|
|
||||||||
Ann Thorac Surg 1997;64:185-192
© 1997 The Society of Thoracic Surgeons
Multidisciplinary Thoracic Oncology Program, University of North Carolina, Chapel Hill, North Carolina
| Abstract |
|---|
|
|
|---|
Methods. From 1988 to 1996, 166 consecutive patients with esophageal cancer were evaluated; 66 entered a protocol of chemotherapy (5-fluorouracil, cisplatin) concurrent with radiation (45 Gy) followed by esophagectomy. Fifty-four patients completed the protocol.
Results. Toxicity associated with induction chemoradiotherapy was minimal. The actuarial survival at 12, 24, and 36 months was 59%, 42%, and 32%, respectively. The pathologic complete response (pCR) rate was 41%, with 12-, 24-, and 36-month survivals of 77%, 50%, and 45%, whereas non-pCR patients had survivals of 46%, 35%, and 23%. The difference in survival between pCR and non-pCR patients was not significant (p = 0.13), but the difference in recurrence-free survival was significant (p = 0.007).
Conclusions. This well-tolerated protocol resulted in a high pCR. Trimodality treatment for esophageal cancer may provide long-term survival in some patients regardless of their pCR status.
| Introduction |
|---|
|
|
|---|
The introduction of a multimodality approach for treatment of esophageal cancer has resulted in cautious optimism for a disease that has been nearly uniformly fatal [17]. The rationale for induction chemoradiotherapy followed by operation for gastrointestinal malignancies was first introduced by Nigro and colleagues [8] in 1974) for carcinomas of the anal canal. Adaptation of this approach for esophageal cancer has resulted in several phase II studies that report 3-year survival rates of 36% to 55% [1,2,46]. In nearly all of these studies, patient survival was increased two- to fourfold compared with operation-only historic controls.
In 1984, we reported our experience with operation alone as the treatment for esophageal cancer [9]. Our 3- and 5-year actuarial survival rates were 10% and 5%, respectively. The initial report by Steiger and colleagues [10] of a trimodality approach to esophageal cancer prompted us to reevaluate our belief that treatment for esophageal cancer was primarily palliative. This report reviews our experience at the University of North Carolina with induction chemoradiotherapy followed by esophagectomy for patients with esophageal cancer.
| Patients and Methods |
|---|
|
|
|---|
Of the 166 patients, 66 were found to meet eligibility criteria and were offered trimodality therapy. Twelve patients did not complete the trimodality protocol. Six patients refused operation after completing chemoradiotherapy, 4 patients had development of liver metastases during chemoradiotherapy, 1 patient refused to complete the chemoradiotherapy, and 1 patient's performance status deteriorated after chemoradiotherapy. Therefore, 54 `s completed the entire protocol, including operation, and these patients are the focus of this report.
Initial Evaluation
Initial evaluation included a detailed history and physical examination, chest radiograph, chest and abdominal computed tomographic scans, barium swallow, and laboratory studies. Bronchoscopy was performed if the tumor was in close proximity to the trachea. Bone and head computed tomographic scans were obtained as warranted by clinical suspicion for metastatic disease. Preoperative mediastinoscopy, thoracoscopic staging, and endoscopic ultrasonography were not performed.
Patient Characteristics
The characteristics of all 66 patients who entered the protocol are shown in Table 1
. Forty-nine patients had squamous cell histology, and 17 patients had an adenocarcinoma. In addition, at diagnosis 8 patients (11%) had no dysphagia, 43 (66%) had solid food dysphagia, 2 (3%) had dysphagia only to liquids, and 13 (20%) had both solid food and liquid dysphagia. The average weight loss was 10.5 kg (range, 0 to 32 kg). Eighteen patients (27%) required enteral nutritional support with nasoenteral feeding tubes or surgically placed feeding jejunostomies (n = 7). Tobacco was used by 83% of the patients, and 55% of the patients used alcohol.
|
Chemotherapy was given concurrently with radiation therapy. 5-Fluorouracil was administered as a continuous infusion over 96 hours (1 g/m2 per day) on days 1 through 4 and 29 through 32 of the irradiation. Cisplatin was administered as an intravenous bolus (100 mg/m2) on day 1 of irradiation only. Patients were hydrated intravenously before and after the cisplatin infusion. Toxicity was defined according to the Cancer and Leukemia Group B expanded common toxicity criteria.
At the completion of their chemoradiotherapy, all patients had a chest and upper abdominal computed tomographic scan to evaluate treatment response of the tumor and associated lymph nodes. In addition, 34 patients had an esophagogastroscopy performed immediately before operation.
Operation
Operation was performed approximately 4 to 6 weeks (mean, 38 ± 5 days) after completion of chemoradiotherapy. The majority of patients had an Ivor Lewis esophagectomy (Table 1
). Esophagogastric continuity was reestablished using the stomach in all patients except 2, in whom colon was used secondary to previous gastric operations. All patients with a gastric interposition had a pyloromyotomy or pyloroplasty performed, and accessible lymph nodes in the celiac, paraesophageal, and subcarinal regions were removed en bloc with the specimen. The anastomosis was stapled or sewn in two-layer fashion, depending on surgeon preference. A feeding jejunostomy was placed and enteral alimentation instituted 24 to 48 hours postoperatively. A barium swallow was obtained on days 5 to 7 to evaluate for an anastomotic leak.
The pathologic stage of the tumor was determined according to the classification of the American Joint Committee on Cancer [11]. The absence of tumor in the resected specimen was considered to be stage 0 and a pathologic complete response (pCR).
Follow-up Evaluation
After completion of the protocol, patients were followed up every 3 months for the first year and at 6-month intervals thereafter. Extensive evaluation for tumor recurrence, either local or distant, was initiated only if the patient had symptoms suggestive of recurrence. Patient death records were obtained from the hospital or through our tumor registry. Specific quality of life issues were not recorded.
Statistical Analysis
Statistical analysis was performed using the Statview 4.01 program (Abacus Concepts, Inc, Calabasas, CA). Actuarial survival and recurrence-free survival times were calculated by the Kaplan-Meier method and compared using the log-rank and Wilcoxon tests. Overall survival was calculated from the date of diagnosis until death or May 1996. All in-hospital deaths were included in the survival analysis. All deaths were presumed to be secondary to esophageal cancer unless the cause of death was known to be unrelated to the malignancy. Recurrence-free survival was calculated from the time of completion of the protocol to the date of recurrence or May 1996.
| Results |
|---|
|
|
|---|
Toxicity associated with the protocol was minimal (Table 2
). No patient died as a result of chemoradiotherapy. One patient had severe thrombocytopenia and suffered an upper gastrointestinal bleed, which required transfusion of two units of packed red blood cells and intensive antiulcer medications. Although 8 patients had grade 3 neutropenia, the clinical significance of this is questionable because only 1 patient required hospital admission for intravenous antibiotics for pneumonia, and he was not neutropenic. The incidence of esophagitis was similar to that reported in other studies; however, no patient required parenteral nutrition. Finally, the degree of nausea and vomiting was mild in nearly all patients.
|
Operation
The majority of patients underwent an Ivor Lewis esophagectomy (Table 1
). The complete resectability rate was 100%. The 30-day operative mortality was 4 of 54 patients (7%), and the in-hospital mortality was 6 of 54 (11%). The causes of death were pneumonia with adult respiratory distress syndrome (4 patients), septic shock and multisystem organ failure (1), and massive hemoptysis in 1 patient who required a tracheal resection and reconstruction in addition to his esophagogastrectomy. The median perioperative transfusion was two units (range, 0 to 12), and 25 patients (46%) required no transfusion. The median intensive care unit stay was 2 days (range, 1 to 54 days). The median hospital stay was 13 days (range, 6 to 80 days). No patient had persistent dysphagia after operation.
Pneumonia developed in 11 patients (20%), and acute respiratory distress syndrome necessitating prolonged treatment in the intensive care unit developed in 5 of them. Other complications included temporary vocal cord paralysis (3 patients), small bowel obstruction (2; 1 required an operation), superficial wound infection (3), acute renal failure (3; 1 required dialysis), myocardial infarction (1), and chylothorax (2 patients). The anastomotic leak rate was 6 of 54 patients (11%). Five of these healed without operation, and 1 patient required reoperation. No difference was found in the leak rate between hand-sewn or stapled anastomoses.
Pathologic Stage
The pathologic stage of the tumors is shown in Table 3
. Forty-one percent of the patients had a pCR (stage 0). Analysis of the incidence of a pCR based on tumor histology was 5 of 15 (33%) for adenocarcinoma and 17 of 39 (43%) for squamous cell carcinoma. Tumor was identified in lymph nodes in 14 of 54 resected specimens (26%).
|
The actuarial survival for patients completing the protocol at 12, 24, and 36 months was 59%, 42%, and 32%, respectively (Fig 1
). Of the 22 patients who had a pCR, the 12-, 24-, and 36-month actuarial survival was 77%, 50%, and 45%, respectively. Actuarial survival at 12, 24, and 36 months for those patients without a pCR was 46%, 35%, and 23% (Fig 1
). Although a trend toward improved overall survival was seen in the pCR group, the difference between the pCR and non-pCR group was not significant (log rank p = 0.13; Wilcoxon p = 0.08). Analysis of survival according to tumor histology failed to demonstrate any difference between adenocarcinoma and squamous cell carcinoma.
|
|
Recurrence Patterns
Radiographic evaluation for recurrence-distant, local, or both-was performed only when clinically indicated. Of the 54 patients who completed the protocol, 17 had clear radiographic or histologically proven recurrence of their malignancy. If local recurrence was documented initially, a systemic evaluation for distant disease was performed. No patient had initial local recurrence only. Tumor recurrence occurred distantly in 15 of 17 (88%) and both locally and distantly in 2 of 17 known recurrences (12%) (Table 4
). The most common sites of distant recurrence were the liver, bone, and lungs. The median time to recurrence was 7 months (range, 1 to 32 months). The mean survival of patients after a documented recurrence was 2.6 months. Of the documented recurrences, 23% had a pCR (4 distant, 1 local and distant) and 38% had non-pCR (11 distant, 1 local and distant). No significant difference was seen in recurrence rates for adenocarcinoma (6 of 15 patients, 40%) versus squamous cell carcinoma (11 of 39 patients, 28%).
|
| Comment |
|---|
|
|
|---|
The induction chemoradiotherapy protocol used in this study is similar to that in other reported phase I/II studies [2,47,14]. The induction protocol was well tolerated. The mortality rate during induction chemoradiotherapy has been reported to be 2% to 6% [2, 4, 5, 7]; however, no patient in this series died. We believe one reason this aggressive chemoradiotherapy protocol was well tolerated by even the most severely ill patients was through early and frequent use of enteral alimentation. We avoid the use of central venous catheters for parenteral nutrition and prefer enteral administration. The fact that cisplatin was given only once, at the beginning of the protocol, may also have reduced the acute toxicity.
Esophagectomy was performed, with an in-house mortality rate of 11% and an operative mortality rate of 7%. These results are the same as those we have previously reported with operation only [9] and are similar to those from other studies using induction chemoradiotherapy [1, 3, 14, 15]. This suggests that the addition of chemoradiotherapy does not appreciably increase the mortality associated with esophagectomy.
The pCR rate of 41% in this study is as high as or higher than rates in recently reported series using a trimodality approach (Table 5
). As pointed out by Forastiere and colleagues [16], the pCR rate can be affected by how diligent the pathologist is in seeking tumor cells (falsely high pCR) or by the timing of pathologic evaluation in relation to completion of radiation therapy (falsely low pCR). Despite these potential problems with defining a pCR, most studies [1, 6, 16], including this one, have identified a pCR as a favorable prognostic indicator.
|
The overall 3-year actuarial survival of 32% in this study is comparable with that in other series (Table 5
). For those patients having a pCR, 3-year survival increased to 45%, compared with 23% in patients without a pCR. Although a trend toward improved survival in the pCR group is apparent, the difference was not significant in our sample size. Studies where survival was evaluated based on the pCR status of the patients have yielded conflicting results. Several researchers [1517] have shown a significant survival advantage for pCR patients, whereas others have failed to make this observation [2, 14]. Even if a patient has residual tumor in the resected specimen, survival appears to be better with a multimodality treatment than with radiation therapy or operation alone, both of which have 5-year survival rates of less than 10% [9, 18]. This study suggests that a pCR may indeed be a favorable prognostic factor, but even in the absence of a pCR there appears to be an improved survival compared to single modality treatment for esophageal cancer.
There is no diagnostic study to predict which patients will have no viable tumor after completing chemoradiotherapy. We have previously reported the inaccuracy of pre-resection esophagogastroscopy for determining the presence of residual carcinoma [19]. This point is reemphasized in this study, where 48% of patients were found to have tumor after resection despite having a "normal" pre-resection endoscopic examination. It has also been our impression that the radiographic response rate does not correlate with a pCR. Similarly, the presence or absence of dysphagia after induction therapy does not appear to be a good indicator of residual tumor.
Therefore, we believe that even in patients in whom pre-resection studies have not demonstrated tumor, surgical resection should be performed because of the real risk of failing to cure or palliate the patient. In addition, several studies [4, 5, 14], including this one, have reported pathologic T0 N1 tumors, suggesting that absence of tumor in the esophagus does not rule out regional disease. It is hard to imagine that patients with residual tumor despite negative endoscopy would be long-term survivors without undergoing resection.
Operation for palliation of dysphagia is also important. Twenty percent (5 of 25) of our patients had no endoscopic evidence of tumor but still had either no improvement or a worsening of dysphagia after chemoradiotherapy. The ability of chemoradiotherapy only to improve dysphagia in patients with esophageal cancer is 58% to 77% in two large series [20, 21]. This is in contrast to a more than 85% complete relief of dysphagia with esophagectomy [9]. Thus, operation as part of a trimodality treatment provides a significant palliative benefit aside from improving the chance for cure.
The local recurrence rate after esophagectomy alone for esophageal cancer has been reported to be 20% to 65% [19]. Chemoradiotherapy for esophageal cancer has local failure rates of 40% or more [20, 21]. The local recurrence rate was 4% (2 of 54 patients) in this study. Other studies using trimodality therapy for esophageal cancer have identified local recurrence rates of 0% to 5% [2, 17, 22]. Le Prise and associates [14] noted a 41% incidence of local recurrence; however, that study used only 20 Gy of preoperative radiation therapy. Despite excellent locoregional control, the majority of patients with documented tumor recurrence in this study developed distant disease. Although some investigators [2] have suggested that induction chemotherapy may kill existing micrometastatic disease, no data are available to substantiate this. In fact, the number of cycles of chemotherapy given in this study and most other phase I/II studies of trimodality therapy for esophageal cancer is less than the number of cycles preferred for systemic therapy. Rather, the chemotherapy may be primarily radiosensitizing. It is not surprising to observe that excellent local control rates do not always translate into improved survival rates because distant metastatic disease develops in patients, and they die from these recurrences. Therefore, if survival rates are to be improved further, one needs to consider treatment strategies for the high rate of distant metastatic disease in patients whose local disease is well controlled.
Although this study demonstrates improved survival compared with historic controls, both at our own institution and in the literature, it was neither randomized nor multiinstitutional. We agree with Naunheim and associates [4] and Mackillop and Dixon [23] that the use of historic controls to evaluate treatment for esophageal cancer is not optimal. Although concerns about stage migration, physician attitude toward treatment, and overall perioperative patient care are all real, it is difficult not to be impressed that our 3-year survival with a trimodality approach is 32%.
Nearly every published study of multimodality treatment for esophageal cancer has concluded that phase III studies should be performed to compare operation only to trimodality therapy. A multiinstitutional, phase III trial (Intergroup 0113) comparing preoperative and postoperative chemotherapy with operation alone has recently reached accrual and the data are now maturing. Another intergroup trial that will likely compare trimodality therapy with operation alone is under discussion; however, no study start date has been set. A prospective, randomized study by Walsh and associates [24] showed a significant survival advantage at 3 years for the trimodality approach compared to operation alone for esophageal adenocarcinoma. Researchers at the University of Michigan have conducted a prospective, randomized trial comparing these two arms, and an abstract of the preliminary results suggests that there is no survival benefit at a median follow-up of 1.8 years [25]. It will be interesting to see whether these results persist with longer follow-up.
In conclusion, this study adds to the increasing number of published reports suggesting that induction chemoradiotherapy followed by esophagectomy is a safe, tolerable, and effective treatment strategy for patients with esophageal carcinoma. When randomized, multiinstitutional phase III data are not available, phase II studies, such as this one, provide the best data. Our data suggest that for patients with esophageal cancer, the trimodality approach offers a high pCR rate, excellent local control and palliation, and improved survival compared to operation alone.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
Address reprint requests to Dr Jones, University of North Carolina, 108 Burnett-Womack Bldg, CB 7065, Chapel Hill, NC 27599-7065.
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
N. P. Rizk, E. Venkatraman, M. S. Bains, B. Park, R. Flores, L. Tang, D. H. Ilson, B. D. Minsky, and V. W. Rusch American Joint Committee on Cancer Staging System Does Not Accurately Predict Survival in Patients Receiving Multimodality Therapy for Esophageal Adenocarcinoma J. Clin. Oncol., February 10, 2007; 25(5): 507 - 512. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Agarwal, S. G. Swisher, J. Ajani, K. Kelly, R. R. Komaki, E. Abu-Hamda, A. M. Correa, and J. A. Roth Differential Response to Preoperative Chemoradiation and Surgery in Esophageal Adenocarcinomas Based on Presence of Barrett's Esophagus and Symptomatic Gastroesophageal Reflux Ann. Thorac. Surg., May 1, 2005; 79(5): 1716 - 1723. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. F. Isik, G. Ozturk, S. Ugras, and M. Karaayvaz Enzymatic dissection for palliative treatment of esophageal carcinoma: an experimental study Interactive CardioVascular and Thoracic Surgery, April 1, 2005; 4(2): 140 - 142. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S. Donington, D. L. Miller, M. S. Allen, C. Deschamps, F. C. Nichols III, and P. C. Pairolero Preoperative chemoradiation therapy does not improve early survival after esophagectomy for patients with clinical stage III adenocarcinoma of the esophagus Ann. Thorac. Surg., April 1, 2004; 77(4): 1193 - 1199. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Koshy, N. Esiashvilli, J. C. Landry, C. R. Thomas Jr., and R. H. Matthews Multiple Management Modalities in Esophageal Cancer: Combined Modality Management Approaches Oncologist, April 1, 2004; 9(2): 147 - 159. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. S. Donington, D. L. Miller, M. S. Allen, C. Deschamps, F. C. Nichols III, and P. C. Pairolero Tumor response to induction chemoradiation: influence on survival after esophagectomy Eur. J. Cardiothorac. Surg., October 1, 2003; 24(4): 631 - 637. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Takeuchi, S. Ozawa, N. Ando, Y. Kitagawa, M. Ueda, and M. Kitajima Cell-Cycle Regulators and the Ki-67 Labeling Index Can Predict the Response to Chemoradiotherapy and the Survival of Patients With Locally Advanced Squamous Cell Carcinoma of the Esophagus Ann. Surg. Oncol., August 1, 2003; 10(7): 792 - 800. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. RIEDEL, H. J. STEIN, L. MOUNYAM, F. ZIMMERMANN, U. FINK, and J. R. SIEWERT Influence of Simultaneous Neoadjuvant Radiotherapy and Chemotherapy on Bronchoscopic Findings and Lung Function in Patients with Locally Advanced Proximal Esophageal Cancer Am. J. Respir. Crit. Care Med., November 1, 2000; 162(5): 1741 - 1746. [Abstract] [Full Text] |
||||
![]() |
M. M. DeCamp Jr, S. J. Swanson, and M. T. Jaklitsch Esophagectomy After Induction Chemoradiation* Chest, December 1, 1999; 116(suppl_3): 466S - 469S. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Chidel, T. W. Rice, D. J. Adelstein, P. A. Kupelian, J. H. Suh, and M. Becker Resectable Esophageal Carcinoma: Local Control with Neoadjuvant Chemotherapy and Radiation Therapy Radiology, October 1, 1999; 213(1): 67 - 72. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |