A new study has found that a joint blood test for detecting cognitive decline has an accuracy rate of up to 90% in determining whether memory loss is due to Alzheimer’s disease.

In comparison, neurologists and other memory specialists were able to accurately diagnose Alzheimer’s in 73% of their cases. Primary care physicians had less success, with an accuracy rate of only 61%, according to the study.

One part of the blood test - called plasma tau 217, or p-tau217 for short - is one of many blood biomarkers being evaluated by scientists for use in diagnosing mild cognitive impairment and early-stage Alzheimer’s.

Dr. Sebastian Palmqvist, an associate professor and senior consultant in neurology at Lund University in Sweden, and a co-author of the study, said the test measures tau protein 217, which is an excellent indicator of amyloid disease.

In an email, Palmqvist wrote: "The increase in tau protein 217 concentrations in the blood is very profound in Alzheimer’s disease. In the dementia stage of the disease, levels are more than 8 times higher than in elderly individuals without Alzheimer’s."

Research published in January found that a similar test for tau protein 217 is accurate up to 96 percent in identifying elevated levels of beta-amyloid and has an accuracy of up to 97 percent in determining tau. The presence of beta-amyloid and tau tangles in the brain are distinguishing features of Alzheimer’s disease.

In the new study, the p-tau217 test was combined with another blood biomarker for Alzheimer’s called the amyloid ratio 42/40, which measures two types of amyloid proteins, another biomarker for Alzheimer’s.

The combination of the amyloid and tau tests, called the amyloid probability score, proved to be the most predictive.

Dr. Maria Carrillo, chief science officer at the Alzheimer’s Association, said, "We would like to have a blood test that can be used in a primary care physician's office, functioning like a cholesterol test but for Alzheimer’s."

Carrillo, who oversees research initiatives at the association that included partial funding for the new study, added: "The p-tau217 blood test has proven to be the most specific for Alzheimer’s and the most accurate. It appears to be the most promising candidate."

She added that highly accurate blood tests, once fully validated, could "change the game in the speed with which we can conduct Alzheimer’s trials and access the next new drug." "These are truly transformative times."

How does the p-tau217 blood test work?

Carrillo said that the p-tau217 peptide is unique because it can only be detected when amyloid plaques are present in the brain.

She continued, "What this means to us scientifically is that when we measure p-tau217, we are measuring the neuronal damage caused by tau very early in Alzheimer's disease, but only when amyloid is already present."

Carrillo said, "You’re not really measuring amyloid, but the test tells you it's there, and that has been supported by objective PET (positron emission tomography) scans that can see amyloid in the brain. It’s a hallmark of Alzheimer’s: If you don’t have amyloid present, you don’t have Alzheimer’s. However, if you have elevated tau in your brain, we know that is a sign of a different type of dementia."

Tau tangles are associated with several other neurological diseases like frontotemporal dementia (FTD) or frontal lobe dementia; tau tangles target the frontal lobe of the brain, causing behavioral and emotional changes and loss of executive functions like planning. Memory loss, if it occurs, comes much later.

In Alzheimer’s, tau tangles accumulate in the part of the brain that controls memory, but amyloid plaques play a key role. Small clusters of plaques can gather at synapses and interfere with the ability of neurons to communicate. Amyloid plaques may also overly stimulate the immune system, leading to inflammation that can cause further damage to the brain.

Experts say some of the latest drugs to treat dementia, like lecanemab and donanemab, target beta-amyloid and are believed to be less effective in people with advanced tau diseases.

Since amyloid deposits can start accumulating in the brain decades before symptoms appear, even when a person is in their thirties or forties, early diagnosis of amyloid in the brain may be crucial for lifestyle modification and preventive drug treatment.

Current screening tests are inconclusive

The study, published on Sunday in JAMA Neurology, followed 1,213 people with an average age of 74 who were undergoing cognitive assessments in both primary care and specialty clinics in Sweden.

Each person's blood was tested using p-tau217, and the results were combined with blood measurements for beta-amyloid 40/42 to develop the final score.

Dr. Richard Isaacson, a preventive neurologist and director of research at the Neurodegenerative Disease Institute in Florida, who did not participate in the study, said: "When you use a combination of the 40 to 42 ratio and p-tau217, it improves the diagnostic accuracy of p-tau217."

The 90 percent accuracy of the joint blood test from the study was confirmed by taking a cerebrospinal fluid sample, which is alongside PET scans the only standard scientific method aside from autopsy to diagnose Alzheimer’s disease. Experts say both tests are costly and require surgical intervention and are not readily available across the United States.

Results from the blood tests were then measured against the patient diagnoses provided by Swedish primary care physicians and specialists. The relatively low accuracy rates - 61 percent and 73 percent - highlight how difficult it is for doctors to correctly identify Alzheimer’s pathology using current tools: a short interview with the patient, a brief cognitive test, and brain imaging.

Dr. Oscar Hanson, a professor and senior consultant in neuroscience at Lund University and a co-author of the study, said: "Overall, traditional tests using pencil and paper and digital cognitive assessments do not have high accuracy in specifically identifying Alzheimer’s."

Hanson said in an email: "Many other conditions and diseases can present similar cognitive symptoms, potentially leading to misdiagnosis or incorrect diagnosis."

Hanson continued that between 20 and 30 percent of patients seen by specialists are on medications or have other medical conditions that can mimic Alzheimer’s. Conditions that may affect cognitive function include vascular dementia, depression, thyroid disease, sleep apnea, and even vitamin B12 deficiency.

Carrillo said that if these mimicking conditions are not detected during initial screening, individuals without Alzheimer’s may clog waitlists for specialists and appointments for spinal taps and amyloid PET scans.

She added that by the time someone with true amyloid diseases sees a specialist, they may "miss out on the eligibility window for the drugs we have, which is terrible."