Treatment of Resectable Gastric Cancer: An Update on the Role of Radiation and Chemotherapy

Review Article

Treatment of Resectable Gastric Cancer: An Update on the Role of Radiation and Chemotherapy

Sarwat Ahmad^* and Nader Hanna
Department of Surgery, University of Maryland School of Medicine, USA

^*Corresponding author: Sarwat Ahmad, Department of Surgery, University of Maryland School of Medicine, 22 S. Greene St, Baltimore, MD 21201, USA

Published: 07 Dec, 2016
Cite this article as: Ahmad S, Hanna N. Treatment of Resectable Gastric Cancer: An Update on the Role of Radiation and Chemotherapy. Clin Surg. 2016; 1: 1223.

Abstract

The operative and perioperative management of gastric cancer has been an area of controversy and ongoing study in both Eastern and Western patient populations. The extent of lymphadenectomy varies across regions. Various combinations of chemotherapy and radiation in the neoadjuvant and adjuvant setting have been studied in large randomized controlled trials. In the West, two major trials, the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial and the SWOG/INT 0116 trial by Macdonald et al. have influenced perioperative therapy, but the efficacy of one versus the other has not yet been determined. Newer studies, such as the CRITICS trial, are investigating the potential benefits of neoadjuvant therapy in combination with adjuvant chemoradiation. This review describes the ongoing controversies in gastric cancer multimodal management and encompasses the evidence to date supporting current practices in the West.

Introduction

At an incidence of 950,000 new cases per year worldwide in 2012, gastric cancer is among the top five most common cancers and ranks as the 3rdand 4thleading cause of cancer death in men and women, respectively [1]. Regional variances persist, with Eastern Asia most heavily affected, partially due to differences in H.pylori infection, smoking, and diet [1]. While the incidence and mortality of gastric cancer has been declining overall, Western countries have seen arise in cancers of the gastric cardia, likely related to increased obesity and its comorbidities such as gastroesophageal reflux disease [2 and 3] as well as improved diagnostic modalities [4]. Compared to the East, gastric cancers in North America and Europe are diagnosed at more advanced stages, leading to poorer prognoses [5]. Nearly 40% of patients still present with metastases, while 50% present with locoregional disease [6]. The overall 5-year survival for patients with resectable gastric cancer ranges from 10-30% [7-9].
Curative therapy involves surgical resection with adequate lymphadenectomy; however, more than 80% of resected cases develop locoregional or systemic recurrence [10].This high rate of recurrence has invigorated researchers over the last twenty years to study various combinations of chemotherapy and radiation given before and after surgery to improve rates of recurrence and survival. Results of these trials have been variable and difficult to interpret given different study designs and patterns of disease in the East versus the West.
In Europe, for example, perioperative chemotherapy is the standard of care, based on the results of the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial in 2006 [11]. In Japan, the recommendation is to give adjuvant chemotherapy with S-1, an oral fluoropyrimidine [12-13]. In North America, adjuvant chemotherapy and radiation are the current standard of care, based on the results of the Southwest Oncology Group (SWOG) 9008/INT 0116 trial by Macdonald et al [14]. Among other limitations, study design and patient characteristics varied significantly among the three trials. The subsequent lack of consensus has led to several new studies such as the CRITICS trial that directly compare multimodal approaches.
This review will highlight management controversies and presents the major prospective phase III clinical trials that have shaped the Western approach to resectable gastric cancer, comparing and contrasting their findings with data from smaller phase II or III trials under investigation and provide clinical considerations drawn from the currently available evidence.

Extent of Lymphadenectomy

The extent of nodal dissection for gastric cancer has been controversial. In Japan, the modified D2 resection (removal of regional perigastric nodes without pancreatico-splenectomy) has acceptable morbidity and mortality [15] and is the standard approach [16]. A D1 resection entails subtotal or total gastrectomy plus greater and lesser omentum (stations 1-6). A D2 lymphadenectomy involves removal of omental bursa along with the front leaf of the transverse mesocolon with clearance of the nodes along the left gastric artery (station 7), common hepatic artery (stations 8), celiac artery (station 9), and splenic artery (stations 10 and 11). Western countries have begun to limit D2 resections since several randomized controlled trials (RCT) showed no survival benefit of D2 over D1 lymphadenectomy, morbidity and mortality of which are summarized in Table 1 [17-19]. In the Dutch cancer group study of 711 patients from 1989 to 1993, patients in the D2 group had a higher rate of complications, postoperative deaths, and longer hospital stays versus the D1 group, with little difference in the risk of recurrence at five years: 43% for D1 and 37% for D2 [20]. A 15-year follow-up study of the Dutch D1D2 trial showed that again there was no difference in overall survival (21% for D1 and 29% for D2, p=0.34); however, locoregional recurrence and gastric cancer-related death rate were significantly higher in the D1 group (48% vs. 37%, p=0.001), and postoperative deaths were higher after D2 (10% vs. 4%, p=0.004). When hospital deaths were excluded, the survival rates improved to 32% for D1 and 39% for D2. The authors concluded that with the availability of a safer, spleen-preserving D2 lymphadenectomy in high volume centers, this approach should be employed when possible to reduce locoregional recurrence in patients with resectable gastric cancer [21]. Nevertheless, practice patterns continue to vary across regions, but most expert surgical oncologist performs D2 lymphadenectomy without distal pancreatectomy and splenectomy with acceptable morbidity and mortality. As such, the extent of lymphadenectomy varies significantly among the major gastric cancer trials discussed in this review, making it difficult to draw valid conclusions regarding therapeutic efficacy.

Table 1

		N	Morbidity rate, p-value		Mortality rate, p-value		5-year survival
Degiuli et al. [18]	D1	76	10.5%	NS	1.3%	NS	Not evaluated
	D2	86	16.3%		0%
Cuschieri et al. [17]	D1	200	28.0%	<0.001	6.5%	0.04	35%	0.72
	D2	200	46.0%		13.0%		33.%
Hartgrink et al. [19]	D1	380	25.0%	< 0.001	4.0%	0.004	30%	0.35
	D2	331	43.0%		10.0%		35%

Table 1: D1 vs. D2 Lymphadenectomy: 3 randomized controlled trials.

Table 1
D1 vs. D2 Lymphadenectomy: 3 randomized controlled trials.

Perioperative Chemotherapy

The MAGIC trial was the first RCT to show a conclusive survival benefit from the use of perioperative chemotherapy for resectable gastric and gastroesophageal junction (GEJ) cancer over surgery alone [Table 2]. The MAGIC trial randomized 503 patients with locally advanced resectable gastric (74%) and GEJ (11%) cancer for surgery alone (n=253) or perioperative chemotherapy (n=250) with three pre-operative and three post-operative cycles of ECF (epirubicin, cisplatin, and infused 5-fluorouracil (5-FU)). Perioperative chemotherapy significantly improved resectability, progression-free and overall survival (PFS and OS, respectively), demonstrating a 5-year survival of 36% for perioperative chemotherapy compared to 23% for surgery-alone (HR 0.75, 95% CI, 0.60-0.93; P=0.009) [11].
As such, perioperative chemotherapy has become the standard of care in the UK and much of Europe. But while the MAGIC regimen improved tumor stage, node negativity, and subsequent curative resection, the efficacy of pre-operative versus post-operative chemotherapy could not be determined. Only 55% of patients randomized to chemotherapy actually received chemotherapy after surgery, and only 42% of them completed all six cycles postoperatively [11]. The most common reason for not enrolling into postoperative chemotherapy was disease progression or early death. Critics of the MAGIC trial also point out that outcomes for these patients were not stratified by disease stage or histopathology, thus making it difficult to decide which patients would benefit from chemotherapy before surgery. The authors argue that accurate preoperative diagnosis of locally advanced disease is difficult in itself, especially since endoscopic ultrasound was not widely available when the MAGIC trial was conducted [22].
Alternative regimens with more intensified chemotherapy have been investigated since MAGIC. The German AIO phase II/III FLOT4 study randomized patients to six cycles of perioperative ECF or four cycles of pre-operative 5-FU, leucovorin, oxaliplatin, and docetaxel (FLOT) followed by four cycles of FLOT postoperatively. Preliminary results show that FLOT led to better pathological complete response (pCR) than ECF (12.8% vs. 5.1%) [23]. These results encourage further study of FLOT to characterize toxicity and long-term survival benefits.
The MAGIC trial also led to the implementation of MAGIC B, or the UK NCRI ST03 trial, which opened accrual in 2007 to investigate the effect of bevacizumab, the monoclonal antibody to vascular endothelial growth factor (VEGF), in advanced gastric cancer. Bevacizumab combined with chemotherapy has shown efficacy in the treatment of various solid tumors [24-26]. VEGF is expressed in gastric cancer [27] and is a negative prognostic factor [28]. In a multicenter phase II trial of 47 patients with unresectable gastric cancer, bevacizumab was shown to be safe and effective at improving time to progression and overall survival when combined with chemotherapy (irinotecan and cisplatin) [29]. The ST03 trial randomized patients to perioperative chemotherapy with ECX (epirubicin, cisplatin, and capecitabine) +/- bevacizumab in HER2- negative patients in addition to two other treatment arms testing the feasibility of lapatinib in HER2-positive patients. Capecitabine given orally is more convenient than the 5FU infusion and thus avoids need for central lines and infusion pumps and their associated infectious complications. Unfortunately, ST03 did not demonstrate an improvement in overall survival (OS), progression-free survival (PFS), or disease-free survival (DFS) for resectable gastric cancer. Moreover, patients treated with bevacizumab had an increased risk of postoperative anastomotic leaks [30], likely due to impaired wound healing from inhibited angiogenesis. Despite these negative results, targeted therapy is still an area of interest in developing chemotherapeutic regimens, and other agents are being explored [31].

Neoadjuvant Chemotherapy

The effect of neoadjuvant chemotherapy on resectability was alsodemonstrated in a French multicenter phase III trial of 224 patients with stage II or greater gastric or GEJ cancer (FNLCC/FFCD trial). Compared to surgery alone, perioperative therapy with infusional 5-FU and cisplatin resulted in a higher likelihood of obtaining an R0 resection (73 vs. 84%). Similar to MAGIC, only half of the patients randomized to receive chemotherapy actually received their postoperative treatment. Thus, authors limited their conclusions to preoperative 5-FU and cisplatin, which improved DFS and OS [32]. In the EORTC 4095 phase III trial, those who received preoperative chemotherapy (cisplatin, leucovorin, infusional 5-FU) had more postoperative complications and no survival advantage over surgery alone, although the study was underpowered due to poor accrual [33]. A meta-analysis of 14 trials and 2271 patients who received neoadjuvant chemotherapy versus surgery alone for locally advanced gastric cancer concluded that neoadjuvant chemotherapy improves OS, PFS, R0 resectability rate, and does not worsen perioperative complications or mortality [34]. Another benefit of neoadjuvant therapy is the ability to assess tumor biology and response to therapy in the time before surgery, which may spare gastrectomy for a patient with particularly aggressive disease [35].

Table 2

	MAGIC	US INT 0116	CRITICS
Years	1994-2002	1991-1998	2007-2015
Region	UK	USA	Europe
Total (n)	503: Control: surgery only, n=253 Experimental: perioperative chemotherapy, n=250	556: Control: surgery only, n=275 Experimental: adjuvant chemoradiation, n=281	788: (All received Neoadjuvant chemo) Control: adjuvant chemotherapy only, 393 Experimental: adjuvant chemoradiation, n=395
Stage of adenocarcinoma	II-IV	IB-IV	IB-IV a
Neoadjuvant therapy	In experimental arm only: ECF x 3 cycles(epirubicin, cisplatin, infusional 5-FU)	Not applicable	ECC x 3 cycles (epirubicin, cisplatin, capecitabine)
Surgery	R0 resection—69.3% Esophago-gastrectomy: 23% D1: 19% D2: 40% Non-curative/Unknown: 40%	R0 resection: 100% D0 – 54% D1—36% D2—10%	Information not yet available.
Adjuvant therapy	ECF x 3 cycles	Bolus 5-FU, leucovorin, radiation with 45 Gy	3x ECC, radiation with 45 Gy
Completion of adjuvant therapy	42%	64%	Control arm: 46% Experimental arm: 55%
5-year overall survival	25% for surgery only 42% for perioperative chemotherapy	23% for surgery only 36% for adjuvant chemoradiation	re41.3% for adjuvant chemotherapy 40.9% for adjuvant chemoradiationsect

Table 2: Summary of three major phase III trials.

Table 2
Summary of three major phase III trials.

Neoadjuvant Chemoradiation

Recent randomized trials from Europe have compared neoadjuvant chemotherapy versus chemoradiation in esophageal and GEJ adenocarcinoma and noted improved histologic complete response, R0 resection rate, and lower frequency of lymph node metastases, but no significant effect on survival [36-37]. It is uncertain, however, whether these results can be applied to noncardia gastric cancers. At this time, no large phase III trials have examined chemoradiotherapy in the neoadjuvant setting for noncardia gastric cancers. The few small studies that have combined neoadjuvant chemoradiation with chemotherapy for locally advanced gastric cancer have been promising, however. The Radiation Therapy Oncology Group (RTOG) 9904 trial demonstrated improved R0 resection rate (70% vs. 30%) and pathological response rate of 26% for neoadjuvant chemoradiation with infusional 5-FU and paclitaxel [38-39]. Similarly, docetaxel-based chemoradiotherapy demonstrated a 24% pathological complete response and was not associated with postoperative complications [40]. As centers around the world continue to build experience with multimodal cancer therapy, neoadjuvant chemoradiation will likely gain popularity and has been identified as a focus of study in upcoming European and Asian clinical trials.

Adjuvant Chemotherapy

The decision to give perioperative chemotherapy largely depends on referral of the patient to an oncologist prior to surgery. Therefore, accrual for patients in FNLCC/FFCD and other trials of neoadjuvant chemotherapy has been low. In contrast, several studies have examined the efficacy of adjuvant chemotherapy versus surgery, but they are generally small, heterogeneous, and underpowered. Thus, a meta-analysis was conducted by the Global Advanced/Adjuvant Stomach Tumor Research International Collaboration (GASTRIC) group which studied 17 trials (3838 patients, median follow-up >7 years) and identified a 6% improvement in overall 5- and 10-year survival and 18% reduction in the risk of relapse, second primary, or death in patients who received adjuvant fluorouracil regimens versus surgery alone [41].
Another large phase III trial to show a survival benefit of chemotherapy with S1 (an oral fluoropyrimidine) over surgery alone was the Adjuvant Chemotherapy Trial for S-1 for Gastric Cancer (ACTS-GC) in Japan, in which S1 was given for a year postoperatively. Overall survival and RFS was significantly higher in the S-1 group (72.2% vs. 59.6%) at 3- and 5-year follow-up. Therapeutic efficacy of S-1 has not yet been definitively shown in non-Asian countries [42], as pharmacokinetics of this drug may differ between Eastern and Western populations [43]. However, a recent phase III RCT of S1 and cisplatin vs. infusional 5-FU and cisplatin showed non-inferiority and a better safety profile for the S1pluscisplatin group [42], encouraging further studies of S1 in non-Asian countries.
All patients in ACTS-GC underwent D2 dissection; in contrast, 40% patients in the MAGIC trial underwent D2 dissection. The CLASSIC trial (Capecitabine and Oxaliplatin Adjuvant Study in Stomach Cancer) studied the effect of adjuvant chemotherapy with capecitabine and oxaliplatin x 8 cycles after D2 lymphadenectomy. This trial also showed a higher 3-year DFS for chemotherapy (83%) vs. surgery alone (78%) [44-45]. Both the ACTS-GC and CLASSIC trials studied Asian populations who underwent D2 lymphadenectomies and had higher overall survival rates than Western trials of adjuvant chemotherapy. It is important to note, however, that results in the East cannot be readily applied to the West due to inherent differences in patient disease, surgical experience and technique. Nevertheless, the results of these studies demonstrate that post-operative chemotherapy prolongs survival in resectable gastric cancer in studies that did not administer pre-operative chemotherapy or radiation.

Table 3

Trial	Identifier	Phase	Country	Experimental Regimen
Neoadjuvant Chemotherapy
Japan Clinical Oncology Group 0501	NCT00252161	III	Japan	Neoadjuvant cisplatin +S1 vs. surgery alone
PRODIGY	NCT01515748	III	Korea	Neoadjuvant docetaxel, oxaliplatin, S1,followed by standard S1 adjuvant therapy
AIO FLOT4	NCT01216644	II/III	Germany	Standard 6 cycles of perioperative ECF or to 4 cycles of 5-FU, leucovorin, oxaliplatin, and docetaxel (FLOT) preoperatively and 4 cycles of FLOT postoperatively
Perioperative Chemotherapy
NEO-FLOT	NCT01160419	II	Germany	Perioperative 5-FU, leukovorin, oxaliplatin, docetaxel
Trial of Adjuvant XELOX Chemotherapy and Concurrent Capecitabine and Radiotherapy for Resected Gastric Carcin	NCT01711242	II	China	Adjuvant chemo alone (4 days of XELOX vs. chemoradiation 45 Gy in 1.8-Gy fractions with concurrent capecitabine, 825 mg/m2 twice daily on days of radiation), then 2 further cycles of XELOX after D2 lymphadenectomy
rections	rections	III	China	Adjuvant oxaliplatin + S1 vs. S1 alone
Neoadjuvant chemoradiation
TOPGEAR, Gastro-Intestinal Trials Group (AGITG)	NCT01924819	II/III	Australia, Canada, Europe	Neoadjuvant. Control = 3 pre, 3 post of ECX/ECF. Experimental arm receive 2 cycles of preoperative ECX/ECF followed by chemoradiation with fluoropyrimidine, then 3 cycles of postop ECX/ECF. , D1 LAD
Targeted therapy
RAMSES trial	NCT02661971	II/III	Germany	Perioperative FLOT vs. Ramucirumab plus FLOT
TOXAG	NCT01748773	II	Turkey	Adjuvant trastuzumab, oxaliplatin, and capecitabine (TOX), then chemoradiation with capecitabine, then mono therapy with trastuzumab
HerFLOT	NCT01472029	II	Germany	Perioperative FLOT plus trastuzumab

Table 3: Select ongoing clinical trials.

Table 3
Select ongoing clinical trials.

Adjuvant Chemoradiation

To further reduce the risk of recurrence, radiation has been studied as an adjuvant therapy in the treatment of gastric cancer. The SWOG/INT 0116 phase III trial randomized 553 patients after resection to either adjuvant chemoradiation or observation [Table 2]. External beam radiation was administered to the tumor bed, regional nodes, and two centimeters beyond the proximal and distal resection margins. The chemoradiation arm demonstrated significant improvement in 3-year relapse-free survival, overall survival, and local recurrence rates (19 vs. 29%) [11]. These advantages persisted at more than 10 years follow-up [46].
Due to treatment toxicity, only 64% completed the adjuvant treatment. However, given the significant survival advantage, adjuvant chemoradiation became the standard of care for resectable gastric cancer in the United States. To reduce toxicity, bolus 5-FU has been replaced by either oral capecitabine or continuous infusion 5-FU [47]. Other attempts at reducing toxicity include using the alternative regimen of ECF plus radiation, as done in the CALGB 80101 trial in the US. Results have shown lower toxicity rates but no change in overall survival [48].
The question remains as to whether adjuvant radiation benefits D2 and D1lymphadenectomy equally. In INT 0116, less than 10% of patients (n=54) underwent a D2 lymphadenectomy and 54% received a D0 lymphadenectomy. Therefore, criticism arose as to whether the survival advantage of adjuvant chemoradiation could be substituted by a more extensive lymphatic dissection. A subgroup analysis in the 10-year updated results showed no difference in survival between D1 vs. D2 lymphadenectomy, but the subgroups were insufficiently powered. Nevertheless, the authors obtained a Maruyama index of unresected disease (MI) on each patient and showed that MI was an independent prognosticator of RFS, OS, and response to chemoradiotherapy. The MI predicts the positivity of remaining lymph nodes and can indicatethe adequacy of a nodal dissection [49-50]. A retrospective study by Dikken et al. [51] controlled for MI and found no advantage for adjuvant chemoradiation after D2 dissection. Therefore, though data remains conflicted, inadequacy of lymph node dissection may be considered an indication for adjuvant chemoradiation.

Adjuvant Chemotherapy vs. Chemoradiation

The value of radiation in the post-operative setting was challenged by the ARTIST trial (Adjuvant Chemoradiotherapy Trial in Stomach Tumors) from Korea. 458 patients with R0 resections and D2 lymphadenectomies were randomized to either adjuvant chemotherapy with capecitabine and cisplatin (XP) vs. XP plus radiation. There was no difference in 3-year DFS. However, subset analysis of those with node-positive disease showed significantly better DFS with chemoradiotherapy vs. chemotherapy alone [52]. This led to the design of the ARTIST-II trial, which will compare adjuvant chemotherapy versus chemoradiation in D2 resected nodepositive patients. The results of ARTIST-II may not be fully applicable to Western populations where D2 lymphadenectomy is not the standard of care, but will address an important question regarding patient selection for adjuvant radiation.

Neoadjuvant Chemotherapy Plus Adjuvant Chemoradiation

It is not possible to directly compare results from the MAGIC and Macdonald (INT 0116) trials as they differ in their study design and eligibility criteria. Therefore, to determine whether adjuvant radiation offers benefit to patients who have undergone perioperative chemotherapy and adequate surgery (D1+), the European CRITICS trial was undertaken in 2007 [53]. The main features of the CRITICS trial compared to MAGIC and INT 0116 are summarized in Table 2. This large, multinational phase III trial randomized 788 patients’ status post neoadjuvant chemotherapy and adequate surgical resection to receive either adjuvant chemotherapy or adjuvant chemoradiation. Preliminary results presented at the World Congress on Gastrointestinal Cancer (WCGC) in 2016 showed no survival benefit for chemoradiation vs. chemotherapy; both groups had a 5-year survival of 40% [54]. This again calls into question the benefit of adjuvant radiation and emphasizes that perhaps only a limited number of patients are good candidates for adjuvant therapy and this population remains to be defined. Secondary endpoints such as toxicity and quality of life and subgroup analysis are pending as well. The study does, however, suggest that intensifying neoadjuvant therapies may be a better approach toward improving survival. Indeed, future directions for the CRITICS investigators may include a new study that will compare outcomes between three different neoadjuvant treatments: 1) four courses of chemotherapy; 2) two courses of chemotherapy combined with chemoradiation; versus 3) chemoradiation only.

Ongoing Trials and Future Directions

Table 3 highlights some of the ongoing clinical trials that are currently studying new multimodal combinations for resectable gastric cancer. Assessing the role of perioperative chemoradiation combined with standard regimens of chemotherapy was the focus of ARTIST-II and CRITICS, final results of which are pending. Other trials are studying intensification and diversification of chemotherapeutic regimens. Because participants in previous trials had relatively low completion rates of post-operative treatment, a goal of these newer trials is to reduce toxicity and maximize efficacy of neoadjuvant therapies. Intensification is this setting is better tolerated than when given post-operatively. In the East, Korean and Japanese groups are conducting phase III trials that add docetaxel, oxaliplatin, and/or cisplatin to standard S-1 therapy [55,56].The phase III CAPITAL study (NCT01795027) in China is studying oxaliplatin + S1 in the adjuvant setting, while another Chinese study (NCT01711242) is studying adjuvant XELOX (capecitabine, oxaliplatin) with and without chemoradiation. The German NEOFLOT trial prolonged preoperative chemotherapy with docetaxel, 5-FU, leucovorin, and oxaliplatin. Phase II results showed an R0 resection rate of 86%, pCR of 20%, and < 10% residual tumor [57]. Intensification of perioperative chemotherapy is represented by the FLOT-4 trial, which has randomized 714 patients with resectable gastric cancer to either the standard six cycles of perioperative ECF/ ECX or 4+4 cycles of perioperative FLOT. Phase II results from this head-to-head trial have shown more pathological remissions with FLOT than with ECF/ECX [23].
Alternative options such as targeted therapy have been explored. Bevacizumab has not yet shown a survival benefit for resectable gastric cancer, but current and ongoing trials are investigating ramucirumab (NCT02443883), which targets VEGF receptor-2 and has shown improved OS in advanced gastric cancer [31]. The Her FLOT trial (NCT01472029) will combine FLOT with trastuzumab to assess for superior response in HER2-positive adenocarcinoma of the stomach or GEJ. Within the next several years, the results of these trials will be eagerly anticipated and applied toward improving outcomes.

Summary and Recommendations

Surgery with adequate lymph node dissection (at least D1, preferably D2 with preservation of distal pancreas and spleen) remains the only curative option for locoregional gastric adenocarcinoma. In both Western and Eastern populations, adjuvant chemotherapy plays a role in reducing recurrence and improving overall survival after surgery; however, standard pharmaceutical regimens vary globally. The MAGIC and INT 0116 trials have influenced subsequent trials significantly, resulting in systematic combinations of neoadjuvant and adjuvant approaches in both the East and the West. For example, pre-operative chemotherapy added to the standard adjuvant chemotherapy is now under investigation in Japan using S-1.The benefit of adjuvant chemoradiation demonstrated by INT 0116 continues to garner controversy. The recent report from the CRITICS trial that adjuvant chemoradiation did not significantly improve survival in those who had received preoperative chemotherapy and adequate resection will challenge the North American approach defined by INT 0116 even further. Final results and subgroup analyses are anticipated and will be critical for this discussion. The study is not likely to change practice in Europe, where radiation is not yet a firstline treatment for resectable gastric cancer.
Patients with resected N1 disease should be considered for adjuvant chemoradiation as opposed to chemotherapy alone. This recommendation is supported by subset analysis of the ARTIST trial and also the INT 0116 trial, which showed that adequacy of lymph node dissection, correlated with response to chemoradiation. For patients with T2N0 gastric cancer, current data supports either observation after surgery or adjuvant therapy.
As multiple studies have shown improved outcomes after neoadjuvant chemotherapy with or without adjuvant treatment, it is our recommendation that all patients with stage II or higher gastric cancer should be evaluated by a multidisciplinary panel of oncologists and radiation oncologists prior to surgery. Pathological response to neoadjuvant therapy correlates with baseline T and N stage [40] and selects patients who will benefit from further intensive therapy. Indeed, if clinical features are suggestive of a high likelihood of metastasis (e.g. bulky tumor, visible nodes, linitis plastica, positive peritoneal cytology), one should certainly consider neoadjuvant chemotherapy. Based on the preliminary results of the CRITICS trial and the observation that MAGIC and other perioperative chemotherapy trials only enrolled about half of their patients into postoperative chemotherapy, the emphasis on developing and refining neoadjuvant therapies is evident in current and future trials.

Review Article