Biomarker analysis of the TRIBUTE Trial of chemotherapy with or without erlotinib in advanced non-small cell lung cancer (NSCLC): FISH positivity predicts outcome: D2-05

Epidermal growth factor receptor (EGFR) gene copy number by fluorescence in-situ hybridization (FISH) is a predictor of survival benefit from EGFR inhibitor monotherapy, as demonstrated in prospective placebo-controlled clinical trials in advanced NSCLC. We aimed to investigate the predictive role of EGFR gene copy number in patients treated in a clinical study testing the combination of erlotinib and chemotherapy (TRIBUTE trial). TRIBUTE was a phase III clinical trial that randomly assigned 1.059 patients to carboplatin and paclitaxel (CP) with erlotinib/placebo followed by maintenance erlotinib/placebo. There was no difference in overall survival or progression-free survival between treatment groups (P=0.95 and P=0.36, respectively). We analyzed 275 tissue samples from study participants (26% of study population) for EGFR gene copy number by FISH using commercially available EGFR/CEP7 probe (Abbott Molecular, Des Plaines, IL, USA) and previously established scoring criteria (Cappuzzo et al., 2005). The results of EGFR FISH test were available for 245 patients (89% of group subjected to analysis, including 121 patients who received erlotinib and 124 who received placebo). In this group of patients, time-to-progression (TTP) was not different between study arms but overall survival (OS) was longer for placebo than for erlotinib (median 13.2 months vs. 9.6 months, respectively, P=0.033). One hundred patients were EGFR FISH positive (FISH+, 40.8%). In the FISH+ subset, objective response rates to CP and erlotinib vs. CP and placebo were 11.6% and 29.8%, respectively (P=0.049). Response rates in FISH- subset of patients were 21.8% vs. 25.4%, respectively (P=0.695). In FISH+ patients, TTP was significantly longer for erlotinib vs. placebo (HR=0.59, 95%CI: 0.35-0.99 log-rank P=0.043) and the difference was observed mainly after six months of treatment. There was no difference in OS (HR=1.52, 95% CI: 0.94-2.46, log-rank P=0.083). In FISH- patients, TTP and OS were not different between treatment arms (HR=1.42, 95%CI: 0.95-2.14, log-rank P=0.089 and HR=1.24, 95%CI: 0.84-1.82, log-rank P=0.278, respectively). The treatment by EGFR FISH interaction test was significant for TTP (P=0.007) but not for OS (P=0.49). EGFR gene copy number assessment by FISH may predict longer TTP in advanced NSCLC patients treated with first-line chemotherapy and erlotinib. This difference is seen mainly after completion of chemotherapy. The predictive value of EGFR FISH test on OS was not confirmed. These exploratory results suggest the benefit from maintenance erlotinib after completion of chemotherapy in FISH+ group of patients.