Introduction: The optimal use of epidermal growth factor receptor (EGFR)-related molecular markers to prospectively identify tyrosine kinase inhibitor (TKI)-sensitive patients, particularly after a previous chemotherapy treatment, is currently under debate.
Methods: We designed a prospective phase II study to evaluate the activity of EGFR-TKI in four different patient groups, according to the combination of molecular (EGFR gene mutations, EGFR gene copy number and protein expression, and phosphorylated AKT expression, pAKT) and clinicopathological (histology and smoking habits) factors. Correlations between molecular alterations and clinical outcome were also explored retrospectively for first-line chemotherapy and EGFR-TKI treatment.
Results: Patients who had progressed during or after first-line chemotherapy were prospectively assigned to EGFR-TKI treatment as follows: (G1) EGFR mutation (n = 12); (G2) highly polysomic/amplified EGFR (n = 18); (G3) EGFR and/or pAKT positive (n = 41); (G4) adenocarcinoma/bronchoalveolar carcinoma and no smoking history (n = 15). G1 and G4 had the best and second-best overall response rate (25% and 20%, respectively), whereas the worst outcome was observed in G2 (ORR, 6%; p = 0.05). Disease control was highest in G1 and G4 (>50%) and lowest in G3 (<20%) (p = 0.02). Patients selected by EGFR mutation or clinical parameters (G1 and G4) also had significantly better progression-free survival and overall survival (p = 0.02 and p = 0.01, respectively). Multivariate analysis confirmed the impact of sex, smoking history, EGFR/KRAS mutation, and pAKT on outcomes and allowed us to derive an efficient predictive model. Histology, EGFR mutations, and pAKT were independent predictors of response to first-line chemotherapy at retrospective analysis, whereas pAKT and human epidermal growth factor receptor 2 expression were the only independent predictors of progression-free survival and overall survival.
Conclusions: Selection of patients based on either EGFR mutation or clinical characteristics seems an effective approach to optimize EGFR-TKI treatment in chemotherapy-pretreated non-small-cell lung cancer patients.