| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Ann Thorac Surg 2000;70:365-366
© 2000 The Society of Thoracic Surgeons
Discussion
Dr. Douglas E. Wood (Seattle, WA): Doctor Kouchoukos and Dr Pairolero, thank you for allowing me the privilege of discussing this year’s J. Maxwell Chamberlain Memorial Paper. This work by Dr Keller and his colleagues stands out as exceptional and truly deserving of the J. Maxwell Chamberlain Award. I would like to offer my congratulations to Dr Keller for an elegant, straightforward, and insightful presentation.
The optimum lymph node dissection during lung cancer surgery remains controversial among thoracic surgical oncologists. Those in favor of lymphadenectomy argue that the procedure improves the accuracy of lung cancer staging and improves results by directing more accurate adjuvant therapy and decreasing locoregional recurrence.
Those opposed feel that lymphadenectomy increases operative time, blood loss, incidence of recurrent laryngeal nerve injury, chylothorax, and bronchopleural fistula, with no evidence of improved oncologic staging or survival. The strengths and weaknesses of the current analysis have been well presented by Dr Keller. The strengths are certainly the size of the study and the fact that it was a prospective multiinstitutional trial. However, the conclusions are weakened by the fact that this was not a randomized or standardized protocol, creating numerous possibilities for selection or treatment bias.
This study does not report the morbidity of lymphadenectomy, which could temper our enthusiasm for the apparent improved survival. Also, patients who are unable to continue with adjuvant therapy were excluded from the trial. This could significantly alter the results if lymphadenectomy patients had more complications, were unable to undergo adjuvant therapy, and then were excluded from statistical comparison.
Doctor Keller, would it be possible to examine the cohort of patients that did not complete the study in order to evaluate the complications of lymph node dissection as well as confirm or exclude this potentially important study bias?
Doctor Keller, I agree with your conclusions regarding the equivalent accuracy of lymph node sampling and lymphadenectomy for lung cancer staging, but I am not sure that I can agree with your conclusions that lymph node dissection improves survival or that these findings should direct our clinical practice. Two randomized trials have failed to show improved survival with mediastinal lymphadenectomy. Fortunately, the ideal trial to definitively answer this question has recently been initiated under the auspices of the American College of Surgeons Oncology Group, or ACOSOG. This trial will prospectively randomize 1,000 patients with clinical stage I and II disease to lymph node sampling versus dissection and will directly examine the key outcome variables of postoperative complications and survival. This and other ACOSOG trials give surgeons the ability to directly enroll patients in surgically relevant clinical trials, with funds provided directly to the surgeon for support of clinical research personnel, facilitating participation by academic and community surgeons alike.
Doctor Keller, I have two questions. In the manuscript you showed no significant difference in locoregional or extrathoracic recurrence between groups. Why do you think that this study was able to show a survival difference yet no difference in disease recurrence between groups? These results seem contradictory to me.
Finally, 85% of patients with stage IIIa lung cancer fail due to metastatic disease, a sign that N2 nodes are a marker of systemic illness, and this is the rationale for all of our treatment strategies in stage IIIa lung cancer. Why do you think that a more radical surgical dissection of the mediastinal lymph nodes would result in improved survival if the majority of these patients have occult metastatic disease at the time of surgical resection?
Doctor Keller, I enjoyed your paper and especially commend you on an elegant and clear presentation. Thank you to the Society for the privilege of the discussion.
DR FREDERIC W. GRANNIS (Duerte, CA): We have been doing a complete and systematic mediastinal node dissection as an integral part of all standard resections for non-small cell lung cancer for the past 12 years at City of Hope National Medical Center. I am glad to see that this work validates our impression of improved survival with mediastinal node dissection. We did not use neo-adjuvant chemotherapy, and did treat with postoperative adjuvant radiation therapy, achieving a 30% actual, 5-year, disease-free survival in N2 patients, including patients with clinical N2 disease recognized before surgery.
Mediastinal lymph node dissection is an easy technique to learn. It can be done in one half of an hour. The morbidity rate is very low, and mortality is not increased. I thank Dr Keller for this important work validating the value of mediastinal node dissection. Thank you.
DR KELLER: Thank you, Dr Woods. The study accrued 488 patients. Approximately 120 of those patients were found to be ineligible, principally because they did not have sufficient documentation of their lymph node dissection. The 373 patients included in this study all had adequate lymph node dissections ensuring adequate staging. Eighty-five percent of the 488 patients received all the assigned radiotherapy and 75% received three or more cycles of the chemotherapy. Therefore, the majority of patients had indeed received the protocol treatment.
The two randomized trials that were cited have been published. One is from Japan, and the second from Germany. Both are quite different from the current study. The trial from Japan was limited to patients whose primary tumors were less than 2 cm in size and there were essentially no patients with lymph node metastases. The second study from Germany included approximately 75 patients with lymph node metastases. The presented today is therefore five times as large and has a follow-up that is twice as long.
The median survival of patients with N1 disease who were entered in this trial was 45 months, and for the patients with N2 disease was 30 months. This is much better than most of us would have imagined and demonstrates the importance of randomized prospective trials rather than relying on historic controls.
Why was the survival so good? I believe that many of the patients with N2 disease do not have systemic disease, but rather local disease. By performing an effective and thorough mediastinal lymph node dissection, we are able to eradicate that local disease and improve long-term survival.
DR WILLIAM I. BRENNER (Hackensack, NJ): Doctor Keller, based on your results, would you advocate a right thoracotomy and lymph node dissection in patients with left-sided lung cancer?
DR KELLER: Japanese investigators have written about very extensive lymph node dissections for a number of years. This involves mobilizing of the arch of the aorta. Some investigators even recommend sternotomy or supraclavicular lymph node dissection. I think that at the current time a complete mediastinal lymph node dissection is indicated on the left side, in order to appropriately stage the patient and identify as many levels of N2 disease as is possible. This will provide additional prognostic information and will help stratify patients who are entered in additional clinical studies. However, it may be time for those of us who do node dissections in the United States to rethink the extent of the lymph node dissections and perhaps to follow our Japanese colleagues and do them in greater detail.
I would like to reiterate Dr Wood’s comments about the American College of Surgeons Oncology Group, which has recently opened a Phase III trial comparing lymph node sampling to complete mediastinal lymph node dissection. The primary objectives of this trial are to evaluate whether complete mediastinal lymph node dissections result in better overall survival when compared with systematic sampling in patients undergoing resection of N0 or N1 cancer.
Patients will undergo intraoperative sampling and, if no N2 disease is found, they will be randomized to either no further lymph node dissection or a complete node dissection. Thoracic surgeons may participate as individuals (this is very important, via the American College of Surgeons) and do not need cooperative group affiliation. I urge all the members of the Society to join through the American College of Surgeons and participate in this study.
Related Article
Ann. Thorac. Surg. 2000 70: 358-365.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
