Esophageal Cancer

Esophageal cancer is the ninth most commonly occurring cancer worldwide and the sixth most common cause of cancer mortality (1). It is highly curable in its earliest stages; however, it usually presents with advanced disease. Despite the last two decades of progress in clinical research, the median survival time for a patient with symptoms of a primary esophageal cancer is less than 18 months. Most research and controversy in the treatment of esophageal cancer currently focuses on the role of chemoradiotherapy as either primary or adjuvant therapy for esophageal cancer.

EPIDEMIOLOGY
United States

• Esophageal cancer was estimated to account for 1% of all malignancies and 6% of all gastrointestinal malignancies in 2008 (2). The age-adjusted incidence from 2000 to 2004 is 4.6 cases per 100,000 population (http://seer.cancer.gov/csr/1975_2004).
  
• Approximately 16,470 new cases and 14,280 deaths were estimated for 2008.
  
• The median age at diagnosis is 69 years. This cancer rarely occurs in patients younger than 25 years.
  
• Esophageal cancer is two to four times more frequent in men than in women. Siewert type 1 tumors (adenocarcinoma [ADC]) are eight to nine times more common in men than in women.
  
• Rates of occurrence of esophageal cancer are approximately threefold higher among blacks than among whites.
  
• Squamous cell carcinoma (SCC) is more common in black men; ADC is more common in white men.
  
• Five-year relative survival rates were 5% from 1975 to 1977, 10% from 1984 to 1986, and 16% from 1996 to 2003 (2).
  

Rest of the World

• There are approximately 500,000 cases of esophageal cancer in the world, but there is marked geographic variation. Regions with clusters of high rates include China (e.g., Linxian), Iran, France, and South Africa.
  
• In the 1970s, approximately 90% of esophageal cancers were SCCs. The incidence of ADCs has increased dramatically and currently accounts for approximately 60% to 70% of new cases—a rate of acceleration greater than that of any other cancer in the United States.
  

ETIOLOGY
Adenocarcinoma

• Barrett’s esophagus
  
• Obesity
  
• Gastroesophageal reflux disease (GERD), which can be caused by obesity and might result in Barrett’s esophagus
  

Squamous Cell Carcinoma

• Tobacco
  
• Alcohol
  
• Predisposing conditions:
  
  • Tylosis (SCC)
    
  • Achalasia
    
  • Esophageal diverticula and webs (SCC)
    
  • Plummer-Vinson syndrome
    
  • Human papillomavirus (HPV)
    
  • Celiac disease
    
• Less significant causes include environmental exposure and dietary factors.
  

BARRETT’S ESOPHAGUS

Barrett’s esophagus, perhaps as a result of GERD, is the most important risk factor (100 times risk increase over other factors) for ADC.
Screening recommendations (no randomized trial data for surveillance practices) are as follows:

• For no dysplasia, endoscopy every 2 to 3 years
  
• For low-grade dysplasia, endoscopy every 6 months for 12 months and then yearly
  
• For high-grade dysplasia, esophagectomy or three monthly endoscopies or photodynamic therapy (PDT)
  

CLINICAL PRESENTATION The most common clinical presentations of esophageal cancer are listed in Table 4.1 and are usually related to local compression or infiltration symptoms or generalized malaise and anorexia. The classic triad for presentation of esophageal cancer is as follows:

• Asthenia
  
• Anorexia
  
• Analgesia (for dysphagia)
  

DIAGNOSIS

• Symptoms
  
  • Dysphagia or odynophagia
    
  • Hematemesis
    
  • Dyspepsia
    
  • Hoarseness
    
  • Dyspnea
    
  • Anorexia
    
• Signs (usually late presentation)
  
  • Horner syndrome
    
  • Left supraclavicular lymphadenopathy (Virchow’s node)
    
  • Cachexia
    
  • Hepatomegaly
    
  • Bone metastases (rare, but paraneoplastic hypercalcemia can occur)
    
• Upper gastrointestinal endoscopy
  
  • This diagnostic procedure is the gold standard. The combination of endoscopic biopsies and brush cytology has an accuracy of greater than 90% in making a tissue diagnosis of esophageal cancer.
    
• Barium contrast radiography
  
  • This diagnostic procedure can document contour and motility abnormalities and unexpected airway fistula and may be useful when the entire esophagus has not been visualized endoscopically. However, a tissue diagnosis is needed for definitive diagnosis.
    

TABLE 4.1. Clinical presentation of esophageal cancer
Symptoms Patients with symptoms (%) Dysphagia (solids usually before liquids) 80-96 Weight loss 42-46 Odynophagia ≤50 Epigastric or retrosternal pain ≤20 Cough or hoarseness ≤5 Tracheoesophageal fistula 1-13

PATHOLOGY

• Most newly diagnosed patients have ADC, but there are contrasting reports on their relative prognosis. Less than 1% of esophageal tumors are lymphoma, melanoma, carcinosarcoma, or small cell carcinoma.
  
• Fifty percent of tumors arise in the lower one-third of the esophagus, 25% arise in the upper esophagus, and 25% of tumors occur in the middle one-third of the esophagus (2).
  

STAGING
The American Joint Commission for Cancer (AJCC) has designated staging of cancer by TNM classification, which defines the anatomic extent of disease (3) (Table 4.2). Of note, cervical adenopathy in tumors in the lower one-third of esophagus is M1 as opposed to N1. The Siewert classification subclassifies gastroesophageal junction tumors into three types according to their anatomic location: type I are distal esophagus tumors, type II are cardia tumors, and type III are subcardia gastric tumors (4). Staging workup can include the following:

• Computerized tomography (CT) scan: CT scan of the chest and abdomen can demonstrate evidence of spread of tumor to lymph nodes or distant metastases to the liver (35%), lungs (20%), bone (9%), and adrenals (5%). CT scan may underestimate the depth of tumor invasion and peri-esophageal lymph node involvement in up to 50% of cases. Magnetic resonance imaging (MRI) provides similar results to CT.
  
• Endoscopic ultrasound (EUS): EUS may be helpful when metastases are not detected by CT or other imaging modalities. EUS is the optimal technique for locoregional staging. A meta-analysis demonstrated greater than 71% sensitivity in staging preoperative depth of invasion (T) and greater than 60% sensitivity for locoregional lymph nodes (N); specificity was greater than 67% and greater than 40%, respectively (5).
  
• Positron emission tomography (PET): PET is useful when CT is negative for metastatic disease, and the diagnosis can change management of cancer in 25% to 40% of patients (6,7). Bronchoscopy is required in tumors less than 25 mm from the incisors to exclude invasion of the posterior membranous trachea or tracheoesophageal fistula.
  

TABLE 4.2. Definition of TNM and stage grouping

TNM stage Primary tumor (T) Distant metastasis (M) Primary tumor cannot be assessed MX: Distant metastasis cannot be assessed T0: No evidence of primary tumor M0: No distant metastasis Tis: Carcinoma in situ M1: Distant metastasis T1: Tumor invades lamina propria or submucosa Tumors of the lower thoracic esophagus T2: Tumor invades muscularis propria M1a: Metastasis in celiac lymph nodes T3: Tumor invades adventitia M1b: Other distant metastasis T4: Tumor invades adjacent structures Tumors of the midthoracic esophagus M1a: Not applicable M1b: Other distant metastasis Regional lymph nodes (N) Tumors of the upper thoracic esophagus N1: Regional lymph node metastasis M1a: Metastasis in cervical nodes NX: Regional lymph nodes cannot be assessed M1b: Other distant metastasis N0: No regional lymph node metastasis Stage grouping Stage 0 Tis N0 M0 Stage I T1 N0 M0 Stage IIA T2 N0 M0 T3 N0 M0 Stage IIB T1 N1 M0 T2 N1 M0 Stage III T3 N1 M0 T4 Any N M0 Stage IV Any T Any N M1 Stage IVA Any T Any N M1a Stage IVB Any T Any N M1b

TREATMENT
Surgery

• Surgery alone remains a standard treatment for esophageal cancer with resectable local or locoregional disease. In 1993, surgery was used as a component of treatment in 34% of patients. Surgery alone was used in 18% of patients (8).
  
• Recent improvements in staging techniques and patient selection have improved surgical morbidity and mortality. Operative mortality rates are now less than 5%. Surgical expertise is a major contributor to survival, with better outcomes in high-volume centers. Resection is possible in approximately 50% of patients (9). Five-year survival in patients with surgical resection is 5% to 25%.
  
• Surgical principles include a wide resection of the primary tumor with the goal of an R0 resection (no residual tumor), including more than 5-cm resection margins plus regional lymphadenectomy. Intraoperative frozen section can assess for residual disease, which, if present, is considered an R1 (microscopic tumor) or R2 (macroscopic tumor) resection.
  
• In general, patients with cervical carcinoma of the esophagus are not considered candidates for surgical resection; chemoradiation is favored in these patients.
  
• Surgical approaches include the following:
  
  • Transthoracic resection: En bloc esophagectomy requires laparotomy and thoracotomy, for example, total thoracic or transthoracic (Lewis) procedures. A three-field lymph node dissection (extended lymphadenectomy) includes superior mediastinum and cervical lymphadenectomy. It is the treatment of choice in Japan, but is associated with increased toxicity and has a questionable survival advantage.
    
  • Transhiatal esophagectomy: This includes laparotomy and cervical anastomosis. This technique avoids thoracotomy.
    

Chemoradiotherapy (Combined-Modality Approach)
Although no large prospective randomized trials have directly compared primary chemoradiation with surgery, definitive chemoradiation for locoregional carcinoma of the esophagus is considered an alternative to surgery.
Inoperable Disease

• The Radiation Therapy Oncology Group (RTOG) 85-01 trial (10) demonstrated a survival advantage (14 vs. 9 months median survival and 27% vs. 0% 5-year survival) in favor of chemoradiotherapy over radiotherapy alone (Table 4.3). A number of randomized trials of chemoradiotherapy versus radiotherapy alone have failed to duplicate the results of RTOG 85-01; however, a recent Cochrane review has confirmed the superiority of chemoradiotherapy versus radiotherapy in fit, motivated patients (11). For operable disease, chemoradiotherapy is an alternative to surgery for patients with operable esophageal cancer. Upper thoracic esophageal (above the aortic arch) tumors and T4 or N1 tumors are usually considered unresectable.
  

Operable Disease
Trials that used surgery after chemoradiotherapy in patients who responded and higher doses of radiation for patients who did not respond failed to show a benefit for the group receiving surgery. A better determinant of the role of surgery after chemoradiotherapy is evident from two recent trials:

• In the first trial, patients with locally advanced but resectable tumors were treated with chemoradiotherapy, and patients with at least a partial response were randomized to continued chemoradiotherapy or surgery (12). There was no difference in overall survival, but early mortality and length of hospital stay were less in the chemoradiotherapy arm (12.8% vs. 3.5% [P = 0.03])
  
• The second trial randomized patients with locally advanced tumors to either definitive chemoradiotherapy or chemoradiotherapy (lower doses of radiation) and surgery (13). There was no significant difference in survival outcomes (17.7 vs. 19.3 months, respectively) between the two groups of patients.
  

Radiation dose escalation has not proved to be beneficial. A recent trial examining this approach was closed after an interim analysis indicated that there would be no advantage with higher doses of radiation.

TABLE 4.3. Radiation Therapy Oncology Group (RTOG) 85-01 trial of chemoradiotherapy versus radiotherapy alone in esophageal cancer

Number of patients Histology Treatment Median survival 5-Year survival P-value 121 SCC = 88% Cisplatin 75 mg/m2 d 1 14.1 mo vs. 9.3 mo 27% vs. 0% 0.0001 ADC = 12% 5-FU 1,000 mg/m2/d d 1-4 50 Gy of radiation wk 1, 5 Cisplatin/5-FU wk 8, 11 vs. 64 Gy of radiation 2 Gy/fx SCC, squamous cell carcinoma; ADC, adenocarcinoma; 5-FU, 5-fluorouracil; Gy, Gray of radiation; fx, fraction of radiation. Source: From Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally advanced esophageal cancer long-term follow-up of a prospective randomized trial (RTOG 85-01). JAMA 1999;281:1623-1627, with permission.

Neoadjuvant Chemoradiotherapy (Trimodality Approach)

• The rationale for preoperative chemoradiotherapy was first studied by Leichman et al. in 21 patients with SCC. Patients were treated with 3,000 cGy of radiation and with two cycles of concurrent 5-fluorouracil (5-FU) and cisplatin (14). An additional 2,000 cGy of radiation was given postoperatively when residual tumor was seen at surgery. The pathologic complete response was 37%, with a median survival of 18 months.
  
• A number of prospective randomized phase 3 trials have addressed the issue of whether preoperative chemoradiotherapy offers any benefit over surgery alone (15,16,17,18,19,20,21,22,23). Much debate exists over interpretation of these trials. Walsh et al. demonstrated a significant benefit in median survival (16 vs. 11 months, P = 0.01) and 3 years survival (32% vs. 6%, P = 0.01) for patients receiving preoperative chemoradiotherapy (15). However, limitations of this trial include poor surgical outcome, small numbers of patients studied, and the fact all patients had ADC. The Cancer and Leukemia Group B (CALGB) 9781 was designed as a prospective randomized Intergroup trial of trimodality therapy versus surgery in 56 patients with stages I-III esophageal cancer. Median survival was 4.48 years versus 1.79 years (P = 0.002) in favor of trimodality therapy (23).
  
• In a meta-analysis of 12 randomized trials, the benefit of chemotherapy and surgery was observed over surgery alone (24). Overall survival in individual patient data from nine trials showed an absolute benefit of 4% (increased from 16% to 20%). The disease-free survival over 5 years showed an absolute benefit of 4% (increased from 6% to 10%). This meta-analysis showed a small but significant benefit for neoadjuvant chemotherapy over surgery alone (P = 0.03).
  
• Neoadjuvant chemoradiation has shown a trend toward superiority versus neoadjuvant chemotherapy alone in patients with locally advanced but resectable tumors. In one randomized trial, patients received chemotherapy followed by surgery or chemotherapy followed by chemoradiation for 3 weeks, followed by surgery. A complete resection was possible in 77% versus 85% in arms A and B, respectively. Complete histologic response was 2.5% with chemotherapy and 17% after chemoradiation (P = 0.06). The median survival was 21.2 versus 32.8, months and 3 years survival was 27% versus 43% respectively (P = 0.14) (25).
  
• A meta-analysis in 2007 clarified the benefits of neoadjuvant chemoradiation or chemotherapy versus surgery alone. The absolute difference in survival between neoadjuvant chemoradiation versus surgery alone was 13%, while absolute difference in survival between neoadjuvant chemotherapy versus surgery alone was 7% (P = 0.05) (26).
  

Adjuvant Chemoradiotherapy There are few data available on the use of postoperative chemoradiotherapy in esophageal cancer.

• A recent trial found a statistically significant survival advantage for postoperative chemoradiotherapy compared to surgery alone in gastroesophageal and gastric cancers (Table 4.4) (27). Based on these data, adjuvant chemoradiotherapy has been recommended for ADC tumors of the lower esophagus.
  It is possible that the survival benefit associated with the use of chemoradiotherapy results from reductions in local recurrences and thus compensates for inadequate surgery (only 10% of patients had the recommended D2 resection).

TABLE 4.4. Intergroup-116 trial of adjuvant chemoradiotherapy versus surgery alone in gastric or gastroesophageal adenocarcinomas

Number of patients Surgery Treatment Median survival 3-Year survival P-value 556 D0 54% Surgery followed by 36 mo vs. 27 mo 50% vs. 41% 0.005 D1 36% 5-FU 425 mg/m2/d d 1-5 D2 10% LV 20 mg/m2/d d 1-5a 4,500 cGy radiation d 28, 180 cGy/d 5 d/wk for 5 wk vs. surgery alone 5-FU, 5-fluorouracil; LV, leucovorin. aTwo further courses of chemotherapy were given 1 month apart after radiation. Additional 5-FU 400 mg/m2 and LV 20 mg/m2 were given on the first 4 days and the last 3 days of radiotherapy.

Radiation Therapy
Radiation therapy alone is generally considered palliative and is used in patients who are unable to tolerate chemoradiotherapy. No randomized trials of preoperative single-modality radiotherapy have demonstrated a survival benefit in patients. Similarly, there is no benefit with single-modality postoperative radiotherapy in patients. Chemotherapy

• Single-agent chemotherapy demonstrates response rates of 15% to 25%. Combination chemotherapy response rates are 25% to 45%, but this has not definitively improved survival in advanced disease states.
  
• Cisplatin with 5-FU is the most frequently used regimen for both combined-modality therapy in locoregional disease and systemic therapy for palliation.
  
• SCC may be more sensitive to chemotherapy, but there is no difference in long-term outcome between SCC and ADC.
  
• New chemotherapeutic agents have demonstrated encouraging response rates.
  

Preoperative (Neoadjuvant) Chemotherapy

• The poor survival, even for patients with clinically localized carcinoma of the esophagus, suggests that occult metastases are present at diagnosis, thereby providing the impetus to add systemic therapy early during patient management.
  
• In the two largest trials examining preoperative chemotherapy, the Intergroup (INT 0113) trial (28) showed no survival benefit, whereas the Medical Research Council (MRC) trial (29) demonstrated a 3-month median survival advantage for chemotherapy over surgery alone.
  
• The following differences in the two studies may have contributed to their different outcomes:
  
  • Chemotherapy was of longer duration and was with higher doses in INT 0113. This therapy may have been detrimental by delaying access to surgery and causing more toxicity.
    
  • Surgery was performed in only 80% of the patients in the chemotherapy arm in INT 0113 compared to 92% in the MRC trial. Outcome for surgery alone was poor in the MRC trial, thereby possibly exaggerating the benefits of chemotherapy.
    
  • Radiation therapy off protocol (equally distributed between treatment arms) was available in the MRC trial.
    
  • A larger sample size in the MRC trial may have facilitated detection of a statistically significant result.
    
• One Japanese study suggests neoadjuvant chemotherapy is highly superior to adjuvant chemotherapy (HR = 0.64, P = 0.014) (30).
  
• A recent trial assessed the use of perioperative epirubicin, cisplatin, and 5-FU (ECF) chemotherapy or surgery alone in esophagogastric cancer. In 503 patients, 15% had esophagogastric cancer and 11% had esophageal cancer. Adjuvant chemotherapy increased progression-free survival and resectability rates and showed a trend toward improved survival (P = 0.06) (31).
  

Palliation

• Palliative options can be split into local or systemic options.
  
• Local therapies include external beam and brachytherapy radiation. This approach can palliate dysphagia in approximately 80% of patients. PDT has also been approved by the U. S. Food and Drug Administration (FDA) for this indication. For rapid palliation, laser or balloon dilatation and stenting is recommended. The placement of a gastrostomy or jejunostomy tube may improve the patient’s nutritional status.
  
• The systemic chemotherapy options in esophageal cancer are improving.
  
• Cisplatin combined with 5-FU is the most commonly used regimen, but this did not demonstrate a statistically significant benefit over cisplatin alone when tested in a randomized trial of patients with advanced SCC of the esophagus (32).
  
• Most data on chemotherapy in advanced esophageal cancer are extrapolated from trials in gastric cancer that often include gastroesophageal tumors. In Europe, ECF is frequently used because of its superior survival compared to 5-FU, doxorubicin, and methotrexate (FAMTX) in advanced esophagogastric cancer (33).
  
• A trial in metastatic or unresectable gastric and gastroesophageal cancer evaluated docetaxel, cisplatin, and 5-FU (DCF) and cisplatin and 5-fluorouracil (CF). Time to disease progression improved from 3.7 months to 5.2 months (hazard ratio 1.704), and median overall survival improved from 8.5 months to 10.2 months (P = 0.0053) in patients receiving DCF compared to those receiving CF (34).
  
• Another trial evaluated capecitabine and oxaliplatin as alternatives to cisplatin and flurouracil. In a two by two design, patients were randomly assigned to ECF (epirubicin/cisplatin/5-flurouracil), ECX (epirubicin/cisplatin/capecitabine), EOF (epirubicin/oxaliplatin/5-flurouracil), EOX (epirubicin/oxaliplatin/capecitabine). The median survival time was 9.9, 9.9, 9.3, and 11.2 months respectively. Overall survival was higher with EOX (P = 0.02) (35).
  
• The distribution, treatment, and survival in patients with esophageal cancer according to stage in the United States are given in Table 4.5.
  

TABLE 4.5. Stage, distribution, treatment, and survival of esophageal cancers in the United States

Stage Distribution (%) Treatment 5-Year relative survival rates (%) Localized (I and II) 25 Surgery 34 Chemoradiotherapy if surgery not possible Adjuvant chemoradiotherapy indicated for GE or lower esophageal ADC tumors Regional (III) 28 Surgery if possible (e.g., T3 N0 lesions) 17 Definitive chemoradiotherapy Neoadjuvant chemoradiotherapy followed by surgery Adjuvant chemoradiotherapy indicated for GE or lower esophageal ADC tumors Distant (IV) 25 Best supportive care/palliation 3 Local: Radiation therapy or brachytherapy Intraluminal intubation or dilatation Laser or endocoagulation Photodynamic therapy Combination chemotherapy: CF (cisplatin, 5-fluorouracil) EOX (epirubicin, oxaliplatin, capecitabine) DCF (docetaxel, cisplatin, 5-fluorouracil) ECF (epirubicin, cisplatin, 5-fluorouracil) Folfiri (5-fluorouracil, irinotecan) ADC, adenocarcinoma; GE, gastroesophageal; 5-FU, 5-fluorouracil.

FOLLOW-UP FOR PATIENTS WITH LOCOREGIONAL DISEASE There is no standard surveillance scheme.

• History and physical examination, complete blood count (CBC), urea, electrolytes, and liver function tests are recommended every 4 months for 1 year, every 6 months for 2 years, and then annually (www.nccn.org).
  
• Chest radiograph should be obtained as indicated.
  
• CT scans of the chest/abdomen should be obtained as clinically indicated.
  
• Upper gastrointestinal endoscopy should be performed as clinically indicated.