To investigate the potential application of thermal therapy in the treatment of prostate cancer, the effects of supraphysiological temperatures (40-70 degrees C) for clinically relevant time periods (approximately 15 minutes) were experimentally studied on attached Dunning AT-1 rat prostate cancer cells using multiple assays. The membrane and reproductive machinery were the targets of injury selected for this study. In order to assess membrane injury, the leakage of calcein was measured dynamically, and the uptake of PI was measured postheating (1-3 hours). Clonogenicity was used as a measure of injury to the reproductive machinery 7 days post-injury after comparable thermal insults. Experimental results from all three assays show a broad trend of increasing injury with an increase in temperature and time of insult. Membrane injury, as measured by the fluorescent dye assays, does not correlate with clonogenic survival for many of the thermal histories investigated. In particular, the calcein assay at temperatures of < or = 40 degrees C led to measurable injury accumulation (dye leakage), which was considered sublethal, as shown by significant survival for comparable insult in the clonogenic assay. Additionally, the PI uptake assay used to measure injury post-thermal insult shows that membrane injury continues to accumulate after thermal insult at temperatures > or = 50 degrees C and may not always correlate with clonogenicity at hyperthermic temperatures such as 45 degrees C. Last, although the clonogenic assay yields the most accurate cell survival data, it is difficult to acquire these data at temperatures > or = 50 degrees C because the thermal transients in the experimental setup are significant as compared to the time scale of the experiment. To improve prediction and understanding of thermal injury in this prostate cancer cell line, a first-order rate process model of injury accumulation (the Arrhenius model) was fit to the experimental results. The activation energy (E) obtained using the Arrhenius model for an injury criterion of 30 percent for all three assays revealed that the mechanism of thermal injury measured is likely different for each of the three assays: clonogenics (526.39 kJ/mole), PI (244.8 kJ/mole), and calcein (81.33 kJ/mole). Moreover, the sensitivity of the rate of injury accumulation (d omega/dt) to temperature was highest for the clonogenic assay, lowest for calcein leakage, and intermediate for PI uptake, indicating the strong influence of E value on d omega/dt. Since the clonogenic assay is linked to the ultimate survival of the cell and accounts for all lethal mechanisms of cellular injury, the E and A values obtained from clonogenic study are the best values to apply to predict thermal injury in cells. For higher temperatures (> or = 50 degrees C) indicative of thermal therapies, the results of PI uptake can be used as a conservative estimate of cell death (underprediction). This is useful until better experimental protocols are available to account for thermal transients at high temperature to assess clonogenic ability. These results provide further insights into the mechanisms of thermal injury in single cell systems and may be useful for designing optimal protocols for clinical thermal therapy.