Gastric cancer presents a daunting health problem globally, accounting for the second leading cause of cancer-related death worldwide. In the United States, more distal gastric cancers are declining in incidence, partly due to eradication of Helicobacter pylori infection. The decrease in gastric cancer has been accompanied, however, by an ongoing increase in cancers of the distal esophagus and gastroesophageal junction.
Aggressive screening strategies in Korea and Japan have led to an earlier stage at diagnosis of gastric cancer, and overall superior outcomes with surgical management. In the West, by contrast, with the absence of effective screening, patients present with more locally advanced disease and have a poorer outcome.
The current AJCC staging of gastric cancer was recently modified to reclassify nodal disease dependent on nodal number, and to define a positive peritoneal cytology (found at laparoscopy) as true stage IV metastatic disease. Gastric cancers are now classified into three categories, which appear to have a distinct histopathology, molecular genetic profile, and etiology: (1) cancers of the gastroesophageal junction, associated with gastroesophageal reflux disease, which by current AJCC staging are now classified as esophageal cancers. More distal gastric cancers fall into two categories: (2) intestinal type pathology, often related to antecedent infection by H. pylori, and (3) diffuse or signet ring histology.
The selection of appropriate therapy for gastric cancer is based on clinical staging, including initial endoscopic biopsy, and CT scan of the abdomen and pelvis to rule out distant metastatic disease. If a CT scan is negative for metastasis, then I proceed with an endoscopic ultrasound for accurate T and to a lesser extent, N staging. A PET scan should be considered. Although up to 30 percent of gastric cancers may not be PET-avid (usually cases with diffuse histology), up to 15 percent of patients may be identified with occult metastatic disease not evident on CT scan imaging.
There are now recognizable familial and genetically predisposed cases of gastric cancer, and all patients should have documentation of a careful family history. Gastric cancer is associated with Lynch syndrome familial colorectal and uterine cancer, and Lynch syndrome patients require periodic upper endoscopic screening in addition to annual colonoscopy.
In families with diffuse gastric cancer occurring at a young age, consideration should be made for familial e-cadherin mutation-related gastric cancer. E-cadherin gene mutation carriers require referral for prophylactic gastrectomy.
Early Stage Disease
Occasionally very early stage gastric cancers are identified in patients undergoing follow-up for gastric ulcer disease, or in the evaluation of anemia, upper GI bleeding, or hemoccult positive stools. T1a tumors, without penetration into the submucosa, have a near zero incidence of nodal spread of disease, and can be treated with endoscopic mucosal resection without surgery. Pathologic T1b-T2 tumors upon EMR have a higher rate of nodal metastasis and should be referred for surgical resection.
Laparoscopic Staging and D2 Gastrectomy
For endoscopic ultrasound T3 or node-positive cancers, laparoscopic staging should be considered prior to definitive surgical resection, or prior to any preoperative or neoadjuvant therapy. PET scan will usually miss small volume peritoneal metastasis, and even with more modern fine-cut CT scan, at laparoscopy up to 20 to 30 percent of patients will be found to have occult peritoneal or abdominal metastases.
As noted above, a positive peritoneal cytology, even in the absence of visible peritoneal metastasis, is now defined as stage IV disease. Such patients should be considered for initial systemic chemotherapy as upfront gastrectomy generally results in poor short-term outcomes. These patients develop recurrent metastatic disease in nearly all cases.
Another area of controversy is the extent of lymph node dissection at the time of surgery. The Dutch trial comparing D1 versus D2 gastrectomy initially failed to show a survival improvement for the more radical D2 resection. However, long-term follow-up on this trial indicated a survival benefit for D2 gastrectomy. Globally D2 resection, modified to avoid splenectomy and pancreatectomy, is now accepted as the standard and optimal cancer operation for gastric cancers.
Neoadjuvant Chemotherapy or Postoperative Chemotherapy or Chemoradiotherapy?
There is ongoing controversy about the optimal adjuvant approach for a T3 or node gastric cancer. The U.S. Intergroup 116 trial compared surgery alone with surgery followed by adjuvant chemotherapy with 5-FU, leucovorin, and radiotherapy. Postoperative chemoradiotherapy improved overall survival by nine percent, and a survival benefit has held up in long-term follow-up.
This therapy became an adopted standard in the U.S. However, the main benefit from adjuvant treatment was to reduce local recurrence of disease, and the quality of surgery on this trial was poor. Only a small minority of patients underwent D2 gastrectomy, and the majority of patients actually had less than a D1 resection. These results were further questioned by trials indicating a similar or superior survival benefit for patients treated with perioperative chemotherapy, treatment before and after surgery with ECF (epirubicin, cisplatin and 5FU) chemotherapy as reported in the MAGIC trial. Survival was improved by 13 percent at five years with pre- and postoperative ECF, without the use of radiotherapy.
Further questioning of the contribution of radiotherapy to postoperative chemotherapy came from Japanese and Korean trials, indicating a survival benefit of up to 10 to 15 percent for a year of adjuvant S-1, and more recently, for six months of adjuvant capecitabine and oxaliplatin, also without radiotherapy.
Acceptable adjuvant therapy options in gastric cancer now include:
1. Postoperative 5-FU and radiotherapy in patients undergoing less than a D1-D2 resection;
2. Pre- and postoperative chemotherapy with ECF; and
3. After D2 resection, adjuvant chemotherapy alone with capecitabine or 5-FU combined with oxaliplatin.
The role of postoperative radiotherapy is now being studied in the CRITICS trial in Europe, in which patients with GE junction or gastric cancer receive preoperative chemotherapy followed by surgery; postoperatively, patients are randomized to complete chemotherapy or to receive postoperative chemotherapy and radiotherapy.
The potential benefit of postoperative radiation therapy, even after D2 gastrectomy, was recently raised by publication of the Korean ARTIST trial. On this trial, patients after D2 gastrectomy received postoperative capecitabine and cisplatin for six months, with or without postoperative radiotherapy. In a preplanned subset analysis, node-positive patients achieved a significant improvement in disease-free survival at three years (5%), representing a 31 percent reduction in the risk of recurrence. The data are provocative and suggest a small benefit for the addition of radiation to adjuvant chemotherapy even after a D2 resection.
Clinical Trials Biomarkers, and Beyond
There are no currently accepted biomarkers recommended for tumor screening in locally advanced disease. HER2 overexpression is present in up to 10 percent of intestinal gastric cancers and six percent of diffuse cancers. Unlike breast cancer, however, HER2 overexpression does not appear to be an independent prognostic factor.
Gastric cancer is rare and the priority is enrollment of patients on appropriate clinical trials when available. The recent CALGB 80101 trial recently studied postoperative therapy comparing 5-FU and leucovorin with ECF chemotherapy, with all patients receiving postoperative radiotherapy with continuous infusion 5-FU. ECF was no better than 5-FU and leucovorin as the adjuvant chemotherapy component.
In development are preoperative chemotherapy trials using early PET scan imaging for response assessment to potentially guide the usage of chemotherapy. In the U.K., where pre- and postoperative chemotherapy are the standard for gastric cancer, the ongoing MAGIC B trial is comparing chemotherapy with or without the addition of bevacizumab as adjuvant therapy.