Alzheimer's disease (AD) is the most common form of dementia in the elderly. AD is characterized by the deposition of two aggregated proteins: Amyloid beta (Aß) and hyperphosphorylated tau. Accumulations of these proteins are thought to be the signature of a pathogenic process causing the loss of neurons and synaptic connections associated with severe cognitive impairments in AD patients. Other important factors are involved in the neurodegeneration occurring in AD, such as neuroinflammation. In the present thesis work, we are investigating the effects of passive anti-Aß immunization. Passive anti-Aß immunization consists in the delivery of antibodies targeting Aß, with the aim to prevent its accumulation and promote the removal of pathological amyloid species. Although the treatment has been shown to clear amyloid pathology in the brain of AD patients, it has so far failed to prevent or slow down cognitive decline. Hence, it is important to determine the effects of anti-Aß immunization when administered either before or during Aß deposition, and explore how the treatment affects the Alzheimerâs manifestations in mouse models combining Aß and tau pathologies. In order to continuously deliver antibodies in AD mouse models, we used encapsulated cell technology (ECT) to implant myoblasts genetically engineered and secrete the recombinant mAb11 mouse IgG2a antibody directed against aggregated forms of Aß. Using this approach, we investigated the following three topics of research: 1) We examined the effect of a preventive administration of anti-Aß antibodies in a slowly progressing mouse model with both Aß and tau (P301L) pathologies. We demonstrated that delivery of anti-Aß antibodies before the onset of the disease dramatically reduces Aß pathology. We also found that the treatment decreases hyperphosphorylated tau accumulation and recruits microglial around Aß plaques. 2) We next assessed the effects of passive anti-Aß immunization in another AD mouse model overexpressing both Aß and wild-type (WT) tau. In this second study, we administered the anti-Aß antibodies when both Aß and tau pathologies were already established. In the hippocampal formation, we showed that WT tau overexpression might affect the microglial response to ongoing pathology by reducing the tight interaction of these cells with the amyloid plaques. This effect, however, is counteracted by passive anti-Aß immunization, which was found to redirect microglia towards Aß deposits. In addition, passive anti-Aß immunization decreases tau spreading throughout the hippocampus, but fails to prevent neither tau phosphorylation, nor the hippocampal degeneration induced by local tau overexpression. 3) In parallel, we developed a human myoblast cell line secreting anti-Aß antibodies for translation of our ECT towards human application. Using C2C12 mouse myoblasts, our flat-sheet device allowed for long-term continuous antibody delivery in the plasma of subcutaneously implanted mice. We were able to generate antibody-secreting human cells using the immortalized C25 cell It was possible to differentiate these cells into myotubes and adapt them to long-term survival inside the capsules maintained in vitro. However, when implanted in the subcutaneous tissue of immunocompromised mice, myoblasts survived in the ECT device as a thin layer of cells and were not able to secrete recombinant antibodies in sufficient amounts to be considered for therapeutic purpose.