**2. Oncogenic pathway combinations predict outcome of gastric cancer patients**

Gastric cancers are molecularly heterogeneous tumors, showing different dysregulated oncogenic pathways such as E2F, K-RAS, p53, and Wnt/β-catenin signaling that occur with varying frequencies in these types of tumors [16, 17]. In contrast to the previous studies that have focused on single pathways [18–20], experimental evidence indicates that, in most cases, carcinogenesis is dictated by complex interactions between multiple pro- and anti-oncogenic signaling pathways [21].

According to the World Health Organization classification, the vast majority of gastric cancers are adenocarcinomas, divided into papillary, tubular, mucinous (colloid), and poorly cohesive carcinomas (including signet-ring cell carcinoma and other variants) [4]. Although it was proposed a long time ago (1965), the Lauren classification is still widely used in clinical practice and subdivides gastric carcinomas into intestinal and diffuse types, associated with different pathogenesis, ways of spreading, and outcome [5]. Unfortunately, these two classification systems have little clinical impact, making the development of classifiers that can

Literature data have demonstrated that the development of gastric cancer is associated in the majority of cases with infectious agents such as the Gram-negative spiral bacterium *Helicobacter pylori* (most often) [6] and Epstein-Barr virus (EBV) (about 9% of all cases of gastric cancer) [7]. Only a small percentage of gastric cancer patients (hereditary cases) are associated with germline mutation in E-cadherin (CDH1) [8] or mismatch repair genes (Lynch syndrome) [9]. In contrast to the familial clustering of gastric cancer, sporadic mismatch repair-deficient gastric tumors present epigenetic silencing of hMLH1 and p16 in the context of a CpG island

Due to a lack of early specific clinical features, most patients with gastric cancer are diagnosed in advanced stages, resulting in poor 5-year survival rates [11], with a median survival of less than 1 year in case of metastatic stage IV patients [12–14]. Nowadays, survival has gained only minor improvement despite the advances in diagnostic techniques, the multidisciplinary therapeutic management, and the development of novel molecular targeted treatment agents. Unfortunately, despite modern treatments, less than a quarter of gastric cancer patients survive longer than 5 years after surgery. Gastric cancer represents a complex disease, showing

Recent data have demonstrated that gastric tumors present a high molecular heterogeneity involved both in the process of carcinogenesis and cancer spread. By identifying the specific molecular patterns of the tumor, it is possible to assess its behavior, the prognosis of the patient, and also to decide the most appropriate treatment, a much more personalized one. It is well known that in the pathogenesis of stomach cancers many different genetic mutations, epigenetic alterations, as well as dysregulated signaling pathways are involved, each of these

Currently, novel therapeutic agents that target some of these aberrant molecular signaling pathways are already part of the standard treatment of gastric cancer, while others remain to

**2. Oncogenic pathway combinations predict outcome of gastric cancer** 

Gastric cancers are molecularly heterogeneous tumors, showing different dysregulated oncogenic pathways such as E2F, K-RAS, p53, and Wnt/β-catenin signaling that occur with varying

major differences in their tumor cell behavior and responses to chemotherapy.

molecular abnormalities acting in a different stage of the disease.

prove their therapeutic efficacy in the setting of clinical trials [15].

define prognosis and guide patient's treatment as an urgent need.

methylator phenotype (CIMP) [10].

110 Gastric Cancer

**patients**

Unlike previous gastric tumor microarray researches relating different gene expression patterns to specific histological features or anatomical location [22], Ooi et al. [23] have succeeded in subdividing gastric cancers into molecularly homogenous subgroups that enable personalizing patient treatments and improving prognosis. It was for the first time when, by using multiple patterns of oncogenic pathway activation, a novel cancer classification approach has been developed, namely a genomic taxonomy of gastric cancer. They developed an in silico technique to define activation levels of different oncogenic pathways implemented in context of complex tumor profiles and validated this classification approach using proof-of-concept examples from breast cancer. Afterward, they have applied this method to gastric tumors and evaluated 11 oncogenic pathways previously known to be involved in gastric tumorigenesis [16–20, 24–27]. They have assessed over 300 primary stomach cancers coming from three independent patient cohorts. The researchers have discovered three oncogenic pathways, nuclear factor-kB (NF-κB), Wnt/β-catenin, and proliferation/stem cell, which were dysregulated in over 70% of gastric cancers and validated the patient stratification in vitro using gastric cancer cell lines. Patient classification by oncogenic pathway combinations revealed significant survival differences, suggesting a major role for pathway combinations in determining gastric cancer behavior. Therefore, gastric cancer can be taxonomized into biologically, molecularly, and clinically significant subtypes.

The authors defined concomitant activation of different oncogenic pathways, such as of E2F, MYC, p21 (repression), and the "proliferation/stem cell" pathway, most likely due to increased cellular proliferation in tumor cells [28], and in stem cells (embryonic stem cells (ESCs)) [29]. Co-activation of different pathways demonstrates the ability of the cancer cell to dictate the activity of multiple pathways.

The study showed that NF-κB signaling may be elevated in a significant proportion of gastric cancers probably due to *H. pylori* infection [30]; therefore, targeted NF-κB inhibitors may represent an appropriate treatment for gastric tumors. Pathway-based taxonomies may be useful in developing potential pathway inhibitors and novel targeted therapies that would be studied on prestratifying patients using molecular or pathologic criteria.
