| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Click here to read this article as a CME activity
J Thorac Cardiovasc Surg 2008;135:1228-1236
© 2008 The American Association for Thoracic Surgery
General Thoracic Surgery |
a Department of Surgery, the Keck School of Medicine, University of Southern California, Los Angeles, Calif
b Department of Cardiothoracic Surgery, the Keck School of Medicine, University of Southern California, Los Angeles, Calif
Received for publication July 2, 2007; revisions received September 15, 2007; accepted for publication October 4, 2007. * Address for reprints: Steven R. DeMeester, MD, Department of Cardiothoracic Surgery, University of Southern California, 1510 San Pablo St, Suite 514, Los Angeles, CA 90033. (Email: sdemeester{at}surgery.usc.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: The charts of all patients with esophageal adenocarcinoma that had neoadjuvant therapy and en bloc or transhiatal esophagectomy from 1992–2005 were reviewed. Patients found to have systemic metastatic disease at the time of the operation or who had an incomplete resection were excluded.
Results: There were 58 patients: 40 had an en bloc resection and 18 had a transhiatal esophagectomy. A complete pathologic response occurred in 17 (29.3%) of 58 patients. Median follow-up was 34.1 months after en bloc resection and 18.3 months after transhiatal resection (P = .18). Overall survival at 5 years and survival in patients with residual disease after neoadjuvant therapy was significantly better with an en bloc resection (overall survival: 51% for en bloc resection and 22% for transhiatal resection [P = .04]; survival with residual disease: 48% for en bloc resection and 9% for transhiatal resection [P = .02]). Survival in patients with complete pathologic response tended to be better after an en bloc resection (en bloc, 70%; transhiatal, 43%; P = .3).
Conclusion: An en bloc resection provides a survival advantage to patients after neoadjuvant therapy compared with a transhiatal resection, particularly for those with residual disease. Similar to patients treated with primary resection, an en bloc esophagectomy is the procedure of choice after neoadjuvant therapy.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
The incidence of adenocarcinoma of the esophagus and esophagogastric junction is increasing faster than any other cancer in the Western world.1
Although survival is excellent in patients with early-stage tumors, unfortunately, most patients present with dysphagia and have locally advanced disease. The prognosis for these patients is poor, and consequently, attention has focused on the use of neoadjuvant therapy in an effort to reduce the local-regional tumor burden and eradicate micrometastatic systemic disease. During the past 2 decades, neoadjuvant chemoradiotherapy has been tested in numerous trials, and although a clear improvement in overall survival compared with that after surgical resection alone has not been demonstrated, subgroup analysis has suggested that patients who obtain a complete pathologic response have improved survival compared with those who have residual disease at the time of resection.2 This finding has prompted some oncologists to recommend eliminating surgical resection after neoadjuvant therapy because it is thought that those with complete pathologic response do not need resection and those with incomplete response have such a poor prognosis that they are unlikely to benefit from the operation.
It has been our policy to reserve neoadjuvant therapy for patients with locally advanced disease based on findings from computed tomographic (CT) scans, endoscopic ultrasonographic analysis, and/or positron emission tomographic (PET) scanning. After neoadjuvant therapy, we have taken fit patients for an en bloc resection, given evidence that local recurrence is minimized and survival is maximized with this procedure in patients who have surgical intervention as the primary therapy for their cancer.3 The aim of this study was to review our experience and determine whether the local control and survival benefits of an en bloc esophagectomy would extend to patients after neoadjuvant therapy and in particular to those with residual disease on final pathology.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The records of all patients who had neoadjuvant therapy followed by an en bloc or a transhiatal esophagectomy for esophageal adenocarcinoma were reviewed. Neoadjuvant therapy was chemotherapy alone or chemoradiotherapy. Chemotherapy consisted of standard agents, including 5-fluorouracil, cisplatin, carboplatin, and/or paclitaxel. Patients were excluded if they were found to have systemic metastatic disease at the time of the operation, had a tumor that penetrated into an adjacent organ (T4), or had an incomplete pathologic resection (R1). In addition, patients who had an operation other than an en bloc or transhiatal esophagectomy were excluded. This study was approved by the Institutional Review Board of the University of Southern California.
En bloc esophagectomy was performed in patients who were younger than 75 years and who were free of substantial medical comorbidities. Transhiatal esophagectomy was performed in patients 75 years or older and those with significant cardiac or pulmonary disease or other major comorbid conditions. Patients were followed by the operating surgeon at regularly scheduled intervals (every 3 months for the first 3 years, every 6 months up to 5 years, and yearly thereafter). CT scans of the chest and abdomen and routine blood chemistry results were obtained at each visit, and PET scans were obtained yearly.
Surgical Treatment
En bloc esophagectomy was performed as described in previous publications.4 In brief, the en bloc dissection was performed through a right seventh-interspace posterolateral thoracotomy to have excellent exposure of the potentially involved mediastinal nodes below the carina, an upper midline laparotomy, and a left neck incision. The thoracic dissection removed en bloc the esophagus, azygos vein, thoracic duct, and surrounding lymph node–bearing mediastinal tissues. The borders of the dissection extend superiorly to above the azygos arch; laterally to include the left and right mediastinal parietal pleura; anteriorly to the membranous trachea, pericardium, and diaphragm; posteriorly to the spine and aorta; and inferiorly to the esophageal hiatus and down into the costal vertebral angle. Above the aortic arch, the dissection is kept adjacent to the esophagus to minimize potential injury to the recurrent laryngeal nerve. A paratracheal node dissection is not routinely performed for adenocarcinoma of the distal esophagus or gastroesophageal junction. The abdominal dissection was performed through an upper midline incision and removed the node-bearing tissue from the lateral surface of the right crus, the porta hepatis, and around the portal vein and the common hepatic, celiac, and left gastric arteries. Additionally, node-bearing tissue was removed from the lateral surface of the left crus, around the splenic artery, medial to the splenic hilum, and anterior to the adrenal gland. The en bloc dissection allowed systematic removal of lymph nodes in the following areas: low paratracheal, subcarinal, perihilar, paraesophageal, parahiatal, costal-vertebral space, porta hepatis, superior retropancreatic, and around the portal vein and the hepatic, celiac, and splenic arteries.
Transhiatal dissection was performed as described by Orringer5 through an upper midline incision and a left neck incision. The lower mediastinal and esophageal dissection was performed through a widened diaphragmatic hiatus by using blunt and sharp dissection. Paraesophageal and mediastinal lymph nodes were removed as exposure allowed, but a complete node dissection in the mediastinum was not possible. The abdominal node dissection was identical to that described for the abdominal portion of the en bloc procedure.
Reconstruction in most patients was with a tubularized gastric pull-up, but if the stomach was unavailable or unsuitable, a colon interposition was performed. Reconstructions were preferentially placed in the posterior mediastinum, and the proximal anastomosis was hand sewn in the neck.
Statistical Analysis
Grouped data were expressed as medians with interquartile ranges. Statistical analysis appropriate for nonparametric data was used. Univariate analyses were performed by using the Fisher exact test for categorical variables and the Mann–Whitney test for continuous variables. Survival was calculated by using the Kaplan–Meier method, with comparisons of survival by means of the log-rank test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 1. The median radiation dose was 40 Gy in each group, and approximately 90% of patients in each group had both chemotherapy and radiotherapy. Furthermore, chemotherapy was standard cisplatin and 5-fluorouracil in 93% of patients undergoing en bloc resection and 83% of patients undergoing transhiatal resection (P = .36).
|
Table 2. The final pathologic stage was similar in both groups, but stage 3 tumors tended to be more prevalent in the en bloc group (37.5% for en bloc resection vs 28% for transhiatal resection, P = .6). The majority of patients in both groups had residual cancer in the resected esophagus. Furthermore, residual cancer was found in both the esophagus and at least 1 lymph node in 67% of patients. In those with no residual cancer in the esophagus (n = 19), only 2 (10.5%) patients had 1 or more involved lymph nodes.
|
Table 3). The prevalence of local-regional recurrence was significantly reduced in patients who had an en bloc resection.
|
Figure 1). There were 4 non–cancer-related deaths during follow-up: 3 in the en bloc group and 1 in the transhiatal group. If survival is compared by using only cancer-related deaths, the improvement with the en bloc resection is even more significant (56% for en bloc vs 25% for transhiatal at 5 years, P = .03). In patients with a complete pathologic response, there was a trend toward improved 5-year survival with an en bloc resection compared with that after a transhiatal resection (en bloc, 70%; transhiatal, 43%; P = .3;
Figure 2). Survival in patients with residual disease was significantly better after an en bloc resection compared with that after transhiatal resection (48% vs 9%, respectively; P = .02;
Figure 3). Among the 30 patients with residual disease after neoadjuvant therapy who had an en bloc resection, there were 17 patients operated on more than 5 years ago, and from this group, 6 were alive, for an absolute 5-year survival of 35%. In contrast, 8 patients had a transhiatal esophagectomy after neoadjuvant therapy more than 5 years ago, and no patient survived for 5 years (longest survival, 38.5 months).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The only randomized trial to show improved survival with preoperative chemoradiotherapy was published by Walsh and colleagues,7 but this trial has been dismissed by many because of the lack of appropriate staging, methodological flaws, and most importantly survival curves that do not match the data in the text.2,7,8 Trials from England have suggested a benefit from preoperative chemotherapy alone, but these trials also suffer from methodological flaws, and similar results were not found in a major US trial.9,10 The most encouraging results of these trials has been that patients that have a complete pathologic response appear to have improved survival, with most series reporting about 50% five-year survival in this subgroup.11,12 Because overall survival is similar to that of patients treated with resection alone, this would imply that patients who do not have a complete pathologic response to therapy have worse survival than similarly staged patients treated with primary esophagectomy, and this has been substantiated in a number of reports.6,13 The survival after resection in patients with residual disease after neoadjuvant therapy has been so poor that some oncologists and surgeons question the role of esophagectomy in this group.
We and others have reported excellent survival in patients with esophageal adenocarcinoma treated with primary en bloc resection, and in similarly staged patients with limited lymph node involvement, we have shown that survival was improved with the en bloc compared with the transhiatal esophagectomy.14-18 However, criticism regarding stage migration and patient selection, as well as the complexity of the en bloc operation, has hampered widespread acceptance of this procedure. The aim of this study was to compare the effect of the type of resection in a group of patients with locally advanced esophageal adenocarcinoma all treated with neoadjuvant therapy. Similar to our experience with primary resection, we found that overall survival after neoadjuvant therapy was significantly better in patients who had an en bloc compared with a transhiatal esophagectomy. Our 22% five-year survival with transhiatal resection after neoadjuvant therapy is similar to what has been reported by others.6,11 This similarity confirms that although our transhiatal group of 18 patients is small, there is no reason to believe that larger numbers would substantially alter our results, and that the significant difference we found in favor of the en bloc resection is valid. The few non–cancer-related deaths in this population of patients indicate that regardless of comorbidities, the most likely cause of death in patients treated for locally advanced esophageal adenocarcinoma is recurrent cancer, and even after neoadjuvant therapy, the extent of resection is still an important determinant of ultimate survival from this disease.
The improved survival with an en bloc resection is likely due to the high frequency of involved lymph nodes both in the mediastinum and abdomen after neoadjuvant therapy and the fact that a systematic lymphadenectomy allows removal of both known and unknown (micrometastatic) disease. This is also the likely explanation for why survival steadily improves with an increasing number of lymph nodes removed in patients with esophageal cancer (data not shown). This same correlation for increasing survival with an increasing number of resected nodes has previously been shown to be true for gastric cancer.19 Resection of potentially involved lymph nodes is important, even after neoadjuvant therapy, because Prenzel and colleagues20 have shown that neoadjuvant therapy does not reliably eliminate lymph node disease. These authors used both routine histology and immunohistochemistry to evaluate for micrometastatic nodal disease and reported that after neoadjuvant therapy, 75% of patients have nodal metastases. The high prevalence of residual nodal disease emphasizes the importance of complete node dissection and almost certainly explains our finding of improved survival with an en bloc resection, particularly because the incidence of systemic disease was similar in our en bloc and transhiatal resection groups.
Given the failure of randomized trials to demonstrate a clear survival advantage for all patients treated with neoadjuvant therapy before surgical resection, attention has focused on the subgroup of patients with complete pathologic response. We found that complete pathologic response on final pathology was present in 29% of patients in this retrospective analysis, and as has been previously demonstrated, these patients had improved survival compared with those with residual disease. The 5-year survival after complete response in patients who had a transhiatal resection was 43%, which is similar to what has been reported in other series.11 However, the 70% five-year survival with an en bloc resection in patients with complete pathologic response sets a new high standard. It is interesting that nodal disease as the only form of residual disease (no tumor left in the esophagus) was rare. Given this, one would expect that in patients with complete pathologic response, the type of resection would not affect survival. However, we found a trend toward improved survival in these patients after an en bloc resection, and with larger numbers, the difference might have been significant (type 2 error). It is likely that the en bloc resection is removing unknown micrometastatic nodal disease in these patients. This concept is supported by studies using immunohistochemistry to identify micrometastases in lymph nodes, such as one by Prenzel and colleagues,20 in which 41% of patients thought to be node negative based on routine histology had micrometastases. This emphasizes the inaccuracy of only using routine histology to evaluate for complete pathologic response and reinforces the importance of a systematic lymphadenectomy to remove microscopic residual disease, prevent recurrence, and maximize survival, even in patients thought to be complete pathologic responders.
The most important finding in this study was that there was a significant improvement in survival with an en bloc resection compared with that with a transhiatal resection in patients with residual disease after neoadjuvant therapy and that the 48% five-year survival with en bloc esophagectomy justifies resection in these patients. Even when absolute survival is considered, we found that at 5 years, 35% of patients with residual disease were alive after an en bloc resection compared with 0 patients in the transhiatal group. The fact that these patients all had residual disease eliminates concerns regarding stage shift or differences in the extent of disease accounting for the improved survival in the en bloc group and confirms that the extent of resection directly affects the likelihood of survival from esophageal adenocarcinoma. The 9% survival after transhiatal resection we observed in patients with residual disease is similar to the 12% survival reported for this group in the University of Michigan trial, and therefore it is not that our patients with residual disease who have a transhiatal resection are doing worse than expected. Rather, the clear message from our data is how well patients do long-term with an en bloc resection after neoadjuvant therapy, even in the setting of residual disease. These data should discourage the dismissal of the group with residual disease as incurable and encourage a systematic lymphadenectomy and complete resection to enable a chance for cure in these patients. However, in patients with residual disease, a transhiatal resection should not be routinely offered, given the poor survival with this operation in these patients.
The primary goal of surgical resection is complete disease removal in an effort to minimize local-regional recurrence, and we have previously shown that the en bloc technique is associated with a 1% local-regional failure rate.14 In contrast, local-regional failure rates with transhiatal resection are 30% to 40% in series from around the world.11,21,22 The major effect of neoadjuvant therapy appears to be an improvement in local regional disease control without a reduction in systemic disease. In the randomized trial from the University of Michigan, local-regional disease recurrence was reduced from 42% in the transhiatal resection–only group to 19% in the neoadjuvant therapy group.11 The prevalence of local-regional failure after neoadjuvant therapy and transhiatal resection in our experience was similar at 16.6%. However, this is significantly higher than the complete absence of local regional failure we observed in patients who had an en bloc resection after neoadjuvant therapy. The similar low local-regional failure rate with the en bloc esophagectomy, whether done as primary therapy or after neoadjuvant therapy, calls into question the utility of neoadjuvant therapy for local control unless a transhiatal operation is the only option for the patient. Similar to reports from the randomized trials of neoadjuvant therapy, we noted that the majority of recurrences were systemic, despite the administration of neoadjuvant therapy, and this frustrating fact represents a clear challenge for the oncology community and should encourage adoption of new approaches for this daunting disease, including consideration of long-term, low-dose adjuvant chemotherapy administration and chemosensitivity-directed chemotherapy protocols.
The shortcomings of this study include that it is retrospective and that the selection of patients for a transhiatal resection was based on advanced age and medical comorbidities. However, several facts show these shortcomings to be of minor significance. First, the survival and local-regional recurrence rates we report for the patients undergoing transhiatal resection are very similar to the results from randomized trials and from a large population-based database.6,11,12 Furthermore, the major cause of death in these patients was cancer, with only a single death in the transhiatal group occurring as a result of a non–cancer-related cause. Thus concerns regarding the significantly increased age and comorbidities in the transhiatal group affecting survival are unfounded, and an analysis of only cancer-related deaths merely strengthens the significance of our findings. Another issue is that we deliberately did not provide preoperative staging information because clinical staging for esophageal cancer is poor. However, our bias is to perform primary en bloc resection in patients with limited local-regional disease. Patients with locally advanced tumors with multiple nodes on endoscopic ultrasonographic analysis or PET scanning are referred for neoadjuvant therapy, and the decision for an en bloc versus transhiatal resection after neoadjuvant therapy is not based on response but is made purely on the basis of the age, comorbid conditions, and cardiopulmonary status of the patient. If anything, in this study the deck was stacked against the en bloc group, where follow-up was longest, fewer patients had a complete pathologic response, and 68% had stage 2 or 3 disease compared with 44% in the transhiatal group after therapy. Thus the survival differences we demonstrated are unlikely to be due to substantial differences in preoperative stage between the en bloc and transhiatal groups or as a result of the selection of patients with the best response to neoadjuvant therapy for an en bloc resection. Lastly, despite the differences between groups in regard to age and comorbidities, we found no evidence that the neoadjuvant therapy differed significantly between groups, with the majority of patients receiving standard chemotherapy and radiotherapy.
In conclusion, in this study of patients with locally advanced adenocarcinoma of the esophagus, all of whom were treated with neoadjuvant therapy, there was a significant improvement in overall survival and survival with residual disease in patients who had an en bloc esophagectomy compared with those who had a transhiatal resection. Clearly, even after neoadjuvant therapy, the extent of resection is an important determinant of long-term survival from esophageal adenocarcinoma. The explanation for this finding relates to the higher local-regional failure rate with a transhiatal resection and likely to the removal of both known and unknown (micrometastatic) disease with the extended lymphadenectomy that is performed with the en bloc procedure. Poor survival with a transhiatal resection in patients with residual disease mandates an en bloc resection in this group, and the improved survival overall should make an en bloc esophagectomy the procedure of choice in all patients after neoadjuvant therapy. Further improvements in survival with this deadly disease will require chemotherapy agents and protocols that are able to reduce the incidence of systemic recurrence and represent a major challenge for the oncology community. However, these compelling data proving the superiority of the en bloc resection in patients who have had neoadjuvant therapy should finally put to rest the question of whether an extended resection is necessary, beneficial, or both for the treatment of patients with adenocarcinoma of the esophagus.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  J. M. Leers, S. R. DeMeester, N. Chan, S. Ayazi, A. Oezcelik, E. Abate, F. Banki, J. C. Lipham, J. A. Hagen, and T. R. DeMeester Clinical characteristics, biologic behavior, and survival after esophagectomy are similar for adenocarcinoma of the gastroesophageal junction and the distal esophagus J. Thorac. Cardiovasc. Surg., September 1, 2009; 138(3): 594 - 602. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.H. Chang and O.J. McAnena En bloc esophagectomy reduces local recurrence and improves survival compared with transhiatal resection after neoadjuvant therapy for esophageal adenocarcinoma. J. Thorac. Cardiovasc. Surg., January 1, 2009; 137(1): 253 - 253. [Full Text] [PDF]
|
|
|
|
|
|
S. R. DeMeester Reply to the Editor: J. Thorac. Cardiovasc. Surg., January 1, 2009; 137(1): 253 - 254. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
