Early diagnosis of Alzheimer's disease (AD) has become a major imperative in recent years, especially since the FDA's approval of treatments such as Aducanumab and more recently Lecanemab, two anti-amyloid therapies. One of the main current research directions aims to facilitate the identification of AD patients at preclinical stages, before the appearance of initial symptoms of cognitive dysfunction and neurodegeneration, in order to offer these new therapies in the future. Three main research areas are being explored: the development of new neuropsychological tests, the refinement of tracers and imaging techniques to detect tau/amyloid protein deposits, and the use of biofluids to detect specific disease biomarkers. Since AD is the leading cause of dementia worldwide, affecting more than 50 million people, the use of biofluids like blood plasma appears to be the preferred candidate for widespread application. The β-amyloid (Aβ) peptide and tau protein can now be quantified in plasma samples using ultra-sensitive techniques such as Single Molecule Array (Simoa). However, these proteins circulate not only in plasma in soluble form but are also present in extracellular vesicles derived from neurons (NDEVs) secreted by neurons and crossing the blood-brain barrier into plasma. The aim of this thesis work is (i) to identify the most promising candidate for AD diagnosis in plasma, (ii) to assess the role of NDEVs as an AD biomarker, notably through an in vitro approach studying vesicles, and (iii) to explore their possible impact on the quantification of soluble plasma biomarkers. A cohort of AD patients and control participants without cognitive disorders was recruited, some of whom had access to PET-Amyloid and PET-Tau imaging techniques. Quantification of amyloid peptides 42–40, total Tau protein, and pTau 181 in plasma and NDEVs was performed via SIMOA, and the DELFIA ELISA immunoenzymatic assay was used for the relative quantification of circulating NDEVs in plasma. The main conclusions of this thesis work indicate that the Aβ42/40 ratio proves to be an effective indicator for AD screening, reducing the probability of AD to 5.2% at age 60 (vs 15.8% without the test) and to 12.5% at age 80 (vs 32.6% without the test) in case of a negative test result, and plasma pTau 181 performed as well, and their association reduced probability of AD to 2% at age 60 and 4% at age 80. An in vitro study conducted on neurons demonstrated that the content of extracellular vesicles secreted reflects the cellular AD alterations occurring. The presence of NDEVs in plasma was identified as a significant confounding factor in plasma biomarker measurements, although their content does not seem to be a relevant biomarker for screening individuals at risk of developing the disease, at least according to quantifications performed by SIMOA.