Antagonism between gefitinib and cisplatin in non-small cell lung cancer cells: Why randomized trials failed?

Introduction: Four phase III randomized trials adding epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors to standard chemotherapeutics in patients with advanced non-small cell lung cancer (NSCLC) have failed to show benefits. The mechanism of these failures was examined. Methods: Fifteen previously untreated NSCLC cell lines were simultaneously treated with gefitinib plus cisplatin. Three exhibited sensitizing-EGFR mutations. Three selected lines were further tested with paclitaxel/cisplatin, paclitaxel/gefitinib, and paclitaxel/cisplatin/gefitinib combinations. The tetrazolium colorimetric assay with application of the classic isobole method was used, and dose-versus-log-response curves (DRCs) were analyzed to evaluate possible resistance mechanisms. Results: Of the 15 cell lines tested, combined gefitinib/cisplatin was significantly antagonistic in 10 wild-type and three sensitizing-EGFR mutant cell lines (group mean combination index = 1.184, 95% confidence interval = 1.12-1.24, p = 0.001). The mean combination index values of paclitaxel/cisplatin/gefitinib were higher than or comparable with those of paclitaxel/cisplatin and paclitaxel/gefitinib. DRC analysis consistently showed nonsaturable passive resistance, suggesting that gefitinib at 0.001 to 0.3 μM can interfere with cisplatin cell entry (at concentrations >1-3 μM) in a dose-dependent manner and lead to antagonism. This antagonism may or may not be schedule dependent in different cell lines. Conclusions: In most EGFR wild-type or sensitizing-mutant NSCLC cells, the concomitant gefitinib/cisplatin combination showed antagonism, likely because gefitinib interfered with cisplatin entry into the cell. The findings that three-drug combination was not better than the two-drug combinations are in accordance with the results of the randomized trials. The EGFR-tyrosine kinase inhibitor/platinum antagonism is a possible reason for the failure of those randomized trials.