| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 1997;64:191-192
© 1997 The Society of Thoracic Surgeons
DR MARK B. ORRINGER (Ann Arbor, MI): I congratulate Dr Jones and his colleagues for their carefully performed and nicely presented study. This report adds to the list of published phase II nonrandomized trials of induction chemoradiotherapy followed by esophagectomy for esophageal cancer. Such reports have generally compared 3-year survivals with combined modality therapy in the range of 20% to 40% with that in historic controls treated with operation alone, usually around 10%, and have concluded that there is a survival advantage with this latest rage in the treatment of esophageal carcinoma. But in the absence of confirmation of these inferences with randomized phase III trials, the argument for uniform adoption of multimodality therapy for esophageal carcinoma is weak at best.
We, at the University of Michigan, have contributed to the current zeal for multimodality therapy. Our phase II trial with 43 patients, half with adenocarcinoma and half with squamous carcinoma, reported an overall 5-year survival in 36 patients with completely resected tumors of 42%. In the 24% of 6patients who were complete responders, there was a gratifying 60% 5-year survival.
On the basis of these encouraging data, we took the next step of a phase III randomized trial comparing transhiatal esophagectomy alone with chemoradiotherapy followed by transhiatal esophagectomy. There were 50 patients randomized to each arm of a 100-patient study. There was a generalized groan as our preliminary data with a median follow-up of 2.3 years showed no significant difference in median, 1-year, or 2-year survival. Hot off the press, however, median follow-up for the 19 living patients now at 5.2 years shows 3-year survival statistics looking more promising, 15% versus 32%. But again, statistical significance is still lacking with a p value of 0.07.
So where are we with multimodality therapy for esophageal carcinoma? We still do not have the proof that is needed to justify this as a standard of care in spite of the wave of enthusiasm and the pressure to do so.
Because roughly 25% of patients receiving preoperative chemotherapy have a complete pathologic response, our oncologic colleagues, for the most part, have decided that if no remaining tumor can be identified endoscopically or with ultrasonography, esophagectomy is not needed. This is completely unjustified. And I commend Dr Jones and his colleagues for pointing out that 25 of their 34 patients who underwent esophagoscopy after preoperative chemotherapy had no evidence of malignancy. In 48% of these (12 patients), there was residual tumor in the resected specimen and this likely would have progressed without resection. The growing number of patients who we are seeing for "salvage esophagectomy" attests to the danger of assuming that there has been a complete pathologic response to chemoradiotherapy simply because tumor cannot be identified at esophagoscopy and biopsy. And an esophagectomy performed months after radiation fibrosis has had a chance to become established is a challenge orders of magnitude beyond the standard esophagectomy.
I will close, Dr Jones, with some questions. Your hospital mortality rate for esophagectomy of 11% is a little high, but it is acceptable. But your morbidity is a little problematic. We now seldom perform transfusion in our transhiatal esophagectomy patients. Your median transfusions were two units per patient, and more than half of the patients received transfusions. Why?
Under the gun of managed care, our transhiatal esophagectomy patients now generally go directly to the floor from the recovery room; rarely is intensive care needed. Your median intensive care unit stay was 2 days. Is this avoidable?
Similarly, our patients are discharged at 7 days now; your median was 13 days. This can also be brought down. Can this be reduced?
Your 20% incidence of postoperative pneumonia, 5 cases with adult respiratory distress syndrome, is very worrisome and needs to be explained. Do you operate on patients still smoking cigarettes? Do your patients use incentive inspirometers preoperatively?
And finally, how do you explain your complete pathologic response rate of 41%, the highest yet reported? Is this a function of your chemotherapy/radiation therapy protocol or the methodology used by the pathologists to study the specimen?
DR GORDON F. MURRAY (Morgantown, WV): I compliment Dr Jones on his good work and certainly an excellent presentation, and to a degree with a sense of nostalgia. Thirteen years ago, when I was contributing to the Chapel Hill group, we presented actually an equal number of patients with esophageal carcinoma who underwent resection alone. There were two concluding points made at that time: First, that esophagectomy does offer excellent palliation for carcinoma of the esophagus; and second, that good results can be achieved by resident surgeons in a well-supervised training program.
Today Dr Jones reports that induction chemoradiation therapy followed by esophagectomy offers both excellent palliation and improved survival compared with operation alone. Indeed, Dr Jones and his colleagues have increased the median survival of their patients by 50% at the University of North Carolina in the last decade. Doctor Orringer has addressed the limitations of Dr Jones's data; however, I feel that the message remains quite powerful, and commend Dr Jones and his coauthors on their efforts.
DR ANTONIO LAUDITO (New York, NY): I did not see the use of transesophageal ultrasound in your preoperative staging system. Don't you think that this modality would have been able to detect the lesions not detected by endoscopy?
DR ROBERT J. GINSBERG (New York, NY): I would like to add to that question. There were no clinical staging data presented, which I think is extremely important, to understand what kind of patients you were selecting. In what stage were the patients who had a complete response; were they all in stage I and stage II clinically? I commend Dr Jones and his associates, but I warn them that they should perform endoscopy in every patient before operation.
I would like to ask Dr Jones and his associates why they kept changing the denominator. Fifty-four patients were entered into the protocol, and all results should include the total denominator.
DR MICHAEL J. O'SULLIVAN (San Diego, CA): Doctor Jones, was there any difference in response on the basis of cell type?
DR JONES: I thank all the discussants for their comments.
Addressing Dr Orringer's questions first, we agree that the incidence of pneumonia in our study is problematic. This study evaluated patients over an 8-year period. We have become increasingly more liberal with our use of epidural catheters, and over the past several years the incidence of pneumonia has decreased significantly. We attribute this, in part, to better postoperative pain management.
We do continue to operate on patients who smoke, although we encourage them to stop smoking. All patients use incentive spirometry postoperatively.
In the past several years, we have reevaluated our blood transfusion practice and have subsequently given transfusions to only 2 patients. We have elected not to perform transfusion in patients who are hemodynamically stable, even though they may have a low hematocrit.
We have yet to achieve such excellent postoperative care that the patient can go to the floor immediately after the operation. However, our intensive care unit stay has been reduced to 1 day for most patients.
The pCR rate seen in this study is as high as, or higher than, that in most other studies. There are inherent problems in pCR determination because it depends, in part, on the degree of diligence on the part of the pathologist in examining the specimen. This can result in a falsely high pCR rate. Alternatively, a falsely low pCR rate can be seen, depending on the time between completion of radiation therapy and surgical resection. Our high pCR rate may be a function of the induction therapy, although this probably does not explain it entirely. Regardless, we were pleased to see such a high pCR rate and to demonstrate its usefulness as a favorable prognostic indicator.
I thank Dr Murray for his unique perspective on this study, as well as his kind comments.
Doctor Ginsberg and Dr Laudito asked about preoperative and clinical staging of the patients. No patient had preoperative thoracoscopic or endoscopic ultrasonography performed. We are considering investigating their use at our institution. We agree with Dr Ginsberg that all patients should have preoperative endoscopy. We did not change the denominator of 66 patients in this study. We agree that all 66 patients should be evaluated, and this is addressed in the manuscript. The median survival of the 12 patients who did not complete the protocol was 8 months. For those patients who chose not to have an operation after completing their induction therapy, the mean survival was 11 months. Although these data are not presented, they are included in the manuscript.
Finally, there were no significant differences in the pCR rate or patient survival with respect to the tumor histology.
Related Article
Ann. Thorac. Surg. 1997 64: 185-192.
| ||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
