In Japan, where gastric cancer is endemic, the Japanese Research Society for Gastric Cancer (former Japanese Gastric Cancer Association) defined and recommended D2 lymphadenectomy for gastric cancer in reference to long-term experience with gastric cancer surgery. Based on a large case review, they mapped the location of metastatic lymph nodes according to the primary tumor site, and classified metastatic lymph nodes into three groups.1314 Briefly, group 1 indicates perigastric lymph nodes; group 2 includes lymph nodes around major vessels in the vicinity of the pancreas and splenic hilum; and group 3 indicates lymph nodes beyond group 2. D1 lymphadenectomy represents the resection of lymph nodes in group 1, and D2 lymphadenectomy refers to the dissection of all lymph nodes in group 2, including group 1. This guideline15 has recommended D2 lymphadenectomy for AGC, and a randomized controlled trial (RCT) from Taiwan proved the benefit of D2 over D1 lymphadenectomy [5-yr overall survival (OS) was 59.5% in D2 and 53.6% in D1, p=0.041] in gastric cancer.16 With belief that a greater extent of surgery can improve the prognosis of gastric cancer (and maybe because there was no additional option for further improving survival of GC at that time), D3 lymphadenectomy (D2+para-aortic lymph node dissection) was widely applied in Japan and Korea. However, a RCT, which compared the outcomes of D2 vs. D3, showed that D3 surgery was related to a tendency toward increased operative complications (20.9% in D2 vs. 28.1% in D3, p=0.07) with no improvement in the prognosis of gastric cancer [5-yr OS was 69.2% in D2 and 70.3% in D3; hazard ratio (HR) of D3 compared to D2 was 1.03, p=0.85].1718 Thus, the extent of lymphadenectomy for gastric cancer was decided at a level of D2, and it has remained standard surgical procedure in East Asia.19

On the other hand, in the West, limited lymphadenectomy had been considered as standard surgery because two RCTs, MRC trial and Dutch Gastric Cancer trial, failed to show the survival benefits of D2 over D1 lymphadenectomy.2021 Even worse, D2 surgery was related to un-acceptably high mortality (10-13%) and morbidity (43-46%) in these trials.2223 Consequently, compared to outcomes from East, which reported a mortality of less than 1% after D2 surgery,1724 mortality over 10% would be difficult to accept in practice. However, 15-year follow-up results of Dutch Gastric Cancer trial showed that despite no benefit to OS for D2 surgery, loco-regional recurrence (41% in D1 vs. 25% in D2) and lower gastric-cancer-related death (48% in D1 vs. 37% in D2) were lower for D2 surgery than D1 surgery.25 Moreover, there have been reports2627 that D2 surgery can be performed safely (mortality rate was 1.7% to 3.6%) if it is conducted by experienced-hands in high-volume centers, even in the West. Consequently, now D2 surgery for gastric cancer is recommended in both East and West guidelines with a precondition of being performed at specialized, high-volume centers where it can be performed safely.152829

Historically, pancreatico-splenectomy was standard surgery for total gastrectomy when cancer is located in the proximal stomach. This was because the extent of D2 lymphadenectomy for total gastrectomy includes lymph nodes around the supra-pancreatic and splenic hilar area; thus, for removing these lymph nodes completely, distal pancreatectomy with splenectomy was thought to be mandatory.30 However, a substantial number of intraperitoneal abscess and pancreatic fistula occurred as unpleasant operative sequels after distal pancreatectomy accompanying total gastrectomy. Maruyama, et al.31 reported the necessity of pancreas-preserving (PP) total gastrectomy with results showing that pancreas preserving total gastrectomy was superior to pancreas resection (PR) in terms of morbidity (39.4% in PR vs. 19.6% in PP), mortality (0.9% in PR vs. 0.3% in PP), 5-yr OS (54.5% in PR vs. 70.5% in PP for stage II, and 36.7% in PR vs. 54.1% in PP for stage II), and newly developed diabetes mellitus (37% in PR vs. 0% in PP). Following results from a RCT supported the advantages of total gastrectomy with PP: distal pancreatectomy with splenectomy for gastric cancer was related to high morbidity and poor prognosis.22 Thus, pancreas preserving total gastrectomy became a standard procedure for advanced proximal gastric cancer.

The next question was whether spleen should be removed as part of lymphadenectomy in total gastrectomy for proximal gastric cancer. Noh, et al. presented the technical feasibility of total gastrectomy with spleen preserving hilar lymph node dissection at the Second International Gastric Cancer Congress.32 His data showed that splenectomy during total gastrectomy is related to increased morbidity, while providing no survival gain.3334 Following studies supported the results: a study from U.S. reported that splenectomy is related to high operative morbidity and mortality after total gastrectomy;35 a RCT from Korea reported that prophylactic splenectomy to remove lymph nodes around spleen hilum for proximal gastric cancer is not recommended because it offers no prognostic advantages.36 The incidence of lymph node metastasis at the spleen hilum reportedly ranges from 9.8% to 15.4%,373839 and prophylactic splenectomy may not be mandatory, except when the primary tumor directly invades the spleen or definite gross lymph node metastases is present at the spleen hilum. Nevertheless, organ preserving surgery should not mean that necessary lymph node dissection can be omitted during radical surgery. Spleen preserving total gastrectomy should represent total gastrectomy with D2 lymphadenectomy, including splenic hilar lymph node dissection, while preserving the spleen. We anticipate that an ongoing RCT, the Japan Clinical Oncology Group (JCOG) 0110-MF trial, will provide more concrete evidence for the oncologic feasibility of spleen preserving gastrectomy.40

Minimally invasive surgery (MIS) including laparoscopic and robotic surgery is now considered a standard operation in most of surgical fields not only for benign surgery but also for cancer surgery. In the field of gastric cancer surgery, Kitano, et al.41 reported the first laparoscopic distal gastrectomy for gastric cancer, and this procedure has been widely propagated for gastric cancer surgery. In the beginning of laparoscopic surgery for gastric cancer, conservative surgeons criticized that it would be impossible to perform adequate lymphadenectomy through laparoscopic devices, so its indication was limited to very early stage gastric cancer, in which the extent of lymphadenectomy is more limited, compared to AGC. However, cumulative experience with better laparoscopic surgical devices has shortened the gap in surgical quality between conventional open surgery and laparoscopic surgery. And its potential advantages [cosmetic benefit due to smaller incision size, better quality of life (QOL), and less pain with shorter hospital stay] and increasing incidence of EGC through nationwide mass screening in Korea and Japan has bolstered the popularity of laparoscopic surgery for gastric cancer. At present, even though RCTs4243 into the oncologic outcomes of laparoscopic surgery for EGC compared to open surgery have not been published yet (KLASS and JCOG0912 trial), laparoscopic gastrectomy is considered as a possible option for clinical EGC, based on the long term result of a large-scale Korean multicenter study.44 However, the question about whether laparoscopic surgery can be applied for AGC requiring D2 surgery remains unanswered. Although reports4546 have demonstrated the possibility of laparoscopic surgery, even for AGC (similar short and long term outcomes compared to open surgery), concerns about its potential risks and whether similar quality of surgery can be achieved by laparoscopic gastrectomy compared to that of open surgery or not (because D2 surgery for AGC is still difficult to perform even through an open technique) are still under debate. In an effort to diminish those concerns and to expand the indication of laparoscopic surgery from EGC to AGC, a multicenter RCT (KLASS II, NCT01456598) is ongoing.

Robotic surgery systems were introduced into the field of gastric cancer surgery in 2005 and have propagated, especially in Korea, because of their potential advantages over laparoscopic surgery: robot systems provide a better three-dimensional view and facilitate fine movements with tremor filtering and articular movement.4748 Most robotic surgeries have been performed by experienced laparoscopic surgeons, and it was expected that robotic surgery systems would help overcome technical difficulties in laparoscopic surgery.49 A meta-analysis,50 which compared the short-term outcomes of robotic surgery to laparoscopic and open surgery for gastric cancer, showed that morbidity and mortality were similar and that robotic surgery offers practical advantages of less blood loss and hospital stay, compared to open surgery, although operative time was longer than other modalities. Even though its high cost was criticized, some surgeons have suggested that robotic surgery would have advantages over laparoscopic surgery in technically difficult and complicated cases (e.g., far advanced cancer in which combined resection should be performed for R0 resection). Now, its indication in gastric cancer surgery are similar to those of laparoscopic surgery,5051 although concerns remain high for whether this expensive approach can be justified over equivalent but cheaper laparoscopic surgery.

As the proportion of EGCs has increased and its prognosis is quite good,52 enough to be considered as cured in most cases after surgery, physicians have now began to focus on improving QOL of patients with EGC. Consequently, questions of whether prophylactic lymphadenectomy for EGC is mandatory have been raised. If there are no metastatic lymph nodes around the stomach in EGC, surgeons may be able to spare the lymph nodes from surgery and resect only the primary tumor.

A greater extent of lymphadenectomy leads to more surgical morbidity and mortality,2223 and gastrectomy itself can decrease QOL of patients. If metastatic lymph nodes could be identified before or during an operation, surgeons could spare lymph nodes that are noncancerous. Thereby, segmental gastrectomy with a safe margin would be possible and would improve QOL of patients without compromising prognosis. Sentinel lymph node biopsy has been widely applied in surgery for breast cancer and melanoma. There have also been attempts to adapt sentinel lymph node biopsy for gastric cancer surgery. Even though the JCOG0302 trial was terminated in the midst of enrollment because of its unexpectedly high false negative rate,53 another phase II trial from Japan showed the practical possibility of sentinel lymph node mapping for gastric cancer.54 Based on the lessons learned from both Japanese trials, a new phase III clinical trial (SENORITA trial, NCT01804998) of using both Tc 99m and indocyanine green injection around tumors through intra-operative endoscopy was initiated in Korea after a quality control study (NCT01544413).55 The result of this trial will guide the possibility of tailored and limited surgery for EGC.

In Japan, endoscopic mucosal resection (EMR) has been accepted for treating EGC with a very low probability of lymph node metastasis and adopted into practice.56 Gotoda, et al.57 reviewed a large cohort (over 5000 cases) of EGC patients who underwent gastrectomy and found that some subgroups show a low incidence of lymph node metastasis. Accordingly, the authors proposed expansion of criteria for local treatment. Development of endoscopic devices has made endoscopic submucosal dissection (ESD) possible, and this new procedure has shown better outcomes in en-block resection with a complete resection rate, although with more complications of bleeding and perforation.58 Nowadays, EMR and ESD have become viable treatment options for EGC within their indications and widely adopted across Japan and Korea. Its oncologic outcomes, however, compared to surgical resection with lymphadenectomy, have yet to reach consensus.59

Macdonald, et al.60 applied combined two strategies, chemotherapy and radiation therapy (fluorouracil and leucovorin with total 4500 cGy of radiation), after surgical resection of gastric cancer for AGC and compared the outcomes thereof with those of surgery alone. A total of 556 patients with resectable gastric cancer or gastro-esophageal junction cancer were enrolled in this RCT [intergroup 0116 (INT-0116) trial], and the results were promising: improved median survival (27 months for surgery alone vs. 36 months for additional chemoradiotherapy group), OS (HR of surgery alone was 1.35, 1.09 to 1.66, p=0.005), and relapse-free survival (HR of surgery alone was 1.52, 1.23 to 1.86, p<0.001) with acceptable incidence of toxic effect from chemoradiotherapy. However, only 10 % of the patients underwent D2 surgery, and 54% of patients received less than D1 surgery, igniting a fierce debate over the clinical utility of the treatment when D2 surgery is completed. Regardless, based on the results, post-operative chemoradiotherapy has become standard treatment for resectable gastric cancer in the United States.

Another RCT from the United Kingdom, the Medical Research Council Adjuvant Gastric Infusional Chemotherapy trial (MAGIC), enrolled 503 patients with resectable stomach cancer, gastroesophageal junction cancer, and lower esophageal adenocarcinoma and compared the outcomes of peri-operative chemotherapy (epirubicin, cisplatin, and infused fluorouracil) in conjunction with surgery with those of surgery alone.61 The results showed the benefit of peri-operative chemotherapy over surgery alone: better OS (HR of chemotherapy group was 0.75, 0.60 to 0.93, p=0.009) and progression-free survival (HR of chemotherapy group was 0.66, 0.53 to 0.81, p<0.001) with similar postoperative complications (46% in chemotherapy group vs. 45% in surgery alone). However, only 67.8% of patients received curative intent surgery, 24.1% received gastrectomy including esophagectomy and 41.4% of patients underwent D2 surgery. Again, similar debates as with the INT-011660 were raised regarding the clinical utility thereof after D2 surgery. Based on the results, however, peri-operative chemotherapy has been applied a standard strategy for AGC in Europe.

Although these two RCTs showed the benefits of additional post-operative chemoradiotherapy and peri-operative chemotherapy for AGC over surgery alone, the 5-year OS thereof was much lower than that of surgery alone in East Asia.18 Thus, a question of whether chemotherapy provides additional benefits even after radical surgery (D2 surgery) or not was raised in East Asia, where radical D2 surgery is a standard surgery for AGC. To answer to this question, two landmark phase III RCTs were conducted to compare the outcomes of chemotherapy as adjuvant treatment after radical surgery to those of surgery alone. The first study to demonstrate the benefits of adjuvant chemotherapy after D2 surgery for AGC was the Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC) in Japan.6263 A total of 1059 patients with stage II and III disease were enrolled in over 100 centers after D2 surgery in this trial. The results showed that the prognosis of S-1 monotherapy after D2 surgery was better than that of D2 surgery alone regarding 5-year OS (71.7% in S-1 group vs. 61.1% in surgery only, HR of S-1 group was 0.669, 0.540 to 0.828) and 5-year RFS (65.4% in S-1 group vs. 53.1% in surgery only, HR of S-1 group was 0.653, 0.537 to 0.793) with acceptable toxicity (less than 6% of over grade 3 toxicity).

Thereafter, the CLASSIC trial (Capecitabine and Oxaliplatin Adjuvant Study in Stomach Cancer), which was conducted at 37 centers throughout Korea, China, and Taiwan, further supported the benefits of adjuvant chemotherapy after D2 surgery.664 In this study, 1035 patients with stage II and III gastric cancer who underwent D2 surgery were enrolled, and capecitabine and oxaliplatin (XELOX) were applied as an adjuvant concept to the XELOX group. The results revealed that adjuvant XELOX improved the prognosis of the patients: estimated 5-year OS was 78% in the XELOX group versus 69% in the surgery alone (HR of XELOX group was 0.66, 0.51 to 0.85, p= 0.0015) and estimated 5-year DFS was 68% in the XELOX group versus 53% in surgery alone (HR of XELOX group was 0.58, 0.47 to 0.72, p<0.001). In addition, the results of a meta-analysis conducted by the Global Advanced/Adjuvant Stomach Tumor Research International Collaboration (GASTRIC) group supported the benefits of adjuvant chemotherapy (pooled HR of chemotherapy for OS was 0.82, 0.76 to 0.90, p<0.01 and that of DFS was 0.82, 0.75 to 0.90, p<0.01).65 Taken all together, all doubts of whether adjuvant chemotherapy after D2 lymphadenectomy is needed or not were resolved,66 and adjuvant chemotherapy has become a standard treatment for resectable AGC in East Asia.

The results of the INT-0116 trial, which showed the benefits of chemoradiotherapy, suggested the potential benefits of radiation therapy in gastric cancer, and additional results showed that adjuvant chemotherapy improves the survival of AGC patients. The next question facing clinicians is whether radiation could save more patients if added to D2 surgery with adjuvant chemotherapy. To address this, the ARTIST trial (Adjuvant Chemoradiation Therapy in Stomach Cancer) was conducted in Korea.6768 This RCT assigned 458 patients who underwent D2 surgery into two groups: XP group (treated by six cycles of capecitabine and cisplatin) and XPRT group (treated by two cycles of XP, followed by chemoradiation therapy with capeci-tabine, and then two additional cycles of XP). After 7 years of follow up, the results revealed no significant benefits of additional radiation therapy (HR of XPRT for OS was 1.130, 0.775 to 1.647, p=0.5272, and HR of XPRT for DFS was 0.740, 0.520 to 1.050, p=0.0922). It is has been argued that radiation therapy is a loco-regional treatment rather than systematic treatment, and D2 surgery would be enough for loco-regional control of gastric cancer. Therefore, the addition of radiation therapy to D2 surgery would have no further benefit. Interestingly, subgroup analyses showed a possible benefit of XPRT over XP regarding DFS in lymph node positive and intestinal-type gastric cancer. The ARTIST-II successor trial is currently evaluating the benefits of adjuvant XPRT in patients with lymph node metastasis after D2 surgery.

One of the problems of traditional chemotherapy is that it acts against all actively proliferating cells, normal and cancerous, causing serious collateral damage. Thus, treatments against targets specific to cancer cells could spare normal cells. Although understanding of the molecular genetic underpinnings of gastric cancer has lagged behind that for other solid cancers, recent efforts to better understand gastric cancer biology has led to the discovery of a handful of genetic alterations specific to cancer cells: overexpressed proteins are present in cancer cells but not in normal cells, and specific mutant proteins drive cancer growth and survival.

The precedent success of target therapy by trastuzumab, a humanized monoclonal antibody interferes human epidermal growth factor receptor type 2 (HER2/neu/ErbB2), for HER2 positive breast cancer encouraged the expanding of its indications to gastric cancer. The Trastuzumab for Gastric Cancer (ToGA) trial, a RCT of patients with HER2 positive mostly metastatic gastric/gastro-esophageal cancer, was conducted in 122 centers in 24 countries.7 A total of 549 patients were randomly assigned into two groups: trastuzumab with chemotherapy and chemotherapy alone. The results showed that trastuzumab improves the prognosis of HER2 positive gastric cancer; the median OS of trastuzumab with chemotherapy was 13.8 months versus 11.1 months in chemotherapy alone (HR of trastuzumab with chemotherapy was 0.74, 0.60 to 0.91, p=0.0046). The success of the ToGA trial suggested that cancer biology-driven target therapies could be possible in gastric cancer and encouraged investigations into additional candidate targets for AGC.

The next target evaluated was vascular endothelial growth factor (VEGF), which is known to be related to angiogenesis in tumorigenesis. Treatment with bevacizumab, a humanized monoclonal antibody that inhibits VEGF, exhibited a positive impact among patients with several types of cancers, such as colorectum,69 lung,70 and recurrent glioblastoma.71 Also, a phase II trial to evaluate the safety and efficacy of bevacizumab with conventional chemotherapy (doxetaxel, cisplatin, and fluorouracil) for advanced gastro-esophageal cancer seems to have been successful.72 Expecting continued success, the Avastin in Gastric Cancer (AVAGAST) trial for evaluating the efficacy of additional bevacizumab to chemotherapy (capecitabine and cisplatin) as the first-line treatment for unresectable AGC was conducted.73 This double blind, placebo-controlled phase III RCT enrolled 774 patients. The primary endpoint of improving OS, however, was not satisfied (median OS was 12.1 months in adding bevacizumab group vs. 10.1 months in placebo plus chemotherapy group; HR of adding bevacizumab was 0.87, 0.73 to 1.03, p=0.1002), although progression free survival and overall response rates were significantly better for adding bevacizumab over the placebo group. A subsequent biomarker evaluation study, which analyzed blood and tumor tissue samples collected for the AVAGAST trial reported that VEGF-A and tumor neurophilin-1 (NRP1) could be potential predictive biomarkers for bevacizumab efficacy.74 The most recently reported REGARD study, an international Phase III RCT for assessing the clinical benefit of VEGF receptor 2 (VEGFR-2) inhibition in gastric and gastroesophageal cancer, showed promising results for angiogenesis blockade therapy in gastric cancer.75 The study showed that patients assigned to treatment with ramucirumab, a fully humanized IgG1 monoclonal antibody inhibitor of VEGFR-2, had significantly improved survival (HR 0.776, 95% CI 0.603-0.998, p=0.047), with a median survival of 5.2 months versus 3.8 months with placebo. Still, there are many questions regarding the results for the AVAGAST and REGARD trials. In AVAGAST trial, it seemed that Asian patients were less benefited from bevacizumab compared to patients in the rest of the world, and in biomarker study of the trial showed that those biomarkers, VEGF-A and NRP1 were not effective to predict responsiveness of bevacizumab. In REGARD trial, however, the proportion of Asian patients enrolled was only around 8%. This is largelry due to the practical reason: 2nd line chemotherapy is considered standard of care in Asia. Togetehr, the resutls of the two pivotal trials on anti-angiogenesis treatment for gastric cancer imply that understanding of heterogeneity in gastric cancer across the globe76 including ethnicity could impact the biologically targeted therapies for this disease. Regardless, the data impose a significant clinical insight of targeted biological options in selected patients' subpopulations with corresponding target presence.