A new study from the University of Southern California adds a structural dimension to one of medicine's long-standing puzzles: why depression so often precedes dementia. Researchers examining brain scans from over two thousand older adults found that those with depression showed measurable shrinkage in a specific hippocampal region called CA23DG — a composite area central to memory, learning, and spatial navigation. Notably, this reduction held even after accounting for the biological signatures of Alzheimer's disease, suggesting depression may carve its own separate path through the brain's m
Depression linked to shrinkage in key brain memory region, study finds
Depression may damage memory through its own distinct biological pathway
Mark
So we're looking at a physical difference in the brains of people with depression. What exactly is the hippocampus doing that makes it so central to this story?
Mimi
The hippocampus is the brain's memory hub—it's where you form new memories, consolidate them, and navigate space. It's not a single structure; it's made up of several subregions, each with slightly different jobs. This study zoomed in on one composite area called CA23DG and found it was noticeably smaller in people with depression.
Mark
And this shrinkage—is it the depression causing it, or something else?
Mimi
That's the honest answer: we don't know yet. The study shows the two things occur together, but it can't prove one causes the other. What complicates it further is that people taking antidepressants also showed this shrinkage, which raises the question of whether the medication itself plays a role, or whether people with worse depression are simply more likely to be on medication.
Mark
So the medication might be protective, or it might be harmful, or it might have nothing to do with it.
Mimi
Exactly. The researchers found that people who'd lived with depression longer—even if their symptoms were currently mild—showed more shrinkage than people with severe depression who were newly diagnosed. That suggests duration matters. But again, those longer-suffering patients might also have been on medication longer, so we're back to the same uncertainty.
Mark
What makes this different from what we already knew about depression and Alzheimer's?
Mimi
The key finding is that this hippocampal shrinkage appears to happen through a separate mechanism from the classic Alzheimer's pathology. When researchers looked at the biological markers of Alzheimer's—amyloid and tau proteins, genetic risk factors—those didn't explain the CA23DG shrinkage. It suggests depression might damage memory through its own distinct biological pathway.
Mark
What happens next?
Mimi
They need longitudinal studies—following people over years, watching their brains before depression, during it, and after treatment. Only then can we separate what depression itself does from what the medications do. Until then, the researchers are clear: people shouldn't change their treatment based on this.
Le Pouls
- A specific hippocampal subregion — CA23DG — appears physically smaller in people with depression, offering the first concrete structural candidate linking depression to dementia risk.
- The shrinkage persists even after controlling for amyloid, tau, and APOE ε4, meaning depression may be damaging memory circuitry through a biological pathway entirely its own.
- A complicating signal emerged: antidepressant users showed volume loss in an additional hippocampal region, raising the unsettling question of whether treatment itself plays a role.
- Duration of depression, more than its severity, correlated most strongly with the shrinkage — suggesting that years lived with the condition may quietly accumulate structural cost.
- Researchers are urging caution: this is correlation, not causation, and they explicitly warn patients not to alter medications based on these findings.
- The path forward requires longitudinal studies tracking brains before, during, and after depression and treatment — a long road before clarity arrives.
A new study from the University of Southern California adds a structural dimension to one of medicine's long-standing puzzles: why depression so often precedes dementia. Researchers examining brain scans from over two thousand older adults found that those with depression showed measurable shrinkage in a specific hippocampal region called CA23DG — a composite area central to memory, learning, and spatial navigation. Notably, this reduction held even after accounting for the biological signatures of Alzheimer's disease, suggesting depression may carve its own separate path through the brain's memory architecture. The finding is a clue, not a verdict, and the researchers are careful to say so.
For years, clinicians have observed that depression raises the risk of dementia and Alzheimer's disease, yet the biological bridge between the two has remained elusive. A new study published in Translational Psychiatry brings that bridge into partial view.
Researchers at the University of Southern California analyzed MRI and PET scans from more than 2,000 cognitively healthy adults aged 50 to 90, roughly a third of whom had experienced depression. They found that a specific hippocampal subregion — CA23DG, a composite of three smaller fields involved in memory, learning, and spatial navigation — was noticeably smaller in people with depression. The association held after controlling for age, sex, BMI, and physical activity.
What made the finding especially striking was its independence from Alzheimer's biology. Even after accounting for amyloid and tau proteins and the APOE ε4 gene variant, the CA23DG shrinkage remained. Depression, the data suggest, may damage the hippocampus through its own distinct pathway.
But the study also surfaced complications. Participants taking antidepressants showed volume loss in an additional hippocampal region, CA1 — a counterintuitive pattern that raises questions about whether medications contribute to the changes, or whether more severely affected patients simply receive more treatment. The researchers suspect the latter, but the data cannot confirm it. Similarly, longer depression duration correlated more strongly with shrinkage than current symptom severity — though here too, medication history may be the hidden variable.
Lead author Danielle Luu was deliberate in framing the limits of the work: the study shows correlation, not causation, and she explicitly urged patients not to change or discontinue medications based on these results. What is needed next, she said, are longitudinal studies that follow individuals before depression develops, through treatment, and into recovery — tracing the brain's memory center across the full arc of the illness.
For years, doctors have noticed that people with depression face a higher risk of developing dementia and Alzheimer's disease, but the biological mechanism linking these conditions has remained largely mysterious. A new study published in Translational Psychiatry offers a concrete clue: in people with depression, a specific region of the brain's memory center appears to physically shrink.
Researchers at the University of Southern California examined MRI and PET scans from more than 2,000 cognitively healthy adults between 50 and 90 years old. About one-third of the participants had experienced depression. What they found was striking. A particular area of the hippocampus—the brain region essential for forming and storing memories—showed noticeably reduced volume in people with depression compared to those without it. This area, called CA23DG, is actually a composite of three smaller hippocampal subfields: CA2, CA3, and the dentate gyrus. Each of these regions plays a distinct role in different types of memory, learning, and spatial navigation.
Neuroscientist Danielle Luu, the study's lead author, emphasized the significance of pinpointing this specific region. "Depression has long been associated with an increased risk of developing Alzheimer's disease, but we still do not fully understand the biological connection between the two," Luu explained. "By looking closely at the individual parts of the hippocampus, we identified a specific area that may be particularly sensitive to depression in older adults." The researchers controlled for numerous variables—age, sex, body mass index, and physical activity levels—and the CA23DG shrinkage remained statistically significant.
What makes this finding particularly intriguing is that the reduced volume in CA23DG appeared independent of the biological hallmarks typically associated with Alzheimer's disease. When researchers factored in known Alzheimer's risk factors—including levels of amyloid and tau proteins in the brain and the presence of the APOE ε4 gene variant—the association between depression and CA23DG shrinkage remained largely unchanged. This suggests that depression may damage the hippocampus through a separate biological pathway, distinct from the mechanisms that drive Alzheimer's pathology.
However, the study also revealed complications that muddy the interpretation. When the researchers accounted for antidepressant use, they found that people taking these medications showed reduced volume not only in CA23DG but also in another hippocampal region called CA1. This counterintuitive finding raises a troubling question: Are the medications themselves contributing to hippocampal shrinkage, or are people with more severe depression simply more likely to be prescribed antidepressants? The researchers suspect the latter, but they cannot be certain from this data alone.
Another unexpected pattern emerged around the duration of depression. People who had experienced depression for longer periods—even if their current symptoms were mild and controlled—showed a stronger association with CA23DG shrinkage than people with more severe depression who were newly diagnosed. This suggests that the length of time someone has lived with depression may matter more than how acutely symptomatic they are at any given moment. Yet here too, the researchers acknowledge uncertainty: the effect could be driven by medication use rather than depression duration itself.
Luu and her colleagues were careful to emphasize what their findings do not show. The study demonstrates correlation, not causation. They cannot prove that depression or antidepressants are actually causing the hippocampal shrinkage. For this reason, they explicitly urged people with depression not to alter their medication regimens based on these results. "Long-term studies that follow people before and after treatment will be essential for understanding these relationships," Luu stated. "Our results should not be interpreted as a reason for anyone to change or discontinue prescribed medication." The next phase of research will need to track individuals over time, observing their brains before depression develops, during treatment, and after recovery, to untangle what is driving these changes in the brain's memory center.
Citations marquantes
Depression has long been associated with an increased risk of developing Alzheimer's disease, but we still do not fully understand the biological connection between the two.— Danielle Luu, neuroscientist, University of Southern California
Our results should not be interpreted as a reason for anyone to change or discontinue prescribed medication.— Danielle Luu
Envie de l'histoire complète ?
Lire l'original sur ScienceAlert ↗
Health & Medicine