Background: Diacetyl-bis(N4-methylthiosemicarbazone), labeled with 64Cu (64Cu-ATSM) has been suggested as apromising tracer for imaging hypoxia. However, various controversial studies highlighted potential pitfalls that maydisable its use as a selective hypoxic marker. They also highlighted that the results may be tumor locationdependent. Here, we first analyzed uptake of Cu-ATSM and its less lipophilic counterpart Cu-Cl2 in the tumor overtime in an orthotopic glioblastoma model. An in vitro study was also conducted to investigate the hypoxiadependent copper uptake in tumor cells. We then further performed a comprehensive ex vivo study to compare64Cu uptake to hypoxic markers, specific cellular reactions, and also transporter expression.Methods: μPET was performed 14 days (18F-FMISO), 15 days (64Cu-ATSM and 64Cu-Cl2), and 16 days (64Cu-ATSM and 64Cu-Cl2) after C6 cell inoculation. Thereafter, the brains were withdrawn for further autoradiography andimmunohistochemistry. C6 cells were also grown in hypoxic workstation to analyze cellular uptake of Cu complexesin different oxygen levels.Results: In vivo results showed that Cu-ASTM and Cu-Cl2 accumulated in hypoxic areas of the tumors. Cu-ATSMalso stained, to a lesser extent, non-hypoxic regions, such as regions of astrogliosis, with high expression of coppertransporters and in particular DMT-1 and CTR1, and also characterized by the expression of elevated astrogliosis. Invitro results show that 64Cu-ATSM showed an increase in the uptake only in severe hypoxia at 0.5 and 0.2% ofoxygen while for 64Cu-Cl2, the cell retention was significantly increased at 5% and 1% of oxygen with no significantrise at lower oxygen percentages.Conclusion: In the present study, we show that Cu-complexes undoubtedly accumulate in hypoxic areas of thetumors. This uptake may be the reflection of a direct dependency to a redox metabolism and also a reflection ofhypoxic-induced overexpression of transporters. We also show that Cu-ATSM also stained non-hypoxic regions suchas astrogliosis.