This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kelsen, D. P. Search for Related Content PubMed PubMed Citation Articles by Kelsen, D. P.

Postoperative Adjuvant Chemoradiation Therapy for Patients With Resected Gastric Cancer: Intergroup 116

From the Memorial Sloan-Kettering Cancer Center, New York, NY.

Address reprint requests to David P. Kelsen, MD, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021-6007; email kelsend{at}mskcc.org

WITH RARE exceptions, a complete surgical resection is necessary to offer potentially curative therapy to patients with adenocarcinoma of the stomach. However, even after a complete resection with negative margins (an R0 resection), many patients, particularly those with more locally advanced tumors, will experience recurrence. In the face of recurrence, in general only palliative therapy is possible. Patterns of recurrence have been studied in autopsy series, in a few second-look laparotomy series, and using clinical means.^1-3 For stages II, IIIA, and IIIB cancers, the risk of local-regional as well as distant failure is high.

Attempts to improve cure rates by using postoperative systemic therapy have been extensively studied in the past two decades. Many individual trials were seriously underpowered; however, several recent meta-analyses suggest that any impact on outcome is modest.⁴ Recently, the Southwest Oncology Group, as the lead institution for the United States Gastrointestinal Intergroup, reported the preliminary results of Intergroup 116.⁵ This trial tested the hypothesis that postoperative chemoradiation therapy would decrease the risk of recurrence and thereby increase the cure rate for patients with resected gastric cancer.

The Intergroup 116 study population involved patients who had an R0 resection of a newly diagnosed gastric cancer. How widely applicable are the results of this trial? That is, how many United States patients would be affected by the results of Intergroup 116? Wanebo et al⁶ reported the results of an American College of Surgeons Patient Care Study that involved 18,365 patients. In this report, data from patients treated in 1982 and 1987 were analyzed. Overall, 72% of recently diagnosed patients underwent an exploratory laparotomy. Fifty-seven percent of patients had any type of resection, but importantly, only 31% of patients had an R0 resection. These data and other similar series indicate that only one third of all patients newly diagnosed with gastric cancer can undergo potentially curative surgery. Intergroup 116 therefore applies to approximately 7,000 American patients per year.

With this in mind, what were the strengths and weaknesses of Intergroup 116? The strengths are clear: Intergroup 116 is a prospective, random assignment trial. The study was both multicenter and multigroup. The trial was adequately powered to address the major end points. It should be noted that during the study, total accrual was increased to 550 patients, above the initial projections, not because an interim analysis suggested the need for more patients, but because accrual rates were actually somewhat greater than projection. Finally, the study has a surgery-only control arm.

There are several potential weaknesses of Intergroup 116. It is possible that the outcome of the surgery-only arm was exceptionally poor. Secondly, there may have been too few patients in any given subgroup (because the trial included patients with stages IB, II, IIIA, IIIB, and IV disease), so that drawing meaningful conclusions about the effectiveness of therapy by stage may be difficult. Lastly, and perhaps most importantly, this treatment plan involved adding a second local-regional modality, radiation therapy, to the primary local-regional modality, surgery. As discussed below, serious questions can be raised as to whether an adequate operation was performed in many of the patients who entered this trial.

The first potential weakness of the study is that the surgery-only control arm may have done exceptionally poorly and that the study was a positive one only because of this factor. The control arm of Intergroup 116 involved 275 patients who had undergone an R0 resection. Their 3-year survival at the time of the preliminary analysis was 40%. In the American College of Surgeons Patient Care Study described earlier, 3,044 patients who had R0 resections were included.⁶ Their 3-year survival was 42%. These data suggest that the surgery-only control arm in Intergroup 116 did not have an exceptionally poor outcome, in that the survival reported is comparable to other United States series. However, as will also be described below, many American patients had undergone less than a recommended (a D1) lymph node dissection.

Most of the patients in Intergroup 116 had stage IIIA or IIIB disease. Only 36 patients had stage IB tumors. Hundahl et al⁷ recently reported the results of the National Cancer Database for gastric cancer. Patients with stage IB disease (T1N1M0 or T2N0M0) had a relatively good 3-year prognosis. Because so few patients with stage IB disease were treated in this study (18 in each arm) and because the prognosis is relatively favorable for patients with stage IB disease, the applicability of postoperative chemoradiation to this group is not clear.

In the preliminary report, it was noted that 54% of patients in Intergroup 116 had undergone less than an accepted cancer operation. That is, the lymphadenectomy, if performed, did not conform to at least a D1 dissection. Lymphadenectomy for gastric cancer has been well described.⁸ A D1 dissection is the removal of perigastric lymph node stations along the greater and lesser curvatures. The more extensive D2 operation is the removal of both perigastric lymph node stations and regional lymph node stations (extended lymphadenectomy). Estes et al⁵ have already reported the preliminary surgical quality data from Intergroup 116. A D0 lymphadenectomy is a dissection in which fewer than (or none of) the six perigastric lymph node stations included in the D1 procedure were removed.

Two recent large-scale random assignment trials have evaluated the survival effect of a recommended standard D1 dissection versus the more extensive D2 dissection. The Dutch Gastric Cancer Group studied 589 patients who underwent an R0 resection and were randomly assigned to have a D1 or D2 operation.⁹ Three-year survival was 56% versus 58%, not statistically significantly different. Operative morbidity and mortality were higher for the D2 group. In a similar study, the Medical Research Council also randomized patients to a D1 versus a D2 dissection.⁹ In this study, which involved 400 patients, 3-year survival was 50% versus 57%. Again, the D2 dissection was not statistically superior.

Although neither of these two studies demonstrated an improvement for more extended D2 lymph node dissections (which are commonly performed in Japan and in certain centers in the United States and Europe), the results of the control arm of these trials (the group that underwent a D1 dissection) were equivalent or superior to those of the experimental arm of Intergroup 116, in which 281 patients who had undergone R0 resections received postoperative chemoradiation therapy. The 3-year survival for the chemoradiation group was 50%. The Dutch Gastric Cancer Group and Medical Research Council studies suggest that patients who have at least a minimal recommended lymph node dissection have a similar outcome to patients who receive chemoradiation therapy.

Why do so many American patients not receive a D1 dissection? There are few if any data to address the increased risk of a D1 versus D0 dissection. Although the random assignment trials described above do not indicate a survival advantage for a D2 dissection, careful removal of D1 perigastric lymph nodes has been accepted as a standard operation by most academic surgical oncologists, and the risk of this procedure in the hands of well-trained surgeons seems to be low. Therefore, strong efforts should be made to encourage surgeons to adopt the general practice of performing a D1 dissection as part of a curative gastric cancer operation, as this may improve survival outcome compared with incomplete lymph node dissection.

Several cautionary notes should be raised about Intergroup 116. Radiation therapy treatment planning as used in this study requires extremely careful attention to detail. Thirty-two percent of patients in Intergroup 116 required a change in radiation planning after central review. It is important to note that in Intergroup 116, radiation therapy quality control was very high, with a single radiation therapist reviewing all treatment plans. Had appropriate changes not been performed, serious, including life-threatening toxicity, may have occurred in up to 10% of patients. Furthermore, serious errors in the target volume’s ability to cover all areas of disease might have been present in up to one fifth of patients. Second, as noted above, too few patients with stage IB disease were treated to make meaningful comments regarding the effectiveness of this therapy.

Intergroup 116 was a well-designed trial with a control arm that had comparable overall survival of 3 years similar to that of other United States series. With careful planning, treatment-related toxicity is acceptable. The preliminary data indicates a clinically meaningful survival advantage to the investigational chemoradiation therapy arm. Thus, if the final report confirms the preliminary analysis, and in particular, if further analysis indicates that D1 dissection does not in itself cause a marked survival advantage, the use of chemoradiation therapy after surgical resection for patients with stages II, IIIA, and IIIB disease should be considered a standard care option for patients who have undergone resections with these stages of gastric cancer.

Why was Intergroup 116 a positive trial? Was it because of better local control resulting from chemoradiation therapy? Or because of better control of distant disease from the systemic component? Although the site of first failure is difficult to accurately assess in resected gastric cancer, local failure was seen in 29% of control patients versus 19% of those who received chemoradiation. Distant failure was seen in 18% of control patients versus 35% of patients who received chemoradiation. This failure pattern data, if accurate, suggest that the positive result of Intergroup 116 may have primarily resulted from the use of radiation in chemoradiation therapy.

These data support the undertaking of several future studies to build on the results of Intergroup 116. Better surgery using a formal D1 dissection may decrease the need for radiation, or when combined with chemoradiation, may further improve on local control as seen in Intergroup 116. Better systemic chemotherapy may improve the control of both regional and distant disease. Clearly, because only one third of all United States patients currently undergo a R0 resection, it is urgent to increase the number of patients who might undergo potentially curative operation. Neoadjuvant chemotherapy or chemoradiation is under active study to address this point. Postoperative intraperitoneal therapy is also being studied to decrease the risk of peritoneal failure. There are data to suggest that better systemic chemotherapy may be available in the near future. Several new active agents have been identified in this disease, including the taxanes, docetaxel and paclitaxel, and irinotecan. These agents are now being studied in combination chemotherapy regimens both in the adjuvant and neoadjuvant approach.

REFERENCES

1. Gunderson LL, Sosin H: Adenocarcinoma of the stomach: Areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 8: 1-11, 1992

2. Ha HK, Kim HH, Kim HS, et al: Local recurrence after surgery for gastric carcinoma: CT findings. AJR Am J Roentgenol 161: 975-977, 1993[Abstract/Free Full Text]

3. Orita H, Matsusaka T, Wakasugi K, et al: Clinicopathologic evaluation of recurrence in early gastric cancer. Surg Today 22: 19-23, 1992[Medline]

4. Earl CC, Maroun JA: Adjuvant chemotherapy after curative resection for gastric cancer: Revisiting a meta-analysis of randomized trials. Proc Am Soc Clin Oncol 17: 263a, 1998 (abstr 100)

5. Estes NC, Macdonald JS, Touijer K, et al: Inadequate documentation and resection for gastric cancer in the United States: A preliminary report. Am Surg 64: 680-685, 1998[Medline]

6. Wanebo H, Kennedy BJ, Chmiel J, et al: Cancer of the stomach: A patient care study by the American College of Surgeons. Ann Surg 218: 583-592, 1993[Medline]

7. Hundahl SA, Menck HR, Mansour EG, et al: The National Cancer Database report on gastric carcinoma. Cancer 80: 2333-2341, 1997[Medline]

8. Macdonad JS, Smalley S, Benedetti J, et al: Postoperative combined radiation and chemotherapy improves disease-free survival (DFS) and overall survival (OS) in resected adenocarcinomas of the stomach and G.E. junction. Results of Intergroup Study INT-0116 (SWOG-9008). Proc Am Soc Clin Oncol 19: 1a, 2000 (abstr 1)

9. Bonenkamp JJ, Herman S, Sasaki M, et al: Extended lymphnode dissection for gastric cancer. N Engl J Med 340: 908-914, 1999[Abstract/Free Full Text]

This article has been cited by other articles:

				B Brenner, M. Shah, M. Karpeh, M Gonen, M. Brennan, D. Coit, D. Klimstra, L. Tang, and D. Kelsen A phase II trial of neoadjuvant cisplatin-fluorouracil followed by postoperative intraperitoneal floxuridine-leucovorin in patients with locally advanced gastric cancer Ann. Onc., September 1, 2006; 17(9): 1404 - 1411. [Abstract] [Full Text] [PDF]

				R. E. Schwarz and D. P. Kelsen Postoperative Adjuvant Chemoradiation Therapy for Patients With Resected Gastric Cancer: Intergroup 116 J. Clin. Oncol., March 15, 2001; 19(6): 1879 - 1880. [Full Text] [PDF]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kelsen, D. P. Search for Related Content PubMed PubMed Citation Articles by Kelsen, D. P.