Special Report: Alzheimer's Disease Dx
Vol 8, Issue 23 | December 7, 2023
In This Issue
We aren’t diagnosing enough pre-Alzheimer’s conditions
We need simpler Dx for Alzheimer’s
Is it Alzheimer’s alone?
DTC Alzheimer’s test raises concerns
Editors’ Note
This year we’ve reported on a number of breakthroughs in the diagnosis of Alzheimer’s disease - fueled in large part by the development of more promising treatments. We’ve seen a flurry of research and soon-to-be-commercial activity around both early diagnosis and diagnostics that differentiate Alzheimer’s from other neurodegenerative diseases. This issue we focus entirely on recent developments and continuing challenges in Alzheimer’s diagnostics. Next week, we'll focus on other neurological diseases.
We still aren’t diagnosing enough pre-Alzheimer’s conditions
We now have FDA-approved drugs for Alzheimer’s disease (AD): Esai/Biogen’s lecanemab (Leqembi™️) and Biogen’s aducanumab (Aduhelm™️), with donanemab from Eli Lilly and gantenerumab from Roche waiting in the wings. All of them demonstrate only marginal cognitive benefits and carry substantial side-effect risk (ARIA, a form of brain inflammation). The leading hypothesis for the relative ineffectiveness of these and all prior drugs that have been tried is that by the time an AD diagnosis is severe enough to justify treatment, it is already too late for the drugs to have much impact. However, that hypothesis hasn’t been tested.
Two recent papers estimate just how big the need for early diagnosis is - and how much potential benefit it would have. One paper, which used data from 226,756 Medicare physicians, found that just 8% of expected cases of mild cognitive impairment (MCI) were identified by primary-care physicians. (MCI is often the first step down the road toward Alzheimer’s). A second paper showed that MCI diagnosis had improved from 6.2% in 2015 to 7.9% in 2019. Given that of the 56 million over-65s in the US, about 20% have MCI (11 million) and a further roughly 9% (5 million) have clinical dementia, these two papers together show that better diagnosis of this age group would identify 8-10 million new and evolving cases each year.
Commentary: The benefit of objective and comprehensive early AD diagnosis cannot be overstated. Not only would it enable us to test whether current drugs would truly be more effective if used at earlier stages of the disease, but it would also allow us to test new treatments aimed specifically at those early stages. The hunt is on.
The pressure mounts: Better Alzheimer’s treatments require simpler diagnostics.
The first potentially curative treatments for Alzheimer’s disease were approved this year (see article above). These developments have dramatically increased interest in developing diagnostics that are less invasive than a CSF spinal tap, cheaper than imaging, and more accurate than either.
According to the Alzheimer’s Research UK report, a simple blood test would likely strike the right balance of acceptance and accuracy. The UK’s “Postcode Lottery” is in the process of awarding a £5 million “Blood Based Biomarker” research grant that will be funded in January 2024. (Trivia note: a postcode lottery is one where everyone living in a local area who bought a ticket shares a win).
Compounding the need for this type of test, subgroup analyses presented at October’s Clinical Trials on Alzheimer’s Disease conference uniformly demonstrated that the earlier treatment is started, the higher the benefit. If started early enough, it could possibly arrest or even revert dementia progression (for further details see Alzforum commentaries 11/8/23 and 11/18/23). The challenge is that the expense of the imaging tests that are the current diagnostic gold standard means they are rarely available to patients in the earliest AD stages outside a clinical trial. An inexpensive blood test would close this gap.
Is it Alzheimer’s alone, or is there more going on?
One of the big challenges of diagnosing Alzheimer’s disease is that not only do we not know the root causes for the cognitive decline and dementia seen in the illness, several other completely different diseases cause similar cognitive decline and are frequently present alongside traditional Alzheimer’s in the same patient (ALS/TDP43, Parkinson’s/α-synuclein, Lewy body disease, central amyloid angiopathy, etc.). A recent paper documented a technique that distinguished cases of pure Alzheimer’s from cases of Alzheimer’s complicated by other diseases with 84% accuracy compared to definitive diagnosis at autopsy. It used an AI model trained on demographic, genetic, clinical, brain imaging, and biomarker data during life.
Commentary: Research is generating more and more data to help diagnosis, from more and more sources, but no one type is conclusive alone. We now know of 80 different associated gene mutations (e.g., APOE ε4) as well as CSF and blood biomarkers (e.g., amyloid beta 42; phosphorylated tau), retinal changes, brain imaging defects, etc. The promise of AI is to integrate all these disparate clues and thereby accelerate effective clinical treatments.
Alzheimer’s DTC test is . . . DTC. And that makes some folks worried.
So . . . remember that direct-to-consumer (DTC) Alzheimer’s blood test we covered back in August? Two important caveats: 1) It’s not formally FDA-approved as a DTC test. The test is, however, based on the company’s physician-ordered test, which has been available since early 2022. 2) Independent physicians are available to discuss your test results at no additional cost.
Right now, there are only (as far as we can tell) two Alzheimer’s blood tests that have US regulatory imprimatur, and they’re each a little different from the consumer-initiated test. One, which is an LDT, is like the new test in that it looks for the ratio of two versions of amyloid protein, Aβ42 and Aβ40, in plasma; but the company notes that it also checks for “the apolipoprotein E protein profile (equivalent to APOE genotype).” The other, which has FDA approval, looks for levels of phosphorylated Tau (pTau) 181 protein along with APOE. (If you know of others, let us know.) Both are available through health-care providers only.
Commentary: According to a report in Medpage Today, the Alzheimer’s Association is concerned about the new DTC test. They worry that since even professionals cannot agree on how predictive of Alzheimer’s this or other tests really are, that lay people will either panic unnecessarily, and/or, at a minimum, won’t know how to interpret their results.
Our opinion: The DTC horse is well out of the stable by now, with many tests with probabilistic results. 1) Acknowledge the importance of people taking control of their own medical information. Someone self-motivated enough to order this test on their own is likely just as capable of finding out what the results mean. 2) Worrisome test result? Yes, you can check Dr. Google first, but you do that when you get results that your doctor ordered, too. And after that, who are you going to call? Your doctor or your health center. 3) In the (rare) case that someone who has hundreds of dollars to pay for a test out of pocket somehow doesn’t also have a health-care provider, the test company can provide you with a consult as part of the cost of the test.
DTC tests are here and they’re not going away. Do they need to be regulated to ensure best quality and minimize bad actors? We believe yes. Do health care providers need to accommodate them? Yes. Rather than wring our hands and complain that clinicians should be the only ones holding the keys, providers need to adapt and provide clients with the best possible explanation of the results they’ve received, wherever those results come from.
How do clinicians get educated on this? In med schools, curriculums are changing (slowly) to add genomics and diagnostics. This needs to speed up and include practical experience in interpreting these DTC tests. For practicing physicians, we believe that continuing medical education needs to be updated to reflect this current reality. For health care professionals - both clinical and industry folk - colleges, universities, and other grad-school programs need to teach the science, technology, and business of diagnostics as an independent discipline. (We are definitely biased here, as Mara is the co-founder of a master’s degree program in biomedical diagnostics.)
Quick Hits
Screening tools for dementia don’t need to be high-tech to be effective. The McCance Brain Care Score, validated in a study published last week in Frontiers in Neurology, is a short, simple questionnaire about routine test results (cholesterol, A1C, blood pressure), lifestyle (e.g., exercise habits, smoking history), and social / emotional status. Testing at least five points better than the median indicates a significantly decreased risk of either dementia or stroke over the next decade or so.