An estimated 14 million older adults in the United States live with mild cognitive impairment (MCI). While many will never progress to dementia — experiencing stable or even improved cognitive function — predicting the likely course of the condition has remained elusive. That is, until now.
Earlier this year, the US Food and Drug Administration (FDA) authorized BrainSee — the first noninvasive test for assessing the likelihood of progression from amnestic MCI (aMCI) to Alzheimer’s disease (AD) dementia — marking a significant leap from diagnosis to prognosis in dementia care.
How Does It Work?
BrainSee is an artificial intelligence (AI)–driven software application indicated as a prognostic tool to aid in the clinical assessment of how likely older adults who are diagnosed with aMCI are to progress to AD dementia within 5 years of assessment.
The BrainSee test and associated platform were developed by San Francisco–based Darmiyan, Inc. The FDA granted BrainSee breakthrough designation in 2021. It’s classified as a class II device.
According to the FDA, BrainSee analyzes inputs of routine brain MRI, patient demographic information, and cognitive assessments to generate a score between 0 and 100.
This score reflects the degree of similarity between the patient and two reference populations of patients with aMCI who either did not progress to clinical AD-dementia within 5 years (non-converters) or progressed to clinical AD-dementia within 5 years (converters).
A score < 30 indicates a low risk of progressing to AD dementia in 5 years, whereas a score of 30-70 denotes an intermediate risk and a score > 70 indicates a high risk.
“What doctors and patients care about most, especially in neurodegenerative diseases, is what’s going to happen in the future. Will the disease progress over time if nothing changes in the patient’s lifestyle or medications? The BrainSee test helps guide the doctor and the patient to decide next steps in a more informed way and be able to intervene early enough to reduce the risk of dementia and monitor risk over time,” Kaveh Vejdani, MD, Darmiyan’s co-founder and chief medical and technology officer, told Medscape Medical News.
“With a low-risk score, the doctor can provide reassurance and evaluate other potential causes of MCI while also recommending proactive and healthy lifestyle habits,” Vejdani explained.
“With an intermediate or higher risk score, they might be more proactive in terms of management of comorbidities and lifestyle habits and also follow up more frequently with additional testing,” he said.
The BrainSee test is not indicated as a stand-alone diagnostic or prognostic tool. It is intended to provide supplemental information for interpretation in conjunction with a standard neurologic assessment and could be combined with other measures such as amyloid and tau.
What’s the Research Behind It?
A preprint posted on bioRxiv.org, along with data presented at Clinical Trials in Alzheimer’s Disease conferences in 2020 and 2021, showed that the BrainSee algorithm can predict AD dementia in adults with aMCI with an accuracy of 88%-91%.
Vejdani noted that in a more recent study assessing performance accuracy of BrainSee in 700 adults with aMCI, sensitivity of the test was 81%, specificity 89%, positive predictive value 92%, negative predictive value 72%, balanced accuracy 85%, and area under the curve 95%. These data have not been published yet but will be shortly.
Other data awaiting publication showed that BrainSee outperforms alternative methods, including Mini-Mental Health Exam and Clinical Dementia Rating Sum of Boxes scores.
The FDA requires postmarket testing be conducted in patients of diverse race, socioeconomic status, and levels of education, who are from community-based care settings and who have comorbidities that may affect the effectiveness of the device.
How It Works
Physicians can access the BrainSee platform via an online portal.
A separate portal allows patients or their family members to book an initial televisit with a BrainSee provider who will facilitate the process. The company also recently launched DarmiSpace, a comprehensive and holistic brain wellness platform.
All the elements of the BrainSee test are covered under Medicare. People with other insurance plans should consult their insurance provider to understand the coverage specifics.
Caveats and Cautionary Notes?
Christopher Weber, PhD, Alzheimer’s Association director of global science initiatives, told Medscape Medical News that FDA approval of this tool “underscores its potential clinical utility, particularly in early intervention and treatment planning, where tools that use emerging technologies, like AI, may support doctors in clinical settings.”
“What will happen next and what should I/we be doing now are very important questions for patients and their families following an Alzheimer’s disease or MCI diagnosis. This is why it’s critical to continue to invest in research that uncovers the earliest signals that indicate disease trajectory,” Weber said.
“As with any technology, it’s important to understand its predictive accuracy across diverse populations, and that it is evaluated and validated in real-world care on large and representative patient populations. We look forward to seeing published data on BrainSee to help answer these questions,” Weber added.
The authors of JAMA Neurology viewpoint article noted that clinicians and patients will likely be attracted to a prognostic tool for AD that provides a “simple yet powerful metric like a percentage.”
However, they said it will be important for clinicians and patients to be cautious about how much weight to place on a prediction while the accuracy of these devices is confirmed, which could take several years.
“At the moment, there is not a great amount of evidence in the public domain about this device and that makes it challenging to judge how well it works,” first author James Hillis, MBBS, DPhil, Department of Neurology, Massachusetts General Hospital, Boston, noted in an interview with Medscape Medical News.
Hillis said a “key task for neurologists will be to explain the implications of the probabilities generated by the test to their patients and help patients translate these numbers into clinical decisions.”
More broadly speaking, Hillis said more AI-powered devices will likely be coming down the pike to aid in both diagnosis and prognosis of AD as well as other conditions.
“That’s incredibly exciting because AI can consider much more data than is possible for clinicians to consider. A key challenge, however, will be figuring out how to use these devices effectively in clinical practice,” Hillis said.
Vejdani is an employee of Darmiyan. Hillis reported receiving institutional grants from Annalise.ai, Cerebriu, GE Healthcare, Heuron, and Viz.ai. Weber had no relevant disclosures. Darmiyan is a member of the Alzheimer’s Association Business Consortium.
