[en] Full text of publication follows. Aim: Glioblastoma (GBM) is the most frequent cancer of the nervous system and therapies currently used cannot treat definitively this disease. By the way of NucSan (Nuclear for health) program which supports research development about the use of radio pharmaceutics in oncology, the aim of the proposed work is to provide evidences that internal radiotherapy through lipid nanocapsules loaded with Rhenium-188 (LNC¹⁸⁸Re-SSS) is an alternative therapeutic strategy for GBM that can be translated to human medicine. Previous works have shown a remarkable efficiency of this tool among syngeneic rats linked with local and peripheral recruitments of the immune system's effectors. In this context, two animal models have been chosen to validate the feasibility of this new innovative therapy design (LNC¹⁸⁸Re-SSS stereotactic injection). Materials and methods: the syngeneic six weeks old C57BL/6J female mice were treated 6 or 12 days after stereotactic GL261 cells implantation, by a single injection of increasing activities of LNC¹⁸⁸Re-SSS (0,925; 1,85 and 2,7 MBq/5μl). MRI was used to follow tumor progression to determine the mass volume through the selection of regions of interest. The increased median survival time (IMST) was also assessed for treated mice versus control mice (stereotactic injection of saline solution). For long time survival animals (3 times the median survival time), they were rechallenged through the same procedure in the other hemisphere. The brachy-cephalic dog bearing spontaneous tumor will lead to additional evidences to specifically highlight the potential of this innovative technology for GBM treatment. In order to validate procedures of intracerebral injections, a stereotactic head frame specially designed for dog has been conceived which allow both images acquisition (MRI-SPECT and PET) and the achievement of biopsies. Results/Perspectives: as previously observed on rat models, the preliminary data show interesting results on treatment efficiency in GL261 mice model with the long time survival animals assessed by the absence of tumor progression (IMST: 41%). Thanks to the dog head frame accuracy, brain biopsy targeted by pre/post-surgery MRI has been validated. These two preclinical models will give us additional evidences for a translational approach with the help of CGO-2012 (IRAD) and ANR-12-EMMA-003301 Radiohead projects. They aim at collecting data on in vivo feasibility and efficiency of LNC¹⁸⁸Re-SSS for GBM treatment, especially in a spontaneous model linked with immune response. Finally, the objective of this work is to set up a phase I/II clinical trial. (authors)