Abstract
Background: The gold standard of adjuvant treatment after surgical resection of adenocarcinoma of the stomach or gastroesophageal junction (GEJ) is chemoradiotherapy. We retrospectively evaluated chemotherapy without radiotherapy in stomach and GEJ adenocarcinoma, using a combination of etoposide, adriamycin and cisplatin (modified EAP). Patients and Methods: Sixty-five patients with completely resected gastric or GEJ adenocarcinoma and positive regional lymph nodes were treated with modified EAP over an 8-year period. Results: Recurrent disease was diagnosed in 38/58 (69%) patients evaluable for analysis. Only two (5%) had locoregional recurrence. The main toxicity was hematological, with 22 (34%) patients developing neutropenic fever and 12 (18%) anemia requiring blood transfusion. The median survival for the entire group was 20 months, with a median time to recurrence of 11 months. Seventeen (26%) patients are alive for a median of 7+ years, with no evidence of recurrent disease. Conclusion: Our data cast doubt on the benefit of radiotherapy adjuvant to chemotherapy.
The 5-year survival rate for all patients diagnosed with gastric carcinoma remains less than 18%, and reaches only 20-30%, even in those who have undergone resection (1, 2). Because the majority of patients who undergo resection experience relapse and ultimately die of their disease, considerable attention has been paid to neoadjuvant and adjuvant strategies to improve surgical outcome. Meta-analysis of all randomized clinical trials of adjuvant chemotherapy before 2000 revealed a small survival benefit for chemotherapy treatment compared to surgery alone, but with many limitations that do not allow the drawing of a firm conclusion (3).
In 2001, the Intergroup-0116 Study (INT-0116), comparing postoperative chemoradiotherapy to observation alone in patients who had undergone potentially curative resection, concluded that this treatment should be considered standard care for all patients at high risk for recurrence, based on improvements in relapse-free survival and overall survival (4). The main benefit of adjuvant therapy in that study was a reduction in the local failure rate rather than preventing the development of distal metastases. The benefit was mainly attributed to the adjuvant radiotherapy which was potentiated by chemotherapy (5).
A different approach, using preoperative chemotherapy without radiotherapy, has proven to be effective in two recent studies (6, 7). Chemotherapy included epirubicin, continuous 5-fluorouracil, and cisplatin (ECF) in the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial (7) and continuous 5-fluorouracil and cisplatin (CF) in the FFCD-9703 trial (6). Both studies confirmed the efficacy of this approach, with a higher rate of R0 resections, and improvement in disease-free survival (DFS) and overall survival (OS). However, surgical resection as a first step in the treatment of gastric cancer remains common practice.
Between 1991 and 1999, the policy at our hospital was to treat patients with positive regional lymph nodes with completely resected adenocarcinoma of the stomach or the gastroesophageal junction (GEJ) by adjuvant chemotherapy, including adriamycin, etoposide and cisplatin (modified EAP), without radiotherapy. Modified EAP was chosen at that time as it was found to be highly active and associated with acceptable toxicity (8). We herein report our experience with this policy with an emphasis on the pattern of relapse, attempting to answer the question of whether adding radiotherapy could have improved the therapeutic outcome.
Patients and Methods
Between January 1991 and October 1999, patients who were referred to our centers following radical surgery for gastric or gastric-esophageal adenocarcinoma defined as complete resection with tumor-free surgical margins and who had regional lymph node involvement were routinely treated with modified EAP. As a rule, D1 resection was the standard. However, patients with D0 resection might have been included in this series also.
Patients with poor performance status (ECOG >2) and/or who had any medical conditions contraindicating the use of any of the three drugs included in the modified EAP regimen were excluded from this policy. Sixty-five consecutive eligible patients were treated with adjuvant modified EAP. All patients had normal serum creatinine and bilirubin, adequate bone marrow function (WBC count ≥4000/mm3, platelet count ≥100,000/mm3) and normal cardiac function. The EAP regimen consisted of etoposide 100 mg/m2, days 1-3; adriamycin 40 mg/m2, day 1; and cisplatin 27 mg/m2, days 1-3. Doses of all three drugs were reduced by 10% in all patients >65 years. Cycles were repeated every 21 days. Primary prophylaxis with granulocyte colony-stimulating factor (GCSF) was given from September 1995. Survival and recurrence-free survival were measured from the beginning of chemotherapy. The initial site of recurrence was recorded. Locoregional recurrence was defined as recurrence within what is considered a standard postoperative radiation field. Other recurrences were defined as distant.
Patients were followed and examined approximately every six months during the first two years. Imaging studies, endoscopy and determination of serum CEA and CA-19 were carried out at the discretion of the physician.
Results
The major characteristics of the 65 patients are presented in Table I. Most (91%) had T3 or T4 tumors and 62% had fewer than seven regional lymph nodes involved (N1). The primary tumor location was proximal (cardiac and/or GEJ) in 14% of the patients. Subtotal gastrectomy was the major operation, performed in about 2/3 of all cases.
The number of EAP cycles administered was 1-6 (median, 6). Forty-five (71%) patients received the full treatment of six cycles and only nine (13%) received three cycles or fewer. The reasons for withdrawal from the six-planned chemotherapy cycles were toxicity in seven (10%) patients, tumor progression in six (9%), patient refusal in three (5%), and other causes in three (5%). Only one patient received radiotherapy after chemotherapy, due to close surgical margins according to the pathological report.
The main treatment toxicity was hematological, especially before the standard use of primary GCSF prophylaxis; 22 (34%) patients developed neutropenic fever and 12 (18%) had anemia that required red blood cell transfusion. Only one patient had grade 4 thrombocytopenia with bleeding that required platelet transfusion. There was no case of drug-related mortality. Peripheral neuropathy was the main non-hematological toxicity encountered in 21 (32%) patients and was the reason for treatment withdrawal in two cases.
Ten patients died less than two years after the onset of chemotherapy, without documented evidence of disease progression; these patients were excluded from analysis of recurrence pattern. Three of these 10 patients died of other causes, without evidence of recurrent cancer, but the disease status was unknown at the time of death in the additional seven patients. Among 55 patients included in the analysis of recurrence, recurrent disease was diagnosed in 38 (69%). Median time for detection of recurrence was 11 months, with a range of 2-57 months. Sites of disease at the time relapse was first diagnosed are shown in Table II. Only two (5%) patients had initial locoregional recurrences, i.e. limited to sites included in the standard postoperative radiation field. One recurred in the anastomosis and the other in celiac lymph nodes. In the former, recurrence occurred in spite of adjuvant radiotherapy that had been given due to close surgical margins. Two patients recurred in the CNS: one had brain parenchymal metastasis and the other had meningeal spread.
Seventeen (26%) patients have remained alive and without evidence of recurrent malignancy for 4.5 to more than 12 years (median, 7+ years). The median survival for the entire group of 65 patients was 20 months.
Discussion
Our policy of using adjuvant chemotherapy with modified EAP and without radiotherapy started before publication of the Intergroup-0116 study. We used a chemotherapy regime based on the original EAP regime (9) but modified due to unacceptable toxicity associated with that combination (8). In our hands, the modified EAP regime was active in patients with advanced gastric adenocarcinoma (8) and was the standard chemotherapy for gastric cancer in our institutions during the study period. The patients population included in the current series represents a non-selective population of R0 resected patients with gastric and GEJ adenocarcinoma with regional lymph node involvement. Long-term follow-up indicates that about one-quarter of these patients might have been cured. However, it is impossible to draw any conclusion from our study regarding the impact of adjuvant modified EAP on outcome, as the study was not randomized. Nevertheless, our retrospective analysis does reveal that isolated local recurrence was very uncommon. Only 2/55 (4%) patients evaluable for analysis of relapse recurred in sites that would have been included in a standard postoperative radiation field (surgical anastomosis in the only patient who received radiotherapy due to close surgical margins, and celiac lymph nodes in the other). Therefore, the addition of radiotherapy would not have been expected to have a major impact on the therapeutic outcome of these patients.
In the Intergroup-0116 study, the main advantage of adjuvant therapy was to reduce the locoregional recurrence rate. In particular, local recurrence was recorded in 51/275 (19%) patients treated with surgery alone, compared to 23/281 (8%) treated with chemoradiotherapy, The rate of regional recurrence was also reduced from 127/275 (46%) in the surgery alone group to 78/281 (28%) in the chemoradiotherapy group, while distant metastases were recorded in similar numbers in both groups (12% and 14%, respectively). Therefore, the improvement in survival after chemoradiation observed in this study arises solely from better locoregional control (10) and could be attributed mainly to radiotherapy. The relatively inactive chemotherapy in that study, based on bolos 5-fluorouracil and leucovorin, probably had a radiopotentiating effect but failed to decrease the rate of distal metastases.
Characteristics of 65 patients treated with adjuvant EAP.
In this regard, it is worth mentioning a recent large randomized trial from Japan indicating that chemotherapy alone, using one year of treatment with S-1, an oral fluoropyrimidne, was found to be highly efficient as adjuvant chemotherapy for patients who underwent a D2 dissection for locally advanced gastric cancer (11). The local failure rate in the S1-treated group was very low (2.8% in the S1 and 1.3% in the surgery-alone arm) in spite of the fact that adjuvant radiotherapy was not used in that study.
Sites of recurrent disease in 38/55 evaluable patients.
The relatively high local failure rate observed in the Intergroup-0116 study in both the surgery-alone arm (19%) and the chemoradiotherapy group (8%), as well as in other series reported in older literature (12, 13) might result from inadequate surgery. Indeed, as reported by Macdonald et al. (4), more than half their patients underwent D0 resection. In the current study, as a rule, D1 resection was planned. Although we do not have data regarding the rate of D0 resection in our series, it is possible that this rate was much lower than that in the Intergroup-0116 study. The very low recurrence rate in the Japanese study in both the control and the chemotherapy arms could be explained by the fact that D2 resection was standard in that trial.
It should be mentioned that intensive chemotherapy might also reduce the rate of local failure, as demonstrated in the MAGIC trial, in which the local recurrence rate was reduced from 20.6% in the surgery-alone group to 14.6% in the chemotherapy group (7).
Recently, two reports using the Surveillance, Epidemiology, and End Results (SEER) database reported a survival improvement in stomach carcinoma patients after surgery recorded in the last few years (14, 15). According to these reports, the improvement is related to the addition of radiotherapy treatment after surgery, which became the standard of care after the INT 0116 trial. The role of adding chemotherapy was not reported, nor was the benefit of chemoradiotherapy over neoadjuvant chemotherapy.
In summary, the current study further suggests that isolated local recurrence in patients with completely resected adenocarcinoma of the stomach or GEJ treated with adjuvant intensive chemotherapy is uncommon. These results further question the role of routine adjuvant radiotherapy in these patients, which might be unnecessary and could impair patient tolerance to chemotherapy. If active chemotherapy is given in the adjuvant setting, adjuvant radiotherapy could probably be reserved for patients with less than D1 resection.
- Received June 28, 2008.
- Revision received December 1, 2008.
- Accepted December 8, 2008.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.