Alzheimer’s Treatment Needs More Than Amyloid Drugs

TL;DR: Lecanemab and donanemab proved that modifying amyloid biology slows clinical decline. They also proved it does not stop it. A 2026 Science China Life Sciences review argues the next era of Alzheimer’s treatment will be biomarker-guided and combination-driven — tau, APOE, aging biology, immune state, and vascular health all in the picture.

Source: Science China Life Sciences (2026) | Jia et al.

Alzheimer’s disease is usually introduced through two famous lesions: amyloid plaques and tau tangles. The shorthand is useful and dangerous in equal measure. Useful, because it captures the two best-studied features of the disease. Dangerous, because it can make Alzheimer’s sound like a two-object cleanup job — remove the plaques, block the tangles, and cognition should return.

The drugs the field has spent two decades chasing finally arrived. Lecanemab and donanemab can slow the disease in carefully selected early patients. They have not restored memory, reversed dementia, or finished the problem. This review takes that gap seriously and asks what it implies for the next round of treatments.

Amyloid Antibodies Opened the Door, Then Showed Its Frame

Amyloid-beta is a small protein fragment that aggregates into plaques outside neurons in Alzheimer’s disease. Antibody drugs like lecanemab and donanemab mark amyloid-related pathology for immune clearance, and the trials demonstrated something the field had been chasing for twenty years: changing amyloid biology really does change the disease’s trajectory.

What it does not do is reverse cognitive decline. The trials showed slower deterioration, not recovery. That gap is not a footnote — it is the most important number in modern Alzheimer’s treatment, because it tells the field that amyloid is necessary to address but not sufficient to solve. Symptoms must depend on more than amyloid load alone.

The review uses that gap as its starting frame. The point is not to abandon amyloid. It is to stop expecting amyloid alone to carry the full treatment burden.

APP Metabolism, and Why Timing May Matter More Than Target

Upstream of amyloid sits amyloid precursor protein (APP), the larger molecule whose enzymatic cleavage can produce aggregation-prone Aβ species. Therapeutic logic offers several intervention points: reduce production, prevent aggregation, accelerate clearance, neutralize toxic species. Antibodies sit mostly at the clearance and neutralization end of that landscape.

What the review emphasizes is that amyloid biology does not run in isolation. By the time a patient has obvious cognitive symptoms, the disease has often expanded into tau pathology, synaptic loss, neuroinflammation, and vascular injury. A treatment that helps in the early window may have little to grab onto once those downstream processes are entrenched.

Timing, in other words, may matter as much as target. The same molecule may be a transformative drug in preclinical or early-symptomatic patients and a marginal one in late-stage disease.

Tau Tracks the Symptoms More Closely

Tau is a protein that normally stabilizes the internal skeleton of neurons. In Alzheimer’s, abnormal modifications — most importantly hyperphosphorylation, the addition of phosphate groups in the wrong places — change tau’s behavior and let it form intracellular neurofibrillary tangles.

The reason tau matters so much for the next phase of drug development is empirical, not theoretical. Tau pathology aligns with neuronal injury and cognitive symptoms more tightly than amyloid plaque burden. Two patients with similar plaque loads can have very different clinical trajectories — and tau status is often the variable that explains the difference.

That makes tau a parallel engine of the disease, not a downstream artifact. The review treats it as the natural combination partner for amyloid antibodies, especially in patients whose disease is already past the earliest amyloid-driven phase.

APOE and the Uneven Genetics of Risk

APOE codes for a lipid-transport and brain-repair protein, and the ε4 variant remains the most influential common genetic risk factor for late-onset Alzheimer’s. Carrying ε4 raises risk meaningfully — but it does not guarantee disease, and not all carriers experience the same trajectory.

Genotype matters for treatment choice as well as risk prediction. APOE status has shaped clinical conversations about amyloid-related imaging abnormalities (ARIA), the brain swelling and microbleed signals that can show up on MRI during antibody treatment. Patients carrying ε4 — especially homozygotes — face higher ARIA risk, which directly influences whether and how an amyloid antibody should be used.

The review also points beyond APOE. Risk biology is not uniform across ancestries, environments, or health histories, and trial design has been slow to reflect that. A future treatment strategy will likely have to integrate genotype, disease stage, amyloid and tau burden, vascular status, immune state, and patient priorities. A single all-purpose Alzheimer’s drug looks less realistic than a staged, biomarker-driven plan.

Aging Biology Sets the Brain’s Tolerance for Pathology

Alzheimer’s is age-related, but “age” in this context is not a number on a chart. It is a constellation of biological shifts that change the environment amyloid and tau act in. With age, protein-clearance systems slow. Microglia tilt toward an inflammatory phenotype. Mitochondria and cerebrovascular reserves degrade. Cellular stress responses lose flexibility.

That is why the same amyloid burden can mean dementia in one brain and cognitive resilience in another. Some people carry striking pathology and remain sharp; others develop symptoms with relatively modest plaque and tangle loads. The difference often lives in this background biology — what neurologists sometimes call brain or cognitive reserve, but which biologically reduces to how well the tissue environment buffers ongoing pathology.

The review’s broader argument follows naturally from this: Alzheimer’s treatment cannot stop at attacking lesions. It also has to consider the tissue environment that lets neurons and synapses keep functioning despite those lesions.

Systemic Health Is Not Optional

The brain is the body’s most metabolically demanding organ. It runs on blood flow, oxygen delivery, glucose regulation, immune balance, and waste clearance. When any of those systems strains, Alzheimer’s pathology gets harder to tolerate.

Vascular disease, diabetes, obesity, sleep disruption, chronic inflammation, infection burden, kidney disease, hearing loss, and untreated depression all show up in the review as background factors that influence whether brain pathology becomes a clinical syndrome. None of them are stand-ins for disease-modifying drugs. They shape the biological terrain in which those drugs operate.

The clinical implication is unromantic but important. A patient with amyloid pathology, uncontrolled hypertension, fragmented sleep, and elevated inflammation needs a different treatment package than a patient with the same amyloid burden but a clean systemic profile.

The Real Argument: Biomarker-Guided Combination Therapy

This is a review, not a trial. It does not measure a drug effect. Its contribution is the map. And the map points toward biomarker-guided combination therapy — using amyloid PET, tau PET, CSF and blood biomarkers, MRI, genetic data, and metabolic and inflammatory measures to decide which target gets attacked first, which combinations make biological sense, and when treatment risk outweighs likely benefit.

An early amyloid-positive patient with ε4 and well-controlled vascular disease needs one strategy. A tau-heavy symptomatic patient with poor systemic health needs a different one. The era of “is this person an Alzheimer’s patient — yes or no” is ending; the era of “which mechanisms are most active in this brain right now” is the one the next round of trials will have to operate inside.

The review’s strongest single contribution is the reframe. Amyloid antibodies opened an important door. The next room contains tau, genes, aging, immune biology, vascular health, and metabolism — and it is large enough that no single drug is going to clear it.