We evaluate the anti-cancer capacity of plasma-treated PBS (pPBS), by measuring the concentrations of NO2- and H2O2 in pPBS, treated with a plasma jet, for different values of gas flow rate, gap and plasma treatment time, as well as the effect of pPBS on cancer cell cytotoxicity, for three different glioblastoma cancer cell lines, at exactly the same plasma treatment conditions. Our experiments reveal that pPBS is cytotoxic for all conditions investigated. A small variation in gap between plasma jet and liquid surface (10 mm vs 15 mm) significantly affects the chemical composition of pPBS and its anti-cancer capacity, attributed to the occurrence of discharges onto the liquid. By correlating the effect of gap, gas flow rate and plasma treatment time on the chemical composition and anti-cancer capacity of pPBS, we may conclude that H2O2 is a more important species for the anti-cancer capacity of pPBS than NO2-. We also used a 0D model, developed for plasma-liquid interactions, to elucidate the most important mechanisms for the generation of H2O2 and NO2-. Finally, we found that pPBS might be more suitable for practical applications in a clinical setting than (commonly used) plasma-activated media (PAM), because of its higher stability.