Iron dyshomeostasis and senescence-like states in neurons: an evidence-graded synthesis of mechanisms, controversies and therapeutic opportunities in Alzheimer's disease

Alzheimer's disease (AD) is a progressive neurodegenerative disorder defined by amyloid-β (Aβ) plaques, tau tangles and synaptic failure. Brain iron accumulates in AD-vulnerable regions and higher iron load tracks with faster cognitive decline, yet whether iron dyshomeostasis initiates, amplifies or merely accompanies proteinopathy remains unresolved. In parallel, post-mitotic neurons can acquire senescence-like states—persistent DNA-damage signaling, mitochondrial dysfunction, lysosomal remodeling, a senescence-associated secretory phenotype (SASP)-like secretome and synaptic loss—that have been described in AD brain tissue. This review synthesizes and explicitly grades the evidence connecting the two. We map neuronal iron uptake, storage, export and mitochondrial handling, emphasizing compartment-specific regulation of net iron flux, and ask how labile iron, through Fenton chemistry, lipid peroxidation, ferroptosis and inflammatory signaling, could convert a reversible stress response into a durable senescence-like state. Evidence is sorted into five evidence-context categories—human AD tissue or cohort data; AD-relevant neuronal models with perturbation–rescue designs; glial evidence; non-AD CNS injury models; indirect analogy from non-neural tissues—which describe biological context, not study quality. The widely cited sequence “iron overload → neuronal senescence → AD progression” therefore remains a working model supported by converging but largely indirect evidence: human data are cross-sectional and confounded by proteinopathy and inflammation, directionality has been shown in only a few neuronal systems, and the only randomized double-blind trial of a brain-permeant chelator in AD to date (deferiprone) reduced hippocampal iron yet was associated with faster cognitive decline on its cognitive endpoint; an earlier randomized desferrioxamine trial in 48 participants was single-blind and lacked biomarker-based diagnosis, and does not establish iron-specific efficacy. Direct multi-feature evidence is still confined to neuronal cell lines and mixed cultures in AD-relevant systems, with a single non-AD in vivo example, and persistence of an iron-induced state after iron withdrawal has not been demonstrated. We define operational, orthogonal criteria for neuronal senescence-like states in post-mitotic cells, distinguish them from oxidative stress, DNA-damage responses, aberrant cell-cycle re-entry, apoptosis and ferroptosis, propose falsifiable predictions, and specify which compartment-, stage- and cell-type-resolved experiments are required before iron-directed therapies can be rationally designed.