Illustration created with AI image tools.
A man in his fifties began losing his memory and was diagnosed with mild cognitive impairment due to Alzheimer's disease. He enrolled in a clinical trial of aducanumab, an antibody that removes amyloid plaque, and received 30 infusions over about four and a half years. He died roughly four years after his last dose, more than a decade after his first symptoms, and he donated his brain to research. Pathologists at the University of Pennsylvania examined it, and their findings were published in JAMA this month.
What the Penn Autopsy Found
The paper, "Clinicopathologic Evaluation of Amyloid Clearance in Alzheimer Disease," comes from the Perelman School of Medicine. The first author is Christopher A. Brown, MD, PhD, and the co-senior authors are David Wolk, MD, who directs the Penn Memory Center, and Edward B. Lee, MD, PhD. The team found that the brain had cleared amyloid unevenly. In the gyri, the ridges of the cortex, plaque was largely gone. In the neighboring sulci, the grooves between those ridges, substantial amyloid remained.
Tau followed the plaque. The regions that had been cleared contained very little tau. The regions that kept their amyloid had much more tau, along with more inflammation and more tissue degeneration. The pattern also matched brain scans taken while the patient was alive, which tells us the autopsy was confirming something the imaging had already shown.
Why One Patient Still Counts
A single case cannot establish a treatment effect, and the authors say so. It cannot tell us how much amyloid has to come out for the protection to last, and it cannot prove that clearing amyloid caused the drop in tau. Trials of these drugs usually run about 18 months, which may be too short to see long-term protection of this kind.
Even so, this brain is an unusually clean comparison. The gyri and the sulci belong to the same person, with the same genes, the same blood pressure history, the same decades of sleep and diet. When two neighboring pieces of cortex differ mainly in how much amyloid was removed, and one of them has far less tau and inflammation, that is hard evidence to wave away. In my interpretation, this is some of the clearest human data we have that anti-amyloid therapy can limit the spread of tau in tissue, and the authors describe it in similar terms.
Where Plaque Removal Stops Helping
Tau tangles track symptoms more closely than amyloid does, and this patient's disease still progressed over the decade he was followed. The cleared regions had little tau, but the disease had already been running for years before treatment began, and the neurons lost earlier were not coming back. This is the practical problem with treating Alzheimer's at the stage of a clinical diagnosis. By the time a patient sits in my office with memory complaints, a good part of the damage is structural.
The open question, which the Penn authors also raise, is whether treating earlier would change the outcome, before tau tangles have formed. Ongoing prevention trials are testing that. Until they report, the reasonable position is that amyloid removal looks biologically real, and that timing is probably what determines whether it matters to the patient.
What This Means for You
Timing is something a person can act on now. Plasma p-tau217 and amyloid PET can detect Alzheimer's pathology years before a person fails a memory test, and a patient who knows his or her biology at 55 has options that a patient diagnosed at 70 does not. A precision workup also looks past amyloid and tau to the factors that decide how well a brain tolerates pathology: vascular health, insulin resistance, sleep, inflammation, and the other drivers of cognitive reserve. That is the logic behind our Intensive Brain Health Program, which combines advanced MRI, biomarkers, and cognitive testing so that the findings can be acted on while the neurons are still there to protect. I wrote about why I practice this way in Why I Became a Neurologist, and Why I Practice Differently.
If you have a family history of Alzheimer's, or you are a caregiver watching a parent decline, the question to ask your doctor is not whether a drug exists. Ask what your biomarkers show right now, and what can be done about them while there is still time.
In the Neuroeconomy, cognition is the asset that everything else depends on. A brain that has lost its neurons cannot be repaired by a good drug, but a brain that is measured early can be protected, and that is where the value is.
This article was drafted with the assistance of AI writing tools, then reviewed, edited, and approved by Dr. Sean C. Orr, M.D., who holds full editorial responsibility for its content.
Source: Brown CA, Wolk DA, Lee EB, et al. Clinicopathologic Evaluation of Amyloid Clearance in Alzheimer Disease. JAMA. 2026. doi:10.1001/jama.2026.13058. Penn Medicine news release, October 15, 2026.