An Algorithm Can Predict When Alzheimer’s Symptoms Will Appear

A PET Scan and an algorithm can help predict when Alzheimer's will become severe
An algorithm using PET imaging can help predict onset of Alzheimer's.

Key Takeaways

  • An algorithm that uses a brain scan and a patient’s age may be able to determine when someone with amyloid plaques will start to show serious signs of Alzheimer’s.
  • Such a test could supply valuable information about an individual case that could allow patients and their families to make plans.
  • The test involves a PET scan, which is expensive, and which insurance may not cover, so research is still underway to create other tests to detect or predict Alzheimer’s.

A new computer algorithm—paired with a brain scanning technique and a patient’s age—may help clinicians predict when the symptoms of Alzheimer's will start.

If someone learns that they have the amyloid plaques in their brain that are associated with Alzheimer’s, one of their first questions is often, "When will I start to have symptoms?"

While it's understandable that patients would want this information, neurologists usually are not able to give them much of an estimate. That's why this new research could be such a game-changer.

Predicting Alzheimer's Symptoms

A team of researchers led by Suzanne E. Schindler, MD, PhD, an assistant professor of neurology at Washington University School of Medicine in St. Louis, created a way to use brain imaging and a patient’s age to determine when the symptoms of Alzheimer’s will likely start to manifest themselves.   

The type of brain imaging the researchers used is called amyloid positron emission tomography (PET). It can reveal amyloid plaques—long thought to be an important cause of Alzheimer's—in the brains of people who still have normal brain function.

What Is Amyloid?

Amyloid is a substance that accumulates and forms plaques or clumps in the brain. It interferes with how the nerve cells in the brain function. The buildup of amyloid plaques, as well as a protein called tau, occurs for decades before the symptoms of Alzheimer's appear.

The researchers analyzed amyloid PET scans from 236 people who were participating in dementia research at the Knight Alzheimer’s Disease Research Center at Washington University.

Schindler tells Verywell that some of the participants “were normal" and some "were known to have high levels of brain amyloid." All of the participants had at least two brain scans, spaced 4.5 or more years apart. At both scans, the researchers estimated the amount of amyloid in the participants' brains.

Next, the researchers used a metric called the standard uptake value ratio (SUVR) to estimate the amount of amyloid in each of the participant's brains at each scan.

The team also looked at more than 1,300 clinical assessments of cognitive function that had been conducted every few years on 180 of the participants (most of whom had normal cognitive function when the assessments started).

The Tipping Point

The researchers determined that there is a threshold, or tipping point, at 1.2 SUVR. Schindler says that amyloid accumulates at a steady and predictable rate in the brain once the threshold has been reached.

People hit the tipping point at different ages. A person's age when they reach the threshold can be factored into the algorithm, which is then able to determine when it's likely that person will start to show symptoms of Alzheimer's.

The symptoms of Alzheimer's will start after the amyloid accumulation in the brain hits a "tipping point."

The age at which a person reaches this threshold can vary but it might be influenced by whether someone carries the APOE4 gene, which is associated with an increased risk of Alzheimer's.

How old someone is when the amyloid buildup hits that tipping point might also influence when the symptoms of Alzheimer's will start.

In the study, the researchers noted that the participants who reached the threshold at a younger age took longer to show signs of Alzheimer’s than the people who were older when they reached it.

The Findings Mean Earlier Diagnosis Is Possible

Schindler says that a major point of the study is that Alzheimer's symptom onset can be estimated accurately. This matters because up until now, there was no way to make predictions and it was unclear if it was even possible to do.

According to Schindler, the algorithm was created to predict the onset of dementia symptoms,—not to diagnose patients with dementia.

Therefore, Schindler cautions that patients with cognitive impairment "would still need to be evaluated to determine whether their symptoms are caused by Alzheimer's disease or something else.” 

Christopher Weber, PhD, the director of global science initiatives for the Alzheimer’s Association, tells Verywell that the study is "a really exciting piece of research" and that "we can utilize this technology to get an early and accurate diagnosis.”

Suzanne E. Schindler, MD, PhD

We are doing clinical trials on these preventative treatments, and this estimate of symptom onset would be helpful to these trials.

— Suzanne E. Schindler, MD, PhD

Knowing when the symptoms of Alzheimer’s could start to impair daily living would let patients and their families plan for the future. For example, Weber says that they could make legal and financial decisions in advance or perhaps choose to take part in a clinical trial for Alzheimer’s treatment.

“If and when we develop medications that prevent or slow onset of dementia symptoms, then this estimate of symptom onset would be helpful in determining who should take these preventative medications,” says Schindler. “Right now, we are doing clinical trials on these preventative treatments, and this estimate of symptom onset would be helpful to these trials.” 

Limitations and Barriers

While exciting, the research is still preliminary. Schindler says that the model is new and that they want other groups to confirm the model's findings.

There are also some limitations. For one, PET scans are also usually not done until later in the course of the disease. Even when they are done, they're costly—the out-of-pocket cost can be around $6,000.

"One of the issues with PET scan technology is that it's not available everywhere, and it's quite expensive because it's not covered by insurance," Webber says.

Schindler acknowledges the barrier and says that “if costs go down or insurance starts paying for the scans, they will become much more widely used."

Future Research

Research to find other ways to detect or predict Alzheimer’s is ongoing. Referring to the work of the Alzheimer's Association, Weber says that “we continue to research new kinds of biomarkers that look at how we can measure amyloid."

Someday, we may have tests that could measure amyloid in the brain by evaluating samples of spinal fluid, blood, or saliva, which would be less expensive and accessible to more people.

What This Means For You

A new algorithm that uses PET scans to look at the number of amyloid plaques in a person's brain paired with a person's age might be able to help clinicians predict when someone will start to show symptoms of Alzheimer's.

This information could give people with Alzheimer's and their caregivers the chance to start planning for how the disease will affect their lives sooner rather than later.

2 Sources
Verywell Health uses only high-quality sources, including peer-reviewed studies, to support the facts within our articles. Read our editorial process to learn more about how we fact-check and keep our content accurate, reliable, and trustworthy.
  1. Schindler S, et al. Predicting symptom onset in sporadic Alzheimer disease with amyloid PET. Neurology. 2021 Sep9:10.1212/WNL.0000000000012775. doi:10.1212/WNL.0000000000012775.

  2. National Institutes on Aging. What happens to the brain in Alzheimer's disease?.

By Valerie DeBenedette
Valerie DeBenedette has over 30 years' experience writing about health and medicine. She is the former managing editor of Drug Topics magazine.