Simple Tool Predicts Individual Alzheimer’s Risk

A simple blood test, coupled with brief memory tests, showed who will develop Alzheimer’s disease in the future with a high degree of accuracy.

Combining plasma phosphorylated tau (p-tau), APOE genotype, and scores from 10-minute executive function and memory tests predicted Alzheimer’s disease onset within 2 to 6 years among people with memory complaints with 90% certainty, reported Oskar Hansson, MD, PhD, and Sebastian Palmqvist, MD, PhD, both at Lund University in Sweden, and colleagues.

When dementia experts examined the same patients, they were about 71% accurate, the researchers noted in Nature Medicine.

“We show for the first time that a time- and cost-effective diagnostic algorithm clearly outperforms the clinical work-up typically done today when it comes to predicting Alzheimer’s dementia,” Hansson told MedPage Today.

To build their prediction algorithms, the researchers studied 340 people with mild cognitive symptoms in the Swedish BioFINDER cohort. Participants were 60 to 80 years old, had a Mini-Mental State Examination (MMSE) score of 24 to 30 points at baseline, and did not fulfill criteria for any dementia. The primary outcome was prediction of progression to Alzheimer’s dementia versus to any other dementia, or to not progressing, in 4 years.

The researchers assessed demographics, APOE4 genotype, cortical thickness, cognitive measures, and plasma and cerebrospinal fluid (CSF) markers, searching first for the best fit, then omitting variables to find simpler algorithms with similar predictive power. They used data from 543 participants in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) as the validation cohort.

Within 4 years, plasma p-tau217 alone predicted Alzheimer’s in the BioFINDER cohort with an area under the curve (AUC) of 0.83. Combining plasma p-tau 217, memory tests, executive function tests, and APOE boosted accuracy to 91% (AUC 0.91, 95% CI 0.88-0.95, P<0.001).

In ADNI, this model had a similar AUC (0.90) using plasma p-tau181 instead of p-tau217. Within 2 and 6 years, similar models had AUCs of 0.90-0.91 in both cohorts. Using CSF p-tau, amyloid beta, and neurofilament light instead of plasma biomarkers did not improve accuracy significantly.

Phosphorylated tau blood biomarkers have great potential in Alzheimer’s disease, said Nicholas Ashton, PhD, of University of Gothenburg in Sweden, who wasn’t involved with the study. “They offer a low-cost, less-invasive alternative to understanding what’s going on molecularly in the brain,” he told MedPage Today.

Blood tests will be useful both for clinicians and to recruit people into therapeutic trials for drugs specifically targeted for Alzheimer’s pathology, Ashton noted.

Both p-tau217 and p-tau181 performed well in this analysis, he added. “There’s been debate about which is superior, but this study shows that — in combination with the cognitive tests and genetic tests — it doesn’t matter what type of phospho-tau biomarker you use: they both gave the same prediction power.”

The algorithm is being developed for clinics without access to advanced diagnostic instruments, Palmqvist said in a statement. “In the future, the algorithm might therefore make a major difference in the diagnosis of Alzheimer’s within primary healthcare,” he noted.

As of now, it’s been tested only on patients who have been examined in memory clinics, he added. “Our hope is that it will also be validated for use in primary healthcare as well as in developing countries with limited resources.”

Last Updated May 30, 2021