For patients with disease that does not extend beyond the muscle layer of the esophageal wall and are without evidence of lymph node involvement, immediate surgery is recommended. For those with disease that extends beyond the muscle layer or with locoregional lymph nodes, neoadjuvant therapy with chemotherapy and radiation should be considered prior to surgery. Chemoradiation therapy is given over 5 to 6 weeks with daily radiation and various acceptable combinations of chemotherapy agents. The data for neoadjuvant therapy are mixed, with multiple trials demonstrating a survival benefit but a few trials not. Multiple meta-analyses also suggest a benefit to chemoradiotherapy prior to surgery. Most thoracic surgeons, medical oncologists, and radiation oncologists favor such an approach for localized advanced disease prior to surgery. Following the completion of neoadjuvant therapy, restaging is recommended followed by surgery approximately 6 weeks after the last dose of radiation. In contrast, studies for adjuvant chemotherapy or radiation after surgery have not shown a benefit (except for patients with tumors at the gastroesophageal junction, who can be treated similarly to those with gastric cancer).
For patients with localized disease who are not surgical candidates because of concurrent medical conditions, or who refuse surgery, a randomized trial of chemoradiation versus radiation alone demonstrated a survival advantage to combined modality therapy. Further, there does seem to be a definitive long-term recurrence-free rate (up to 25%) in patients treated with chemoradiation alone.
Patients with metastatic disease should be considered for palliative therapy. Chemotherapy can palliate symptoms related to swallowing and can prolong overall survival. There is no single optimal first-line chemotherapy regimen, although in general patients are offered combination regimens that include a platinum agent. Other options for relief of dysphagia in patients with metastatic disease include radiation (either alone or with chemotherapy) or endoscopically placed esophageal stenting.
The prognosis of patients with esophageal cancer is stage-dependent. Patients diagnosed with disease limited to the submucosal layer of the esophageal wall and no lymph node involvement experience 50% to 80% 5-year survival. If the disease is more extensive within the wall but not involving lymph nodes, 5-year survival is 30% to 40%. Patients with nonmetastatic disease but positive locoregional lymph nodes have a 10% to 30% 5-year survival. Patients with involvement beyond locoregional lymph nodes or distant metastases have a 5-year survival of <5%. The median survival with metastatic esophageal cancer treated with palliative chemotherapy is 8–10 months.
GASTRIC CANCER
An estimated 21,600 new cases of gastric cancer and 11,000 related deaths occur in the United States annually. The incidence of gastric cancer in the United States has markedly decreased over decades, likely related to near elimination of certain risk factors for the disease. In 1930 there were 33 new cases of gastric cancer per 100,000 men and 30 new cases per 100,000 women; in 1990 incidence rates dropped to 10 and 5 per 100,000 men and women, respectively. However, gastric cancer remains the fourth most commonly diagnosed cancer (estimated 934,000 new cases annually) and the second leading cause of cancer-related death (700,000 deaths annually).
HISTOLOGY TYPES OF GASTRIC CANCER
The vast majority of tumors in the stomach are adenocarcinomas. Other, markedly less frequent, histologies are lymphomas, carcinoids, leiomyosarcomas, and gastrointestinal stromal tumors (GISTs). There are two subtypes of gastric adenocarcinomas: an intestinal type with cohesive neoplastic cells forming gland-like tubular structures, and a diffuse type in which individual cells infiltrate and thicken the stomach wall. Intestinal-type lesions occur in the distal stomach more often than the diffuse type and are often preceded by a prolonged precancerous phase. Diffuse carcinomas are detected more often in young patients, develop throughout the stomach (particularly the cardia) and are associated with a worse prognosis.
RISK FACTORS FOR GASTRIC CANCER
Multiple factors have been associated with the risk of developing gastric cancer (box 12.1). Chronic atrophic gastritis and its associated condition, intestinal metaplasia, can result in reduced gastric acid production and progression to metaplasia, dysplasia, and eventually adenocarcinoma. Infection with Helicobacter pylori has been associated with gastric cancer. Prospective studies have demonstrated between threefold and sixfold increased risk of gastric cancer in patients serologically positive for H. pylori. Nonetheless, the majority of patients who are H. pylori positive will not develop gastric cancer.
Box 12.1 RISK FACTORS FOR GASTRIC CANCER
Nutritional
Low fat or protein consumption
Salted meat or fish
High nitrate consumption
Environmental
Poor food preparation (smoked)
Lack of refrigeration
Poor drinking water (well water)
Occupation (Rubber, Coal Workers)
Smoking
Low social class
Medical
Gastric atrophy and gastritis
Helicobacter pylori infection
Pernicious anemia
Prior gastric surgery
Hereditary
E-cadherin mutation families
The marked decline in gastric cancer in the United States is presumed to be related to dietary and environmental exposure. The introduction of refrigeration has led to reduced use of salting, smoking, and pickling of food and to improved food preservation—factors that have been associated with atrophic gastritis. Before refrigeration, nitrates and nitrites were used to preserve meat, fish, and vegetables. Foods rich in nitrates, nitrites, and secondary amines can form N-nitroso compounds that have been associated with gastric tumors in animal models. Further, anaerobic bacteria, which colonize in areas with atrophic gastritis or intestinal metaplasia, can convert nitrates and nitrites to carcinogenic nitroso compounds.
CLINICAL PRESENTATION AND MANAGEMENT OF GASTRIC CANCER
Early detection of gastric cancer is difficult because most early-stage lesions do not cause symptoms. In the United States most patients are diagnosed with locally advanced or metastatic disease. In contrast, given the higher incidence rates of gastric cancer in Asia, screening endoscopy programs lead to more frequent detection of localized disease. The most common symptoms leading to medical attention are unexplained weight loss, abdominal pain, fatigue, nausea, anorexia, dysphagia, early satiety, and melena. Initial evaluation of symptoms suspicious for gastric cancer includes barium swallow and/or upper endoscopy. Once a diagnostic biopsy demonstrates adenocarcinoma, staging with CT is recommended. However, imaging is limited in detecting peritoneal metastases, which can be present in up to 10–30% of patients who appear to have localized disease. At the time of surgery an initial exploratory laparotomy is necessary, and detection of peritoneal disease or distant metastases should lead to either a palliative resection or bypass gastrojejunostomy.
The pathological stage is the most important determinant of prognosis and determines treatment strategy. Patients with metastatic disease should be considered for palliative chemotherapy. Multiple randomized trials have shown that chemotherapy prolongs survival in patients with metastases and maintains or improves quality of life compared to best supportive care only. No single regimen is considered standard. For patients with a good performance status, combination regimens that include a platinum agent are reasonable first-line choices.
For the patient with nonmetastatic disease, surgery can be curative. Cancers in the proximal and distal stomach are surgically approached differently, but both have the principle of wide margins and adequate lymph node dissection. The extensiveness of lymph node dissection remains controversial. In general, patients in Asia undergo considerably more extensive removal of lymph nodes than those in the United States and Europe. Randomized trials in Western populations have not demonstrated a survival benefit to removing lymph nodes beyond 3 cm from the tumor. The more problematic controversy is that many surgeries in the United States have an inadequate nodal resection, which likely affects outcomes.
Following surgical resection, nonmetastatic patients whose disease extended beyond the muscle layer of the gastric wall or with positive lymph nodes should be considered for adjuvant chemoradiotherapy. A large, randomized North American trial demonstrates a survival advantage to a program of chemotherapy and combined chemotherapy and radiation lasting approximately 5 months after surgery. Most trials for adjuvant chemotherapy alone have not demonstrated a statistically significant advantage, although meta-analyses of these trials suggest some modest benefit. Although combined modality adjuvant therapy is considered preferable after surgery, for patients who have a contraindication to radiation (most commonly due to radiation for a different cancer that included some of the stomach field) or who refuse radiation, adjuvant chemotherapy alone may be considered an option.
An alternative approach for nonmetastatic gastric cancer has been validated in Europe. Patients deemed surgically resectable are treated with 3 months of combination chemotherapy followed by surgery followed by further chemotherapy. This schema showed a statistically significant survival advantage over surgery alone. Survival rates from these two approaches (surgery followed by adjuvant chemoradiotherapy and neoadjuvant/adjuvant chemotherapy) are not comparable given the difference in timing of when the survival clock starts, as well as the bias inherent in selecting out patients with peritoneal or other undetected metastases who underwent upfront surgery.
Survival is dependent on stage. Five-year survival is 65–80% for patients with disease with either T1 N0-1 disease (limited to submucosa and either no positive lymph nodes or fewer than seven positive nodes) or T2 N0 disease (disease into the muscularis propria but node-negative). Patients with more advanced but nonmetastatic disease have considerably worse outcome, with 5-year survival ranging from 10% to 40%. Metastatic disease is not considered curable, and median survival with chemotherapy is 8–10 months.
PANCREAS CANCER
Pancreas cancer is one of the most fatal cancers because it is rarely detected at an early stage. Although it is the 10th most common cancer in incidence in the United States, it is the fourth most common cause of cancer-related deaths (behind lung, colorectal, and breast cancer). An estimated 45,000 new cases and 38,000 deaths occur annually in the United States. Despite much research in pancreatic cancer, outcomes have not dramatically changed in the last several decades.
RISK FACTORS FOR PANCREATIC CANCER
Pancreatic adenocarcinoma has been associated with various hereditary syndromes (box 12.2). Hereditary nonpolyposis colorectal cancer (HNPCC) results from mutations of mismatch repair genes and is most commonly associated with colorectal and gynecological cancers, although there is an increased risk of pancreatic cancer. Inherited mutations of p16 result in familial atypical multiple-mole melanoma syndrome associated with melanomas and pancreatic cancer. Other syndromes in which the risk of pancreatic cancer is increased are BRCA2, ataxia-telangiectasia, Peutz-Jeghers syndrome, and hereditary pancreatitis.
Box 12.2 RISK FACTORS FOR PANCREATIC CANCER
Convincing Evidence for Risk
Hereditary syndromes (hereditary nonpolyposis colorectal cancer, Peutz-Jeghers, hereditary BRCA2 mutations, p16 syndrome, ataxia-telangiectasia, hereditary pancreatitis)
Tobacco
Diabetes mellitus
Likely Risk
Chronic pancreatitis
Cystic fibrosis
Pernicious anemia
Obesity
Diabetics appear to have an increased risk of pancreatic cancer. Initially, the association was primarily reported for recently diagnosed diabetics, which is likely more a reflection of a symptom of pancreatic cancer rather than a cause. However, more recent observational studies have shown that long-time diabetics have a modestly increased risk of pancreatic cancer compared to nondiabetics. Tobacco is the one modifiable risk factor most consistently associated with development of pancreatic cancer; however, obesity and certain dietary factors may increase the risk as well.
CLINICAL PRESENTATION AND MANAGEMENT OF PANCREATIC CANCER
Nearly three-quarters of pancreatic cancers derive from the exocrine pancreas ductal system and are adenocarcinomas. The other histologies are neuroendocrine tumors that arise in the islets of Langerhans, lymphomas, or metastatic disease. Adenocarcinomas most commonly arise in the head of the pancreas (65%) and less commonly are restricted to the body or tail (15%) or appear diffusely throughout the pancreas (20%). Whereas treatment strategies are similar regardless of origin (although surgical approach is different), the likelihood of curing patients with body or tail lesions is nearly 0%.
Although painless jaundice is a classical presentation of pancreatic cancer, patients more commonly present with unexpected weight loss, back pain wrapping to the right upper quadrant, anorexia, and nausea. Laboratory tests may show elevated levels of total bilirubin and other liver-function tests (alkaline phosphatase more than transaminases). Patients with suspicious symptoms are usually evaluated by CT when a pancreatic mass is detected. For patients with metastatic disease at presentation, the liver is the most common site of metastases, although distant lymph nodes, peritoneum, and lungs are frequent areas of spread. Patients who present with jaundice should have an endoscopic retrograde cholangiopancreatography (ERCP) with stent placement and cytology by brushings and/or biopsy. Alternatively, percutaneous biopsy of the primary pancreatic mass or metastases can be done with the assistance of ultrasound or CT.
Although a TNM (tumor, node, metastases) system that is utilized for solid organ tumors exists for pancreatic cancer, the more practical classification of pancreatic cancers divides them into three stages—local disease, locally advanced, and metastatic (Figure 12.2). Local disease implies surgically resectable and is the only potentially curable stage of pancreatic cancer. Unfortunately, only 15% of patients have local disease at diagnosis. For surgery to be considered, preservation of fat planes around the major blood vessels in the area is required including the superior mesenteric artery, celiac axis vessels, superior mesenteric vein, and portal vein. Typically either pancreatic protocol CT or magnetic resonance imaging (MRI) can determine the status of these vessels. However, ultimately the determination of resectability is based on a surgeon’s judgment at the time of laparotomy. For lesions at the head of the pancreas a pancreaticoduodenectomy (Whipple) operation is performed, with resection of part of the pancreas and duodenum, common bile duct, gallbladder, and distal stomach (Figure 12.3). For lesions of the body or tail of the pancreas, a distal pancreatectomy with or without splenectomy is performed. Resection of body or tail lesions is considerably less common since most such cancers are metastatic at time of diagnosis. Following resection, adjuvant therapy is considered with either chemotherapy alone or combination of chemotherapy and radiation. Only 15–20% of patients who undergo surgical resection will not have recurrences; most recurrences will be detected within the first 2 years after surgery.