Before a brain shrinks, it dims. That is the sequence the imaging suggests: a fall in cerebral glucose metabolism visible on FDG-PET often shows up before structural change is obvious on MRI, which is to say the lights go down in the room before the walls come in. The organelle responsible for keeping those lights on is the mitochondrion, and a review published on 24 March 2026 in Frontiers in Aging Neuroscience argues that its failure in Alzheimer's disease does not look the same in everyone. It looks like two different failures, sorted by sex.
The paper is by Afzal Misrani, Conelius Ngwa and Fudong Liu of the Department of Neurology at McGovern Medical School, The University of Texas Health Science Center at Houston. It is a review: no new patients were scanned, no new tissue was cut, nothing was randomised. What the authors did was assemble published work on oxidative stress, energy metabolism and the regulatory pathways around them, and ask whether the sexes diverge. Their summary of the clinical picture is blunt. Women show more rapid cognitive decline and accelerated brain atrophy during mild cognitive impairment and early dementia. Men more often carry cardiovascular comorbidities, show earlier mitochondrial dysfunction, and present with more neuropsychiatric symptoms, including agitation and aggression, which the review links to higher mortality. Alzheimer's, they write, accounts for roughly 70 percent of dementia cases and still has no disease-modifying treatment.
The grammar underneath is molecular, and the review names its nouns. Mitochondrial trouble in Alzheimer's can be read off a short list of markers: more Drp1, the protein that drives mitochondrial fission; less PGC-1alpha, the master regulator of mitochondrial biogenesis; and rising oxidative stress products including malondialdehyde, nitrotyrosine and protein carbonyls. The authors argue that the sex split in these markers traces back to two sources, sex hormones and sex chromosomes, and that sex chromosomes are emerging as biological determinants in their own right rather than a footnote to estrogen.
APOE4, the strongest common genetic risk factor for late-onset Alzheimer's, gets its own section, and here the evidence is explicitly laboratory evidence. In mouse neuroblastoma (N2a) cells engineered to express ApoE4, mitochondrial respiration and ATP production were diminished, a reduced capacity to meet rising energy demand. In ApoE4 astrocytes, the review reports decreased mitochondrial fission and impaired Parkin-mediated mitophagy, meaning the cell's quality-control system for clearing damaged mitochondria works less well. These are cells in dishes and animals in cages. They are not people, and the review does not claim otherwise.
It is worth following one citation to the end, because it shows how thin a confident sentence can be. The claim that women atrophy faster than men during mild cognitive impairment is referenced in the review to a longitudinal cohort study of older men and women published in Psychological Medicine in 2015, a study whose primary question was depression. It reported that a history of depression was associated with 0.2 percent faster annual hippocampal volume loss in women, with a confidence interval running from 0.01 to 0.36 percent. That is a real, carefully measured, observational association. It is not a measurement of mitochondria, and it is not proof that being a woman causes faster atrophy in Alzheimer's.
So what does a family do with this? Today, nothing. There is no sex-specific mitochondrial test to ask for, no supplement that this review endorses, no dose, no protocol. Its value is upstream, in the design of the next round of studies: if the energy failure really runs on different timetables in men and women, then trials that pool the sexes may be averaging away the very effect they are hunting. For anyone living with a diagnosis now, the practical content of this paper is a question for the clinic, not an instruction from a newspaper.