BACKGROUND Although activating epidermal growth factor receptor (EGFR) mutations are excellent predictors of gefitinib outcome in non-small-cell lung cancer (NSCLC), most patients become resistant to gefitinib. Despite our knowledge of the molecular basis of acquired resistance, clinical predictors have not been well elucidated. This study was undertaken to evaluate predictors of clinical outcome in patients with NSCLC and with EGFR mutations treated with gefitinib.

PATIENTS AND METHODS A total of 170 patients with NSCLC and with EGFR mutations received gefitinib as a first-line (n = 50) and a second-line or more (n = 120) treatment at Seoul National University Hospital. Treatment outcomes were compared between groups based on clinicopathologic factors, such as treatment line, metastatic site, and mutation subtype.

RESULTS Survival outcomes were similar between first-line and second-line or greater gefitinib treatment (overall response rate, 2P = .832; progression-free survival [PFS], 2P = .373; and overall survival [OS], 2P = .290). When the number of metastatic sites was at least 3, significantly reduced survival was observed (median PFS 8.5 vs. 14.0 months, 2P < .001; median OS 21.4 vs. 25.6 months, 2P = .002). In addition, the presence of at least 3 organs with metastases was an independent predictor of PFS (hazard ratio [HR] 1.97 [95% CI, 1.37-2.85]; 2P < .001) and OS (HR 2.00 [95% CI, 1.18-3.39]; 2P = .010). Patients who failed to respond to gefitinib within 6 months of treatment had more lymph node metastases and more sites of metastasis than those who responded later.

CONCLUSIONS Tumor burden, expressed as the number of metastatic sites, is predictive of inferior survival in patients with NSCLC and with activating EGFR mutations who are treated with gefitinib.