Essential updates 2017/2018: Recent topics in the treatment and research of gastric cancer in Japan

Abstract Recent developments in diagnostic technology, accumulated clinical effort and established evidence have boosted early detection and drastically improved early and long‐term outcomes of gastric cancer. However, gastric cancer continues to be one of the most aggressive and life‐threatening malignancies among all cancers and is a global health problem. Between January 2017 and December 2018, various fascinating reports of managements and research were published, including the new 15th Japanese Classification of Gastric Carcinoma reflecting the 8th American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) tumor, node and metastasis (TNM) classification (October 2017) and the new Gastric Cancer Treatment Guidelines version 5 (January 2018). Moreover, pivotal molecular features of gastric cancer were clarified by the worldwide cancer genome project, and various treatment targets and biomarkers such as circulating DNAs and microRNAs were detected. Novel treatment options using programmed cell death protein 1 immune checkpoint inhibitors have been started. In this review, we summarize the recent topics of classification, guidelines, and clinical and basic research in order to bring new insights to gastric cancer treatment.


| INTRODUC TI ON
Gastric cancer is considered to be the fifth most common can-  4 In this review, we summarize the topics of these reports related to new classifications, treatment guidelines, clinical research such as chemotherapy, nutrition and pathophysiology and basic research such as molecular features and liquid biopsy in gastric cancer.

| New 15th JGCA and 8th AJCC/UICC TNM staging related to definition of N3
Tumor-node-metastasis (TNM) classification systems have been the most important tool for cancer treatment and evaluation of patient outcomes worldwide. Although many studies have confirmed that the 7th edition of the AJCC/UICC TNM classification had a higher prognostic predictive ability than its previous TNM systems, 5 some limitations remained because N3a (7-15 metastatic lymph nodes) had the same staging as N3b (≥16 metastatic lymph nodes). 6 However, the greatest change in the recently released 8th edition of AJCC/UICC TNM and the 15th edition of JGCA staging systems for gastric cancer was the separate inclusion of N3a and N3b. 7 Therefore, the new 15th JGCA 3 and the 8th AJCC/UICC TNM classifications differed in that T1-T3 disease was upstaged with N3b, T4aN3a was downstaged from IIIC to IIIB, and T4bN0 and T4aN2 were downstaged from IIIB to IIIA (Figure 1). 3 The 8th edition was validated as showing better prognostic ability than the 7th edition and had a good concordance index (Cindex, 0.719), which was comparable to that of the 8th edition developed from the International Gastric Cancer Association (IGCA) data (0.775). 8 Particularly, these tendencies were precisely proven in patients with >15 retrieved lymph nodes (RLN), because a stage migration may occur if a low number of lymph nodes are retrieved, which may underestimate the malignant potential of gastric cancer. 8,9 Number of RLN is influenced by various factors: extent of lymphadenectomy, enthusiasm to examine more lymph nodes pathologically and the surgical situation, such as fat volume and the innate number of lymph nodes in each patient. Although at least 16 RLN are recommended for staging in the JGCA and AJCC/UICC TNM classifications, a significant prognostic difference has been reported even between RLN <16 and RLN ≥16 in patients with pStage II-III gastric cancer. 10 Previous studies have identified that RLN ≥25 could eliminate the prognostic effect 10 and presented better prognostic staging. 11 Indeed, the German S3 guidelines also recommend ≥25 RLN as a criterion for a D2 gastrectomy. Therefore, RLN ≥25 may be needed for precise nodal staging in the new classification.
Nevertheless, a low number of RLN may still be influenced by various situations and thereby be imperative. Positive lymph node ratio (PLNR), which is obtained by dividing the metastatic lymph node count by the RLN count, has been proposed as an alternative and putative superior staging method to avoid stage migration [12][13][14] and could evaluate the quality of TNM staging. 15 RLN ≥8 might be sufficient as an appropriate use of the PLNR system. 13

| New definition of the esophagogastric junction (EGJ) and distribution of esophageal cancer and gastric cancer in EGJ adenocarcinoma
Definition of EGJ and staging of the carcinoma of EGJ have not been standardized internationally. A unified definition was decided in both the 15th JGCA and the 11th Japan Esophageal Society (JES) classifications, compared to the previously published AJCC/UICC TNM classification. Specifically, the 15th JGCA 16 and the 11th JES classifications 17 defined the EGJ as the border between esophageal and gastric muscles to be defined from the distal end of longitudinal palisading vessels in the lower esophagus to the proximal end of the longitudinal gastric folds.
The priority of the EGJ definition was the distal end of longitudinal palisading vessels by the endoscopic findings. Regarding the definition of EGJ cancer, the Nishi classification defined EGJ carcinoma as a tumor located in the area from 2 cm above to 2 cm below the EGJ irrespective of histological type. Whereas, the Siewert classification covering 5 cm above and below the EGJ is widely used to classify EGJ adenocarcinoma, although a strict definition of EGJ itself has not been described.
In the 7th AJCC/UICC TNM classification, adenocarcinoma arising within 5 cm below the EGJ, which is defined as Siewert I to III, was staged as an esophageal cancer. However, in the new 8th AJCC/ UICC TNM classification, EGJ adenocarcinoma such as Siewert I and II was defined as an esophageal cancer, for which the tumor is located from 5 cm above to 2 cm below the EGJ, whereas EGJ adenocarcinoma such as Siewert III was defined as a gastric cancer, for which the tumor is located from 2 cm below to 5 cm below the EGJ. 7 2.3 | New definition of subclassified station no. 6 lymph nodes and new allocation of station no. 13 regional lymph nodes as a regional lymph node in duodenal invasion of gastric cancer Infrapyloric station no. 6 lymph nodes were anatomically subclassified into three regions as follows: no. 6a, lymph nodes along the right gastroepiploic artery; no. 6i, lymph nodes along the infrapyloric artery; and no. 6v, lymph nodes on the anterior surface of the pancreatic head along the right gastroepiploic vein and the infrapyloric vein. 16,18 This subclassification is anatomically useful to carry out antrum-preserving gastrectomy. Based on this definition, a prospective observation study identified that the metastasis rate to no. 6i nodes was 2.1% in early lower-third tumors and 19.5% in advanced tumors. 19 Moreover, in patients with positive no. 6i nodes, the distance from the distal tumor border to the pyloric ring was proven to be within 44 mm. In contrast, no early middle-third gastric cancers had no. 6i metastasis. 19 According to the TNM classification, in the rare occurrence in which a tumor involves more than one organ or structure, the regional nodes have been reported to include those of all involved structures, even if the nodes of the primary site are not involved.
Specifically, in the esophageal invasion of gastric cancer, station no.
19, 20, 110 and 111 lymph nodes were considered to be regional lymph nodes. 16 Also, in the duodenal invasion of gastric cancer, a station no. 13 lymph node, which was defined as a lymph node located in the posterior surface of the pancreatic head, was considered to be a regional lymph node but not a distant lymph node. 16

| Miscellaneous
One of the biggest changes in the new 15th JGCA and 8th AJCC/UICC TNM staging is a separation of clinical TNM and pathological TNM stages.
Clinical stage is simplified (Table 1)

| Recent overview of topics from the latest version
The latest version of the Japanese Gastric Cancer Treatment Guidelines (January 2018) contains a novel algorithm for gastric cancer. 4 Briefly, the committee added novel treatment indications for stage IV gastric cancer, EGJ carcinoma, standard D2 gastrectomy,

| Clinical indications of stage IV gastric cancer
Regarding the novel algorithm of version 5 (2018), new treatment guidelines of stage IV gastric cancer were depicted in greater detail than in the previous version. In advanced gastric cancer with a single stage IV non-curable factor except for cytology positive factor, a recent phase III study (REGATTA study) showed that cytoreductive gastrectomy followed by chemotherapy did not show any survival benefit compared with chemotherapy alone. Therefore, cytoreductive gastrectomy followed by chemotherapy cannot be justified for the treatment of patients with stage IV factors (CQ1). 20 However, some clinical concerns may remain unanswered for elderly or highrisk patients who cannot tolerate standard chemotherapy.
Although the REGATTA study indicated demerit for the strategy of palliative gastrectomy followed by chemotherapy, it remains unclear whether conversion surgery with curative intent for stage IV gastric cancer would be justified after complete response for stage IV factors following intensive chemotherapy. 21  Regarding patients with only CY1 factor (CQ20) or minute peritoneal metastasis, which is recognized as a clinical status of micrometastasis, and only a low tumor burden at the peritoneal cavity, perioperative S-1 based chemotherapy for CY1 26 or HIPEC (hyperthermic intraperitoneal infusion chemotherapy) for CY1 27 or i.p. and i.v. paclitaxel plus S-1 28 in addition to curative gastrectomy could be considered. Intraperitoneal and i.v. paclitaxel plus S-1 appears to be effective not only for CY1 or low-tumor-burden peritoneal metastasis but also for highly advanced peritoneal metastasis cases with malignant ascites. Also, recent studies suggest that i.p. taxane may be useful to control highly advanced peritoneal metastases. 29

| New guidelines for the extent of lymphadenectomy and type of gastrectomy in EGJ carcinoma
Gastric cancer located in the cardia or the EGJ has drastically increased in Asia and South America as well as in the USA and Europe. 30 There is considerable controversy as to whether EGJ adenocarcinoma in the Nishi classification, which is similar to Siewert type II carcinoma, is actually esophageal cancer or gastric cancer.
From the viewpoint of lymphatic spreading, total gastrectomy is apparently a more common procedure than proximal gastrectomy in this population. Nevertheless, in EGJ adenocarcinoma of less than 40 mm, a recent nationwide study identified that the therapeutic value of lymphadenectomy was high at station nos 3, 1, 2 and 7, and the incidences of metastasis at station nos 4sa, 4sb, 4d, 5 and 6 were <1% even in patients with high dissection rates. 31 Previous studies also showed that the incidences of metastasis at station nos 4d, 5 and 6 were low, and that the benefits of prophylactic nodal dissection of these regions were questionable 32,[34][35][36][37][38] (Table 2). The new guidelines indicated in CQ6 that complete nodal clearance along the distal portion of the stomach offers only a marginal survival benefit, and total gastrectomy is not essential for local control in EGJ adenocarcinoma of <40 mm.

| New guidelines for standard D2 gastrectomy for advanced proximal gastric cancer
In Japan, total gastrectomy with splenectomy is carried out as a standard D2 procedure for complete removal of the splenic hilar lymph nodes, which are defined as a station no. 10 lymph node.
However, a recent Japanese multicenter phase III trial (JCOG0110) compared splenectomy with spleen-preserving surgery and con- Regarding bursectomy for resectable cT3-T4a gastric cancer, it has been carried out as a standard component of D2 gastrectomy during curative distal or total gastrectomy. However, it remains controversial as to whether bursectomy can prevent peritoneal metastasis. 40,41 A recent prospective JCOG1001 trial showed that bursectomy did not provide a survival advantage over non-bursectomy. 42 Therefore, the new guidelines suggested that bursectomy should be avoided as a standard component of D2 gastrectomy.
However, D2 dissection with omentectomy is still warranted as standard surgery for resectable cT3-T4a gastric cancer due to a lack of evidence against it.

| New categories for the indications and curability of endoscopic treatment
In the new treatment guidelines, two major revisions were made in the treatments using endoscopy. Indication for endoscopic resection was divided into three categories, which are absolute indication, expanded indication, and relative indication. Also, endoscopic Therefore, a real-world validation analysis using the National Clinical Database (NCD), which includes resection data by non-expert endoscopists, is also needed to investigate whether the eCura system is indeed useful in practical clinical settings. Recently, patients who required radical surgery after endoscopic submucosal dissection for early gastric cancer were investigated. 44 The eCura system consists of five clinicopathological factors, which are scored as follows: one point each for tumor size >30 mm, positive vertical margin, venous invasion and SM2 (depth of tumor invasion into the submucosa is ≥500 μm from the muscularis mucosa), and three points for lymphatic invasion. Although the rate of lymph node metastasis is already presented in the new guidelines, caution is needed when applying this system because these data have a selection bias using a small number of patients and include only 14.8% of undifferentiated-type early gastric cancer. 44 Also, the status of the positive vertical margin may indicate the presence of advanced cancers. Nevertheless, these data might be useful for elderly or high-risk patients with comorbidities to avoid additional surgery with lymphadenectomy.

| New guidelines for chemotherapy for advanced or recurrent gastric cancer patients
Recent topics concerning neoadjuvant chemotherapy for scirrhoustype gastric cancer such as type 4 or large type 3 gastric cancer (JCOG0501) 45 and adjuvant chemotherapy for stage III gastric cancer (JACCRO GC-07) 46 were previously nicely reviewed. 47 In the present study, we summarized the new guidelines for chemotherapy for advanced or recurrent gastric cancer patients, and chemotherapeutic regimens were classified into two categories as follows: (i) recom-

| Recent topics of preoperative nutrition and absorptive disorders following gastrectomy
Regarding preoperative nutrition, prealbumin concentration, which is a rapid turnover protein and a real-time and more sensitive nutritional indicator than albumin, was proven to be independently correlated with overall morbidity (OR, >22 mg/dL vs <15 mg/L: 1.0 vs 4.5). 55 Prealbumin could be a pivotal indicator to improve preoperative nutrition and avoid complications.
A major postoperative concern is weight loss, which is associated with marked deterioration in quality of life (QOL), reduced tolerance to chemotherapy 56 and worsening of the final prognosis. 57 A recent multicenter randomized control trial (RCT) identified that a regular diet plus an oral elemental nutritional supplement for 6-8 weeks using 300 kcal/day of Elental (Ajinomoto Pharmaceuticals, Tokyo, Japan), which contains essential amino acids with a low fat content, could significantly attenuate body weight loss following gastrectomy, especially following total gastrectomy. 58 However, another recent RCT could not show the significance of using immunonutrition based on an eicosapentaenoic acid-enriched oral nutritional supplement for 3 weeks perioperatively using 600 kcal/day. 59 The reason for the negative result is that an oral nutritional supplement may give rise to a decrease in oral intake. In contrast, recent studies have suggested that home enteral nutrition for 6 weeks using jejunostomy feeding did not affect oral intake and improved postoperative nutrition after esophagectomy and total gastrectomy in esophagogastric cancer. 60 Forced nutrition using jejunostomy home feeding may be one of the strategies in high-risk weight loss patients after surgery. Regarding

| Nucleic acids as liquid biopsy: Circulating DNA as a clinical biomarker and companion diagnosis in gastric cancer
The concept of liquid biopsy has become widely accepted in a clinical setting. Liquid biopsy is a less invasive approach for obtaining genetic and epigenetic aberrations that are closely associated with cancer initiation and progression. Moreover, liquid approaches allow for repeated sampling, and this makes it possible to evaluate the longitudinal evolution of a tumor and its heterogeneous characteristics, which single sampling may fail to capture. 76 Recently, a liquid biopsy using nucleic acids such as cell-free DNAs and microRNAs in blood could be realized in clinical settings. We clarified the utility of the digital polymerase chain reaction (PCR)-based HER2 copy number assay as a liquid biopsy to detect ERBB2 amplification in blood cell-free DNA for tumors having heterogeneities. 77 We identified the potential utility of circulating cell-free DNA to detect EBV-DNA.
Identification of the EBV subtype using liquid biopsy could be useful for real-time monitoring of tumor progression and treatment response. 78 Also, a recent study identified a blood test, Cancer SEEK, that can detect eight common cancer types including gastric cancer through assessment of the levels of circulating proteins and mutations in cell-free DNA. 79 4.5 | Nucleic acids as liquid biopsy: Circulating microRNAs as a clinical biomarker and future perspectives in gastric cancer MicroRNAs (miRNA), which are small non-coding RNAs, regulate the translation of specific protein-coding genes. Altered expressions of miRNAs contribute to the development of gastric cancers. In gastric cancer, various cancer-related miRNAs and their target genes were detected (Table 3). Also, various circulating miRNAs have already been proven to have the potential to enable diagnosis of gastric cancer at an early stage, predict prognosis and recurrence, evaluate patient status and therapeutic efficacy and provide optimal, individualized treatment strategies in gastric cancer. 80 As a next-generation liquid biopsy biomarker reflecting tumor dynamics, particularly, the upregulated oncogenic miRNAs in blood such as miR-20a, miR-21, miR-25, miR-106a, miR-106b, miR-199a-3p, miR-221, miR-223 and miR-421, might be useful candidates as blood-based biomarkers for gastric cancer.
Regarding the downregulated tumor suppressor miRNAs in blood, let-7a, miR-101, miR-181b, miR-203, miR-204, miR-218, miR-375, miR-451 and miR-486 might be useful candidates as blood-based biomarkers and oligonucleotide therapeutics for gastric cancer. Liquid biopsy using circulating tumor cells and cell-free nucleic acids such as cell-free DNAs and miRNAs in gastric cancer patients could provide valuable new insights into prognosis and treatments in the near future. However, in the present review, we could summarize only some of the major topics and had to exclude topics on less invasive surgeries such as laparoscopic 81,82 and robotic gastrectomy and clinical study regarding the surgical approach based on the sentinel node concept for early gastric cancer. 83 However, additional updates on major topics including these topics could be summarized in the near future.