6 Cohort analysis links elevated baseline p-tau217 to increased 5- and 10-year risk, brings the pre-symptomatic risk model closer and reveals why routine screening is not yet ready.
Study: Prognostic value of blood-based P-Tau217 levels on progression to cognitive impairment. Image credit: Lightspring / Shutterstock
In a recent study published in American Medical Association JournalResearchers found that baseline concentrations of plasma phosphorylated tau-217 (p-tau217), a blood-based protein biomarker, were associated with risk of future cognitive impairment in cognitively intact older adults.
This study combined data from six international cohorts (n = 2,684 participants) and found that higher baseline p-tau217 levels were associated with long-term clinical progression after adjusting for APOE ε4 status and amyloid PET imaging. Although further validation in unselected representative populations and development of assay-specific thresholds are required before p-tau217-based estimates can guide individual prognosis, these findings may support the development of prognostic models and the design of preclinical Alzheimer’s disease trials.
background
Despite decades of neurobiological research aimed at understanding the mechanistic basis of cognitive decline, scientists still believe that early identification of individuals at high risk for Alzheimer’s disease (AD) and dementia remains a major diagnostic challenge.
Alzheimer’s pathology can build up in the brain long before clinical symptoms such as memory loss or confusion appear.
Current biomarker-based diagnostic approaches, such as positron emission tomography (PET) brain scans and post-lumbar puncture cerebrospinal fluid (CSF) analysis, can help detect Alzheimer’s disease-related brain lesions and identify individuals at high risk for future cognitive decline, but these methods can be costly, resource-intensive, or invasive, necessitating the development of more accessible alternatives.
Furthermore, although these approaches are useful in estimating the risk of future cognitive decline, they cannot pinpoint exactly when impairment will occur.
Recent advances in ultrasensitive biomarker assay technology have enabled researchers to develop blood tests that can measure AD-associated proteins such as p-tau217. However, p-tau217 testing is currently not recommended for people without cognitive impairment outside of research studies or clinical trials.
About research
This study aimed to address these diagnostic needs by conducting a harmonized multicohort analysis to determine whether baseline plasma p-tau217 measurements in cognitively intact older adults can be used to estimate the absolute risk of progression to cognitive impairment and rate of cognitive decline over 2, 5, and 10 years.
The pooled sample consisted of standardized data from six international observational and clinical trial cohorts, with a total of 2,684 cognitively unimpaired older adults, median age 69.6 years, interquartile range 66.2 to 74.2 years, and 63% of participants were female. Participants in this study had blood drawn at enrollment to quantify baseline levels of phosphorylated tau-217 (p-tau217). At baseline, 43% of participants were amyloid positive.
Participants also underwent amyloid beta PET scans within two years of baseline to measure brain amyloid burden and genetic testing to identify the apolipoprotein E epsilon 4 (APOE epsilon) allele, a well-known genetic risk factor for dementia.
Participants then underwent annual cognitive follow-up for up to 13.5 years to track progression from no cognitive impairment to cognitive impairment (defined herein as mild cognitive impairment (MCI), dementia, or two consecutive cohort-specific diagnoses with two consecutive Global Clinical Dementia Clinical Assessment scores of 0.5 or higher).
Research results
Long-term follow-up of this study revealed that 478 of 2,684 enrolled participants progressed to clinical cognitive impairment over a median of 5.4 years. Using a Cox proportional hazards model, this study found that higher baseline plasma p-tau217 levels were associated with an increased risk of clinical progression. Elevated p-tau217 was also associated with accelerated cognitive decline in the latent preclinical Alzheimer’s cognitive complex. However, the estimated 2-year absolute risks remained low across all p-tau217 categories, ranging from 1% to 4%.
Specifically, each standard deviation increase in p-tau217 was associated with a 38% increase in the hazard of progression, with a hazard ratio (HR) of 1.38 and a 95% confidence interval (CI) of 1.30 to 1.46, p < 0.001.
This association remained statistically significant after additional adjustment for amyloid PET imaging, with a HR of 1.32 (95% CI, 1.24 to 1.41; p < 0.001).
Model-based group estimates showed that participants with baseline p-tau217 concentrations in the “very high” category (defined as at least 2.5 standard deviations above cohort-specific norms derived from amyloid-negative participants younger than 70 years) faced an estimated 38% absolute risk of developing cognitive impairment within 5 years, with a 95% CI of 33% to 43%.
The estimated risk was higher at 10 years, reaching 78% in the very high group, with a 95% CI of 69% to 84%. However, these long-term estimates should be interpreted with caution, as only 5% of participants had at least 10 years of follow-up, most from one cohort. This finding indicates that baseline p-tau217, although not yet suitable for individual prediction, provides clinically relevant prognostic information beyond demographic factors, APOE ε4 status, and amyloid PET.
conclusion
This study extends earlier single-cohort studies by providing harmonized time-specific absolute risk estimates across multiple selected cohorts and assay platforms. Although a single baseline p-tau217 measurement was associated with long-term progression risk, this study did not evaluate continuous monitoring or establish that Alzheimer’s disease was the cause of any disability event.
The results of this study suggest that plasma p-tau217 may contribute to future models that estimate group-level 5- and 10-year risks together with clinical and biomarker information. However, these estimates are not yet accurate enough to guide individual prognosis or clinical decision-making.
An important caveat is that current clinical practice guidelines do not recommend p-tau217 testing in people without cognitive impairment outside of research studies or clinical trials, while disease-modifying treatments are still being evaluated in the preclinical phase.
Beyond these current limitations, this study may inform future research on presymptomatic risk stratification and support the design of prevention trials. Because the selected cohorts differed in recruitment, assay platform, and result ascertainment, and had limited demographic and socioeconomic diversity, further validation in population-representative cohorts, with assay-specific thresholds and clear consideration of comorbidities and mortality as competing risks, is required before p-tau217-based estimates can guide routine care.
