We evaluated conventional pulse exposure versus continuous exposure models of 5-fluorouracil (5-FU) radiosensitization in HT-29 (human colon adenocarcinoma) and DU-145 (human prostate cancer adenocarcinoma) cell lines. Cell survival following treatment with drug and/or radiation was determined by colony formation assays. Radiation was delivered either by itself, approximately midway through a 1-hr exposure to 5-FU (10 micrograms/ml), or at various times following initiation of exposure to 5-FU (0.5 microgram/ml) present throughout the entire period of incubation. Drug concentrations were selected to approximate those achieved in vivo in humans. HT-29 cells showed a plating efficiency of 87% and similar cytotoxicity (survival reduced to 0.57-0.71) for all 5-FU conditions. The Do's of the radiation survival curves were not different for 1 hr of 5-FU exposure versus radiation alone. However, continuous exposure conditions demonstrated statistically significantly different Do's from radiation alone and pulse 5-FU exposure. DU-145 cells displayed a plating efficiency of 17% and cytotoxicities of 0.10-0.91 for the 5-FU conditions. DU-145 cells showed different radiation 5-FU interactions: 5-FU produced statistically significant changes in Do well as the differences between cell lines insofar as their radiosensitization by 5-FU underscore the caution required in extrapolating these radiobiologic models to the clinical setting.