GC can present at an early or advanced disease stage. “Early gastric carcinoma” (EGC) is defined as a carcinoma which has infiltrated the mucosa or submucosa regardless of the presence or absence of lymph node metastases [19, 20]. Conversely, GCs infiltrating into the muscularis propria and beyond are defined as “advanced.”

EGCs are classified into three types based on the endoscopic appearance according to the Paris classification: type I (protruded), polypoid growth (subcategorized into Ip (pedunculated) and Is (sessile)); type II (superficial), non-polypoid growth (subcategorized into type IIa (slightly elevated), type IIb (flat), and type IIc (slightly depressed)); and type III, excavated growth [21] (Figs. 6.2 and 6.3).

The macroscopic appearance of advanced GC is classified using the Borrmann classification [22] which divides GC into four distinct types: type I, polypoid type; type II, fungating; type III, ulcerated; and type IV, diffusely infiltrative (Fig. 6.4).

While the macroscopic appearances are different between early and advanced GC, the histological appearances are similar. Two major histological GC subtypes (intestinal type GC and diffuse type GC) have been described by Laurén [23] which have different clinicopathological profiles and molecular pathogenesis and often occur in distinct epidemiologic settings.

According to the World Health Organization (WHO) [1], GCs are classified as tubular, papillary, mucinous, poorly cohesive (with or without signet ring cells), and mixed (Fig. 6.5). Tubular and papillary carcinomas roughly correspond to the intestinal type and poorly cohesive carcinomas correspond to the diffuse type according to Laurén’s classification (Table 6.1). The Laurén and WHO classifications are the ones most commonly used outside of Japan. In Japan, the recommended histological typing is similar but not 100 % identical to the WHO classification [24].

Nakamura’s classification into differentiated and undifferentiated subtype is used together with the size of the lesion and presence or absence of ulceration to decide whether a lesion can be treated endoscopically [25, 26]. Apart from the classifications based on tumor morphology, GC can be classified on the basis of the presence or absence of cell differentiation markers – MUC5AC and trefoil peptide TFF1 (markers of surface gastric epithelium (foveolar cells)), MUC6 and trefoil peptide TFF2 (markers of mucus neck cell, pyloric gland, and Brunner’s gland cells), and MUC2, CDX-2, and CD10 (markers of intestinal goblet cells) – into four phenotypes: (1) gastric, (2) mixed gastric and intestinal, (3) intestinal, and (4) unclassifiable or null phenotype which does not express any of these markers [27–29].

The staging for carcinoma of the stomach was substantially modified in 2009 as detailed in Table 6.2. Major changes included the subdivision of T1 cancers into T1a (mucosa) and T1b (submucosa), the renaming of T2a (muscularis propria) as T2 and T2b (subserosa) as T3, and the subdivision of T4 (serosa) into T4a (penetrates serosa) and T4b (invades adjacent structures). Consequently, the categorization of the T (depth of invasion) is now uniform throughout the gastrointestinal tract, whereas differences remain for the categorization of the N (presence or absence of regional lymph node metastases). The N categories for GC are N0 (no regional lymph node metastasis), N1 (1 to 2 lymph node metastases), N2 (3 to 6 lymph node metastases), N3a (7 to 15 lymph node metastases), and N3b (metastases in 16 or more regional lymph nodes) [19, 20].

Gastric carcinomas can spread by (i) direct extension to adjacent organs, (ii) lymphatic invasion, (iii) blood vessel invasion, and/or (iv) peritoneal dissemination. Intestinal type GCs preferentially metastasize hematogenously to the liver, whereas diffuse type GCs preferentially metastasize to peritoneal surfaces [1]. GCs with mixed histological phenotype exhibit the metastatic patterns of both types.

A recent meta-analysis comparing survival rates after gastrectomy between GC patients from the West and the East from patients recruited into large randomized controlled clinical trials showed an association between type of surgical resection performed in the East and improved survival [30]. The known difference in surgical techniques between the East and the West is one potential variable that may be responsible for discrepancy in outcomes. Noguchi et al. [31] reported a survival difference between high-volume centers in the USA and Japan which was no longer apparent after adjusting for tumor location. Verdecchia et al. [32] demonstrated that the survival of Italian GC patients was inferior to that of Japanese GC patients and that this survival difference disappeared after adjusting for stage. Bollschweiler et al. [33] compared the survival of Japanese and German GC patients and concluded that the country itself was a prognostic factor. Higher frequency of early stage GC and more accurate staging have also been associated with improved survival in Japan compared with Western nations [34].

Early and advanced GCs differ in prognosis. Japanese patients with EGC have an excellent prognosis with a 5-year survival rate exceeding 90 % after surgical treatment. Nevertheless, approximately 2 % of EGC recur after curative resection and lymph node metastases occur in 2–3 % of intramucosal carcinomas [35, 36] and 20–30 % of submucosal carcinomas [37]. Risk factors for lymph node metastasis in EGC include age at time of diagnosis (the younger, the more frequent the lymph node metastases), size greater 20 mm, depressed macroscopic type, grade of differentiation, presence of an ulcer or scar, lymphatic channel invasion, and submucosal invasion by more than 500 μm [35, 37].

Five-year survival rate of advanced GC, the most frequent type in the West, is around 23 % when treated by surgery alone and around 36 % when treatment includes perioperative chemotherapy [38]. For advanced GC, depth of infiltration into the wall (T category of the TNM classification), number of lymph node metastases (N category of the TNM classification), and presence of distant metastases (M category of the TNM classification) remain the strongest prognostic indicators [19, 20]. Lymphatic and venous invasion are also predictors of poor survival in GC. Perineural invasion correlates with T stage and tumor size and may serve as a marker of advanced disease [39].

On the basis of clinical criteria, the International Gastric Cancer Linkage Consortium defined families with HDGC syndrome as families meeting one of two criteria: (i) two or more documented cases of diffuse type GC in first- or second-degree relatives with at least one of them diagnosed before the age of 50 years or (ii) three or more documented cases of diffuse GC in first- or second-degree relatives independent of the age at diagnosis [41]. Women in these families have an elevated risk of lobular breast cancer. The criteria for genetic testing were updated in 2010 [42]. In several HDGC families, a higher incidence of orofacial clefts has been noted [43, 44].

Alterations of the CDH1 gene, which encodes E-cadherin, constitute the genetic causal event in HDGC patients [45]. In clinically defined HDGC patients, CDH1 germline mutations are detected in 30–40 % of cases [42]. Seventy-five to eighty percent of CDH1 mutations are truncating mutations, and the remaining are missense mutations. In addition, large germline deletions have also been found in HDGC families which tested negative for point mutations [46].