Blood Test Clock Model Predicts Alzheimer's Symptom Onset Within 3-4 Years

A new study published in Nature Medicine on February 19 shows that a single blood test can assess not only a person's risk of developing Alzheimer's disease but also predict approximately when symptoms will start, with an average margin of error of about 3-4 years. The study was developed and launched by the Foundation for the National Institutes of Health's Biomarkers Consortium.

Researchers in the study analyzed p-tau217, a protein linked to Alzheimer's disease, in blood samples collected over a period of up to 10 years from more than 600 adults (ages 62-78 years) who were initially free of cognitive symptoms. The team then built statistical clock models that related blood p-tau217 changes over time to future symptom onset.

The test is based on an abnormal form of a protein called tau that circulates in the blood, and begins to accumulate in the brains of people with Alzheimer's well before symptoms such as memory loss appear. Abnormal tau proteins can form tangled fibres that disrupt communication among the brain's nerve cells.

People with higher levels of plasma p-tau217 tended to develop Alzheimer's symptoms sooner, and older individuals developed symptoms more quickly after reaching elevated levels of this protein. The model was able to estimate how many years away a person might be from developing memory and thinking problems related to Alzheimer's, with an accuracy of within 3-4 years.

Notably, the time from p-tau217 positivity to onset of symptoms was markedly shorter in older individuals. For example, if a person had elevated p-tau217 in their plasma at age 60, they developed symptoms 20 years later. If p-tau217 wasn't elevated until age 80, symptoms appeared much sooner, suggesting that younger people's brains may be more resilient to neurodegeneration and that older people may develop symptoms at lower levels of Alzheimer's pathology.

The study was part of a project developed and launched by the Foundation for the National Institutes of Health Biomarkers Consortium, a public-private partnership. Participants included 603 older adults who lived independently in the community from two independent long-running Alzheimer's research initiatives: the Washington University School of Medicine Knight Alzheimer Disease Research Center and the Alzheimer's Disease Neuroimaging Initiative, based at multiple sites in the U.S.

Plasma p-tau217 was measured with PrecivityAD2, a clinically available diagnostic blood test for Alzheimer's disease from C2N Diagnostics, a Washington University startup. Plasma p-tau217 was also measured in the ADNI cohort using blood tests from other companies, including one cleared by the U.S. Food and Drug Administration. Similar models were constructed with data from one p-tau217/Aβ42 immunoassay and four plasma p-tau217 immunoassays.

Plasma p-tau217 has previously been shown to correlate strongly with the accumulation of amyloid and tau in the brain as shown on PET scans. The key hallmarks of Alzheimer's disease, amyloid and tau are misfolded proteins that begin building up in the brain many years before Alzheimer's symptoms develop. Plasma measures of p-tau217 have high associations not only with amyloid PET but also with tau PET, brain volumes and cognition.

This level of accuracy could prove especially helpful for clinical trials, helping researchers select participants most likely to develop symptoms within a trial's timeframe—making studies more efficient and powerful. The models could allow clinical trials of potentially preventive treatments to be performed within a shorter time period.

The research team also developed a web-based application that allows scientists to visualize how levels of plasma p-tau217 change over time and how they relate to Alzheimer's symptoms. The interactive tool helps researchers to explore complex relationships between plasma p-tau217, age, and symptoms.

The senior author stated that the study shows it is possible to use blood tests to predict not only if individuals are likely to develop Alzheimer's symptoms but also to estimate when the symptoms will begin, adding that researchers are working to make these models even more accurate. However, at this point, cognitively unimpaired individuals are not recommended to have any Alzheimer's disease biomarker test. Both Alzheimer's biomarker testing in people without cognitive symptoms and use of the web-based application should be limited to research settings, the study authors wrote.

The study was supported by private-sector partners AbbVie Inc., the Alzheimer's Association, The Alzheimer's Drug Discovery Foundation's Diagnostics Accelerator, Biogen, Johnson & Johnson, and Takeda. More than 7 million Americans live with Alzheimer's disease, with health and long-term care costs for Alzheimer's and other forms of dementia projected to reach nearly $400 billion in 2025, according to the Alzheimer's Association.

The study builds on data and insights generated from previous and ongoing FNIH research partnerships committed to advancing understanding of Alzheimer's and other neurodegenerative diseases. Beginning with the Alzheimer's Disease Neuroimaging Initiative in 2004 and continuing through the Accelerating Medicines Partnerships and the Biomarkers Consortium, the FNIH has spearheaded numerous collaborative efforts focused on identifying and validating biomarkers for early detection and effective treatment of Alzheimer's.