Amyloid-Beta as an Alzheimer's Biomarker
Peptide Biomarkers
AUC 0.94-0.97
Plasma p-tau217/amyloid-beta ratio achieves near-perfect diagnostic accuracy for Alzheimer's amyloid pathology, rivaling invasive CSF testing.
Palmqvist et al., Alzheimer's & Dementia, 2025
Palmqvist et al., Alzheimer's & Dementia, 2025
View as imageAmyloid-beta (A-beta) is a 36-43 amino acid peptide fragment that sits at the center of Alzheimer's disease diagnosis, research, and drug development. For decades, measuring this peptide in cerebrospinal fluid (CSF) required a lumbar puncture. Amyloid PET imaging could visualize plaques in living brains but cost thousands of dollars per scan. In 2025, blood-based biomarker tests entered clinical guidelines for the first time, making amyloid-beta measurement accessible at a scale previously impossible. This article examines what amyloid-beta is, how it becomes a biomarker, the diagnostic technologies that measure it, the therapeutic implications of anti-amyloid drugs, and what this peptide tells us about the biology of neurodegeneration. For the broader landscape of peptide biomarkers in clinical medicine, see Chromogranin A: The Neuroendocrine Tumor Peptide Marker.
Key Takeaways
- Amyloid-beta accumulates in brain plaques 15-20 years before cognitive symptoms appear, making it one of the earliest detectable biomarkers of Alzheimer's pathology
- CSF amyloid-beta 42 decreases as brain plaque burden increases, because the peptide is trapped in plaques instead of flowing into spinal fluid (inversely proportional biomarker)
- Blood-based p-tau217/amyloid-beta ratio achieved AUC of 0.94-0.97 for detecting amyloid pathology in 2025 clinical validation studies, approaching the accuracy of CSF testing
- Lecanemab reduced amyloid plaque burden and slowed cognitive decline by 27% over 18 months in the Clarity AD trial (van Dyck et al., NEJM, 2023)
- Donanemab slowed clinical decline by 35% in patients with low-to-moderate tau burden in the TRAILBLAZER-ALZ 2 trial (Sims et al., JAMA, 2023)
- The amyloid hypothesis remains debated: clearing plaques slows decline but does not stop or reverse it, suggesting amyloid is necessary but not sufficient for disease progression
What Is Amyloid-Beta?
Amyloid-beta is not a rogue molecule. It is a normal product of cellular metabolism. Every neuron in the brain continuously produces amyloid precursor protein (APP), a transmembrane protein whose function is still debated but may involve synaptic plasticity and iron export. APP can be cleaved by two different enzymatic pathways.
In the non-amyloidogenic pathway, alpha-secretase cuts APP within the A-beta sequence, preventing intact A-beta formation. In the amyloidogenic pathway, beta-secretase (BACE1) cuts APP at one end of the A-beta sequence, and gamma-secretase cuts at the other, releasing an A-beta peptide fragment into the extracellular space.[1]
The precise length of the released peptide depends on where gamma-secretase makes its cut. A-beta 40 (40 amino acids) is the most abundantly produced form. A-beta 42 (42 amino acids) is less abundant but far more prone to aggregation. The two extra amino acids at the C-terminus of A-beta 42 make it more hydrophobic, causing it to misfold, oligomerize, and eventually deposit as insoluble fibrils in brain plaques. Gao and colleagues (2025) identified a relationship between the anti-aging protein alpha-Klotho and BACE1, finding that Klotho can modulate BACE1 cleavage activity, providing a potential mechanistic link between aging and amyloid production.[2]
The Amyloid Cascade Hypothesis
In 1992, John Hardy and Gerald Higgins proposed that amyloid-beta accumulation is the initiating event in Alzheimer's disease, triggering a cascade of tau phosphorylation, neuroinflammation, synaptic loss, and neuronal death. This "amyloid cascade hypothesis" has dominated Alzheimer's research for three decades.[1]
Selkoe and Hardy (2016) reviewed the hypothesis at its 25-year mark and found that genetic evidence strongly supported it: every known familial Alzheimer's mutation either increases total A-beta production, shifts the A-beta 42/40 ratio toward the more aggregation-prone form, or alters A-beta clearance. Conversely, a protective variant of APP (A673T, the "Icelandic mutation") reduces BACE1 cleavage and is associated with reduced Alzheimer's risk and preserved cognition in old age.[1]
The hypothesis has survived repeated challenges, but with important modifications. The original version emphasized insoluble plaques as the toxic species. Current evidence points to soluble oligomers of A-beta (small aggregates of 2-12 peptide molecules) as the primary neurotoxic form. Plaques may actually serve as reservoirs that sequester these toxic oligomers, which is why plaque burden and cognitive decline are only loosely correlated. Mele and colleagues (2026) demonstrated that protein disulfide isomerase can dissolve and detoxify oligomeric assemblies of amyloid-beta, further supporting the oligomer toxicity model.[3]
CSF Biomarkers: The Established Standard
The first biomarker application of amyloid-beta came from measuring it in cerebrospinal fluid. The key finding: as A-beta 42 accumulates in brain plaques, its concentration in CSF decreases. The peptide is being trapped in the brain instead of being cleared into the spinal fluid. Low CSF A-beta 42 (or a low A-beta 42/40 ratio) indicates brain amyloid pathology.
Combined with elevated CSF phosphorylated tau (p-tau) and total tau (t-tau), the A-beta 42 measurement forms the "ATN" biomarker framework (Amyloid, Tau, Neurodegeneration) that the National Institute on Aging and Alzheimer's Association adopted for research classification in 2018.
CSF biomarkers achieve high diagnostic accuracy (sensitivity and specificity both exceeding 85%) for identifying Alzheimer's pathology. The limitation is the lumbar puncture: invasive, uncomfortable, and impractical for screening large populations or monitoring treatment response over time.
Blood-Based Biomarkers: The 2025 Breakthrough
The field shifted decisively in 2024-2025 when blood-based biomarker tests reached clinical-grade accuracy.
Plasma p-tau217 (phosphorylated tau at threonine 217) emerged as the strongest single blood biomarker for Alzheimer's amyloid pathology. The p-tau217/A-beta 42 ratio in plasma achieved areas under the curve (AUC) of 0.94-0.97 for detecting amyloid PET positivity in multiple independent cohorts. This performance is clinically equivalent to, and in some comparisons superior to, established CSF testing.
In 2025, the Alzheimer's Association published its first clinical practice guideline for blood-based biomarkers, recommending their use in specialized care settings for the diagnostic workup of suspected Alzheimer's disease. The guideline specified performance thresholds: tests with 90% or greater sensitivity and 75% or greater specificity can serve as triage tools, while tests meeting 90% or greater for both metrics can substitute for amyloid PET or CSF testing.
The practical implications are profound. A blood draw replaces a lumbar puncture. Cost drops from thousands of dollars (PET imaging) to hundreds (blood test). Screening becomes feasible for primary care, clinical trials, and population studies. However, the guideline limited recommendations to specialized care settings, acknowledging that performance in unselected primary care populations remains less well validated.
Amyloid PET Imaging
Amyloid PET uses radioligands (florbetapir, florbetaben, flutemetamol, or the newer [18F]NAV4694) that bind to fibrillar A-beta deposits and visualize plaque burden in living brains. A positive amyloid PET scan confirms the presence of amyloid pathology with high specificity.
PET imaging has been the reference standard against which CSF and blood biomarkers are validated. It provides spatial information that fluid biomarkers cannot: where plaques are deposited (cortical versus subcortical) and how burden changes over time.
The limitation of amyloid PET is cost (typically $3,000-6,000 per scan) and availability (requires a specialized facility with cyclotron access or delivery of short-lived radiopharmaceuticals). Blood-based biomarkers are positioned to serve as gatekeepers, identifying patients who warrant PET confirmation and monitoring.
Anti-Amyloid Therapeutics: Biomarker-Guided Treatment
The diagnostic use of amyloid-beta biomarkers became clinically urgent when anti-amyloid antibody therapies received FDA approval.
Lecanemab (Leqembi)
The Clarity AD trial (van Dyck et al., 2023) randomized 1,795 participants with early Alzheimer's disease to lecanemab or placebo over 18 months. Lecanemab reduced brain amyloid plaque burden by a mean of 59 centiloids (from 77.9 to 17.5) and slowed clinical decline by 27% on the Clinical Dementia Rating-Sum of Boxes (CDR-SB). Amyloid PET confirmed nearly complete plaque clearance in many participants.[4]
Donanemab (Kisunla)
The TRAILBLAZER-ALZ 2 trial (Sims et al., 2023) randomized 1,736 participants with early symptomatic Alzheimer's disease. Donanemab slowed clinical decline by 35% in participants with low-to-moderate tau burden. Remarkably, 47% of donanemab-treated participants reached amyloid-negative status on PET by 12 months, at which point the drug was discontinued. Some participants maintained amyloid-negative status for over a year after stopping treatment.[5]
Both drugs require biomarker confirmation of amyloid pathology before treatment initiation. Lecanemab's prescribing information requires either a positive amyloid PET scan or an amyloid-positive CSF test. This creates an unprecedented situation: a peptide biomarker is now a prerequisite for prescribing a drug that targets the same peptide.
For how GLP-1 receptor agonists may independently affect amyloid pathology, see the emerging evidence: Qi et al. (2026) showed that liraglutide reduced amyloid-beta accumulation and improved cognitive function in a transgenic mouse model of Alzheimer's disease.[6]
Beyond Plaques: Amyloid-Beta in Systemic Health
Amyloid-beta is not exclusively a brain molecule. It is produced in peripheral tissues (platelets, skeletal muscle, intestinal epithelium) and circulates in plasma. Sopova and colleagues (2026) found that A-beta (1-40) levels in plasma were associated with systemic metabolic health parameters, suggesting that amyloid-beta participates in metabolic regulation beyond the central nervous system.[7]
Yong and colleagues (2025) reviewed evidence that amyloid-beta functions as an antimicrobial peptide, capable of killing bacteria, fungi, and viruses through mechanisms similar to other innate immune peptides like defensins. This "antimicrobial hypothesis" proposes that A-beta production is part of the brain's innate immune response, and that chronic infection or microbiome dysbiosis could trigger pathological overproduction. The hypothesis does not replace the amyloid cascade model but may explain why amyloid-beta exists at all: it may be a protective peptide that becomes pathological when chronically overproduced or inadequately cleared.[8]
Amyloid-Beta Interactions with Other Peptides
Amyloid-beta does not aggregate in isolation. Its behavior is modulated by interactions with other peptides and proteins in the brain.
Pham and colleagues (2025) investigated how somatostatin and its derivatives interact with amyloid-beta aggregation. They found that certain somatostatin analogs could modulate A-beta oligomerization, suggesting a potential mechanistic link between somatostatin deficiency (which occurs in Alzheimer's disease) and increased amyloid pathology.[9]
Jeong and colleagues (2018) showed that MK-0677, a ghrelin receptor agonist, alleviated amyloid-beta-related pathology in a transgenic Alzheimer's mouse model (5XFAD), reducing both plaque burden and neuroinflammation through mechanisms involving enhanced microglial phagocytosis and reduced BACE1 expression.[10]
These interactions are significant because they suggest that the amyloid system does not operate independently. Other peptide signaling systems (somatostatin, ghrelin, GLP-1) can modulate amyloid production, aggregation, and clearance. This opens therapeutic angles beyond direct anti-amyloid antibodies. For how other peptide biomarkers are used in clinical diagnosis, see the sibling articles on BNP and NT-proBNP, C-Peptide, Copeptin, and Procalcitonin.
The Evidence Landscape
The diagnostic utility of amyloid-beta as a biomarker is no longer debatable. CSF A-beta 42/40 ratio and amyloid PET imaging are validated, guideline-endorsed tools. Blood-based p-tau217/A-beta ratios are reaching clinical-grade accuracy. The remaining questions are about interpretation and context.
What we know well: Amyloid-beta accumulation precedes symptoms by 15-20 years. CSF and blood biomarkers reliably detect amyloid pathology. Anti-amyloid antibodies can clear plaques and modestly slow cognitive decline. Amyloid-beta 42 is more pathogenic than A-beta 40 due to its increased aggregation propensity.
What remains uncertain: Whether amyloid-beta is a cause of neurodegeneration or a response to it. Why plaque clearance produces only modest clinical benefit. Whether blood-based biomarkers perform equivalently in diverse populations and primary care settings. The role of peripheral amyloid-beta in systemic health. Whether earlier intervention (treating amyloid-positive, cognitively normal individuals) would produce larger clinical effects. The long-term safety profile of chronic anti-amyloid therapy, particularly regarding amyloid-related imaging abnormalities (ARIA).
The relationship between this peptide and disease continues to evolve. Amyloid-beta went from a laboratory curiosity to the defining biomarker of Alzheimer's disease, and from a failed drug target to a viable (if modest) therapeutic success. The story of this 42-amino-acid peptide is far from over.
For related neuroscience content, see Amyloid-Beta: The Peptide Fragment at the Heart of Alzheimer's and BDNF: The Brain Peptide That Builds New Neural Connections.
The Bottom Line
Amyloid-beta is the most clinically consequential peptide biomarker in neurology. Its measurement in CSF, blood, and PET imaging defines Alzheimer's disease pathology and now gates access to the first disease-modifying therapies. Blood-based biomarker tests reaching clinical guidelines in 2025 represent a turning point for accessible diagnosis. The amyloid cascade hypothesis, while modified by the recognition that soluble oligomers rather than plaques are the primary toxic species, remains the dominant framework for understanding Alzheimer's pathogenesis.