Factorial phase III randomised trial of rofecoxib and prolonged constant infusion of gemcitabine in advanced non-small-cell lung cancer: the GEmcitabine-COxib in NSCLC (GECO) study

Background: The addition of cyclo-oxygenase-2 (COX-2) inhibitors and prolonged constant infusion (PCI) of gemcitabine to treatment for advanced non-small-cell lung cancer (NSCLC) might improve treatment efficacy. We aimed to assess whether the addition of rofecoxib or PCI gemcitabine could improve overall survival compared with first-line treatment with cisplatin plus gemcitabine given by standard infusion. Methods: Patients with stage IV or IIIb (with supraclavicular nodes or pleural effusion) NSCLC who were under 70 years of age and who had performance status 0 or 1 were eligible for this multicentre, prospective, open-label, randomised phase III trial with 2×2 factorial design. Patients were randomly assigned to one of four treatment groups: group A, gemcitabine 1200 mg/m² in a 30-min intravenous infusion on days 1 and 8 and intravenous cisplatin 80 mg/m² on day 1, every 21 days for six cycles; group B, the same treatments as group A plus oral rofecoxib 50 mg/day until disease progression; group C, intravenous PCI gemcitabine 1200 mg/m² in a 120-min infusion on days 1 and 8 and intravenous cisplatin 80 mg/m² on day 1, every 21 days for six cycles; group D, the same drugs as group C plus oral rofecoxib 50 mg/day until disease progression. The primary endpoint was overall survival; secondary endpoints were progression-free survival, response rate, quality of life, and toxicity. Analyses were intention-to-treat. This trial is registered on the clinical trials site of the US National Institutes of Health website http://clinicaltrials.gov/ct/show/NCT00385606. Findings: Between Jan 30, 2003, and May 3, 2005, 400 patients were enrolled. Median age was 60 years (range 29-71). PCI gemcitabine did not improve overall survival (median 47 weeks [95% CI 40-55] vs 44 [36-52], with standard gemcitabine infusion, hazard ratio (HR) of death 0·93 [0·74-1·17], p=0·41), progression-free survival, nor any other secondary endpoint. Vomiting and fatigue were significantly worse with PCI gemcitabine. The two rofecoxib groups were closed early (on Oct 1, 2004) due to withdrawal of the drug because of safety issues. With intention-to-treat statistical analyses limited to 240 patients (ie, those randomised before July 1, 2004) who had at least 3 months of treatment, rofecoxib did not prolong overall survival (median 44 weeks [CI 36-55] vs 44 [40-54] without rofecoxib, and HR of death 1·00 [0·75-1·34], p=0·85), or progression-free survival, but did improve response rate (41% vs 26%, p=0·02), global quality of life, physical, emotional and role functioning, fatigue, and sleeping. Rofecoxib significantly increased the incidence of diarrhoea and decreased constipation, fatigue, fever, weight loss, and pain, and analgesic consumption. Severe cardiac ischaemia was more frequent with rofecoxib than without; however, the difference was not statistically significant in the primary analysis (p=0·06) and became significant when patients who were randomised between July 1, 2004, and Sept 30, 2004, were included in the analysis (p=0·03). Interpretation: Neither PCI gemcitabine nor rofecoxib prolonged survival in the patients in this study. Rofecoxib improved response rate and several quality-of-life items, including pain-related items and global quality of life. Further studies with less cardiotoxic COX-2 inhibitors are needed in NSCLC.