Abstract: Neurodegenerative diseases are a growing burden on healthcare systems. Patients with Alzheimer’s or Parkinson’s diseases (AD or PD) are desperately waiting for innovative solutions that are slow to come, despite several decades of research worldwide. In 2021 and again in 2023, two monoclonal antibodies, aducanumab and lecanemab, have been approved by the U.S. Food and Drug Administration, and a third, donanemab, is currently under review.  However, these treatments have very limited efficacy on cognitive functions and are accompanied by major side effects: amyloidrelated imaging abnormalities, microhemorrhages, and accelerated brain volume loss (Høilund-Carlsen et al., 2024). The paradigm underlying these treatments is the amyloid cascade leading to the accumulation of amyloid plaques in the brain of patients (Fantini et al., 2020). Even if this strategy remains favored by most pharmaceutical companies, it suffers from major contradictions that could eventually lead to its abandonment. If the involvement of the Alzheimer’s amyloid-β (Aβ) protein remains undeniable, the debate has been open for several years to unequivocally identify the neurotoxic form of this protein. In fact, the presence of amyloid plaques is not systematically associated with AD, and we know of at least one deletion in the Aβ protein (Osaka mutation noted E22del or E22Δ) associated with the disease but without amyloid plaque (Fantini et al., 2020).