PURPOSE: This study aimed to determine the extent of paclitaxel-induced cytotoxicity and cell-cycle perturbations when used alone and in combination with radiation in human glioma cells. METHODS AND MATERIALS: The effect of paclitaxel alone on three human glioma cells lines--SF-126, U-87 MG, and U-251 MG--was assessed after 24, 48, 72, or 96 h treatment. For experiments in combination with radiation, cells were exposed to either a long (48-h) or short (8-h) duration of paclitaxel treatment prior to irradiation. Cell survival was determined by clonogenic assay. Cell cycle perturbations were assessed by using flow cytometry to measure the proportion of cells in G1, S, and G2/M phases. RESULTS: When cells were treated with paclitaxel alone for > or = 24 h, cytotoxicity increased up to a threshold dose, after which it plateaued. When treatment duration was < or = 24 h, cytotoxicity was appreciably greater in U-251 MG cells than in SF-126 and U-87 MG cells. After 24 h of paclitaxel treatment, cells in plateau phase growth had increased survival compared to cells in log phase growth. In contrast, after 8 h paclitaxel treatment, mitotic cells had reduced survival compared to cells from an asynchronous population. Cell-cycle perturbations were consistent with the presence of a mitotic block after paclitaxel treatment, although changes in other cell-cycle phase fractions varied among cell lines. For experiments in combination with radiation, cytotoxicity was increased when cells were irradiated after 48 h of paclitaxel treatment but not after 8 h of treatment. CONCLUSION: The duration of paclitaxel treatment and the location of cells in the cell cycle modify the degree of radiation cytotoxicity. The mechanisms of paclitaxel cytotoxicity are likely to be multifactorial because varying effects are seen in different cell lines. Furthermore, it is clear that simply increasing the number of cells in G2/M is insufficient in itself to increase the response of cells to radiation.