Gastric Adenocarcinoma is the fourth-most common cancer worldwide and the tenth-most common malignancy in the United States. Its incidence has decreased dramatically over the last 60 years, perhaps secondary to improvements in refrigeration and diet. In addition, the anatomic pattern of gastric cancer is changing, with proximal or cardia cancers comprising a greater proportion of gastric cancers. Approximately one third of gastric cancers are metastatic at presentation. The overall 5-year survival rate is 15%.
The etiology of gastric cancer is complex and multifactorial, involving a combination of genetic, environmental, and infectious risk factors. Risk factors for gastric cancer include male gender, family history, polyposis syndromes, diets high in nitrates, salts, or pickled foods, adenomatous gastric polyps, previous gastric resection, Ménétrier disease, smoking, H. pylori infection, and chronic gastritis. Aspirin, fresh fruits and vegetables, selenium, and vitamin C may be protective against the development of gastric cancer.
- Ninety-five percent of gastric cancers are adenocarcinomas arising from mucus-producing cells in the gastric mucosa. The Lauren classification system is most widely used and divides gastric cancers into two subtypes:
- Intestinal-type cancers (30%) are glandular and arise from the gastric mucosa. Occurring more commonly in elderly men and in the distal stomach, they are associated with H. pylori and other environmental exposures that lead to chronic gastritis, intestinal metaplasia, and dysplasia.
- Diffuse-type cancers (70%) arise from the lamina propria and are associated with an invasive growth pattern with rapid submucosal spread. They occur more commonly in young patients, females, and in the proximal stomach. Transmural and lymphatic spread with early metastases are more common, and diffuse-type cancers have worse overall prognosis.
Presentation of gastric cancer
generally involves nonspecific signs and symptoms such as epigastric abdominal pain, unexplained weight loss, nausea, vomiting, anorexia, early satiety, and fatigue. Dysphagia is associated with proximal gastric cancers, whereas gastric outlet obstruction is more typical of distal cancers. Perforation and upper GI bleeding are the presenting manifestations in a minority of patients (1% to 4%) and generally portend advanced disease with poor prognosis.
Classic physical findings in gastric cancer represent metastatic and incurable disease and include the following:
- Enlarged supraclavicular nodes (Virchow’s node).
- Infiltration of the umbilicus (Sister Mary Joseph’s node).
- Fullness in the pelvic cul-de-sac (Blumer’s shelf).
- Enlarged ovaries on pelvic examination (Krukenberg’s tumor).
- Hepatosplenomegaly with ascites and jaundice.
Diagnosis can be made by
Double-contrast upper GI barium contrast studies or by EGD. Endoscopy is generally the diagnostic method of choice because it permits direct visualization and biopsy of suspicious lesions. Screening examination by endoscopy or contrast studies is not cost-effective for the general U.S. population, given the low incidence, but may be warranted in high-risk individuals, such as patients more than 20 years post–partial gastrectomy, patients with pernicious anemia or atrophic gastritis, immigrants from endemic areas (Russia, Asia), and patients with familial or hereditary gastric cancer. Mass screening in Japan, a country with high incidence of gastric cancer, resulted in an increase in the detection of gastric cancer confined to mucosa and led to improvements in 5-year survival rates.
Staging is important in determining prognosis and appropriate treatment.
The American Joint Committee on Cancer and International Union against Cancer (AJCC/UICC) jointly developed a staging system that is most widely used worldwide (Table 9-1). The distribution of stage at presentation in the United States is 20% stage I, 19% stage II, 34% stage III, and 27% stage IV (J Gastrointest Surg 2005;9:718). Once the diagnosis of gastric cancer is established, computed tomography (CT) and endoscopic ultrasonography (EUS) are the primary modalities employed for staging.
CT scan of the abdomen and pelvis is the best noninvasive modality for detecting metastatic disease in the form of malignant ascities or hematogenous spread to distant organs, most commonly the liver. Overall accuracy for tumor staging is 60% to 80% depending on the protocol used, but accuracy for determining nodal involvement is more limited and variable
EUS adds to the preoperative evaluation of gastric cancer in several ways. It is superior to CT in delineating the depth of tumor invasion in the gastric wall and adjacent structures and identifying perigastric lymphadenopathy. EUS is the most accurate method available for T staging of gastric cancer, and accuracy for N staging approaches 70%. Addition of fine needle aspiration (FNA) of suspicious nodes increases accuracy even further and brings specificity to near 100%.
Positron emission tomography (PET)/CT combines the spatial resolution of CT with the contrast resolution of PET. It is most useful for its specificity in detecting nodal and distant metastatic disease not apparent on CT scan alone. Preliminary studies suggest that the use of PET/CT in staging patients with gastric cancer leads to upstaging in 6% and downstaging in 9% of patients.
Laparoscopy significantly enhances the accuracy of staging in patients with gastric cancer. Routine use of laparoscopy has been shown to detect small-volume peritoneal and liver metastases in 20% to 30% of patients believed to have locoregional disease, thereby avoiding unnecessary laparotomy in these patients (Br J Surg 1985;72:449, Ann Surg 1997;225:262). Although laparoscopic ultrasound enhances the accuracy of staging in other gastrointestinal cancers, its role in gastric cancer awaits further study. Laparoscopy is not indicated in patients with T1 and T2 lesions, given the low incidence of metastases with these tumors (J Am Coll Surg 2003;196:965).
TABLE 9-1 TNM (Tumor, Node, Metastasis) Staging of Gastric Cancer
Surgery is the mainstay of curative therapy in the absence of disseminated disease.
Extent of surgical resection generally involves a wide resection to achieve negative margins with en bloc resection of lymph nodes and any structures involved by local invasion. In general, gross margins of 6 cm, confirmed to be negative intraoperatively with frozen section, are usually required to ensure microscopically negative margins on final histologic analysis.
Proximal tumors of the stomach comprise up to half of all gastric cancers and can be resected by total gastrectomy or proximal subtotal gastrectomy. Total gastrectomy with Roux-en-Y esophagojejunostomy is generally the preferred option to avoid postoperative morbidity of reflux esophagitis and impaired gastric emptying associated with proximal subtotal gastrectomy. Tumors of the GE junction may require esophagogastrectomy with cervical or thoracic anastomosis.
Midbody tumors comprise 15% to 30% of tumors and generally require total gastrectomy to achieve adequate margins.
Distal tumors may be resected by distal subtotal gastrectomy or total gastrectomy with no difference in overall survival (Ann Surg 1989;209:162, Ann Surg 1994;220:176). However, nutritional status and quality of life are superior following subtotal gastrectomy, making it the preferred option when adequate margins can be obtained while maintaining an adequate gastric remnant (Ann Surg 1997;226:613, Ann Surg 1999;230:170).
Early gastric cancers,
defined as tumors confined to the mucosa, have limited propensity for lymph node metastasis and may be treated by limited gastric resections or endoscopic mucosal resection. Experience outside of Japan with early gastric cancers is limited.
Laparoscopic gastric resections
have been reported for the treatment of gastric cancer, with advantages of reduced pain, shorter hospitalization, and improved quality of life. Long-term outcome with respect to cancer recurrence awaits further study in a randomized, controlled fashion.
Extent of lymphadenectomy has long been a controversial
issue in the surgical management of gastric cancer. Early retrospective Japanese studies showed improved survival with radical lymph node dissections. A standard (D1) lymphadenectomy entails removal of perigastric nodes, whereas an extended (D2) resection includes removal of nodes along the left gastric, hepatic, splenic, and celiac arteries. Although the results of major trials attempting to answer this question have yielded confounding results, it is generally agreed on that, at high-volume centers, D2 lymphadenectomies that preserve the distal pancreasand spleen can be performed without increased morbidity, improve staging accuracy, and yield a survival advantage in patients with stage II and III gastric cancer.
Adjuvant therapy for gastric cancer
is important because the majority of patients with locoregional disease (all patients except those with T1-2N0M0 disease) are at high risk for local or systemic recurrence following curative surgery.
Adjuvant combined modality therapy. Although adjuvant chemotherapy or radiation therapy alone has not shown much benefit in studies, a recent landmark trial was able to demonstrate significant improvement in overall and disease-free survival rates in patients with completely resected gastric cancer treated postoperatively with 5-fluorouracil (5-FU)/leucovorin chemotherapy coupled with radiation therapy (N Engl J Med 2001;345:725).
Neoadjuvant chemotherapy for gastric cancer has the potential
for improving patient tolerance, resectability rates (downstaging) and overall patient survival. A recently reported European trial demonstrated significant improvement in 5-year survival rates in patients with gastric cancer who were treated with six cycles of chemotherapy (three preoperatively and three postoperatively) compared to surgery alone (N Engl J Med 2006;355:11). Chemotherapy regimen in this trial consisted of epirubicin, cisplatin, and 5-FU. Furthermore, preoperative chemotherapy improved curative resection rates.
Palliative therapy of gastric cancer
is important due to overall low cure rates. Generally, patients with peritoneal disease, hepatic or nodal metastases, or other poor prognostic factors benefit most from endoscopic palliation. Laparoscopic or open palliative surgical resection can be considered in patients with better prognosis and good performance status to prevent bleeding, obstruction, and perforation in patients with metastatic or otherwise unresectable cancer. Palliative surgical resections appear to provide superior relief of symptoms compared to surgical bypass. Palliative chemoradiation therapy also prolongs survival in patients and improves symptoms and quality of life when it can be administered safely