Bay Area researchers find Alzheimer’s disease diagnostic technique

Infographic describing study visually interpreting tau-PET to detect Alzheimer's
Joseph Casey/Senior Staff

Related Posts

Researchers from UC Berkeley, Lawrence Berkeley National Laboratory and UCSF looked at 274 unique brain scans to develop a method to diagnose Alzheimer’s disease, the leading cause of dementia, while patients are still alive.

Alzheimer’s disease is defined as the buildup of the tau and amyloid beta proteins in the brain, according to paper co-author and UCSF researcher Renaud La Joie. Researchers, and potentially physicians, can use the study’s methods to diagnose Alzheimer’s disease based on the amount of tau protein accumulation in a patient’s brain, according to Ida Sonni, a co-author of the paper and former researcher at the Lawrence Berkeley National Laboratory. 

“There are not a lot of people in the world who have looked at this many scans,” La Joie said. “There is something exciting and nice about sharing this expertise with other people.”

Researchers in the study analyzed tau protein accumulation by injecting patients with a partially radioactive chemical tracer, called flortaucipir, that binds to tau protein and is visible on positron emission tomography, or PET, scans, Sonni said.

Sonni and Orit Lesman-Segev, another co-author of the paper, then looked at 274 unique PET scans to determine which ones exhibited tau protein accumulation indicative of Alzheimer’s disease, according to La Joie.

“The publication of this paper is important in light of the approval of flortaucipir by the FDA,” Sonni said. “It can now be used clinically, so we need to give the tools to the clinicians to interpret, correctly, the scan.”

One research method for diagnosing Alzheimer’s disease is to use “complicated methods” to quantitatively measure tau protein accumulation, according to La Joie.

Clinicians, however, couldn’t quantify the amount of tau tracer when looking at PET scans — they needed a way to visually judge the amount of tau protein in a patient’s brain to diagnose Alzheimer’s disease, La Joie said.

“The framework for the images is to make sure that two people read the images the same and that the same person identifies them the same,” La Joie said. “The visual read performed as well as the quantification.”

Currently, physicians give a probabilistic Alzheimer’s disease diagnosis, according to La Joie. He added that, until recently, the only way a patient could be definitively diagnosed with Alzheimer’s disease was in an autopsy.

Flortaucipir is one tool which can help differentiate Alzheimer’s disease from other forms of dementia, according to Sonni. Additionally, she said it is crucial neurologists accurately diagnose their patients with Alzheimer’s disease so they can choose the appropriate therapeutic management strategy.

“We are definitely excited that this paper was published,” Sonni said. “We wanted to get this out in the medical community — we are happy and excited this is finally out.”

Eric Rogers is the lead research and ideas reporter. Contact him at [email protected] and follow him on Twitter at @eric_rogers_dc.