, et al. Optimum lymphadenectomy for esophageal cancer. Ann Surg 2010;251:46 50. 27. Altorki NK, Zhou XK, Stiles B, et al. Total number of resected lymph nodes predicts survival in esophageal cancer. Ann Surg 2008;248: 221 6. 28. Raut CP, Tseng JF, Sun CC, et al. Impact of resection status on pattern of failure and survival after pancreaticoduodenectomy for pancreatic adenocarcinoma. Ann Surg 2007;246:52 60. 29. Marijnen CAM, Nagtegaal ID, Kapiteijn E, et al. Radiotherapy does not compensate for positive resection margins in rectal cancer patients: report of a multicenter randomized trial. Int J Radiat Oncol Biol Phys 2003;55:1311 20.