It is recommended that perioperative chemotherapy with an ECF-like schedule is offered to patients with a resectable gastric carcinoma (more than stage I) who are eligible in terms of condition and comorbidity.
In situations in which preoperative chemotherapy has not been administered, adjuvant chemoradiotherapy after resection with curative intent may be considered in case of a pT3-4 and/or N+ disease or after an R1 resection.
Surgery is the only curative treatment method for gastric carcinoma, but nevertheless many patients die after surgery with curative intent, due to a recurrence or metastases of their disease. The high risk of recurrence of the disease after surgery has lead to research on treatments other than surgery to reduce the chance of recurrence and/or metastasis and therefore increase the chance of survival. In this chapter, additional treatment modalities besides curative surgery are discussed.
Neoadjuvant treatments Neoadjuvant chemotherapy The goal of neoadjuvant (preoperative) chemotherapy is to treat any microscopically small distant metastasis present and possibly also to reduce the size of the primary tumour to increase the chance of a radical (microscopically complete) resection. A benefit of chemotherapy prior to surgery is a greater tolerance to the treatment because surgery-related problems/complications can play a role directly after surgery, making chemotherapy impossible or delaying chemotherapy at that moment 54. However, there may also be disadvantages to neoadjuvant therapy. It may result in overtreatment of patients with an early form of gastric carcinoma (stage 0 and I). These patients run the risk of unnecessary morbidity, which may reduce the success of surgery 189.
Three systematic reviews 16361100and an RCT 88 reported on neoadjuvant chemotherapy; referring to a total of 6 RCT's. Of these, 2 are relevant for this question and conducted in Western countries: the English MAGIC trial (see perioperative chemotherapy) and the RCT of the Dutch Gastric Cancer Group 88.
The Dutch Gastric Cancer Group (2004) researched the effect of preoperative 5-Fluorouracil, doxorubicin and methotrexate (FAMTX). While the 5-year survival in the chemotherapy group was lower than in the control group, 21% versus 34% respectively, the difference was not significant (p=0.17). Furthermore, this study was ended early due to insufficient inclusion, so the power of this study is small.
Neoadjuvant radiotherapy One systematic review researched the effect of neoadjuvant radiotherapy on survival in patients with curative surgery 68. Four RCT's with 832 patients were incorporated. The largest study with 370 patients was from China and only included patients with a cardia carcinoma. The remaining three studies were from Russia and Ukraine and were of moderate quality. Results from the meta-analysis showed that both the 3-year mortality and 5-year mortality were lower in the group of patients receiving preoperative radiotherapy (3-year mortality OR 0.57 (95% BI 0.43-0.76); 5-year mortality OR 0.62 (95% BI 0.46-0.84).
Neoadjuvant chemo-radiotherapy Randomised studies on neoadjuvant chemoradiotherapy have not been found. However, a positive effect on pathologic response has been suggested in prospective patient series 4.
Perioperative chemotherapy The largest randomised study in which chemotherapy before and after surgery was researched is the British MAGIC study 47. In this study, 503 patients were divided randomly in a perioperative chemotherapy group (n=250) and a group undergoing surgery only (n=253). Chemotherapy consisted of 3 pre- and 3 postoperative cycles with epirubicin, cisplatin and 5-fluorouracil (ECF). After a median follow-up of four years, 149 patients from the chemotherapy group had died and 170 from the group only undergoing surgery. The HR for general survival of the chemotherapy group in comparison to the surgery-only group was 0.75 (95% BI 0.60-0.93). The HR for disease-free survival was 0.66 (95% BI 0.53-0.81). The removed tumours were smaller in the chemotherapy group than in the group with only surgery (3 cm versus 5 cm, p<0.001). It must be noted that only 42% of the patients were able to complete the planned courses, mainly due to the fact that the postoperative chemotherapy was heavy.
Adjuvant treatments Adjuvant chemotherapy The goal of adjuvant therapy is to reduce the chance of recurrence and metastasis after surgery. There are substantially more studies on the efficacy of adjuvant therapy than of neoadjuvant therapy. Six meta-analyses 906013910094244 reported the effects of adjuvant chemotherapy. The methodological quality of the studies included varied from moderate to high. The results of these meta-analyses are shown in table 1. All studies outlined results with the help of an OR. However, because the OR overestimates the clinical effect in populations with a high basic risk, the absolute risk reduction is of high importance. The absolute reduction in risk by adjuvant chemotherapy is estimated at 2 to 4% (see table 1). A seventh meta-analysis was identified but could not be obtained 168. On the basis of the results of the abstract, this meta-analysis does not change the conclusion.
In the most recent meta-analysis by Zhao, subgroup analysis was also performed and the effect in the Western subgroup was considerably smaller than in the Asian subgroup (OR 0.92 versus 0.71) 244. Remarkably, the effect appeared to be the greatest with mitomycin-containing schedules.
Table 1 Results of meta-analyses on the effect of adjuvant chemotherapy on survival of patients with gastric carcinoma
Twelve recent randomised studies examined the effect of adjuvant chemotherapy; 7 were European studies. Five of these compared the effects of adjuvant chemotherapy with that of surgery only 54241316042. These studies included between 209 and 397 patients. While all studies show a reduction in the risk of mortality within 5 years, the differences found are not significant (see table 2). Furthermore, the absolute risk reduction is limited to a maximum of 4.7%. The HR for 5-year survival varies from 0.74 to 0.98 and for disease-free survival from 0.70 to 0.9.
Two RCT's compared two types of chemotherapy; the first examined a treatment with fluorouracil (FU), irinotecan (CPT-11) and docetaxel plus cisplatin (combination) with a treatment consisting of mitomycin only in a group of 166 patients 55. The 3-year disease-free survival was 67.4% in the group with the combination chemotherapy compared to 50.2% in the mono-chemotherapy group (p=0.045). The 3-year disease-free survival was 62.4% in the group with the combination chemotherapy compared to 50.2% in the mono-chemotherapy group (p=0.045). The treatment with combination chemotherapy was feasible and well tolerated, but a larger study is necessary to demonstrate the possible advantage of one treatment over the other. The other RCT compared the effects of the PELFw schedule (cisplatin, epidoxorubicin and glutathione, supported by filgrastim) with that of 5-fluorouracil and leucovorin in a group of 397 patients with a high risk of recurrence 36. The 5-year survival was 52% in the PELFw arm and 50% in the 5-FU/LV arm. The PELFw schedule did not reduce the risk of death (HR 0.95, 95% BI 0.70-1.29). Less than 10% of the patients in both groups experienced severe side effects (grade 3 or 4). The treatment was moderately tolerated; the number of patients receiving the treatment as planned was only 9.4% in the PELFw group and 43% in the 5-FU/LV group. The majority of patients required dose reduction or larger rest periods.
Table 2 Results of recent European RCT's on the effect of postoperative chemotherapy in comparison with surgery only in patients with gastric carcinoma
More positive results are achieved in Asian studies; 4 RCT's were found 18415615838. Sakuramoto treated 529 patients after surgery with S-1 oral chemotherapy (a 5-FU derivative) and 530 patients with surgical treatment only 184. After three years, 80.1% (95% BI 76.1-84.0%) of the patients with chemotherapy were still alive. This was 70.1% (65.5-74.6%) for the group undergoing surgical treatment only. The HR for mortality in the S-1 group, in comparison with the surgery group, was 0.68 (95% BI 0.52- 0.87). Nakajima also found chemotherapy to provide an advantage after curative surgical treatment in comparison with a group that had only undergone surgery 156. The HR for mortality in the group treated with uracil-tegafur, in comparison with the group that had only undergone surgery, was 0.48 (95% BI 0.26- -0.89). Sakuramoto and Nakajima conclude that treatment with chemotherapy after surgery leads to a higher chance of survival. Nashimoto did not find a significant effect of treatment with mitomycin, fluorouracil (FU) and cytarabine 158. The 5-year survival in the chemotherapy group was higher (91.2% with 95% BI 86.2-96.2%) than in the surgery group (86.1% 95% BI 79.9-92.2%) but this difference was not significant (p=0.13).
Chang, the only Asian RCT of moderate methodological quality, compared three different chemotherapy treatments 38. They found that the addition of MMC (mitomycin C) and/or doxorubicin to 5-fluorouracil only did not make a difference in relation to general survival after 5 years (p=0.97) and disease-free survival after 5 years (p=0.83).
The effect of intra-arterial chemotherapy was examined in a small RCT 211. This treatment, compared to surgery only, did not improve survival in a group of patients with locally advanced tumours (survival after 5 years: chemotherapy 52% and control 54%, p=0.90).
Adjuvant peritoneal chemotherapy Intraperitoneal chemotherapy could be a possible adjuvant treatment for gastric carcinoma because recurrences often develop in the peritoneal cavity. The effect of adjuvant peritoneal chemotherapy versus surgery alone was researched in two meta-analyses of RCT's 235232. Yan included 13RCT's and pooled the results for every type of intervention separately 235. HR's for survival varied from 0.45 (95% BI 0.29-0.68) for ‘hyperthermic and early chemotherapy' to 0.89 (95% BI 0.51-1.55) for ‘delayed postoperative chemotherapy'. Results for 2 of the 5 types of interventions were significant in favour of peritoneal chemotherapy. Xu included 11 RCT's and reported an OR for risk of mortality of 0.51 (95% BI 0.40-0.65) in favour of peritoneal chemotherapy232. Both meta-analyses suggest a possible positive effect of intraperitoneal adjuvant chemotherapy, but all studies except 2 RCT's are small and have been conducted in Asian countries. Two small Austrian RCT´s included in both meta-analyses did not show an effect of this form of chemotherapy on general survival (HR 1.00, 95% BI 0.55-1.82 (Rosen 1998) and 0.89, 95% BI 0.51-1.55 191.
Adjuvant radiotherapy Hallissey compared postoperative chemotherapy and radiotherapy with surgery only in a three-arm study with 436 patients, but did not find a difference in 5-year survival between patients with only surgical treatment, radiotherapy or chemotherapy, respectively 17%, 11% and 17% 82.
Adjuvant chemoradiotherapy A meta-analysis 68 and a systematic review 162 reported on the effects of adjuvant chemoradiotherapy. Five RCT's with 868 patients were included in the meta-analysis. Four small RCT's were published more than 15 years ago. The last and largest study was the SWOG trial, which was also incorporated in the systematic review 133. In this RCT by the Southwest Oncology Group (SWOG 9008)(n=556), a significant improvement in the disease-free period and survival after surgery was found in patients treated with chemoradiation 133. Macdonald used fluorouracil and leucovorin as chemotherapy and the radiation consisted of 25 fractions of 1.8 Gy, 5 days per week for a period of five weeks, for a total dose of 45 Gy 133. The HR for mortality associated with surgery only in comparison with chemoradiation was 1.35 (95% BI 1.09-1.66) and for recurrence of the disease after the surgery 1.52 (95% BI 1.23-1.86). However, the study has been criticised because of the fact that only 10% of the patients received the prescribed D2 lymph node dissection. The advantage of postoperative chemoradiation may then be explained by a form of correction of suboptimal surgery. In a small subgroup analysis, Macdonald did not find an advantage of chemoradiation in patients who were undergoing a D2 resection 133. An observational study by Kim (n=990) does suggest a clinical advantage of chemoradiotherapy after a D2 dissection 111.
Adjuvant chemo-immunotherapy A meta-analysis 183, a systematic review 61 and 2 RCT´s 190173researched the effects of adjuvant chemo-immunotherapy. The majority of studies were conducted in Asian countries. Only 6 RCT's were analysed in a meta-analysis by Sakamoto 183. These results are therefore not taken into consideration. In the systematic review by Earle, three Western studies were incorporated; two European studies found no difference between the effects of only surgery and chemo-immunotherapy, while a third Polish study did find chemo-immunotherapy to have a positive effect.
In a recent Polish RCT of moderate quality, the effects of adjuvant chemo-immunotherapy were compared with that of adjuvant chemotherapy and only curative surgery in 156 patients. The chemo-immunotherapy increased the chance of survival compared to the chemotherapy group (p=0.037) and the control group (p<0.0006). In these studies, the three groups were not comparable in all characteristics; furthermore it must be noted that the survival in the surgery only group was very poor, amongst other things, in comparison with many other studies.
Clear conclusions cannot be drawn from literature about the value of neoadjuvant chemotherapy. Level 2: B Oehler 2006163; Earle 200261; Januger 2002100
It is plausible that neoadjuvant radiotherapy increases the chance of survival in comparison with surgery. Level 2: B Fiorica 200768
A conclusion cannot be drawn from literature about the efficacy of neoadjuvant chemo-radiotherapy. Level 3: B Ajani 20044
It is plausible that perioperative treatment with chemotherapy has a positive effect on survival and reduces the size of the tumour. Level 2: A2 Cunningham 200647
While all studies and meta-analyses showed a positive trend, it has not been demonstrated that adjuvant chemotherapy significantly improves the chance of survival (3 A2 RCT´s and 2 B). Furthermore, clinical advantage is limited with the chemotherapy schedules used in the studies. Level 1: A2 Zhao 2008244; Januger 2002100; Mari 2000139
It is plausible that intraperitoneal chemotherapy does not improve survival. Level 2: B Rosen 1998176; Sautner 1994191
It is plausible that adjuvant radiotherapy does not improve the chance of survival in comparison with surgery only. Level 2: A2 Hallissey 1994
It is plausible that adjuvant chemoradiotherapy reduces the chance of mortality or chance of recurrence in patients with gastric carcinoma. Level 2: A2 Macdonald 2001133
A clear conclusion cannot be drawn from literature about the efficacy of adjuvant immuno-chemotherapy. Level 3: B Sakamoto 2002183; Earle 200261
Most studies have been performed on adjuvant chemotherapy. These generally show a small difference and one that is clinically of little relevance in Western studies. These Western studies are small in size and are furthermore conducted with chemotherapy that is less effective in the metastasised setting. A greater effect is seen in Japanese and other Asian studies, but it is uncertain if these results can be extrapolated to the Western situation.
The SWOG study on adjuvant chemo-radiotherapy shows clear survival advantage, but it is questionable if these results also apply within a better surgical setting. This should first be researched more closely.
Perioperative chemotherapy with more effective cytostatics appear to give the best results, also in the setting of Western patients, taking adequate surgery and also the existing but limited efficacy of adjuvant chemotherapy into consideration. The results from the study by Cunningham are confirmed by the data in the abstract of Boige on the ASCO in 2007, in which chemotherapy was administered with cisplatin and 5-FU 47.
It is unclear to what extent the results of neoadjuvant radiotherapy may be extrapolated to Western patients because these are based on a large Chinese study with cardia tumours, two Russian studies and a study from the Ukraine.
