The succinic dehydrogenase inhibition (SDI) test has been widely used to evaluate the sensitivity of cancer cells to anticancer agents. Recently, the techniques for tissue culture have progressed; the ability to make formazan has improved with the use of methylthiazol tetrazolium bromide (MTT) instead of triphenyl tetrazolium chloride (TTC) in this assay. Therefore, we modified the SDI test in order to evaluate the drug response to gastrointestinal cancer and applied it in a clinical study. In this assay, the optimal concentration levels of mytomycin-C and Adriamycin for the clinical materials are 10 micrograms/ml, that is, 10 times higher than for the cancer cell line, and that of Cisplatin is 30 micrograms/ml, for Cisplatin is unstable in the culture medium. To evaluate its antitumor effect, it was found necessary to expose the cancer cells to 5-fluorouracil (5-FU) for 3 days. We applied this assay for the clinical materials; the evaluable rate of this assay is 83.3%. The peculiarity of this assay is its high evaluable rate. These results suggest that this test may be useful for clinical application.