Modeling Alzheimer’s disease through the integration of exposome, inflammasome, and connectome

Over a century ago, the first clinical and neuropathological insights into major neurodegenerative diseases began to emerge: the description of Alzheimer’s disease (AD) by Alois Alzheimer in 1906, frontotemporal dementia by Arnold Pick in the same years, and Lewy bodies by Friedrich Lewy in 1912. These foundational studies laid the groundwork for the classification of what we now recognize as distinct neurodegenerative entities (Allali, 2024). Since then, decades of research have progressively deepened our understanding of the underlying pathophysiological mechanisms, drawing from in vitro models, animal studies, and human clinical observations. In recent years, these efforts have culminated in the development of the first pharmacological interventions targeting core pathological processes in AD. The approval of two anti-amyloid-β (Aβ) monoclonal antibodies, lecanemab and donanemab, in the United States, and more recently the approval of lecanemab in Europe, represents a milestone in translating biological discoveries into clinical applications. However, the clinical benefits observed thus far remain modest, and several critical challenges persist, notably the trade-off between limited efficacy and the risk of adverse effects. Moreover, for other conditions, such as frontotemporal dementia, the second most common cause of early-onset dementia, no effective treatments are currently available. This disparity, along with increasing scrutiny of the real-world impact of current AD therapies, underscores the need to integrate mechanistic paradigms with novel strategies that can help define new therapeutic priorities. In this article, we pursue a twofold aim: (i) highlighting open questions in AD research and (ii) proposing that the interplay among different factors (exposome, inflammasome, and connectome) offers an integrated framework to reconcile inconsistencies in current AD models.