| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Department of Surgery, Ioannina University School of Medicine, Ioannina, Greece.
Correspondence: Address correspondence to: Dimitrios H. Roukos, MD, Department of Surgery, Ioannina University School of Medicine, 45110 Ioannina, Greece; Fax: 30 26510 97094; e-mail: droukos{at}cc.uoi.gr
During the past several decades, survival of patients with gastric cancer in the United States has remained poor. Less extensive surgery consisting of gastrectomy with limited D0/D1 lymph node dissection has been the routine clinical practice in the treatment of gastric cancer.1,2 This surgical undertreatment is a serious problem in the treatment of gastric cancer in the United States and may partially explain1 the overall 5-year survival rate of only 23%.3 The corresponding rate in Japan, with extensive D2 dissection as the standard of care, is over 50%.4,5 Limited surgery, particularly in advanced-stage cancer, may be associated with high residual disease and recurrence rates,6 as it has recently been demonstrated in the Intergroup study (INT-0116).2 In this United States multicenter randomized controlled trial (RCT), the rates of local (29%) and regional (72%) recurrence were very high in the surgery-alone group with a limited D0/D1 node dissection. With more extensive D2/D3 node dissection, lower local and nodal recurrence rates have been reported following nonrandomized studies not only in Japan4,5 but also in Europe.7 Could residual disease, recurrence, and mortality be improved with appropriate, more extensive, D2 surgery?
On the basis of the better survival with postoperative chemoradiotherapy (fluorouracil plus leucovorin and local-regional radiation) observed in the INT-0116 study conducted by the Intergroup/South West Oncology Group,2 current research efforts in the United States have focused on the development of a more effective regimen using newer agents such as cisplatin and taxanes and similar preirradiation and/or postirradiation regimens. Since the two large European RCTs have failed to demonstrate a survival benefit in favor of D2 over D1 node dissection, 8,9 limited D0/D1 surgery continues to be the most common surgical procedure in several ongoing phase II and III trials. Is this the best waynamely, using as a platform a limited D0/D1 surgeryfor the development of a more effective adjuvant treatment? Or should priority be given to more extensive D2 surgery, which ensures minimal residual disease and may increase the effectiveness of systemic chemotherapy and radiotherapy?
Indeed, on the basis of credible evidence of the safety and effectiveness of extensive D2 dissection provided from new RCTs1012 and the longer survival revealed in the Dutch trial,13,14 reevaluation of the extent of surgery in both clinical practice and research appears to be an important step toward realization of the efforts to improve survival in the United States and Europe.
There are two major arguments, based on the findings of two previous European RCTs,8,9 against wider clinical use of D2 dissection in the United States and Europe. First, D2 versus D1 node dissection increases in-hospital morbidity and mortality. Second, it does not significantly improve recurrence and survival rates.8,9 The safety of D2 dissection has become clearer. The Japanese experience,4,5,15 Western experts opinions,16,17 and most recent RCTs1012 reflect worldwide agreement that D2 node dissection, performed by high-volume surgeons, is a safe procedure. In general, recent scientific evidence indicates that high-risk cancer surgery, including pancreatic resection and esophagectomy18 as well as D2 dissection,19 is associated with low rates of hospital mortality, provided that high-volume surgeons perform the surgical procedure. The message is so clear that patients who are undergoing such surgical procedures are advised to carefully choose20 experienced surgeons with technical skill, to ensure the lowest possible operative mortality risk.20
The second argument against D2 dissection involves the question as to whether D2 dissection provides better locoregional control and longer-term survival than D1 dissection and is much more complicated. The comparison of the impact of D2 and D1 dissections on long-term survival is confounded by several factors, but most important is the so-called Will Rogers phenomenon. With a more extensive node dissection, tumor staging is more accurate.21 Therefore, in a comparison of D1 and D2 surgical techniques, the stage migration phenomenon leads to higher stage-by-stage survival rates but no true overall survival benefit with the more extensive D2 surgery.22 Thus, many western surgeons are skeptical about whether D2 dissection truly improves survival or simply improves tumor staging. The results of the European trials8,9 support this opinion. However, multiple, but nonrandomized, prospective or retrospective studies in Japan and Western specialized institutions suggest a clear, absolute survival advantage of D2 over D1 node dissection.4,5,17
In an effort to develop an objective method that might eliminate multiple factors and bias in the comparison of D1 and D2 techniques, a new concept for evaluating the impact of D2 dissection on long-term survival was formulated.23 Such factors involve stage migration, which cannot be avoided even in RCTs,24 surgeon volume,1620 completeness of D2 dissection,24,25 magnitude of absolute survival benefit expected and sample size needed to detect a significant difference,26 and eventual difference in biological behavior of gastric tumor in Western and Eastern patients.
In a prospectively defined protocol, evaluation of the effectiveness of D2 surgery focused on patients with pN2 node-positive cancer because the gain of D2 over D1 dissection is the additional removal of the extraperigastric N2 nodes around the celiac axis in the Japanese anatomical nodal classification system.27 In a prespecified analysis, the 5-year survival rate in the subgroup of patients with pN2 nodal stage after an R0, D2 dissection was 20%. The validation of an R0, D2 dissection was ensured by quality control of all clinical, surgical, and pathohistological findings.23 Since it was hypothesized that all patients with pN2 disease die after D1 dissection, the absolute survival benefit of D2 dissection was estimated to be 20% for N2 patients. However, the hypothesesthat (1) pN2 disease with no clinical evidence of distant metastasis at diagnosis is a localized disease and that appropriate local treatment, such as D2 lymph node dissection, may result in cure for some patients and that (2) all pN2 patients die after a D1 "R0" resection because of residual positive N2 nodesclearly require confirmation by other studies. Now, 5 years later, both hypotheses are confirmed by the 11-year survival findings in the Dutch trial.13,14 There were 20% long-term survivors among N2 patients who had undergone an R0 D2 dissection, versus 0% among those who had an "R0," D1 dissection.13,14 Another recent Italian study also confirmed the benefit of D2 dissection, demonstrating a 30% survival rate among N2 patients after a D2 dissection.28
According to the pN2 nodal-stage specific concept, the absolute long-term survival advantage of D2 dissection ranges between 20%13,14,23 in the Western world and 40% in Japan.4,5 This difference may partially reflect the higher expertise and surgeon volume of Japanese versus Western surgeons. Indeed, there is accumulating evidence that surgeons experience and skill play an important role not only in short-term outcomes1620 but also in long-term survival.25,26,29,30 A rational explanation is that high-volume surgeons can remove completely all lymph nodes around the superior border of the pancreas from the liver hilus to the celiac axis and to the splenic hilus (D3 dissection), ensuring minimal residual nodal disease in this area.
Nodal status prediction before or during surgery is inaccurate. Therefore, although the benefit of D2 dissection appears to be limited to N2 disease, all patients with curable disease, including those with N0 or N1 disease, should undergone extensive node dissection. Since pN2 disease accounts for 30% of all R0 cases,6,21,24 our estimates30 indicate an overall absolute survival benefit of approximately 6% (30% N2 incidence; 20% N2 survivors).13,14,23 This survival gain may be larger if micrometastatic disease in N2 nodes that are left behind after a D1 dissection have a prognostic impact,17 but this is presently unknown.30
The survival benefit of D2 surgery seems to be particularly enormous for patients with serosa-negative stage cancer,12 because most of them have a low risk of developing distant or peritoneal recurrence and therefore mostly benefit from adequate D2 surgical treatment.31 If the overall benefit of D2 dissection is truly only 6%, then it is unlikely that the two ongoing RCTs,10,11 with small sample sizes of 162 and 221 enrolled patients, will have the power to detect a significant difference in overall survival between D1 and D2 groups. In this case, these trials should also be focused on comparison of survival of patients with pN2 disease after D1 and D2 dissection. It should be noted, however, that following the strict criteria of evidence-based medicine, further confirmation of pN2 nodal-stage-specific benefit of D2 dissection is required. Specific pN2 subgroup analysis was not a primary or secondary endpoint in the initial design of the Dutch trial.
With estimations of an overall survival benefit of 6% with D2 dissection and a marginal (approximately 2%5%) but significant survival benefit with adjuvant treatment, according to recent meta-analyses,32 it appears that both D2 surgery and adjuvant perioperative treatment warrant research funding for the establishment of a multimodality therapy. Despite the encouraging results of the Intergroup study (INT-0116)2 involving chemoradiation after limited D0/D1 resection, a larger survival benefit may be expected with D2 surgery alone.33 Promising data have emerged from several new phase II studies, which, on the basis of the chemoradiotherapy response observed in the Intergroup study, are investigating whether using newer, more effective agents such as cisplatin and taxanes and similar preirradiation and/or postirradiation regimens may lead to improved survival. There are also encouraging preliminary results from ongoing phase III RCTs (MAGIC trial, EORTC trial, Swiss trial) with preoperative and postoperative adjuvant treatment and promising findings with the use of intraperitoneal chemotherapy for reducing peritoneal dissemination after curative D2 surgery, 34 but toxicity is the major challenge. All these data warrant research funding for the establishment of effective, safe adjuvant treatment.
In summary, the goal of treatment is to reduce the risk of recurrent disease. D2 dissection appears to be more effective than D0/D1 surgery in achieving this goal.5,35 Given the surgical undertreatment in the United States1 and the most recent proof of the safety of D2 dissection, priority should to be given toward increasing the incidence of appropriate D2 surgery. Through the training of young surgical oncologists in specialized institutions, this goal can soon be reached. Using safe D2 dissection as a platform might increase not only the survival rate but also the effectiveness of perioperative adjuvant treatment by minimizing residual disease.
The advent of microarrays has opened new avenues toward prediction of lymph node status and survival.36 The technique of sentinel node biopsy (SNB) also provides promise for node status prediction and guided lymphadenectomy.37 Prediction of both nodal status and clinical outcomes before initial treatment on the basis of molecular profiling of a tumor will guide clinical judgment about individual patients with regard to both extent of surgery and preoperative or postoperative use of adjuvant therapy.38 Many patients who are currently undertreated and overtreated in the West and East, respectively, will benefit from this treatment concept. Reports on exciting clinical research with DNA microarrays for gene expression profilebased prediction of outcomes are already available for breast cancer39,40 and other solid tumors.41 A combination of both conventional and cancer genomicsbased criteria will allow personalized management of gastric cancer.
In the future, multimodality therapy consisting of adequate surgery, conventional adjuvant treatment, and new targeted agents aimed at molecular abnormalities involved in the pathogenesis of cancer, while sparing normal cells and avoiding the toxic side effects common with chemotherapy and radiotherapy, will improve outcomes for patients. There is already a growing list of targeted agents approved by the U.S. Food and Drug Administration for the treatment of lung cancer (gefitinib [Iressa]) and breast cancer (trastuzumab [Herceptin]), and there is hope that such novel agents will also be developed for the treatment of gastric cancer.
Until then, surgery with adequate complete resection of both the primary tumor and the positive lymph nodes will continue to be the mainstay of treatment for patients with curatively resectable gastric cancer.
Received for publication May 3, 2004. Accepted for publication May 25, 2004.
REFERENCES
This article has been cited by other articles:
![]() |
E. Briasoulis, D. Ziogas, and M. Fatouros Prophylactic Surgery in the Complex Decision-Making Management of BRCA Mutation Carriers Ann. Surg. Oncol., June 1, 2008; 15(6): 1788 - 1790. [Full Text] [PDF] |
||||
![]() |
T. Liakakos Minimal Residual Disease in Breast Cancer: Can It Be Used as a Prognostic Marker? Ann. Surg. Oncol., June 1, 2008; 15(6): 1793 - 1794. [Full Text] [PDF] |
||||
![]() |
M. Fatouros and D. Ziogas Controversy in the Treatment of Gastric Cancer Ann. Surg. Oncol., June 1, 2008; 15(6): 1795 - 1797. [Full Text] [PDF] |
||||
![]() |
T. Liakakos and E. G. Lykoudis Local Control, Aggressive Surgery, and Overall Survival for Breast Cancer Ann. Surg. Oncol., May 1, 2008; 15(5): 1544 - 1546. [Full Text] [PDF] |
||||
![]() |
T. Liakakos and D. H. Roukos More Controversy than Ever - Challenges and Promises Towards Personalized Treatment of Gastric Cancer Ann. Surg. Oncol., April 1, 2008; 15(4): 956 - 960. [Full Text] [PDF] |
||||
![]() |
T. Liakakos Laparoscopic Gastrectomy: Feasibility, Safety and Efficacy Ann. Surg. Oncol., April 1, 2008; 15(4): 1249 - 1250. [Full Text] [PDF] |
||||
![]() |
E. Hanisch, D. Ziogas, D. Roukos, and C. Hottenrott Advances in Laparoscopic Gastrectomy Expand Clinical Use Ann. Surg. Oncol., April 1, 2008; 15(4): 1251 - 1252. [Full Text] [PDF] |
||||
![]() |
H. Harissis and M. Fatouros Nodal Micrometastases Status--Will It Influence Decisions on Surgical and Adjuvant Treatment for Breast Cancer? Ann. Surg. Oncol., April 1, 2008; 15(4): 1256 - 1257. [Full Text] [PDF] |
||||
![]() |
E. Lykoudis, N. Xeropotamos, D. Ziogas, and M. Fatouros Breast Conservation Therapy: Multiple Reexcisions or Subcutaneous and Nipple-Sparing Mastectomy? Ann. Surg. Oncol., March 1, 2008; 15(3): 943 - 944. [Full Text] [PDF] |
||||
![]() |
L. Theodore Reexcisions in Breast-Conserving Surgery for Breast Cancer: Can They Be Avoided? Ann. Surg. Oncol., March 1, 2008; 15(3): 945 - 946. [Full Text] [PDF] |
||||
![]() |
T. Liakakos and G. Baltogiannis Reducing Local Recurrence after Breast-Conserving Surgery for Breast Cancer Ann. Surg. Oncol., March 1, 2008; 15(3): 949 - 950. [Full Text] [PDF] |
||||
![]() |
T. Liakakos Peritoneal Recurrence for Gastric Cancer: Can It Be Prevented? Ann. Surg. Oncol., January 1, 2008; 15(1): 382 - 382. [Full Text] [PDF] |
||||
![]() |
T. Liakakos and D. H. Roukos Does Lymphadenectomy Improve Survival of Patients with Solid Tumors? Ann. Surg. Oncol., January 1, 2008; 15(1): 385 - 385. [Full Text] [PDF] |
||||
![]() |
D. Ziogas, G. Baltogiannis, and M. Fatouros Lymphadenectomy in Surgical Oncology Ann. Surg. Oncol., January 1, 2008; 15(1): 386 - 387. [Full Text] [PDF] |
||||
![]() |
E. Briasoulis, M. Fatouros, and D. H. Roukos Level I Evidence in Support of Perioperative Chemotherapy for Operable Gastric Cancer: Sufficient for Wide Clinical Use? Ann. Surg. Oncol., October 1, 2007; 14(10): 2691 - 2695. [Full Text] [PDF] |
||||
![]() |
W.-J. Lee No Therapeutic Effect of Extended Lymph Node Dissection for Gastric Cancer J. Clin. Oncol., March 1, 2005; 23(7): 1592 - 1593. [Full Text] [PDF] |
||||
![]() |
W.-J. Lee, A. M. Kappas, M. Fatouros, and D. H. Roukos It Is Still Not the Time to Change Surgical Strategy for Gastric Cancer Ann. Surg. Oncol., February 1, 2005; 12(2): 194 - 196. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |