Carotenoids have been suggested to have either anti- or pro-oxidative effects in several cancer cells, and those effects can trigger an unbalanced reactive oxygen species (ROS) production resulting in an apoptotic response. Our study aimed to evaluate the effect of the well-known carotenoid 3, 3′-dihydroxy-β, β’-carotene-4, 4-dione (astaxanthin, AXT) on glioblastoma multiforme (GBM) cells, especially as a pretreatment of tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL), that was previously shown to increase ROS and to induce apoptosis in cancer cells. We found that AXT by itself did not trigger apoptosis in four investigated GBM cell lines upon a 24 h treatment at various concentrations from 2.5 to 50 µM. However, in U251-MG and T98-MG GBM cells, pretreatment of 2.5 to 10 µM AXT sensitized cells to TRAIL treatment in a statistically significant manner (p < 0.05) while it did not affect CRT-MG and U87-MG GBM cells. We further compared AXT-sensitive U251-MG and -insensitive CRT-MG response to AXT and showed that 5 µM AXT treatment had a beneficial effect on both cell lines, as it enhanced mitochondrial potential and TRAIL treatment had the opposite effect, as it decreased mitochondrial potential. Interestingly, in U251-MG, 5 µM AXT pretreatment to TRAIL-treated cells mitochondrial potential further decreased compared to TRAIL alone cells. In addition, while 25 and 50 ng/mL TRAIL treatment increased ROS for both cell lines, pretreatment of 5 µM AXT induced a significant ROS decrease in CRT-MG (p < 0.05) while less effective in U251-MG. We found that in U251-MG, superoxide dismutase (SOD) 2 expression and enzymatic activity were lower compared to CRT-MG and that overexpression of SOD2 in U251-MG abolished AXT sensitization to TRAIL treatment. Taken together, these results suggest that while AXT acts as an ROS scavenger in GBM cell lines, it also has some role in decreasing mitochondrial potential together with TRAIL in a pathway that can be inhibited by SOD2.