Gefitinib for oesophageal cancer progressing after chemotherapy (COG): a phase 3, multicentre, double-blind, placebo-controlled randomised trial

Background Evidence is scarce for the effectiveness of therapies for oesophageal cancer progressing after chemotherapy, and no randomised trials have been reported. We aimed to compare gefitinib with placebo in previously treated advanced oesophageal cancer. Methods For this phase 3, parallel, randomised, placebo-controlled trial, eligible patients were adults with advanced oesophageal cancer or type I/II Siewert junctional tumours, histologically confirmed squamous-cell carcinoma or adenocarcinoma, who had progressed after chemotherapy, with WHO performance status 0-2, and with measurable or evaluable disease on CT scan. Participants were recruited from 48 UK centres and randomly assigned (1:1) to gefitinib (500 mg) or matching placebo by simple randomisation with no stratification factors. Patients, clinicians, and trial office staff were masked to treatment allocation. Treatment continued until disease progression, unacceptable toxicity, or patient choice. The primary outcome was overall survival, analysed by intention to treat. This trial is registered, number ISRCTN29580179. Findings Between March 30, 2009, and Nov 18, 2011, 450 patients were randomly assigned to treatment groups (one patient withdrew consent; 224 patients allocated gefitinib and 225 allocated placebo included in analyses). Overall survival did not differ between groups (median 3 73 months, 95% CI 3 23-4 50, for gefitinib vs 3 67 months, 95% CI 2.97-4.37, for placebo; hazard ratio [HR] 0.90, 95% CI 0.74-1 09, p=0.29). Among the prespecified patient-reported outcomes (110 patients on gefitinib and 121 on placebo completed both baseline and 4 week questionnaires and were included in analyses), odynophagia was significantly better in the gefitinib group (adjusted mean difference -8.61, 95% CI -14.49 to -2 73; n=227; p=0.004), whereas the other outcomes were not significantly improved compared with placebo: global quality of life (2.69, 95% CI -2.33 to 7.72, n=231, p=0.293), dysphagia (-3.18, 95% CI -8.36 to 2.00, n=231, p=0.228), and eating (-4.11, 95% CI -9.96 to 1.75, n=229, p=0.168). Median progression-free survival was marginally longer with gefitinib than it was with placebo (1.57 months, 95% CI 1.23-1.90 in the gefitinib group vs 1.17 months, 95% CI 1.07-1.37 in the placebo group; HR 0.80, 95% CI 0.66-0.96, p=0.020). The most common toxicities were diarrhoea (36 [16%] of 224 patients on gefitinib vs six [3%] of 225 on placebo) and skin toxicity (46 [21%] vs two [1%]), both mostly grade 2. The commonest grade 3-4 toxicities were fatigue (24 [11%] vs 13 [6%] patients) and diarrhoea (13 [6%] vs two [1%]). Serious adverse events were reported in 109 (49%) of 224 patients assigned to gefitinib and 101 (45%) of 225 on placebo. 54 (24%) of patients in the gefitinib group achieved disease control at 8 weeks, as did 35 (16%) of patients on placebo (p=0.023). Interpretation 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.