The following module was designed to supplement the clinical experience of medical students. It covers the anatomy, epidemiology, screening, presentation, diagnosis, staging, prognosis, management, and follow up care of gastric cancer.
This module covers the following objectives:
The stomach is a hollow, muscular organ situated in the left upper abdomen. It is bound proximally by the gastroesophageal junction (GEJ) and distally by the pyloric sphincter. Adjacent structures include the diaphragm, liver, spleen, left kidney and adrenal, pancreas, and transverse colon.
The stomach consists of four main anatomical regions which differ histologically and functionally:
The stomach is also described as having lesser and greater curvatures. The lesser curvature, which forms the medial border of the stomach, attaches to the liver by the lesser omentum. At the lateral edge, the greater curvature of the stomach is attached to the diaphragm, spleen, and transverse colon by the greater omentum.
The wall of the stomach is divided into five main layers. From inside to outside, these are:
Adenocarcinomas account for 95% of gastric cancers. In fact, the term “gastric cancer” used without qualification is understood to refer to gastric adenocarcinoma . Rarer gastric cancers include lymphomas (second most common), gastrointestinal stromal tumors, carcinoid tumors, adenocanthomas, squamous cell carcinoma, small cell carcinoma, and leiomyosarcoma.
Gastric adenocarcinomas can be classified histologically as either diffuse or intestinal type, as described by Laurén . The intestinal type is named for its tendency to form glands resembling those found in the colon . Intestinal type adenocarcinomas usually develop in the distal stomach and arise from precursor lesions. Risk factors include H. pylori infection, male sex, and older age. In contrast, diffuse type tends to be poorly differentiated and has a greater propensity to invade and spread. They are usually found in the proximal stomach and do not have precursor lesions. Compared to intestinal type, the diffuse type is less strongly linked to H. pylori and more prevalent in women and younger patients [4, 5]. The prognostic value of Laurén classification independent of TNM stage is uncertain .
Tumor site is also important in the discussion of gastric cancer because it affects lymphatic drainage and treatment. A common distinction is cardia versus noncardia gastric cancer. Adenocarcinoma near the cardia are further grouped by Siewert classification :
Siewert types I and II adenocarcinomas are staged as esophageal cancer while Siewert type III adenocarcinomas are staged as gastric cancer . Only Siewert type III is covered in this module. Please refer to the module on esophageal cancer for Siewert types I and II cancer.
Gastric cancer is the 11th most commonly diagnosed cancer in Canada with 3500 new cases and 2100 deaths in 2017 . Worldwide, gastric cancer ranked 6th in total diagnoses with 1,034,000 new cases in 2018. It was responsible for 783,000 deaths, making it the second-leading cause of death after lung cancer .
The incidence of gastric cancer is in steady decline in Canada and worldwide, a trend attributed to improved hygiene, dietary changes, and the eradication of H. pylori [10, 11]. A notable exception is gastric cancer originating from the cardia, whose incidence has remained stable or increased in the US and Europe against the overall trend [12, 13, 14].
Infection by H. pylori is strongly linked to non-cardia gastric cancer and gastric lymphoma, increasing risk by around sixfold [14, 15]. Once infected, the lifetime risk of developing gastric cancer is 1% . Importantly, the eradication of H. pylori reduces this risk .
Obesity increases the risk of gastric cancer of the cardia, but its association with non-cardia gastric cancer is weaker [18, 19].
Smokers are more likely to develop gastric cancer, with dose and duration both contributing to risk [20, 21]. The association is especially strong in intestinal type gastric cancer and gastric cancer involving the cardia . After quitting, the risk of developing gastric cancer decreases over time, with an incidence approaching that of non-smokers after 10 years .
Heavy alcohol consumption has consistently been shown to increase the risk of gastric cancer, but the effects of lighter use are less clear [22, 23]. A nonlinear dose-response relationship may explain this finding .
It appears that high salt intake increases the risk of gastric cancer. In contrast, fruits and vegetables may have a protective role, possibly mediated by vitamin C .
Adjusted for age, men are approximately twofold more likely to develop gastric cancer [10, 26]. Men are projected to account for 2200 of the 3500 annual cases of gastric cancer in Canada .
The incidence of gastric cancer increases with age. Among those newly diagnosed with gastric cancer, 93% are at least 45 years old .
Geographically, the incidence of gastric cancer is highest in East Asia and Eastern Europe . Some of this risk persists in the offspring of parents who migrate to the West .
Positive family history appears to increase the risk of gastric cancer by 1.5 to 3.5 fold in most studies, independent of H. pylori infection and dietary contributions . However, fewer than 3% of gastric cancers are considered hereditary. The most common of the hereditary gastric cancers is hereditary diffuse gastric cancer, in which germline mutations of the CDH-1 gene are most strongly implicated. Rarer hereditary causes of gastric cancer include Lynch syndrome, familial adenomatous polyposis, Li-Fraumeni syndrome, and Peutz-Jeghers syndrome .
The two primary screening modalities for gastric cancer are upper endoscopy and radiography, with upper endoscopy being more effective in reducing mortality from gastric cancer .
Owing to the low incidence of gastric cancer in Western populations and the uncertain value of screening on health outcomes in the West, screening for gastric cancer in average-risk asymptomatic patients is generally discouraged .
Screening programs exist in areas where gastric cancer is more endemic, and they appear to reduce gastric cancer mortality; however, the effect on overall survival is less established . Endoscopic screening may be considered on this basis for recent immigrants from high-risk areas .
Patients at risk for hereditary cancer syndromes should be considered for confirmatory testing and screen on a case by case basis. A hereditary cancer program can be helpful in the management of these patients .
Early stage gastric cancer is usually asymptomatic, and its symptoms are usually nonspecific even when present. As a result, only 28% of patients in the US are diagnosed with localized (N0M0) disease .
The most common symptoms at the time of diagnosis are abdominal pain (50-65%) and weight loss (40%) . In addition to weight loss, other so-called alarm symptoms include dysphagia, persistent vomiting, early satiety, and signs of upper GI bleeding. All of these are generally manifestations of advanced disease but still have limited predictive value owing to the rarity of gastric cancer and the prevalence of these symptoms in the general population . On the other hand, positive findings on the physical examination such as organomegaly or lymphadenopathy necessarily require very advanced disease with dismal prognosis. A summary of presenting signs and symptoms is below.
Deciding on which patients to investigate with endoscopy is challenging given the limited value of the history and physical examination. To aid risk stratification, the patient’s risk factors and past medical history should be considered.
Without a clear alternative explanation, any patient with alarm symptoms in Table 1 should undergo further assessment. Workup is also reasonable for patients with persistent or progressive abdominal pain and those over 55 with persistent or progressive dyspepsia . Upper endoscopic examination with biopsy is the diagnostic test of choice for gastric cancer.
Once the diagnosis of gastric cancer is established, the following tests should beconsidered for staging and to guide management:
TX: primary tumor cannot be assessed
T0: no evidence of primary tumor
Tis: tumor confined to mucosa without lamina propria invasion
T1a: tumor invades lamina propria or muscularis mucosa
T1b: tumor invades submucosa
T2: tumor invades muscularis propria
T3a: tumor penetrates subserosa but not serosa or adjacent organs
T4a: tumor invades serosa
T4b: tumor invades adjacent structures
NX: lymph nodes cannot be assessed
N0: no regional* or common iliac lymph node metastases
N1: 1-2 positive regional lymph nodes
N2: 3-6 positive regional lymph nodes
N3a: 7-15 positive regional lymph nodes
N3b: 16+ positive regional lymph nodes
*regional lymph nodes refer to left/right paracardial, perigastric, celiac axis, left gastric artery, common hepatic artery, splenic, and hepatoduodenal nodes
M0: no distant metastases
M1: distant metastases are present
Stage 0: Tis N0 M0
Stage I: T1-2 N0 M0
Stage IIA: T1-2 N1-3 M0
Stage IIB: T3-4a N0 M0
Stage III: T3-4a N1-3 M0
Stage IVA: T4b N0-3 M0
Stage IVB: M1 disease
Post-neoadjuvant therapy (ypTNM)
Stage I: T1-2 N0 M0 or T1 N1 M0
Stage II: T3-4a N0 M0 or T2-3 N1 M0 or T1-2 N2 M0 or T1 N3 M0
Stage III: T4b N0 M0 or T4a-b N1 M0 or T3-4b N2 M0 or T2-4b N3 M0
Stage IV: M1 disease
Stage 0: Tis N0 M0
Stage IA: T1 N0 M0
Stage IB: T2 N0 M0 or T1 N1 M0
Stage IIA: T3 N0 M0 or T2 N1 M0 or T3 N0 M0
Stage IIB: T4a N0 M0 or T3 N1 M0 or T2 N2 M0 or T1 N3a M0
Stage IIIA: T4b N0 M0 or T4a N1 M0 or T3-4a N2 M0 or T2 N3a M0
Stage IIIB: T4b N1-2 M0 or T3-4a N3a M0 or T1-2 N3b M0
Stage IVA: T4b N3a M0 or T3-4b N3b M0
Stage IVB: M1 disease
SEER  reports that the 5-year survival rates for gastric cancer (based on patients diagnosed between 2009 and 2015) are 68% for localized disease, 31% regional disease, and 5% for distant metstatic disease.
Factors that affect prognosis aside from disease stage include performance status, ALP level, and race . Asians in particular appear to have more favorable outcomes than Caucasians, Blacks, and Hispanics regardless of tumor location . Histologic grade and tumor location have not consistently been found to affect prognosis independent of stage [7, 34, 35].
Approximately 30% of patients treated surgically with curative intent will experience recurrence (with follow-up varying from 2-14 years). Recurrence is usually early with a median duration to failure of 10.8 months. The failure pattern is predominantly distant .
Surgery, radiotherapy, and systemic therapy are all routinely used in the management of gastric cancer. Treatment plans should be tailored to each patient and benefit from multidisciplinary discussion.
Surgery is the mainstay of gastric cancer treatment with curative intent, with the procedure of choice depending on patient and tumor characteristics.
Endoscopic mucosal resection or endoscopic submucosal dissection is appropriate for the treatment of favorable early-stage gastric cancer based on depth of invasion (Tis or T1a), size (2cm maximum), pathology (well or moderately-well differentiated), and the absence of ulceration or lymphovascular invasion . Patients meeting these criteria have a <5% chance of LN metastases, so LN dissection may be omitted . Those with less favorable early-stage disease who are unfit for surgery are also candidates for endoscopic resection .
Because gastric cancers are usually more advanced at the time of diagnosis in Western countries, gastrectomy with lymph node dissection is far more commonly offered to those treated with curative intent. Multimodal therapy should be considered for T2+ or node-positive disease.
For distal tumors that are at least 6cm removed from the gastric cardia, distal gastrectomy is preferred for being less morbid while maintaining similar survival rates to total gastrectomy . The morbidity benefits of proximal gastrectomy for more proximal tumors over total gastrectomy are less established. As such, proximal tumors are typically managed with total gastrectomy, although proximal gastrectomy may be considered in select patients . Regardless of technique, an attempt should be made to achieve negative margins at least 4cm from the gross tumor.
The optimal extent of lymph node dissection during gastrectomy is controversial. A discussion of the merits of the various levels of lymph node dissection is beyond the scope of this module. Readers interested in details are referred to the guidelines published by the National Comprehensive Cancer Network (NCCN) and the Japanese Gastric Cancer Association. The number of nodes that are dissected is important for accurate staging and possibly survival, with a recommended minimum of 15-16 lymph nodes [7, 37, 39].
The effectiveness of multimodal therapy has been demonstrated across multiple trials. Supplementing surgery with chemotherapy or radiation therapy should be strongly considered for all T2+ or node-positive disease . Regimens that have shown superiority to surgery alone include:
Neoadjuvant chemoradiotherapy, whose use was extrapolated from esophageal cancer, has also shown promising early results for gastric cancer [37, 46]. Recently, perioperative fluorouracil, leucovorin, oxaliplatin, and docetaxel (FLOT) has gained traction after it was demonstrated to improve survival over ECF . Finding the optimal sequence and combination of multimodal therapy remains an area of active research, with many variations currently under investigation.
Literature on the treatment of locally advanced disease is limited. Fluorouracil-based chemoradiotherapy appears to be more effective than either chemotherapy or radiotherapy alone in two older trials [48, 49]. For medically fit patients, chemoradiation may increase the likelihood of conversion to resectable disease . The alternative is combination chemotherapy alone, as with metastatic disease discussed henceforth.
Chemotherapy is the cornerstone of treatment for patients with advanced gastric cancer. Compared to supportive care, chemotherapy has been found to improve quality of life and prolong survival from 4.3 to 11 months . Two-agent chemotherapy, which extends survival compared to single-agent chemotherapy, is recommended for most patients [37, 50]. The combination of a fluoropyrimidine (e.g. 5-fluorouracil or capecitabine) and a platinum agent (e.g. cisplatin or oxaliplatin) is considered first line, although several alternative regimens have also shown promising results . Combination chemotherapy with three cytotoxic drugs further prolongs survival and increases response rates, albeit with greater toxicity; these regimens may be appropriate as first-line treatment for select medically fit patients with good performance status [37, 50].
A number of targeted therapies have emerged with encouraging results in specific patient populations over the past decade [51, 52, 53]. Health Canada has approved trastuzumab and ramucirumab for HER2-overexpressing and previously-treated patients with advanced gastric cancer, respectively. Other targeted therapies are also showing promising early results, including FDA-approved pembrolizumab for select MSI-H/dMMR tumors .
Radiotherapy is effective for local palliation, with more than two-thirds of patients experiencing bleeding, pain or obstruction responding to therapy. Concurrent chemoradiotherapy is associated with significantly higher toxicity than radiotherapy alone, but there is a lack of evidence comparing local outcomes . The benefits to chemotherapy on survival are, however, well established. Other options used to control local symptoms include endoscopic treatment, interventional radiology, medical therapy (e.g. proton pump inhibitors), stenting, and stenting . In view of the significant heterogeneity in this population, there are unsurprisingly few studies comparing the effectiveness of different modalities. A multidisciplinary approach is especially critical for this diverse group of patients.
Recommendations on the treatment of gastric cancer are summarized below. This is not a comprehensive account of all treatment options for gastric cancer. The NCCN provides up-to-date guidelines on the management of gastric cancer.
Side effects to gastric radiation are common and sometimes necessitate breaks or discontinuation of therapy, with deleterious effects on survival . Concurrent chemotherapy appears to compound toxicity . For example, in the landmark INT-0116 trial which established the use of adjuvant chemoradiation, 54% of patients experienced grade 3+ hematologic toxicity and 33% experienced grade 3+ gastrointestinal toxicity. A total of 17% stopped chemoradiotherapy altogether as a result of unacceptable toxicity . IMRT, which has the potential to lower the dose delivered to end organs, is especially preferable in cases where there is a concern for high-grade toxicity. We list some common and serious complications below .
These complications often resolve within 1-2 weeks of completing therapy but demand close follow up and aggressive management.
The value of aggressive surveillance is dependent on the stage and modality of treatment offered. Overall, there is a lack of prospective studies guiding follow up and no expert consensus on what constitutes best practice. An example of a surveillance schedule for patients treated curatively, depending on the stage of disease and selected treatment, is below . Follow up should be tailored to each patient based on their unique circumstances.