Physical Address

304 North Cardinal St.
Dorchester Center, MA 02124

Cancer of the Esophagus


Summary of Key Points

  • Classification

  • Adenocarcinomas and squamous cell carcinomas (SCCs) comprise 98% of esophageal tumors.

  • The Siewert classification is now commonly used to categorize gastroesophageal junction (GEJ) tumors, although many studies, especially those for advanced disease, have enrolled patients with esophageal, GEJ, and gastric cancers.

  • Incidence

  • In the United States, approximately 17,290 patients will be diagnosed with esophageal cancer in 2018, with 15,850 deaths.

  • Historically and internationally, SCC is the most common histologic type; however, a dramatic increase in the incidence of adenocarcinoma has been documented in the United States, the United Kingdom, and Western Europe.

  • Pathogenesis

  • Esophageal cancers arise as a result of chronic irritation from a wide variety of sources, including gastric contents in chronic reflux and known carcinogens.

  • A strong association between Barrett esophagus and adenocarcinoma is seen, but a benefit for screening endoscopy for those at risk for or with known Barrett esophagus is unknown because the overall risk of cancer-related mortality is low. Barrett esophagus arises from gastroesophageal reflux disease (GERD) and other risk factors, including obesity and smoking.

  • SCC is associated with smoking and with alcohol use, and the declining incidence has paralleled the decline in smoking.

  • Diagnosis and Staging

  • Symptoms and demographics will strongly suggest the diagnosis.

  • Endoscopy is the best screening examination, with diagnosis made with endoscopy with biopsy and cytologic assessment of tumor.

  • Endoscopic ultrasonography (EUS) should be used to assess T and N stage to guide optimal definitive therapy.

  • Computed tomography (CT) of the chest and abdomen is useful in screening for metastatic disease.

  • Positron emission tomography (PET) scan is useful to detect additional cases of metastatic disease before costly and toxic definitive therapy. It may be superior to endoscopic EUS in detecting intraabdominal lymph nodes, but not periesophageal nodes adjacent to the primary tumor.

  • Additional studies include laparoscopy, thoracoscopy, bone scan, and CT of the brain when indicated by clinical circumstances.

  • Treatment

  • Treatment of premalignant dysplasia is guided by grade of histology. Low-grade dysplasia should be treated with endoscopic ablative techniques or observation. High-grade dysplasia is treated with endoscopic ablation or esophagectomy, although close follow-up may be appropriate for selected patients.

  • Selection of appropriate treatment for carcinoma depends on tumor stage and patient performance status.

  • Chemotherapy is now routinely used in perioperative strategies and is the mainstay of treatment for recurrent or metastatic disease, in which response rates and duration of response are modest for most patients. The addition of targeted therapies incrementally improves outcomes. The evaluation of immune checkpoint inhibitors reveals a modest but notable benefit in the metastatic setting and is of high priority.

  • Surgery is an accepted single-modality therapy for patients with early localized disease (T1–2N0M0) or for patients who may not tolerate combined modality therapy. The selection of surgical approach depends on location and experience, but no approach has been demonstrated to lead to superior cure rates.

  • Combined chemoradiation leads to prolonged median survival and long-term survival compared with radiation alone when used as a definitive nonoperative approach. This represents a potentially curative alternative to surgery for SCCs and is appropriate for unresectable lesions of either histologic type.

  • There are now established data showing that preoperative chemoradiation followed by surgery improves survival versus surgery alone for both locally advanced (clinically staged T2–T4 or node-positive) adenocarcinoma and SCC. There is less certainty about the value of preoperative chemotherapy alone.

  • Randomized trials have not confirmed a survival benefit with surgery added to potentially curative chemoradiation in SCC, but there is a local control benefit. Therefore definitive chemoradiation for patients with SCC is an option, with surgery reserved for locally persistent or recurrent disease. There are limited data regarding this approach for patients with adenocarcinomas, but it can be considered in patients who are borderline candidates for esophagectomy.

  • Postoperative adjuvant chemotherapy or chemoradiation is less well studied in locally advanced esophageal cancer, but trials in gastric cancer including GEJ adenocarcinoma have demonstrated a benefit.

  • Endoscopic palliative therapy includes laser or electrical fulguration, mucosal resection, photodynamic therapy (PDT), or stenting. Excepting very superficial lesions, these therapies are not alternatives to surgery because they do not address deeper disease or lymphatic spread.

  • Radiation and/or chemotherapy may be used to palliate local symptoms.

  • Algorithm

  • T1is or T1a Tumors

    • High-grade dysplasia or T1a tumors are managed with endoscopic ablation or esophagectomy.

  • T1b–2N0 Tumors

    • Surgery is an accepted single-modality therapy for patients with early localized disease (T1–2N0M0) or for patients who may not tolerate combined modality therapy. The selection of surgical approach depends on location and experience, but no approach has been demonstrated to lead to superior cure rates.

  • T2–4N any or T any N+ Tumors

    • Combined chemoradiation leads to prolonged median survival and long-term survival compared with radiation alone when used as a definitive nonoperative approach. This represents a potentially curative alternative to surgery for SCCs and is appropriate for unresectable lesions of either histologic type.

    • There are also established data that preoperative chemoradiation followed by surgery improves survival compared with surgery alone for both locally advanced adenocarcinoma and SCC. There is less certainty about the value of preoperative chemotherapy alone.

    • Randomized trials have not confirmed a survival benefit with surgery added to potentially curative chemoradiation in SCC, but there is a local control benefit. Therefore definitive chemoradiation for patients with SCC is an option, with surgery reserved for locally persistent or recurrent disease. There are limited data regarding this approach for patients with adenocarcinomas, but it can be considered in patients who are borderline candidates for esophagectomy.

    • Postoperative adjuvant chemotherapy or chemoradiation is less well studied in locally advanced esophageal cancer, but trials in gastric cancer including GEJ adenocarcinoma have demonstrated a benefit.

  • Metastatic or Recurrent Disease

  • Chemotherapy is the mainstay of treatment for recurrent or metastatic disease, where response rates and duration of response are modest for most patients. The standard doublet consists of a fluoropyrimidine and a platinum drug. Adding docetaxel to this doublet slightly improves outcomes, but at the expense of significant additional toxicity.

  • The addition of trastuzumab to first-line chemotherapy for HER2-positive disease incrementally improves outcomes. Similarly, ramucirumab with or without paclitaxel in the second-line setting is a validated strategy.

  • The evaluation of immune checkpoint inhibitors against the PD-1/PD-L1 axis reveals a modest but notable benefit in the metastatic setting and is of high priority.

  • Endoscopic palliative therapy includes laser or electrical fulguration, mucosal resection, PDT, or stenting. Excepting very superficial lesions, these therapies are not alternatives to surgery because they do not address deeper disease or lymphatic spread.

  • Radiation and/or chemotherapy may be used to palliate local symptoms.

Esophageal cancer is a devastating disease associated with poor survival outcome and adverse effects on swallowing and quality of life (QoL). Although it is an aggressive malignancy that usually manifests in a locally advanced stage, significant progress has been made in the treatment of this disease, including expanded treatment options, decreased surgical morbidity and mortality, confirmation of the benefit of combined modality therapy, and improvements in identifying patients at risk. These advances are resulting in incremental but still quite meaningful improvements in outcome, but there remains considerable controversy over the optimal management in individual scenarios. The emphasis of this chapter is on selecting the appropriate options in the curative and palliative management of esophageal cancer.

The revised American Joint Committee on Cancer (AJCC) seventh edition staging system refines the categorization of patients. In particular, there is now consideration of location within the esophagus, number of positive lymph nodes, and histologic findings and a changed categorization of extent of primary disease. Adenocarcinoma and squamous cell carcinoma (SCC) histologic types are also considered separately. Survival data based on this staging system are shown in Fig. 71.1 .

Figure 71.1
Survival according to American Joint Committee on Cancer/International Union Against Cancer seventh edition staging system for an international cohort of 4627 patients treated surgically for esophageal neoplasms, including adenocarcinoma patients (A) and squamous cell carcinoma patients (B).

(From Rice TW, Rusch VW, Ishwaran H, et al. Worldwide Esophageal Cancer Collaboration. Cancer of the esophagus and esophagogastric junction: data-driven staging for the seventh edition of the American Joint Committee on Cancer/International Union Against Cancer cancer staging manuals. Cancer. 2010;116[16]:3763–3773.)

Histologic and Molecular Classification and Location

Esophageal cancer is classified based on histologic appearance and cell of origin, as follows: (1) epithelial tumors, (2) metastatic tumors, (3) lymphomas, and (4) sarcomas. Cancers of epithelial cell origin, predominantly SCC and adenocarcinoma, are the most common. Although SCC and adenocarcinoma were previously treated as similar entities and grouped together, there is increasing recognition that they should be studied as separate entities whose optimal therapy may diverge in the era of multimodality treatments.

There is now clear-cut molecular evidence that SCCs and adenocarcinomas are entirely distinct entities. Data from The Cancer Genome Atlas showed different molecular aberrations between them. Specifically, SCCs showed frequent genomic amplifications of CCND1 and SOX2 and/or TP63, whereas ERBB2, VEGFA. and GATA4 and GATA6 were more commonly amplified in adenocarcinomas. Equally notable is the fact that esophageal and gastroesophageal junction (GEJ) adenocarcinomas strongly resembled the chromosomally unstable variant of gastric cancer (which is most commonly found in the cardia ), suggesting that these tumors should be treated as a single entity.

SCC usually occurs in the middle third of the esophagus, whereas adenocarcinomas are most common in the lower third of the esophagus. The Siewert classification is now routinely used to further subtype GEJ tumors based on their location. Siewert type I tumors arise from the distal esophagus and infiltrate the GEJ from above, whereas type III tumors are gastric cardia tumors that infiltrate the GEJ from below; type II tumors are true tumors of the GEJ.

Incidence

Because SCCs and adenocarcinomas account for 98% of all esophageal tumors, this chapter focuses on these two histologic types. Epidemiologic data show that the incidence of esophageal cancer varies considerably from one country to another and often within a single country. This geographic diversity underscores the multifactorial etiologies of esophageal cancer worldwide, where more than 90% of tumors are SCCs. The epidemiologic factors responsible for the geographic variability in incidence of esophageal cancer, including potential dietary and environmental carcinogens, remain under active investigation.

Although cases of SCC have steadily declined in the United States because of a decrease in tobacco and alcohol abuse, the incidence of adenocarcinoma of the distal esophagus, GEJ, and gastric cardia has increased 4% to 10% per year among US men since 1976, so it now comprises 75% of all esophageal tumors. The absolute incidence in the United States has increased from 3.8 per million in 1973 to 1975 to 23.3 per million in 2001 based on the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database. This rate of increase exceeds that of all other cancers, including lung, breast, and prostate cancers and melanoma. Adenocarcinoma is much less common in African Americans but has increased from 0.4 per 100,000 to 0.9 per 100,000, and 0 to 0.2 per 100,000 in females. Even as it appears that the rate of increase may be slowing in the United States, the incidence of esophageal adenocarcinoma globally continues to increase.

Changing epidemiologic factors account for the increasing incidence of adenocarcinomas, which are now more common because of an increased incidence of gastroesophageal reflux disease (GERD) and obesity. Helicobacter pylori, implicated in peptic ulcer disease and associated with an increased risk of gastric cancer, has not been implicated in the pathogenesis of esophageal adenocarcinoma. In fact, because infection with H . pylori may lead to a reduction in gastric acidity in association with atrophic gastritis, there has been speculation that a decline in the prevalence of H . pylori infection may predispose to an increase in GERD and therefore in the incidence of GEJ adenocarcinomas. Similarly, infection with human papillomavirus, while implicated in oropharyngeal SCC, has not been observed in esophageal SCC.

Pathogenesis

Data support the hypothesis that esophageal cancer arises as a result of chronic irritation and inflammation of the esophagus, leading to a sequence of genetic alterations in the damaged epithelium and histologic changes of dysplasia to carcinoma. Factors that are associated with an increased risk of SCC include caustic lye ingestion and radiation therapy, and the interval between injury and the development of cancer may be considerable. In addition, a history of underlying esophageal disease such as achalasia, Plummer-Vinson syndrome, and tylosis is also linked to SCC. In adenocarcinoma, gastroesophageal reflux of acid and bile are major risk factors that cause chronic inflammation, which then leads to carcinogenesis.

There are important differences between the risk factors and histologic progression that support the concept that SCC and adenocarcinoma are separate entities with the potential to respond differently to newer treatment approaches.

Clinical Risk Factors

Two major risk factors for esophageal SCC are alcohol use and tobacco smoking, according to epidemiologic studies from various countries worldwide. This relationship is dose dependent, and there is a multiplicative interaction with alcohol intake and tobacco use. In a prospective cohort study from the Netherlands involving more than 120,000 people with 16 years of follow-up, the strongest risk factor for SCC was alcohol consumption with a 4.6-fold increased risk, whereas combined exposure with smoking increased the risk more than 8-fold. The risk of SCC decreased with smoking cessation and alcohol abstinence, but only after one to two decades.

In adenocarcinoma, smoking but not alcohol is associated with increased risk. A prospective study of tobacco, alcohol, and the risk of esophageal cancer in the United States found an increased risk of SCC among current smokers compared with nonsmokers (hazard ratio [HR] 9.27; 95% confidence interval [CI], 4.04–21.29) and also an increased risk for adenocarcinoma (HR, 3.70; 95% CI, 2.20–6.22). For people who consumed more than three alcoholic drinks a day compared with one drink, there was increased risk of esophageal SCC but not adenocarcinoma.

Nutritional and dietary factors have been found to contribute to the risk of both SCC and adenocarcinoma. Pickled vegetables, processed meat, and other foods containing nitrates have been linked to an increased risk of SCC. Drinking very hot liquids or eating hot foods such as stewed meat frequently is postulated to cause thermal injury to the esophagus and has been found to be a risk factor for SCC. Low consumption of fruits and vegetables is a risk factor for both SCC and adenocarcinoma.

Obesity, defined as a body mass index of 30 or higher, is a strong risk factor for esophageal adenocarcinoma, as demonstrated in a meta-analysis of epidemiologic studies (HR, 2.78; 95% CI, 1.85–4.16). This has clear implications regarding the increasing obesity rates in the United States and could partly explain the rising incidence of esophageal adenocarcinoma. The exact mechanism for this association of obesity is not understood but might reflect an increased propensity for gastroesophageal reflux. Obesity does not appear to be a risk factor for SCC.

Adenocarcinoma: Role of Gastroesophageal Reflux Disease and Barrett Esophagus

There is a clear relationship between GERD and the development of Barrett esophagus. Esophageal adenocarcinoma frequently arises in Barrett esophagus, a condition in which the normal stratified squamous mucosa of the esophagus is replaced by a metaplastic columnar-lined epithelium that extends upward from the GEJ, generally as a result of injury from chronic reflux. Barrett esophagus may progress to dysplasia and then malignancy as the dysplastic epithelial cells accumulate genetic alterations.

GERD is a well-established risk factor for esophageal adenocarcinoma. A population-based case-control study in Sweden demonstrated an odds ratio (OR) of 7.7 (95% CI, 5.3–11.4) for the development of esophageal cancer in patients with chronic GERD. With long-standing, severe symptoms, the OR for esophageal adenocarcinoma was 43.5 (95% CI, 18.3–103.5). There was no association seen between reflux and SCC.

The prevalence of Barrett esophagus is estimated to be 1% to 2% of the general population. The length of time for progression from Barrett esophagus to dysplasia to adenocarcinoma is unknown, although several reports of endoscopic surveillance programs have suggested that progression of Barrett esophagus to adenocarcinoma is uncommon. A nationwide population-based cohort study in Denmark followed more than 11,000 patients with Barrett esophagus for a median of 5.2 years and found that the annual risk of adenocarcinoma was 0.12% (95% CI, 0.09–0.15). Meanwhile, a meta-analysis of 57 studies involving more than 11,000 patients with nondysplastic Barrett esophagus found a pooled annual incidence rate of adenocarcinoma of 0.33%. Patients with low-grade dysplasia had an incidence rate of adenocarcinoma of 0.5% per year, whereas those with high-grade dysplasia has an incidence rate of adenocarcinoma of 3% to 5% per year.

Diagnostic and Staging Evaluation

The most common presentation of esophageal carcinoma is solid food dysphagia and weight loss of several months’ duration. Other presentations that occur with esophageal adenocarcinoma in particular include chest pain in the absence of myocardial ischemia and anemia from chronic gastrointestinal bleeding from the mucosal lesion. These clinical signs and symptoms should prompt endoscopic evaluation and diagnostic imaging. The diagnosis is usually evident by the characteristic narrowing of the esophagus on barium studies, but endoscopy and biopsy are essential for histopathologic diagnosis. Endoscopic biopsies and brushings of the lesion will yield the diagnosis in more than 90% of patients. Multiple biopsies may be necessary for diagnosis of an invasive malignancy that is submucosal or necrotic. A diagnosis of in situ carcinoma or dysplasia in the face of a large lesion seen on endoscopic or radiographic studies should not be accepted, and biopsy should be repeated to confirm the extent of invasion and guide optimal management.

Once the pathologic diagnosis has been established, evaluation to determine the extent of disease should include a computed tomography (CT) scan of the chest and complete abdomen. The accuracy of abdominal CT for identification of metastases to the liver and celiac axis depends on the bulk of the disease. Small liver metastases, peritoneal studding, and abdominal nodes will often be undetectable. For the patient with an adenocarcinoma of the distal esophagus, GEJ, or cardia, a complete abdominopelvic CT scan is recommended because these tumors are more likely to metastasize early to periaortic lymph nodes. A complaint of back pain may signal the presence of enlarged retroperitoneal nodes.

Positron emission tomography (PET) scanning enables the identification of metastatic disease in patients who might otherwise inappropriately receive definitive local therapy and therefore is now considered a standard staging test. Studies demonstrated that PET will detect unsuspected metastatic disease in approximately 15% of patients after all other staging tests have been completed, although it is not as useful as other techniques in identifying involved regional nodes (based on uptake from the primary tumor obscuring peritumoral lymph nodes). A prospective study of 79 patients found that the specificity and sensitivity of PET for identifying stage IV disease were 90% and 74% versus 47% and 78%, respectively, for the combination of CT and endoscopic ultrasonography (EUS), with an overall accuracy of identification of stage IV disease of 82% versus 64% ( P = .004). Furthermore, when PET was added to CT and EUS, 22% of patients had a change in stage that altered their planned treatment (15% upstaged to incurable stage IV disease and 7% downstaged to a stage at which curative therapy would be appropriate). A consensus panel has confirmed the recommendation for PET imaging in possibly localized esophageal cancer to better detect metastatic disease.

Accurate determination of the extent of disease has a major impact on therapeutic decision making for single-modality versus multimodality treatment or curative versus palliative intent, and therefore it is essential that comprehensive staging be performed. A substantial body of literature now also exists regarding EUS and laparoscopy. The largest and earliest experience was with EUS. A pooled analysis of the literature indicates a sensitivity and specificity, respectively, for stage T1 of 81.6% and 99.4%, T2 of 81.4% and 96.3%, T3 of 91.4% and 94.4%, and T4 of 92.4% and 97.4%. EUS has not been considered accurate in distinguishing in situ cancer from invasive (T1) superficial lesions. For nodal staging based on morphologic features (size, shape, border, and echo characteristics), the observed sensitivity and specificity were 84.7% and 84.6%, respectively, which with the addition of fine-needle aspiration improved to 96.7% and 95.5%. EUS is not a reliable technique for diagnosing liver and peritoneal metastases because of the limited depth of penetration of ultrasound waves.

The indications for minimally invasive surgical staging techniques are not fully defined. Laparoscopic evaluation of abdominal lymph nodes can be achieved with minimal risks and can allow for the identification of node-positive disease not detected with cross-sectional imaging and EUS, although whether the presence of locoregional lymphadenopathy will affect the overall management strategy is unclear. Unsuspected findings such as liver metastases or peritoneal studding that would clearly alter treatment occur in approximately 15% of patients. However, there is not a consensus regarding the use of laparoscopy as part of baseline staging. Whereas the Society of Thoracic Surgeons recommends staging laparoscopy for locally advanced (T3 and T4) GEJ adenocarcinoma that infiltrates the gastric cardia, National Comprehensive Cancer Network guidelines consider laparoscopy an optional part of the staging process in patients with GEJ tumors.

Choice of Therapeutic Options: Barrett Esophagus and Dysplasia

Because of the premalignant nature of Barrett esophagus, it has been recommended that patients with this condition undergo regular endoscopic surveillance, primarily to assess for dysplasia. As mentioned in a previous section, it is estimated that the risk of progression from Barrett esophagus to adenocarcinoma is 0.12% to 0.33% annually. This relatively small risk calls into question the value of routine endoscopic surveillance for Barrett metaplasia, especially in the absence of dysplasia or specialized intestinal metaplasia, given the significant cost and small risk of injury from the follow-up endoscopies.

The focus in the future may be on better defining the risk of progression for individual patients through use of biomarkers and genetic and epigenetic abnormalities. For example, a retrospective assessment of a large multicenter database of patients attempted to develop a model of risk assessment based on age and methylation of eight genes previously observed to be associated with progression of Barrett esophagus: p16, HPP1, RUNX3, CDH13, TAC1, NELL1, AKAP12, and SST. A model was developed with a sensitivity of 90% and specificity of 50%, which could create a low-risk group with a 1.7% rate of progression to high-grade dysplasia or adenocarcinoma over 5 years, a high-risk group with a 27% risk, and intermediate-risk groups of patients who would also be good candidates for close follow-up. Any study like this requires prospective validation.

The most recent American Gastroenterological Association medical position statement on the topic recommends follow-up endoscopy every 3 to 5 years for patients with Barrett esophagus without dysplasia. If low-grade dysplasia is identified, both endoscopic ablative therapy and endoscopic surveillance every 12 months are options. Endoscopic ablation should be offered for patients with high-grade dysplasia. Validated options include endomucosal resection (EMR), radiofrequency ablation, or photodynamic therapy (PDT). The relative benefits and disadvantages of these approaches in this context are beyond the scope of this review.

Esophagectomy is the only treatment option that allows a patient to safely stop periodic endoscopic biopsy surveillance. However, other alternatives may be optimal for many patients. A meta-analysis including 1874 patients who underwent esophagectomy for high-grade dysplasia suggested a risk of unsuspected lymph node involvement of only 1% to 2%, further suggesting that more localized therapies are indeed appropriate. However, the patients who choose endoscopic ablative therapy must be willing to undergo endoscopic biopsy surveillance indefinitely.

Chemotherapy

Single-Agent Chemotherapy

Up to one-half of patients with esophageal cancer have metastatic disease at presentation, and, despite multimodality therapy in the localized setting, the majority of patients will eventually develop recurrence. Therefore most patients with esophageal cancer will receive chemotherapy at some point during their disease course.

Until the early 1990s, chemotherapy drugs commonly used to treat esophagogastric cancers (EGCs) included 5-fluorouracil (5-FU), cisplatin, and mitomycin, with single-agent response rates (RRs) ranging from 10% to 25%. Newer drugs with single-agent activity that have since been evaluated include the oral 5-FU prodrugs (capecitabine and S-1 ), the taxanes (paclitaxel and docetaxel ), and irinotecan, with RRs of 15% to 45%.

A Cochrane meta-analysis by Wagner and colleagues that combined three trials of chemotherapy versus best supportive care (BSC) revealed a clear survival benefit for treatment (HR, 0.37; 95% CI, 0.24–0.55). This translates into an improvement in median overall survival (OS) from 4.3 to 11 months with chemotherapy.

Given the modest activity of single agents in esophageal cancer, combination chemotherapy of two and even three drugs has been extensively studied. A Cochrane meta-analysis of 13 trials including 1914 patients (which mostly evaluated older chemotherapy regimens) did show modest improvements in RRs (35% versus 18%; OR, 2.91; 95% CI, 2.15–3.93), median time to progression (TTP; 5.6 versus 3.6 months; HR, 0.67; 95% CI, 0.49–0.93), and median OS (8.3 versus 6.8 months; HR, 0.82; 95% CI, 0.74–0.90) with combination over single-agent chemotherapy.

Despite incremental advances, the duration of response to both modern single agents and combination regimens is only 4 to 6 months, in general, with a median OS of 10 to 12 months.

Combination Chemotherapy

Fluoropyrimidine/Platinum Doublet

The combination of infusional 5-FU–cisplatin has been studied extensively since the 1980s, and the doublet of a fluoropyrimidine with a platinum compound remains a reference regimen in many contemporary trials. Of note, contemporary studies have generally enrolled patients with adenocarcinomatous histology, irrespective of whether the tumor is found in the lower esophagus, GEJ, or stomach.

More contemporary trials have evaluated substitutions of both of these drugs with either an oral 5-FU prodrug (capecitabine or S-1 ) and/or the newer platinum compound oxaliplatin. Regimens such as S-1–cisplatin, capecitabine-cisplatin, infusional 5-FU–oxaliplatin, and capecitabine-oxaliplatin (along with the anthracycline epirubicin) appear to have at least comparable efficacy compared with 5-FU–cisplatin and are also mostly associated with decreased toxicity and increased ease of administration. In fact, an individual patient data meta-analysis of two randomized trials that compared capecitabine-based with infusional 5-FU–based regimens—the capecitabine-cisplatin versus 5-FU–cisplatin trial and the REAL-2 study, mentioned earlier and discussed in more detail later—suggested that capecitabine-based treatments are associated with superior RR and OS than infusional 5-FU regimens.

The other platinum analogue—carboplatin—has been associated with low single-agent activity in esophagogastric (EG) adenocarcinomas. However, phase II trials of carboplatin combined with taxanes have indicated promising activity, and the combination of carboplatin-paclitaxel with concurrent radiation has emerged as an international standard in locally advanced disease.

One of the only studies that has shown a benefit for a fluoropyrimidine doublet compared with fluoropyrimidine alone, although the former is routinely used, is the phase III SPIRITS (S-1 Plus Cisplatin Versus S-1 in RCT in the Treatment for Stomach Cancer) study. S-1 is a mixture of tegafur (an oral 5-FU prodrug), gimeracil (a dihydropyrimidine dehydrogenate inhibitor that may potentiate the effect of 5-FU), and oteracil (which may reduce the gastrointestinal toxicity of 5-FU). In this Japanese trial in which 298 patients with advanced gastric cancer were randomized to receive S-1 with or without cisplatin, the doublet was associated with a higher RR (54% versus 31%; P = .002) and superior OS (13.0 versus 11.0 months; P = .04). Although this study helped to establish S-1–cisplatin as a standard therapy in Japan, a trial performed in US and European patients comparing S-1–cisplatin versus 5-FU–cisplatin failed to demonstrate superiority of S-1 over infusional 5-FU. Accordingly, S-1 is not widely used outside of East Asia.

Moving Beyond 5-Fluorouracil–Cisplatin

Despite the widespread use in the 1990s of high-dose cisplatin (100 mg/m 2 ) with a 4- or 5-day infusion of 5-FU (at 1000 mg/m 2 /day) as a standard therapy, two seminal trials in the late 1990s directed the development of chemotherapy regimens in EGC away from this schedule of therapy. A European trial compared 5-FU–cisplatin with the FAMTX (bolus 5-FU–doxorubicin-methotrexate) and ELF (etoposide–leucovorin–5-FU) regimens. All regimens performed poorly, with high rates of toxicity, RRs of 20% or lower, and median OS of only approximately 7 months. These disappointing results led investigators in continental Europe to pursue better-tolerated colorectal cancer–like schedules of biweekly infusional 5-FU with a platinum compound and to investigate taxane- and irinotecan-based chemotherapy.

The second key trial was performed in the United Kingdom and compared FAMTX and the ECF regimen, which combines epirubicin with a lower dose of cisplatin (60 mg/m 2 ) and a 21-day infusion of low dose 5-FU (200 mg/m 2 /day). The ECF arm achieved superior RRs (45% versus 21%; P = .0002), median OS (8.9 versus 5.7 months; P = .0009), and QoL at 24 weeks compared with FAMTX. This trial established ECF as the reference regimen in the United Kingdom.

Anthracyclines

After the validation of the ECF regimen, a subsequent study compared it with the MCF regimen, in which mitomycin was substituted for epirubicin (and the infusional 5-FU was administered at a higher dose of 300 mg/m 2 /day). Although designed to test the superiority of MCF, the study reported no significant differences in RR or median survival but found that QoL was better maintained with ECF. Accordingly, ECF remained the preferred regimen, even as some investigators questioned whether either of these triplet regimens is actually superior to the 5-FU–cisplatin doublet.

Subsequently, the REAL-2 (Randomized ECF for Advanced and Locally Advanced Esophagogastric Cancer 2) study compared the ECF regimen and the ECX regimen (which involves the substitution of 5-FU with capecitabine), the EOF regimen (substitution of oxaliplatin for cisplatin) and the EOX regimen (a double substitution of both capecitabine and oxaliplatin) in patients with advanced EG adenocarcinomas or SCCs. All the combinations had similar RRs (40%–48%) and toxicities, and the EOX regimen was associated with improved median OS compared with the ECF regimen (11.2 versus 9.9 months; P = .02), leading the authors to propose that the EOX regimen could replace ECF in future trials.

Despite the standard use of ECF or one of its derivates in the U.K., the clear superiority of this triplet over a fluoropyrimidine-platinum doublet has never been demonstrated in a randomized fashion. One piece of evidence frequently cited to support the incorporation of an anthracycline comes from the previously cited Cochrane meta-analysis, which analyzed three individually negative trials (including the negative evaluation of ECF versus MCF described earlier). Combining all three trials revealed a survival benefit for the addition of epirubicin (HR, 0.77; 95% CI, 0.62–0.91), which translates into an approximate 2-month survival advantage. However, this conclusion comes largely from the comparison of ECF versus MCF because that trial contributed two-thirds of the patients to the meta-analysis. Given the greater toxicity noted on the MCF arm and the fact that the comparison is not purely between a 5-FU–cisplatin-only arm at identical doses and ECF, a determination of the relative merits of adding epirubicin remains difficult to make.

Continuing questions regarding the benefit of an anthracycline were raised by the results of a randomized phase II trial performed by the US Cancer and Leukemia Group B (CALGB) and Eastern Cooperative Oncology Group (ECOG). The CALGB 80403/ECOG 1206 trial randomized 245 patients (222 with adenocarcinomas) to one of three chemotherapy regimens—ECF, FOLFOX (biweekly bolus and infusional 5-FU–leucovorin–oxaliplatin), or cisplatin-irinotecan—along with cetuximab, a monoclonal antibody against the epidermal growth factor receptor (EGFR), for patients with advanced esophagogastric adenocarcinomas or SCCs. In the patients with adenocarcinomas, both the ECF and FOLFOX regimens plus cetuximab produced RRs of greater than 40% (61% and 54% respectively), which met the primary objective of the trial. However, survival outcomes were very similar between the cetuximab-ECF and cetuximab-FOLFOX arms. Median progression-free survival (PFS) was 7.1 versus 6.8 months, and median OS was 11.6 versus 11.8 months, respectively. Overall, cetuximab-FOLFOX appeared to be the least toxic of the three regimens.

Of course, the randomized phase II nature of this study was not designed to detect a survival difference between these regimens, and the contribution of cetuximab, now thought to be either neutral or even detrimental, cannot be determined. Nevertheless, the results of this trial do support the contention that any benefit of an anthracycline, if there is any benefit at all, is likely to be small. Another notable finding of this study is that cetuximab-cisplatin-irinotecan had the lowest RRs and survival outcomes and the highest toxicity rates, which has contributed to a significant decline in the use of this chemotherapy regimen in the first-line setting. Correspondingly, the favorable toxicity profile and activity of the FOLFOX arm have reinforced its established role as the standard first-line regimen in the United States.

Taxanes

In comparison to the unclear benefit of adding an anthracycline to a fluoropyrimidine-platinum doublet, there are randomized data to support the addition of a taxane. The phase III V325 randomized trial in GEJ and gastric adenocarcinomas compared the DCF regimen (docetaxel–cisplatin–infusional 5-FU) and infusional 5-FU–cisplatin. The addition of docetaxel improved RRs (37% versus 25%; P = .01) and TTP (5.6 versus 3.7 months; P < .001), but OS was only slightly improved (median OS 9.2 versus 8.6 months; 2-year OS, 18% versus 9%; P = .02). In addition, the three-drug regimen was associated with significantly more toxicity, including a grade 3/4 neutropenia rate of 82% (versus 57%) and febrile neutropenia in 29% of patients (versus 12%). Fifty percent of patients came off treatment because of either severe adverse events or consent withdrawal. Despite these significant toxicities, the authors reported a slower decrement in QoL measurements in the DCF arm. On the basis of this study, docetaxel was approved by the US Food and Drug Administration (FDA) in 2006 for use with 5-FU–cisplatin in this context.

Because the toxicities seen with the parent DCF regimen are significant and may outweigh its small survival advantage over 5-FU–cisplatin, it has not been widely adopted. It is also unclear if similar survival benefits would be accrued by the sequential use of first-line 5-FU–cisplatin followed by subsequent docetaxel (or paclitaxel) at progression.

Several investigators have attempted to modify the regimen to increase tolerability. For example, our group performed a randomized phase II trial of parent DCF (with prophylactic growth factor support) versus a modified DCF (mDCF) regimen (consisting of reduced doses of docetaxel and cisplatin administered with bolus and 2-day infusional 5-FU and leucovorin every 14 days). mDCF was associated with decreased toxicity compared with parent DCF (neutropenic fever rate, 9% versus 16%; grade 3/4 nausea/vomiting rate, 2% versus 23%), and median OS appeared superior in the mDCF arm (18.8 versus 12.6 months; P = .007). Nevertheless, 22% of the patients receiving mDCF (who had a median age of 59 years) required hospitalization in the first 3 months, mostly for treatment-related toxicities (febrile neutropenia and gastrointestinal toxicities), reinforcing the notion that this remains a relatively difficult regimen to administer and is an option only for healthy, motivated patients with frequent access to medical evaluation.

Another extensively evaluated and commonly used modification is the German FLOT regimen, which consists of the substitution of oxaliplatin for cisplatin and is built around a 1-day infusion of 5-FU every 14 days. Encouraging activity was noted in a phase II study of 59 patients, which reported an RR of 58% and a median OS of 11.1 months. The FLOT regimen was subsequently compared with the 5-FU–oxaliplatin doublet (FLO) in the randomized phase II FLOT65+ study, which enrolled patients who were 65 years of age or older and had either locally advanced or metastatic disease. No benefit was seen for the triplet combination in patients aged 65 years or older with metastatic disease or in any patient aged 70 years or older. This study again emphasizes the need for strict patient selection for such three-drug regimens.

Irinotecan

Irinotecan is another active agent in EGC that has been combined with mitomycin, 5-FU–leucovorin, and cisplatin with or without docetaxel in phase II evaluations, with RRs ranging from 30% to 65%. Toxicities in some of these trials (e.g., with cisplatin-docetaxel-irinotecan) have been substantial.

A randomized phase II trial compared the FUFIRI regimen (weekly infusional 5-FU–leucovorin–irinotecan) with cisplatin-irinotecan in patients with advanced GEJ and gastric adenocarcinomas. FUFIRI was associated with superior outcomes and less neutropenia than cisplatin-irinotecan. This led to a subsequent phase III trial of FUFIRI versus 5-FU–cisplatin. Both regimens had comparable efficacy, but there was less neutropenic fever and grade 3/4 stomatitis and nausea in the FUFIRI arm. Only the incidence of grade 3/4 diarrhea was increased in the FUFIRI arm, although more patients withdrew from the 5-FU–cisplatin arm than the FUFIRI arm (22% versus 10%; P = .004) for drug-related adverse events. Although there was no clear benefit for FUFIRI over 5-FU–cisplatin, the favorable toxicity of this combination supports its use as a front-line option, especially in patients who are not candidates for a platinum compound.

The use of first-line 5-FU–irinotecan is now further supported by the results of a phase III French Fédération Francophone de Cancérologie Digestive (FFCD) group study, in which 416 patients were randomized to the FOLFIRI regimen (biweekly bolus and infusional 5-FU–leucovorin–irinotecan) versus ECX in patients with advanced gastric cancer. At progression, patients received therapy with the alternate regimen. This study revealed a superior time to treatment failure for FOLFIRI versus ECX (5.1 versus 4.2 months; P = .008) and comparable PFS (5.3 versus 5.8 months; P = .96) and OS (9.5 versus 9.7 months; P = .95). Toxicities were significantly less for FOLFIRI (e.g., overall grade 3/4 toxicity rate of 69% versus 84%; P < .001). Time to treatment failure was selected as the primary end point because it captured discontinuation of a regimen for both efficacy and toxicity reasons.

Finally, Boku and colleagues also performed a phase III trial in which Japanese patients were randomized to infusional 5-FU versus cisplatin-irinotecan (a third arm was designed to evaluate and did confirm noninferiority of S-1 versus infusional 5-FU). When compared with infusional 5-FU, cisplatin-irinotecan was associated with an improved RR (38% versus 9%) but only a nonsignificant trend toward improved median OS (12.3 versus 10.8 months; P = .055), at the expense of significantly more grade 3/4 toxicities. As noted earlier, results of the CALGB 80403/ECOG 1206 study, which randomized patients to cetuximab plus ECF versus FOLFOX versus cisplatin-irinotecan, revealed that the cisplatin-irinotecan arm had the lowest RR and shortest median OS.

Taken together, these results have led many oncologists to move away from using cisplatin-irinotecan as a first-line regimen for advanced EGCs, although FOLFIRI has recently emerged as a viable first-line option. Some uncertainty about the superiority of irinotecan in the first-line setting was reinforced by the Cochrane meta-analysis of four clinical trials, which revealed a non–statistically significant trend toward a small survival benefit for an irinotecan-containing regimen (HR, 0.86; 95% CI, 0.73–1.02). Notably, the FFCD study discussed earlier was not included in this meta-analysis.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here

Related Posts