Brain metastases of lung cancer are associated with a poor prognosis. Little research has been conducted to directly compare erlotinib with gefitinib regarding the frequency of central nerve system recurrence. This is the first study to directly compare erlotinib with gefitinib in terms of brain metastases recurrence rates in patients with EGFR-mutant NSCLC who had no brain metastasis at the time of starting TKI treatment. This was a singlecenter retrospective study. Advanced or recurrent

more »

... all cell lung cancer patients with no brain metastases at the time of starting initial tyrosine kinase inhibitor treatment who received either gefitinib or erlotinib monotherapy were selected. The primary endpoint was the incidence of brain metastases, and secondary endpoints included the objective response rate, progression-free survival, overall survival, and Post-Progression Survival in subgroups based on tyrosine kinase inhibitor treatment and the occurrence of brain metastases. There were 119 patients in the gefitinib group and 13 patients in the erlotinib group. Brain metastases at disease progression were observed in 16 patients in the gefitinib group, and in no patients in the erlotinib group (13.5% vs. 0%, p=0.37). The median overall survival was 29.2 months in the gefitinib group and was not reached in the erlotinib group (p=0.14). The median PPS was 15.5 months in the gefitinib group and 23.7 months in the erlotinib group (p=0.11). Based on the occurrence of brain metastases, Post-Progression Survival was significantly longer in the no brain metastases group (8.0 months vs. 17.9 months, p=0.01). These data showed the possibility of a lower centralnerve-system recurrence rate with erlotinib compared with gefitinib. Post-Progression Survival in patients with brain metastases was significantly shorter than that of patients without brain metastases.