We established a drug sensitivity panel consisting of 24 human lung cancer cell lines. Using this panel, we evaluated 26 anti-cancer agents: three alkylators, three platinum compounds, four antimetabolites, one topoisomerase I inhibitor, five topoisomerase II inhibitors, seven antimitotic agents and three tyrosine kinase inhibitors. This panel showed the following: a) Drug sensitivity patterns reflected their clinically-established patterns of action. For example, doxorubicin and etoposide were shown to be active against small cell lung cancer cell lines and mitomycin-C and 5-fluorouracil were active against non-small cell lung cancer cell lines, in agreement with clinical data. b) Correlation analysis of the mean graphs derived from the logarithm of IC50 values of the drugs gave insight into the mechanism of each drug's action. Thus, two drug combinations with reverse or no correlation, such as the combination of cisplatin and vinorelbine, might be good candidates for the ideal two drug combination in the treatment of lung cancer, as is being confirmed in clinical trials. c) Using cluster analysis of the cell lines in the panel with their drug sensitivity patterns, we could classify the cell lines into four groups depending on the drug sensitivity similarity. This classification will be useful to elucidate the cellular mechanism of action and drug resistance. Thus, our drug sensitivity panel will be helpful to explore new drugs or to develop a new combination of anti-cancer agents for the treatment of lung cancer.