Predictive Model Accurately Tracks Alzheimer's Disease Progression

Researchers from the University of Texas at Arlington (UTA) have developed a model to predict how a patient’s Alzheimer's will progress over time, according to a study published recently in Pharmacological Research.

The National Institute on Aging (NIA) estimates that more than six million adults in the United States may have Alzheimer's disease. The condition is the most common cause of dementia in older adults, and is the seventh leading cause of death in the US.

Because Alzheimer’s is characterized by a loss of cognitive functioning over time, it is also a major cause of disability. As the disease worsens, patients require increasing levels of caregiver support.

Helping Alzheimer’s patients and their caregivers anticipate these additional care needs can ease some of the physical, psychological, economic, and social challenges that come with the disease’s progression.

However, predicting that progression can be difficult.

To address this, the research team built a novel learning-based framework to pinpoint where a patient is within the Alzheimer's disease-development spectrum. Doing so could enable patients, caregivers, and clinicians to better predict the timing of the later, more severe disease stages, which would bolster future care planning.

“For decades, a variety of predictive approaches have been proposed and evaluated in terms of the predictive capability for Alzheimer’s disease and its precursor, mild cognitive impairment,” explained Dajiang Zhu, PhD, an associate professor in computer science and engineering at UTA, in a news release. “Many of these earlier prediction tools overlooked the continuous nature of how Alzheimer’s disease develops and the transition stages of the disease.”

The framework is designed to code five stages of Alzheimer’s disease development – normal cognition, significant memory concern (SMC), early mild cognitive impairment (EMCI), late mild cognitive impairment (LMCI), and Alzheimer’s disease – using a “disease-embedding tree” (DETree).

DETree represents the five clinical stages and their trajectory as a tree to represent Alzheimer’s progression. By projecting an individual patient’s data onto the tree, their clinical status and trajectory can then be predicted.

To test the framework, the researchers pulled data from 266 patients in the Alzheimer’s Disease Neuroimaging Initiative cohort. The tool was then compared to existing models designed to predict Alzheimer’s progression.

DETree was able to efficiently and accurately predict all five of the clinical stages across the cohort, whereas other models were limited to two clinical stages. The DETree framework also provided more in-depth information about each patient’s Alzheimer's status by projecting where within the disease trajectory they will be in the future.

“We know individuals living with Alzheimer’s disease often develop worsening symptoms at very different rates,” Zhu said. “We’re heartened that our new framework is more accurate than the other prediction models available, which we hope will help patients and their families better plan for the uncertainties of this complicated and devastating disease.”

Moving forward, the researchers plan to investigate the tool’s possible utility in predicting the progression of other multi-stage diseases, such as Creutzfeldt-Jakob disease, Huntington’s disease, and Parkinson’s disease.

Predicting Alzheimer’s disease progression is just one piece of the puzzle. Researchers are also attempting to detect signs of mild cognitive impairment long before the onset of late-stage neurodegenerative diseases.

Experts from Indiana University Health and the Davos Alzheimer’s Collaborative (DAC) sat down with HealthITAnalytics last year to discuss a pilot the two are running at the health system: a program using artificial intelligence (AI)-based digital screening tools to support the early detection of cognitive impairment in the primary care setting.

The pilot is designed to shift cognitive care from reactive to proactive by making screenings more accessible and effective for both patients and clinicians.