RESUMO
BACKGROUND: Targeting the epidermal-growth-factor-receptor (EGFR) in non-small cell lung cancer (NSCLC) is an established treatment option with less toxicity compared to conventional chemotherapy. This study was undertaken to determine whether Erlotinib is non-inferior compared to chemotherapy as a first-line therapy in unselected elderly patients. MATERIALS AND METHODS: Patients ≥ 70 years with untreated, metastatic NSCLC were randomized to Erlotinib (E), 150 mg/day or Carboplatin (AUC5) plus Vinorelbine (25mg/m(2) on days 1 and 8) every three weeks (CV). Primary endpoint was progression-free survival (PFS). After progression, crossover was strongly recommended. Secondary endpoints were duration of response, 1-year survival, overall survival (OS), response rate (RR), quality of life (FACT-L), assessment of comorbidities by simplified comorbidity score (SCS) and Charlsons' comorbidity score, safety and assessment of molecular markers. RESULTS: Between June 2006 and August 2008 284 pts were randomized to E (144) and CV (140). PFS was significantly inferior with E (median PFS 2.4 versus 4.6 months [HR 1.6, 75% CI 1.22-2.09, p: 0.0005]) as well as RR (7.8% v 28.3%, p: 0.0001). No significant difference in OS appeared (median E: 7.3 months versus CV: 8.4 months, HR: 1.24 [75% CI 0.9-1.71]). In never smokers PFS (median PFS: 3.7 v 4.3 m, E v CV, HR 0.72, 75% CI 0.35-1.48) and OS (median: 16.5 versus 17 months, HR 0.99 [75% CI 0.38-2.57]) were comparable. More skin toxicity and diarrhea was seen with E compared to more myelotoxicity, neurotoxicity and constipation with CV. Less severe adverse events were observed with E (81 v 102, E v CV). CONCLUSION: CV had an increased efficacy compared with E in an unselected population of elderly patients with advanced NSCLC.