This article is part of “Innovations In: Alzheimer’s Disease” an editorially independent special report that was produced with financial support from Eisai.
The first hints that Gregory Nelson might be having cognitive troubles were subtle. So subtle, in fact, that his doctor assured him nothing was wrong. “Everyone who hits a certain age just misses words,” Nelson remembers him saying. When Nelson got home, he regretted not pushing harder for a referral. His entire family had noticed changes. Nelson, who is 70, scheduled another appointment and convinced his physician to send him to a neuropsychologist.
Nelson’s greatest fear was that he was in the beginning stages of Alzheimer’s—his father, though never formally assessed, had probably died of the disease. But Nelson didn’t get a diagnosis, at least not right away. He waited months to be seen by the neuropsychologist his physician referred him to. That specialist performed a cognitive assessment, which indicated mild cognitive impairment, but the doctors couldn’t yet tell him the cause. The neurologists in his area were completely booked, so there was another lengthy wait between his primary-care visit and his diagnosis of Alzheimer’s.
On supporting science journalism
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
The delay gave Nelson ample time to prepare for the worst, but it also created a lot of uncertainty and anxiety. “The bummer about waiting is that your brain just goes all over the place,” he says. It would have been nice to have a definitive answer sooner, he says—to be able to say, “This is it, man. This is the diagnosis.”
A quicker, simpler path to early diagnosis may be in sight with recently developed blood tests that can indicate whether someone has known markers for Alzheimer’s. Unlike other diagnostic tools, such as spinal taps and positron-emission tomography (PET) scans, blood tests are relatively cheap and simple to perform. And they can help differentiate Alzheimer’s from other neurodegenerative conditions and medical problems, potentially hastening access to specialists and therapies. The sooner someone knows they have Alzheimer’s, the sooner they can plan for the future and assess possible interventions: Anti-amyloid treatments have been shown to modestly slow disease progression when given early in the course of the disease. Lifestyle changes also seem most effective when adopted before symptoms get too advanced.
But some experts worry that because these blood tests are so simple to perform, physicians might order them right away to provide a quick diagnosis without carrying out a fuller workup on the patient. The current tests come with several caveats that make interpretation tricky, especially for people who have no symptoms of dementia. A positive result doesn’t guarantee that the person will develop cognitive problems. And there aren’t any approved therapies for people without symptoms. “This is not a mass screening test,” says Nathaniel Chin, a geriatrician and medical director for the Wisconsin Alzheimer’s Disease Research Center in Madison. “This isn’t something we just give willy-nilly.”
Alzheimer’s is named for German psychologist Alois Alzheimer, who first identified the hallmarks of the illness in the brain of someone with dementia: abnormal clumps of a peptide called amyloid beta wedged between brain cells, along with tangled filaments of a protein called tau. Although some tests can detect amyloid and tau while patients are alive, examining the brain postmortem is the only definitive way to diagnose the disease.
In living patients, neurologists typically rely on a battery of tests. Chin starts with a clinical history and cognitive testing. “You have to diagnose mild cognitive impairment or dementia first,” he says. Cognitive testing can reveal patterns suggestive of Alzheimer’s, but it’s not proof positive. Vitamin deficiencies, certain medications, and multiple other factors can cause dementialike symptoms. “So truly, to know if it’s Alzheimer’s, you’d want to confirm it biologically,” Chin says.
That kind of confirmation typically involves detecting amyloid beta and tau either in the brain with a PET scan or in cerebrospinal fluid with a lumbar puncture. Compared with the general population, people with Alzheimer’s tend to have lower levels of amyloid beta and higher levels of a dysfunctional form of tau called phosphorylated tau, or p-tau, in their spinal fluid. Such tests are expensive and can’t be done in a regular doctor’s office. As a result, more than half of Alzheimer’s cases—between 50 and 70 percent—are missed or misdiagnosed.
That’s why there is so much excitement around blood tests. They’re cheap, simple to perform and scalable. Researchers have spent more than a decade trying to figure out which blood biomarkers work best for disease detection, and to date p-tau217 seems to align most closely with results from PET imaging and spinal fluid analysis. When researchers compared several blood tests, they found that the level of p-tau217 had the strongest correlation to PET scan measurements of amyloid and tau levels in the brain.
Roughly a dozen blood tests are currently available to physicians, but only one, Lumipulse, has been approved by the U.S. Food and Drug Administration. The rest are available as lab-developed tests, which can be marketed without fda approval. Accuracy varies, but some appear to perform as well as spinal fluid tests.
Amyloid beta can begin accumulating in the brain a decade or more before symptoms of Alzheimer’s emerge, so many people hope that blood tests could eventually be used to spot the disease far earlier than is possible today. “There’s this whole 10 or 20 years where interventions could potentially be done,” says Zaldy Tan, a memory and aging specialist at the Cedars-Sinai Medical Center in Los Angeles. There is some evidence that higher-quality education and improved diet, exercise and social stimulation could help ward off symptoms [see “Cultivating Resilience”]. And preventive therapies are in the works. Anti-amyloid medications that have reached the market over the past couple of years are currently approved only for those with an Alzheimer’s diagnosis, but ongoing trials aim to test whether they might also help stave off symptoms in people who have biological but not behavioral signs of disease [see “A Dangerous Silver Bullet”]. “What we’re really trying to do is extend someone’s functional independence and their ability to maintain a high standard of living,” says Joel Braunstein, president and CEO of C2N Diagnostics, a St. Louis–based company that sells blood tests for Alzheimer’s. “If we can forestall people from developing what we call clinical Alzheimer’s disease, we will have made a profound impact.”
Today, however, most experts agree that only symptomatic people should get blood tests. Lumipulse, for example, is approved for patients 55 years or older who have signs of Alzheimer’s. In that group, the test has proven accuracy: about 92 percent of people with positive results had evidence of amyloid plaques on a PET scan or in their spinal fluid, and more than 97 percent of those who tested negative had no evidence of plaques.
For those who are asymptomatic, however, it’s not clear whether blood tests would be useful. In this group, “we know we will have more false positives because the prevalence of the disease is lower,” says Alicia Algeciras-Schimnich, a clinical chemist at the Mayo Clinic in Rochester, Minn. In fact, the field doesn’t have a standard way to interpret blood test results for people who don’t show signs of cognitive decline.
Does a positive test equal Alzheimer’s? It depends.
Last year the Alzheimer’s Association published revised criteria that rely on positive biomarkers—in the blood or elsewhere—to diagnose disease, regardless of whether the patient is symptomatic. Heather Snyder, the Alzheimer’s Association’s senior vice president of medical and scientific relations, says this approach echoes the criteria used for other diseases with known mechanisms. “Defining a disease by its biology rather than symptoms has been the status quo for years in other areas of medicine such as cancer, heart disease and diabetes,” she says. “The Alzheimer’s field is now making similar progress.”
Many neurologists find this change problematic. Although abnormal blood levels of tau and amyloid put someone at risk of developing symptoms, they say little about the level of risk or the timeline for progression, says Nicolas Villain, a neurologist at the Pitié-Salpêtrière Hospital’s Institute of Memory and Alzheimer’s Disease in Paris. “It’s an increased risk, but we cannot be more specific,” he says. “It’s not very informative for the individual to have this information.”
In one recent study, researchers analyzed results from more than 2,100 adults who did not have dementia but had taken blood tests to measure amyloid beta, p-tau, and a few other biomarkers. Over the course of 16 years, 212 of the subjects developed Alzheimer’s. The researchers found that the tests did an excellent job of predicting who would not develop the disease but a poor job of predicting who would. Less than 25 percent of those who tested positive developed Alzheimer’s during the follow-up period.
Second, no proven preventive treatment exists. And although exercising more, eating healthier and staying connected to others have been linked to reduced risk of cognitive decline, there’s little evidence that such interventions work by acting on amyloid and tau in the brain. It’s more likely, Villain says, that these factors boost the brain’s ability to stay sharp even when amyloid and tau are present.
An international working group that included Villain and 45 other Alzheimer’s experts from 17 countries developed a more nuanced assessment. The group split asymptomatic individuals with positive biomarker tests into two categories. Those with no symptoms and a low lifetime risk of developing them were classified as “asymptomatic at-risk” for the disease. Those with a genetic background that increased their risk or with imaging results suggesting their brain already had tau buildup were designated as having presymptomatic Alzheimer’s. “It may seem to be only semantics,” Villain says. “It’s important semantics, however.”
Even in symptomatic older adults, the tests aren’t perfect. Measuring biomarkers in blood is trickier than doing so in spinal fluid. Because blood concentrations of amyloid and tau are far lower, the tests have to be more sensitive. And because there are so many other proteins in blood, the signal is harder to detect. What’s more, p-tau can be elevated for reasons other than the presence of Alzheimer’s. In patients with kidney disease, for example, p-tau levels might be high because the kidneys aren’t able to properly clear the protein.
Some research also suggests that the tests might not be as accurate in certain racial groups. In one study, William Hu, director of the Center for Healthy Aging Research at Rutgers University, and his colleagues studied data from about 200 older Americans who had undergone detailed clinical evaluations and had given blood and spinal fluid samples. The researchers found that whereas a positive blood test was 87 percent accurate for white patients, it was only 58 percent accurate for Black patients. “So there’s a high chance that they actually won’t have Alzheimer’s,” Hu says. Other studies, however, failed to find racial disparities for blood biomarkers, and no one yet understands why such a discrepancy might exist.
For now, unless the patient has undergone cognitive testing and been diagnosed with cognitive impairment, most neurologists suggest forgoing the blood tests. “That’s a conservative position, but that’s what many think is appropriate right now,” Chin says. There’s good reason for the recommendation. Family doctors aren’t always educated on all the caveats that go along with these blood tests, nor are they able to explain them in the little time they have with a patient. “I worry that this easy-to-click test is going to replace what clinicians are good at: talking to people and hearing their complaint,” Chin says. If someone comes in because of memory problems and takes a blood test, the test could come back positive even if the person’s lapses aren’t the result of amyloid in the brain. Maybe “it’s actually sleep apnea, depression or a medication side effect,” he adds. “That elevated amyloid is just a red herring.”
One of Chin’s patients first visited his primary-care doctor with complaints of forgetfulness. The patient’s lab work and magnetic resonance imaging came back normal, so, without performing any cognitive testing, the doctor did a blood test for Alzheimer’s. The result was positive, and the patient assumed he had dementia. But additional testing at Chin’s clinic revealed that the man had only very mild cognitive impairment. “He came to me thinking, ‘I’m going to be talking about hospice and end-of-life stuff,’” Chin says. But he was probably years away from developing dementia.
In some cases, Hu says, multiple tests from different companies can also seed confusion. “One of my patients said, ‘My tests tell me definitively I have Alzheimer’s but also definitively I do not have it. So what does this all mean?’” he asks.
The Alzheimer’s Association is working to develop clinical practice guidelines for blood-based biomarker tests. In July the group released the first two recommendations. “These are really focused on the specialty-care setting in patients who are living with cognitive impairment,” says Rebecca Edelmayer, vice president of scientific engagement at the Alzheimer’s Association. In those individuals, tests with 90 percent or greater accuracy can be used to diagnose the disease with no further confirmatory testing. That means patients wouldn’t need to undergo a PET scan or lumbar puncture. Less accurate tests can be used to triage people for further testing as long as they have sensitivity of at least 90 percent for detecting disease and specificity of at least 75 percent for identifying people without disease.
Edelmayer says the group is still working on guidance for primary-care physicians and for people who are not cognitively impaired. “We really needed to evaluate all the evidence around individuals who are cognitively impaired first before moving into individuals who are cognitively unimpaired,” she says.
Getting the diagnosis right is crucial, says neurologist Shauna Yuan of the University of Minnesota Medical School. “We have to remember that this is a terminal disease,” she says. “It is not something we can take lightly.”
Who Should Get Tested?
For those who are unsure about whether they or a family member should get a blood test, neurologists say the tests should be reserved for people experiencing memory loss or other cognitive problems. A blood screen can help differentiate Alzheimer’s disease from other types of dementia or other health issues. If it comes back positive, a neurologist might still want to confirm the diagnosis with a lumbar puncture or a PET scan. “It’s a serious diagnosis that I don’t take lightly, so I want it to be as reliable as possible,” says Zaldy Tan of the Cedars-Sinai Medical Center in Los Angeles.
