ABSTRACT
PURPOSE The anti-epidermal growth factor receptor (EGFR) antibody cetuximab is efficacious in multiple tumor types. Patient selection with markers predictive of benefit may enhance its therapeutic index. This retrospective, correlative analysis of the phase III trial BMS099 of cetuximab in advanced non-small-cell lung cancer (NSCLC) was conducted to identify molecular markers for the selection of patients most likely to benefit from cetuximab. METHODS In BMS099, 676 chemotherapy-naïve patients with stage IIIB (pleural effusion) or stage IV NSCLC of any histology or EGFR expression status were randomly assigned to taxane/carboplatin (T/C) with or without cetuximab. Biomarkers analyzed included K-Ras and EGFR mutations by direct sequencing, EGFR protein expression by immunohistochemistry (IHC), and EGFR gene copy number by fluorescent in situ hybridization (FISH). Relationships between biomarker status and progression-free survival (PFS), overall survival (OS), and overall response rate (ORR) were assessed by log-rank tests per treatment arm for treatment-specific effects and across the total evaluable population. Results Tumor samples were available from 225 randomly assigned patients. K-Ras mutations were found in 17% of evaluable patients (35 of 202 patients), EGFR mutations were found in 10% (17 of 166 patients), EGFR positivity by IHC was found in 89% (131 of 148 patients), and FISH positivity was found in 52% (54 of 104 patients). No significant associations were found between biomarker status and PFS, OS, and ORR in the treatment-specific analyses. CONCLUSION In contrast with colorectal cancer, and within the limitations of the data set, efficacy parameters did not appear to correlate with K-Ras mutation status or with any of the EGFR-related biomarkers evaluated. Additional exploratory analyses are essential to identify predictive markers and to optimize patient selection for cetuximab therapy in NSCLC.
