No Overall Survival Improvement but Some Palliative Benefit With Gefitinib in Esophageal Cancer Progressing Postchemotherapy

Get Permission

Gefitinib in Esophageal Cancer

Susan J. Dutton, MSc, and colleagues

Future research should focus on identification of predictive biomarkers to identify this subgroup of benefiting patients. —Susan J. Dutton, MSc, and colleagues

—Susan J. Dutton, MSc, and colleagues

In what may be the first randomized trial of systemic therapy in this setting, Susan J. Dutton, MSc, of University of Oxford, United Kingdom, and colleagues evaluated the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor gefitinib vs placebo in patients with esophageal cancer progressing after chemotherapy.1 As reported in the Lancet Oncology, the COG trial showed no survival benefit with gefitinib but provided evidence of palliative benefit of the drug in odynophagia and other patient-reported outcomes.

EGFR is expressed in most esophageal cancers, and increased EGFR copy number is found in 5% to 10% and is more common in esophageal cancer than in other gastrointestinal cancers. Both of these characteristics are associated with poor survival in esophageal cancer. Further, activating EGFR mutations have been identified in esophageal cancer.

Phase II trials of EGFR tyrosine kinase inhibitors have shown objective responses in approximately 8% of patients including those with adenocarcinoma and squamous cell carcinoma. In contrast, no responses to such treatment have been observed in gastric cancer. These observations formed the rationale for the phase III COG trial.

Study Details

In this double-blind trial, 449 adults at 48 UK centers with advanced esophageal cancer or type I/II Siewert junctional tumors and histologically confirmed squamous cell carcinoma or adenocarcinoma who had progressed after chemotherapy were randomly assigned between March 2009 and November 2011 to receive gefitinib at 500 mg/d (n = 224) or placebo (n = 225). Patients had to have measurable or evaluable disease on computed tomography. All patients received best supportive care in addition to study treatment. The primary endpoint was overall survival in the intention-to-treat population.

The gefitinib and placebo groups were generally balanced for age (median, 65 years in both), sex (82% and 84% men), original diagnosis (adenocarcinoma in 77% and 75%, squamous cell carcinoma in 22% and 25%), disease site (esophageal in 76% and 80%, type I junctional (distal part of esophagus) in 12% and 9%, type II junctional (adenocarcinoma of the real cardia, within 1 cm above and 2 cm below the esophagogastric junction) in 12% and 10%), World Health Organization performance status (0 in 25% in both, 1 in 52% and 56%, 2 in 22% and 20%), previous treatments (3 in 5% and 4%, 2 in 35% and 33%, 1 in 61% in both), brain metastases (no in > 99% and 96%), and body mass index (median, 24.01 kg/m2 in both).

Survival and Disease Control

Median overall survival was 3.73 months (95% confidence interval [CI] = 3.23–4.50) in the gefitinib group vs 3.67 months (95% CI = 2.97–4.37) in the placebo group (hazard ratio [HR] = 0.90, P = .29). Patients were not stratified by performance status; performance status was a significant predictor of overall survival, with median overall survival of 6.1, 3.9, and 2.2 months in patients with performance status of 0, 1, and 2 (P < .0001 for trend).

Median progression-free survival was significantly but minimally prolonged with gefitinib (1.57 vs 1.17 months, HR = 0.80, P = .020). In subgroup analyses, hazard ratios consistently favored gefitinib, with significant or borderline significant benefit observed in subgroups with performance status of 0 (HR = 0.67), those with esophageal disease site (HR = 0.80), males (HR = 0.82), those with one previous treatment (HR = 0.79), and those with adenocarcinoma (HR = 0.81).

Disease control at 8 weeks was present in 24% of patients in the gefitinib group (including 6 partial responses) and in 16% of the placebo group (including 1 partial response; P = .02). Among gefitinib patients, disease control rates were 23% in those with adenocarcinoma and 28% in those with squamous cell carcinoma. Objective response to gefitinib was rapid, with 7 of 8 observed by week 4. Responses persisted for 1.2 to 7.3 months.

In total, 22% of patients received further treatment, including chemotherapy in 9% of the gefitinib group vs 10% of the placebo group, palliative radiotherapy in 8% vs 11%, and stents in 6% vs 8%.

Patient-Reported Outcomes

Health-related quality of life was assessed by the generic European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 and the esophageal, junctional, and gastric cancer–specific EORTC QLQ-OG25. Questionnaires for prespecified patient-reported outcomes were completed at baseline and 4 weeks by 110 gefitinib patients and 121 placebo patients. Among these patients, gefitinib patients had significantly improved odynophagia (P = .004), with improvement from baseline observed in the gefitinib group. No significant differences were found in the other prespecified outcomes, consisting of global quality of life, dysphagia, and eating restriction, although all changes favored gefitinib.

Among exploratory patient-reported outcomes, the gefitinib group had significantly better outcomes in social functioning (P = .013), pain (P = .035), constipation (P = .0001), cough (P = .013), and speech function (P = .0004), with improvement or no worsening from baseline in each, whereas the placebo group reported less worsening of diarrhea (P < .0001). Virtually all other changes (eg, in hair loss, dyspnea, anxiety, sleep, saliva, reflux) favored gefitinib.

Adverse Events

The most common adverse events of any grade in the gefitinib group were diarrhea (16% vs 3% in the placebo group) and skin toxicity (21% vs 1%). The most common grade 3 or 4 adverse events were fatigue (11% vs 6%) and diarrhea (6% vs 1%). In total, grade 3 or 4 gastrointestinal adverse events occurred in 21% vs 23%, and grade 3 or 4 respiratory adverse events occurred in 14% vs 13%.

Serious adverse events were reported in 49% and 45% of patients. Three fatal serious adverse events (2 in the placebo group and 1 in the gefitinib group) were considered possibly related to study treatment. Dose reduction from 500 mg to 250 mg occurred in 4% and 1% of patients.

The investigators concluded, “The use of gefitinib as a second-line treatment in oesophageal cancer in unselected patients does not improve overall survival, but has palliative benefits in a subgroup of these difficult-to-treat patients with short life expectancy. Future research should focus on identification of predictive biomarkers to identify this subgroup of benefiting patients.” ■

Disclosure: The study was funded by Cancer Research UK. For full disclosures of the study authors, visit


1. Dutton SJ, Ferry DR, Blazeby JM, et al: Gefitinib for oesophageal cancer progressing after chemotherapy (COG): A phase 3, multicentre, double-blind, placebo-controlled randomised trial. Lancet Oncol 15:894-904, 2014.


Related Articles

EGFR as a Therapeutic Target for Gastroesophageal Cancer—or Is It Really?

The epidermal growth factor receptor (EGFR) gene is often amplified and its protein overexpressed in upper gastrointestinal cancers—and overexpression has prognostic value. With the advent of monoclonal antibodies and tyrosine kinase inhibitors against EGFR, we have witnessed a rash of randomized...