We describe our experience with succinic dehydrogenase inhibition (SDI) test for solid tumors as a chemosensitivity test using 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide. Specimens were obtained from 76 surgical resected tumors, including 32 colon cancers, 24 stomach cancers and 16 lung cancers. Following enzymatic dissociation of scissors-minced tumors, viable cells were cultured in serum free medium (S-Clone SF-B) for 4 days with eight drug concentrations obtained by 2-fold dilution of drugs. Among 76 specimens tested, 48 specimens including 15 colon cancers, 18 stomach cancers and 15 lung cancers were successfully evaluated. For the purpose of judgement, 50% inhibitory concentration (IC50) was calculated in each case. Tumor specimen was regarded as sensitive to a given agent when the IC50 value was the same or smaller than the cut-off concentrations (1 microgram/ml for mitomycin C, 5 micrograms/ml for cisplatin, 2 micrograms/ml for adriamycin and 50 micrograms/ml for 5-fluorouracil), and was regarded as resistant when it was larger than these levels. In vitro vs in vivo drug sensitivity was successfully evaluated in 23 cases. The overall predicting accuracy rate was 78% (18/23), with one true positive, 5 false positive and 17 true negative cases. This test appeared to be useful to tailor effective agents for patients because of its relatively high successful and predictive rates.