Clinical Observation of Gefitinib with Pericardial Perfusion for Advanced Non-small Cell Lung Cancer / 中国肺癌杂志

ABSTRACT

BACKGROUND@#Epidermal growth factor receptor (EGFR) mutation non-small cell lung cancer (NSCLC) is an important subtype of lung cancer. The incidence of malignant pericardial effusion (MPCE) in EGFR-mutant NSCLC patients is high. However, there are few researches on the treatmentof this type of patients.@*METHODS@#We collected data on clinical characteristics and treatment of advanced NSCLC patients who harboring EGFR mutants and MPCE between January 2010 and December 2016. The treatments were divided into three groups oral gefitinib combined with pericardial perfusion of hydroxycamptotheci (HCPT) group (gefitinib/HCPT); intravenous chemotherapy combined with pericardial perfusion of HCPT group (chemotherapy/HCPT) and gefitinib monotherapy group. And we retrospectively analyzed patients' outcomes in three groups.@*RESULTS@#In 273 advanced NSCLC patients with EGFR mutations, 29 cases had pericardial effusion, among which 6 patients with small amount of pericardial effusion were excluded, and 23 patients were analyzed. Median pericardium progression free survival (PFS) was 247 days. PFS for gefitinib/HCPT group (460 days) was superior to PFS for chemotherapy/HCPT group (94 days, P=0.008) and gefitinib monotherapy group (131 days, P=0.032). As for the efficacy of primary pulmonary lesions, the efficacy in gefitinib/ HCPT group was superior to chemotherapy/HCPT group [objective response rate (ORR) 33.3% vs 12.5%; disease control rate (DCR) 86.7% vs 62.5%]. There is no difference of ORR and DCR between gefitinib/HCPT group and gefitinib monotherapy group. No obvious adverse reaction was observed in all three groups.@*CONCLUSIONS@#First-line gefitinib therapy combined with pericardial perfusion of HCPT can improve pericardium PFS for advanced NSCLC patients who harboring EGFR mutants andmalignantpericardial effusion. This finding should be confirmed further through multicenter, prospective clinical trials with large sample size.