| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Section of General Thoracic Surgery, Department of General Surgery, Emory University School of Medicine, Atlanta, Georgia
Accepted for publication November 26, 2007.
* Address correspondence to Dr Daniel L. Miller, Section of General Thoracic Surgery, Emory University Clinic, 1365 Clifton Rd NE, Atlanta, GA 30322 (Email: daniel.miller{at}emoryhealthcare.org).
Presented at the Fifty-third Annual Meeting of the Southern Thoracic Surgical Association, Tucson, AZ, Nov 8–11, 2006.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Methods: A retrospective review was performed of all patients (281) who underwent an esophagectomy for cancer at our institution from July 1995 through June 2005; 122 patients (43%) had neoadjuvant treatment and form the basis of this study.
Results: Preoperative chemoradiation (CR) was administered in 64 patients and chemotherapy only (CO) in 58 patients. Operative mortality was 6% (4 patients) in the CR group and 0% in the CO group (p = 0.12). Overall postoperative complications rate was 48% in CR patients and 33% in CO patients (p = 0.09). Complete pathologic response occurred in 11 CR patients (17%) and in 2 CO patients (4%; p = 0.02). There was no difference in recurrences between the two groups (p = 0.43). Median survival was 17 months in the CR patients and 21 months in the CO patients (p = 0.14). One-, 3-, and 5-year survivals were 76%, 46%, and 41%, respectively, in the CR patients and 70%, 40%, and 31%, respectively, in the CO patients (p = 0.31).
Conclusions: Although neoadjuvant chemoradiation resulted in a significantly better complete pathologic response rate when compared with chemotherapy alone, that did not translate into a long-term survival advantage. Chemotherapy alone may be the preferred neoadjuvant modality to expedite resection, decrease operative mortality and postoperative complications, and improve survival in patients with locally advanced esophageal cancer.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Two hundred eighty-one patients were identified; 122 patients (43%) had neoadjuvant treatment and form the basis of this manuscript. Neoadjuvant therapy was administered by university-based or community-based medical and radiation oncologists and supervised by our general thoracic surgeons. The induction treatment consisted of platin-based therapy in the majority of patients (96%; Tables 1 and 2).
Radiation therapy ranged from 45 to 60 Gy: 60 Gy in 2 patients, 50.4 Gy in 14, and 45 Gy in 48. All but the 2 patients who received the 60 Gy underwent concurrent single-dose radiation with the chemotherapy. After a recovery period of 3 to 6 weeks, patients underwent an esophageal resection, mostly through an Ivor Lewis approach. All patients were staged clinically before treatment and pathologically at the time of resection according to the TNM classification of the American Joint Committee for Cancer Staging [7]. Complete pathologic response was defined if there was no histologic evidence of viable tumor in the resected specimen (esophagus, stomach, and lymph nodes).
|
|
Follow-up consisted of examinations every 6 months for the first 2 years and annually thereafter. Postoperative imaging studies included chest radiograph, barium swallows if necessary, and computed tomography as well as esophagoscopy. Extensive tumor evaluation was initiated earlier if the patient had symptoms suggestive of recurrence. Deaths were verified from hospital or clinic records, tumor registry, and the Social Security Death Index.
Disease-free and overall survivals were estimated by the Kaplan-Meier method using the date of esophageal resection as the starting point and the date of death or last follow-up as the endpoint [8]. Patients with resected margins positive for cancer or metastatic disease found at time of resection were assigned a disease-free survival of zero time. Discrete data were analyzed with Fisher’s exact test. The influence of variables on survival was analyzed using the log-rank test for discrete variables [9] and the Cox proportional hazards model for continuous variables and multivariate models [10]. All statistical tests were two-sided with threshold of significance set at p less than 0.05. All statistical analyses were conducted using Statistical Analysis Systems software (SAS Institute, Cary, North Carolina).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
|
|
|
|
The median survival for all patients who underwent neoadjuvant therapy followed by esophagectomy for locally advanced esophageal cancer was 18.6 months, with overall 1-, 3-, and 5-year survivals of 72%, 43%, and 35%, respectively (Fig 1). The 1-, 3-, and 5-year survivals after neoadjuvant chemotherapy only was 70%, 40%, and 31% and did not differ significantly from the survivals of 76%, 46%, and 41%, respectively, observed after neoadjuvant chemoradiation therapy (p = 0.305), as shown in Figure 2. The median survival was 20.7 months in the CO group and 17.2 months in the CR group (p = 0.14).
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Theoretically, neoadjuvant chemotherapy offers early treatment of micrometastatic and locally advanced disease that can facilitate surgical resection by downstaging the tumor. The majority of our patients underwent a platin-based chemotherapy regiment. There was on difference between the two groups with respect to the chemotherapy agents used. The additional advantage of combining chemotherapy with radiation is the increased incidence of complete pathologic response, reportedly only 5% with chemotherapy alone [15] versus 16% to 51% with chemoradiation therapy [11, 14, 16, 17]. In our study, the complete pathologic response rates were similar to the literature for both the CR and CO patients. Our study did show that there was a statistically significantly increase in complete pathologic response rate of 17% in patients who underwent neoadjuvant chemoradiation therapy as compared with 3% in patients who underwent neoadjuvant chemotherapy alone.
The adverse effects of preoperative chemotherapy include immune suppression and nutritional depletion which may lead to impaired healing and infectious complications. The addition of preoperative radiotherapy leads to higher intraoperative risks due to increased bleeding, technical difficulty specifically related to increased edema, inflammation, and fibrosis, and postoperative respiratory complications related to lung parenchymal damage [3, 16, 18]. Preoperative radiotherapy also prolongs the preoperative induction time that could be harmful by delaying definitive, albeit modestly. effective treatment with surgery. In this series, the induction treatment time was almost 2 months longer when radiation therapy was used. Preoperative chemoradiation has been associated with an increase in postoperative complications when compared with surgery alone, particularly pulmonary complications such as ARDS [19]. Radiation doses in our series ranged from 45 to 60 Gy; only 2 patients underwent the higher complication-associated dose of 60 Gy. The overall postoperative complication rate was almost 50% in our patients who received chemoradiation compared with one third in the chemotherapy only patients. The CR group had more cardiac (19%), pulmonary (14%), and infectious (14%) complications; however, the CO group had more surgical complications (10%), including the only two vocal cord injuries and also twice as many patients with strictures who required dilatation. There was no difference in the anastomotic leak rates between the groups, which were 6% and 5%, respectively.
Several reports have shown an increase in operative mortality among patients undergoing induction chemoradiation therapy when compared with surgery alone [6, 12, 14]. The Urschel and Vasan [14] meta-analysis of randomized controlled trials showed a nonsignificant trend toward increased operative mortality (p = 0.07) and all treatment mortality (p = 0.05) in the group of patients who underwent neoadjuvant chemoradiation as compared with patients treated with surgery alone. In contrast, another similar meta-analysis by Urschel and coworkers [15] found no difference in operative mortality (p = 0.76) and treatment mortality (p = 0.22) among patients who underwent neoadjuvant chemotherapy as compared with patients treated with surgery alone [15]. All 4 operative deaths (3%) in our series occurred in the CR group and there was none in the CO group; all the deaths were related to postoperative pulmonary complications. Although all of these deaths occurred in the chemoradiation group, no significant difference in mortality was observed between the groups (p = 0.12), but the trend was approaching significance.
Previous studies have shown that pulmonary complications in patients who receive preoperative chemoradiation have been associated with increased postoperative mortality after esophagectomy for esophageal cancer [19–21]. Our findings support this observation, since twice as many patients in the CR group suffered postoperative pulmonary complications as in the CO group, and all of the deaths in this series were related to acute respiratory failure, 3 of those involving ARDS.
There was no difference in overall survival between the CO and CR groups. Although our study is retrospective and nonrandomized, these survival data are consistent with survival results by several randomized trials [3, 5, 19–21]. Most studies have shown improved survival with neoadjuvant therapy among patients with a complete response. Our study showed that the CR group had a 17% complete pathologic response rate with an associated median survival of 32.7 months and a 5-year survival rate of 53%, compared with patients with residual tumor in the CR who had a median survival of 16.9 months and a 5-year survival of 41% (p = 0.28). Similarly, Donington and colleagues [6] showed a 22% complete pathologic response rate in patients who underwent preoperative chemoradiation. These patients had an overall 3-year survival of 64% compared with 34% for patients with residual tumor. Although the trend toward improved survival was observed, the difference between the two groups was not statistically significant (p = 0.17) [6]. In our study, no matter what the neoadjuvant treatment was, there was not a survival advantage between the groups even for the complete responders. Also, the absence of radiation did not affect the local recurrence rate. There was no difference in local recurrence rate between the two groups. Therefore, surgical resection alone may be the only modality necessary for local control in patients with locally advanced disease, whereas chemotherapy is warranted to reduce the risk of distant disease, which usually is the most common cause of death among patients with locally advanced disease [6, 21]. Unfortunately, the ideal chemotherapy agent currently does not exist.
In summary, neoadjuvant chemoradiation leads to a delay in surgery, which is the best modality for local control, and an amplification of operative and postoperative complications. Although the combination of chemotherapy and radiation resulted in significantly more complete pathologic responses when compared with chemotherapy alone, that did not translate into a survival advantage. Thus, chemotherapy alone may be the preferred neoadjuvant modality to expedite resection, reduce incidence of postoperative complications, and improve survival of patients with locally advanced esophageal cancer. A large multicenter, randomized, prospective trial is needed to answer this question.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Now, let’s look closely at your finding that a complete pathologic response (pCR) did not translate into a long-term survival advantage. It has been consistently shown in many other studies that complete responders fare far better than nonresponders. In a recent meta-analysis of induction chemotherapy analyzing 2,051 patients, the pCR rate was only 3%, similar to yours of 4%. In a similar meta-analysis of 1,116 patients undergoing trimodality, the pCR was 21%. In your study, 47% of patients undergoing chemotherapy alone had a postsurgical stage of III or IV compared with 22% of patients receiving induction chemoradiation. Intuitively, pCR and downstaging should make a difference as shown by others. This leads me to question number two. Could you comment on your interpretation of these results? Could your findings that complete responders did not have a surgical advantage simply be due to underpowering of the study?
As you know, there have been two large randomized trials of chemotherapy followed by surgery versus surgery alone. The Kelsen study showed no difference in survival, and the British MRC study did reveal a survival advantage. Question three: Did you look at a matched cohort who underwent surgery alone at your institution? Interestingly, patterns of case studies have shown an increase in trimodality therapy from about 10% in 1992 through 1994 to 27% in 1996 and 1999.
Although I disagree with the overall conclusions in this retrospective and underpowered study, the take-home message is very important: Clinical trial evidence for either induction chemotherapy only or chemoradiation is conflicting and no definitive recommendation can be made. Unfortunately, until we agree to come together and perform well-constructed prospective trials with enough patients to be meaningful, we will remain in an esophageal cancer quagmire. Thank you very much.
DR LUU: Thank you, Dr Reed, for the comments, your excellent questions, and thoughtful insight. Your first question is how can we not have a single p value of ours significant and still make the conclusion that chemotherapy alone may be the preferred treatment. We believe that most of our results did not attain statistical significance because of small sample size, and that is why you are correct that a randomized prospective trial with a larger sample size will determine the ideal treatment regiment.
In regard to the patients with a complete pathologic response, there have been several studies in the literature that do support improved survival. Doctor Miller, when he was at Mayo, conducted a study that did show an early survival advantage at 3 years, however, a subsequent Mayo study showed that this was not true. Again, in our study, there was not a statistically significant difference in 5-year survival because there was not a large enough sample size—only 11 of our patients in the chemoradiation group had a complete pathologic response.
We did not perform a matched control cohort with our surgery-alone patients because we did not have accurate preoperative clinical staging information in the surgery only patients to determine their outcomes.
DR MARK J. KRASNA (Baltimore, MD): I want to congratulate Dr Luu on a beautiful presentation and standing there under Dr Reed’s withering attack. You did a very good job. I think the important message, as Dr Reed and you have now alluded to in your discussion, is that there are several centers with large experiences with neoadjuvant chemorads that have had excellent results. The data from M.D. Anderson, the data from Maryland would suggest in a combined series now of probably more than 300 cases that you could achieve between a 40% and 50% 5-year pCR rate and more than a 40% 5-year survival in patients well selected who get high-dose chemotherapy and radiation therapy.
One of my questions is perhaps in the manuscript, but we don’t know exactly the dose that was given, and I do think it is important to note what the chemotherapy dose was, if you could tell us, as well as the radiation dose.
The other question I have goes to something you alluded to a moment ago, and that is, you did show us that there was no bias between the two groups, but you just noted that they were not able to be matched. The biggest question that came up from several of the prospective randomized trials was that they were not balanced in terms of their staging. So in your current series, how many of these patients had at least esophageal ultrasound and possibly lymph node staging, and if you have those data, were there an equal number of stage III and stage II patients pretreatment in both of those arms? Again, I think you did a great presentation; however, I think the answer is still out there. Thank you.
DR LUU: Thank you, Dr Krasna. At our institution, the majority of our patients came to us after being treated by community-based oncologists and radiation oncologists. We do not know the chemotherapy dosages used or the amount of radiation that was given. We will try to get that information.
Your second question on the use of endoscopic ultrasound—over the last 3 years, more patients underwent endoscopic ultrasound for preoperative staging, but since that number was small, we decided not to include those data in our study.
DR LINDA W. MARTIN (Mountain View, CA): I wanted to congratulate you on an excellent presentation. A question for you is how you calculated your survival time. Was that from time of treatment initiation or from the date of surgery, because if it is from the date of surgery, you are introducing significant bias against the chemoradiation group, because, as you mentioned, the pretreatment time was quite a bit longer. That could make a very big difference in your median survival data.
DR LUU: We calculated our survival starting from the date of surgery. So you are right, the pretreatment time does effect the survival.
DR MARTIN: I would recommend that you go back and look at calculating survival using treatment initiation dates because it might completely alter your reported outcomes.
DR LUU: I think that is an excellent idea. Thank you, Dr Martin.
DR KING F. KWONG (Baltimore, MD): Doctor Luu, you and your colleagues are to be congratulated for tackling a very difficult task of exploring this very difficult cancer. I just have two comments that very much echo those of Dr Reed and Dr Krasna. First off, given all the caveats of a retrospective analysis, did you analyze some of your results looking at a stage-specific, group-by-group comparison? As you have shown in your demographics, these groups were fairly well matched, but did you look at it at least from a stage-by-stage comparison? That might give you at least a hint of a trend that there is a difference between the two treatment regimens.
And lastly, I will add one more comment before sitting down and listening to your answer. I would echo the fact that the small sample sizes may be very difficult in teasing out the fact that your higher complete pathologic response rate with the chemoradiation and neoadjuvant treatment might be much more worthwhile, and that type of downstaging, as Dr Reed had alluded to, will probably carry into longer survival rates. I would like to thank you for your presentation. It was very nicely done.
DR LUU: Again, can you repeat your second question?
DR KWONG: The first question was did you analyze a group-by-group analysis between your different treatment arms stage by stage?
DR LUU: In regard to the first question, we did not analyze our results specifically by stage. Preoperatively, we did not have a good reliable pretreatment stage, so we did not specifically correlate it with our postsurgical stage. I am sorry; I did not catch your second question.
DR KWONG: The second question was essentially a comment. I think that if you were to look at larger sample sizes, as Dr Reed alluded to, you might see a difference in survival between the two groups, as other groups have done as well. Thank you.
DR LUU: I agree. I think that if we did have a larger sample size we may have showed a significant difference.
DR STEPHEN CASSIVI (Rochester, MN): I have more of a comment toward the future. In your very well presented study, I was interested to find that a lot of the complications were respiratory complications. I know that the majority of your esophagectomies are done by Ivor Lewis with a transthoracic component. Conventional external beam radiation is well known to result in some degree of pulmonary injury. I am wondering whether you feel we will be able to minimize these radiation-related respiratory complications with the use of photon radiation or heavy ion radiation. This technology is coming down the line and seems to be promising for esophageal cancer, with the potential of reducing the amount of collateral radiation damage to the lungs and heart.
DR LUU: Preoperative radiation can result in acute lung injury due to lung parenchymal damage, which is a concern. I know Dr Reed’s group showed that radiation does cause an increased incidence of pulmonary complications and subsequently may increase operative mortality. At our institution we use traditional radiotherapy instead of IMRT radiation, but I think your idea of a less aggressive type of radiation should be evaluated.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. E. Merritt, R. I. Whyte, N. T. D'Arcy, C. D. Hoang, and J. B. Shrager Morbidity and Mortality After Esophagectomy Following Neoadjuvant Chemoradiation Ann. Thorac. Surg., December 1, 2011; 92(6): 2034 - 2040. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Logue and S. Griffin Road Map to Esophagectomy for Nurses Crit. Care Nurse, August 1, 2011; 31(4): 69 - 86. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. YOSHIDA, M. KOCHI, M. FUJII, N. KANAMORI, T. KAIGA, Y. MIHARA, T. FUNADA, H. TAMEGAI, M. WATANABE, and T. TAKAYAMA Complete Response to Chemoradiotherapy in a Patient with Synchronous Double Gastric and Esophageal Cancer Anticancer Res, June 1, 2011; 31(6): 2339 - 2342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Solomon, D. Mezentsev, I. Reis, M. Lima, J. Rios, E. Avisar, D. Franceschi, A. Livingstone, L. Podolsky, and B. Ardalan A Phase II Study of Neoadjuvant and Adjuvant Chemotherapy with 5-Fluorodeoxyuridine, Leucovorin, Oxaliplatin and Docetaxel in the Treatment of Previously Untreated Advanced Esophageal Adenocarcinoma Jpn. J. Clin. Oncol., April 1, 2011; 41(4): 469 - 476. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Stiles, P. Christos, J. L. Port, P. C. Lee, S. Paul, J. Saunders, and N. K. Altorki Predictors of survival in patients with persistent nodal metastases after preoperative chemotherapy for esophageal cancer J. Thorac. Cardiovasc. Surg., February 1, 2010; 139(2): 387 - 394. [Abstract] [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 |
