SENSible Question: A piece from the LA Times is critical of a proposal to test people for accumulated beta-amyloid and then provide interventions for those who test positive, even if they are still cognitively normal. The article claims a majority of people with normal cognition who test positive “may never be diagnosed with dementia.” What’s your take on this criticism?
Short summary: A recent article criticizes the diagnosis of Alzheimer’s in its earliest stages based on the presence of beta-amyloid or other aging damage in the brain, on the basis that many people with normal cognitive function who test positive “may never be diagnosed with dementia.” We explain why this is a naïve way to think about Alzheimer’s and other diseases of aging that — if followed through — would condemn many people to the senseless loss of their independence and identity.
The first thing to understand about this claim is that many (and until relatively recently, all) of the studies that people point to as evidence that people can have substantial levels of beta-amyloid in their brains and “never be diagnosed with dementia” only addressed the question indirectly. These studies looked at the brains of people after their death and measured the most readily-detected form of beta-amyloid aggregates: the insoluble beta-amyloid plaque. One problem with such studies is their retrospective, cross-sectional design: because they were performed on dead people, you can’t test their cognitive function and then follow them up to see what happens to it over time. Instead, because they often found people who had died with plaque in their brains had never been diagnosed with dementia during their lifetimes, researchers drew the same conclusion about living people walking around in the community.
In addition to the problem of the studies being retrospective, beta-amyloid plaque is not likely to be the critical driver of dementia in Alzheimer’s. It may even have a partially protective role as a kind of “sink” for the more toxic forms. Instead, as I’ve emphasized in a number of blog posts, it’s the soluble species of beta-amyloid, such as oligomers and protofibrils, that interfere with neuronal function and drive downstream damage. So if these older studies were all there were to the claim that people with beta-amyloid in their brains “may never be diagnosed with dementia,” it would be a rather shaky foundation to reject the efforts of a serious group of scientists and patient advocates.
But the claim at issue is still broadly speaking correct when we look at more recent studies that detect beta-amyloid or aberrant tau in living people and then follow them up to see what happens to them years later. This involves looking for beta-amyloid or aberrant tau in the brain using PET scan tracers, or collecting these damaged proteins in spinal tap fluid, from which researchers can infer what’s happening in the brain. And as the article says, many people do test positive for Alzheimer’s-driving molecular damage and yet never fall into dementia in their lifetimes. For example, a 60-year-old person who tests positive for beta-amyloid using these biomarkers but who is otherwise at typical risk will have “only” a 31% chance of developing Alzheimer’s dementia during his or her lifetime.
So the Times piece is correct on this narrow point — at least under current life expectancies. But the article is misguided in its largely negative slant on the Alzheimer’s Association Working Group proposal. In a bit more detail, the Working Group proposes that doctors begin testing aging people for beta-amyloid burden or aberrant tau whether they have symptoms of major cognitive decline and classify those who test positive as being in one of various stages of “prodromal Alzheimer’s” (that is, stages of the neurodegenerative process that precede frank Alzheimer’s dementia). A particular pain point for the Times is the proposal to diagnose people who have “normal” cognitive function for their age but have Alzheimer’s-driving molecular damage in their brains as being in “Alzheimer’s disease Stage 1,” even though (as the Times piece emphasizes) most of these people will not be expected to become demented within current life expectancies.
In brief, the Working Group proposal is solid. It aligns with what we’ve learned about neurodegenerative aging of the Alzheimer’s type over recent decades and the lessons of the recent Phase III clinical trials for lecanemab, donanemab, and other AmyloSENS therapies, and it prepares the healthcare system and each of us aging people for the fruits of trials that are in process today. Failure to seize the moment and implement this proposal will leave untold numbers of aging people to lose their independence, their memories, and their very identities.
Let’s dig into two separate underlying questions provoked by the Times piece, the first of which will inform the second. First: why would it be that many people with detectable beta-amyloid in their brains would develop dementia? And second: armed with that information, what should we do about it? As we shall see, the latter turns on a number of additional provocative points raised in the Times article.
Damage and Dementia
Within the first question, I will highlight three reasons rooted in the biology of aging and the specifics of neurodegenerative aging of the Alzheimer’s type.
The first and most brutally straightforward reason why a person can test positive for accumulated beta-amyloid or aberrant tau and never develop dementia is that people may simply die before they succumb to it. A paper that the Times links to argue its case lays this out clearly: at every step along the way between the first appearance of beta-amyloid or aberrant tau in the brain and outright dementia, there is a chance that a person will be killed by something else first. This is particularly likely to happen granted that the drivers of both dementia and the overwhelming majority of deaths are aging processes, and that so either one can play stalking horse for the other — and even more so since Alzheimer’s itself is the fifth leading cause of death in people over the age of 65, and some evidence suggests that people enter into a dangerous wasting state in the years leading up to a formal Alzheimer’s diagnosis.
Of course, few people would choose early death as their preferred method of “not developing dementia.” More importantly, that brutal math would change drastically if we were to make substantial headway against other causes of age-related death and didn’t simultaneously seize the moment to press the fight against Alzheimer’s. As the source that the Times piece cites for its core claim itself notes, “if mortality rates decline in the future, and all other factors remaining the same, then the lifetime risks reported here would underestimate actual risks.” Bringing other rejuvenation biotechnologies to bear in addition to beta-amyloid clearing AmyloSENS therapies would be such a game-changer, but even lengthening life expectancy by way of widely-anticipated metabolism-based medical therapies against cardiovascular disease such as new lipoprotein(a)-lowering drugs, wider use of SGLT2 inhibitors, and a new CETP inhibitor increase the “lifetime risk” of Alzheimer’s dementia — if we do nothing to prevent that eventuality.
Equally, making real progress against neurodegenerative aging of the Alzheimer’s type would also increase life expectancy, precisely because Alzheimer’s kills so many people. And the Working Group proposal is a key tool in achieving that much better solution to the despair over Alzheimer’s that saturates the Times piece.
Putting aside these fatal “escapes,” the second and third reasons why a person may never develop dementia despite having measurable beta-amyloid or aberrant tau in their brain both relate to the nature of the causal relationship between Alzheimer’s-driving molecular damage and dementia. Something can be causal for something else without it being the case that the cause always results in the effect. Smoking causes lung cancer, and 87% of people with lung cancer are smokers, yet most smokers never develop lung cancer in their lifetimes.
A closer analogy is atherosclerotic plaque. Having plaque in one’s arteries certainly causes the great majority of heart attacks and strokes, but many people with atherosclerotic lesions in their arteries never wind up suffering either. And strikingly, when hospitals and medical examiners autopsy men who died of other causes, they find that approximately one in ten men in their 20s, 30% of men in their 30s, 40% of men in their 40s, and so on up to 80% of men in their 70s have microscopic cancers in their prostates — yet the lifetime risk of clinically-diagnosable prostate cancer is only 10%, and the risk of dying of it is only 3%.
So putting aside people with amyloid in their brains who are “never diagnosed with dementia” because of an intervening death from a heart attack or a hip fracture from which they never recover, what might prevent this particular cause (beta-amyloid) from leading to this particular effect (dementia) in a given person? To put it in logical terms: if beta-amyloid is necessary to the development of Alzheimer’s dementia specifically, why isn’t it sufficient?
The first of two reasons for this split between a cause and its effect turns on how beta-amyloid drives a person into dementia. Although many studies report that beta-amyloid oligomers can impair neuronal function directly, the critical step in the “Amyloid Cascade” occurs when it drives damaged forms of the protein tau to spread beyond the neurons of the medial temporal lobe into the neocortex, the seat of memory and identity. So if you detect beta-amyloid in a person’s brain at a stage before it has triggered this outspread of tau damage, he or she will likely still have normal cognitive function for their age, and will continue to do so for a decade or more.
And many things entirely unrelated to beta-amyloid contribute to any given person’s propensity to accumulate abnormal tau, such as having high blood pressure variability (all the more so if they have the Alzheimer’s risk variant APOE ε4) or carrying genetic variants that predispose a person to form it. This implies that different people’s neurological “guns” vary in the degree to which they are “loaded” with propensity to spread tau pathology when beta-amyloid comes on the scene to pull the “trigger.”
The other causality-related reason why one person with Alzheimer’s-typical levels of beta-amyloid may fall prey to dementia while another person with similar levels never develops it is all the other aging damage that the brain of the first person on top of beta-amyloid and tau. Although people (including neurologists) talk about “Alzheimer’s disease” as if it were a “disease” with a single, straightforward cause like the measles, the reality is that people fall into dementia when their brains have suffered so much cellular and molecular aging damage of all types that they can no longer carry out normal cognitive and related functions. The vast majority of people develop neurodegenerative aging of the Alzheimer’s type as a consequence of degenerative aging, and the fearsome reality of aging is that multiple kinds of aging damage assault the brain all at once, interacting with and accelerating each other. People with particularly high levels of beta-amyloid will typically have the symptoms that align with the clinical diagnosis of Alzheimer’s dementia, but it’s rarely beta-amyloid alone that got them there.
Depending on their genetics and environmental exposures, any given aging person will not only vary in the amount of beta-amyloid they have in their brain, but also the amount of aberrant tau, alpha-synuclein aggregates, senescent cells, neurons overtaken by mitochondria bearing deletion mutations, strokes (including transient ischemic attacks and “silent strokes”), damage to the small blood vessels, infectious microbes, dysfunction of the protective blood-brain barrier, and degeneration of the insulating material for the “wiring” that links our neurons together into functional circuitry. People also vary in the amount of alcohol they consume, which causes atrophy of their hippocampi (an area key to new memory formation and other brain structures, even at relatively modest intakes. And so on.
Thus, the great majority of diagnosed Alzheimer’s dementia* is actually a kind of mixed dementia, resulting from the cumulative ruin wrought on their brains from multiple kinds of cellular and molecular aging damage. The same is true for Lewy body dementia, vascular dementia, frontotemporal dementia, LATE-NC, and other age-related dementias. Conversely, people with high beta-amyloid but intact cognition will generally have low levels of other kinds of aging damage, and may have a high level of redundancy due to having been born with a high density of neurons or to having benefited from advanced education and a lifetime of cognitive stimulation to buffer them against age-related losses (so-called “cognitive reserve”).
In one study, for example, researchers examined the donated brains of volunteers in the Rush Memory and Aging Project, testing them for the levels of just four kinds of aging lesions: beta-amyloid plaques, aberrant tau in the form of “tangles,” alpha-synuclein aggregates, and evidence of strokes (not including strokes that they judged to have killed the person). Moreover, they only looked for these lesions in some of the “bad neighborhoods” of the aging brain. Even with these narrow criteria, only 28.6% of the older volunteers (average age 88) who did not have dementia were free of the molecular aging damage typical of Alzheimer’s, Lewy body, or vascular dementia — and conversely and only 30% of those with dementia met the criteria for Alzheimer’s disease exclusively.
The most striking finding comes not from these yes/no analyses of individual forms of molecular aging damage, but from looking at all four forms at once. What separated the people who died with dementia from people who suffered only “normal” age-related cognitive decline was often not the presence of any one form of molecular aging damage, but the interface of several of them. While 80% of those who died free of dementia met the criteria for only one such diagnosis or none at all, none of the people who died with dementia were free of all four of these molecular aging lesions, and over half had molecular aging damage that would simultaneously qualify them for more than one different dementia diagnosis (Alzheimer’s, Lewy body, or vascular). In fact, after they adjusted for the fact that people with dementia in their sample were somewhat older than people without it, people with molecular aging damage typical of more than one different kind of dementia were nearly three times as likely to actually have clinical dementia than people who met the criteria for only one of them.
Similarly, in a study of 382 subjects who died with dementia and had donated their brains to the State of Florida Brain Bank, two-thirds of subjects who died diagnosed with Lewy body dementia also had enough beta-amyloid plaque and tau tangles in their brains sufficient to diagnose them with Alzheimer’s; so did 77% of vascular dementia patients and 66% of patients with hippocampal sclerosis.
The fact that biomarkers of beta-amyloid and aberrant tau don’t immediately correspond with or fully predict whether any given person will eventually develop dementia is thus not an argument against testing for this damage, but an argument for using additional tests to capture other kinds of cellular and molecular damage to further winnow down risk, and for developing additional tests to assess damage in the brain that we can currently only look at after a person has died. Indeed, the draft report has a table of existing followon tests for the former purpose, and scientists are working to develop new and better tests every day.
Recognizing All the Harms
The other thing to note about the Times’ objection that most people with “normal” cognitive function for their age will not develop dementia in their lifetime is that it implicitly suggests that beta-amyloid and aberrant tau are harmless if they don’t cause a person to meet the clinical criteria for dementia. This just isn’t true.
First, beta-amyloid and other forms of aging damage are also implicated in driving “normal” age-related cognitive decline. Because this decline is universal, it is treated as both inevitable and irrelevant. In reality, anyone whose profession or leisure-time activities involves the mind loses quality of life as his or her verbal fluidity, abstract reasoning, mathematics, spatial relations, and ability to focus on a task and task-switch declines with age. “Normal” age-related cognitive decline is also doubtless costing the economy and individual knowledge workers in lost productivity, although in most jobs, those losses are counterbalanced during the currently-standard working years of 20-65 by more crystalized intelligence and changing job responsibilities, making it hard to parse out the effect of the fluid intelligence losses. It may even affect the administration of justice.
Secondly, people who are cognitively normal for their age but have signs of amyloid damage in their brains based on fluid from a spinal tap suffer faster cognitive decline over the next few years. And if they have an elevated ratio of a particular form of aberrant tau to beta-amyloid, they are nearly half again as likely to develop so-called “mild” cognitive impairment as are people with a normal ratio, and it’s likely to happen within the next seven years. (Despite the name, MCI is a substantial and disabling level of cognitive disability).
The predictive power of these signs of molecular aging damage is even stronger in people with subjective cognitive decline (SCD) — that is, people who score as normal on standard tests of cognitive function for a person of their age, sex, and education, but have the nagging sense that their thinking and planning is going downhill and find themselves upsettingly confused at times. People with SCD who have biomarker evidence of beta-amyloid accumulation alone are at nearly six times the risk of progressing into MCI or dementia as are people with SCD who are free of this biomarker — and if they have both beta-amyloid and abnormal tau biomarkers, then they are at a shockingly excess eleven times greater risk. Conversely, people who are free of beta-amyloid evidence, with or without evidence of abnormal tau, are at very low near-term risk.
In turn, people who already have MCI who have both beta-amyloid and abnormal tau in their fluid are twelve times as likely to progress into dementia of the Alzheimer’s type as people with “all-clear” biomarkers using the criteria established by the National Institute of Aging (NIA) and the Alzheimer Association.
So testing Alzheimer’s-related molecular damage detected in biological fluids or on PET scans gives insight to the degenerative aging of the brain and adds useful medical information, even if people don’t have dementia and some may never suffer it.
Misguided Medical Misgivings
The article appears to have two main kinds of rational concerns about testing people who have “normal” cognitive function for their age for signs of Alzheimer’s-driving molecular aging damage and the “Alzheimer’s Stage 1” diagnosis for people who have beta-amyloid aggregates or aberrant tau in their brains but are cognitively normal for their age. One set of such concerns is medical. For instance, the article cites the American Geriatrics Society’s (AGS’s) worry that the Alzheimer’s Association’s Working Group proposal “will place many older and multimorbid people at risk of overdiagnosis, which in turn could lead to initiation of treatments with limited benefit and high potential for harm in this population.” The article notes that “While a study of Leqembi’s effects on asymptomatic people has begun, there is currently no evidence that giving it to people without cognitive impairment can reduce the risk of dementia or delay the onset of Alzheimer’s symptoms.”
While it’s true that the Leqembi® (lecanemab) Alzheimer’s prevention trial and others are only getting started, the article’s grave qualms are not just premature, but run contrary to available evidence and are ultimately likely to prove counterproductive. Deeper dives into the Phase III trials for the AmyloSENS therapies lecanemab and donanemab have clearly shown that clearing beta-amyloid works better the earlier therapy is begun. The younger a subject is when he or she first gets beta-amyloid-clearing treatment, and the lower their burden of downstream aberrant tau, the more effectively amyloid-clearing therapies slow cognitive decline and the progression of downstream non-amyloid damage in the brain.
Conversely, the fact that a person’s brain is already a multifactor trainwreck by the time he or she is diagnosed with dementia makes it all the more remarkable that removing beta-amyloid alone at this late stage is enough to slow the rate of further cognitive decline substantially. As we’ve noted before, in addition to beta-amyloid and aberrant tau and the other molecular damage we discussed above, people suffering with AD also have elevated numbers of senescent neurons in their brains — a burden even higher than that of age-matched control subjects. The same is true for the number of their neurons that have been hobbled by mitochondria that bear deletion mutations. And shockingly, they have already lost roughly a third of the neurons outright in the region of the brain that transfers short-term memory traces up from the hippocampus to the neocortex for long-term storage.
A decade ago, this kind of evidence mobilized a groundswell of scientists to call on regulators and companies to recruit people into Alzheimer’s trials before it’s too late for the therapies to be effective on their own. And the fact that Eisai, Biogen, and Lilly enrolled people into their lecanemab and donanemab trials when the volunteers were at an earlier stage in the disease process than were the people in any previous AmyloSENS trial clearly seems to have been key to their success.
So while it will be some years before we have gold-standard Phase III trial evidence that “people without cognitive impairment can reduce the risk of dementia or delay the onset of Alzheimer’s symptoms” by taking AmyloSENS therapies to clear beta-amyloid, all the evidence points that way.
But arguably, the critics are getting ahead of themselves. The AGS worries that testing people who are cognitively normal for their age for beta-amyloid aggregates or aberrant tau would lead doctors to indiscriminately treat those of them that are positive with amyloid-removing therapies before such trials are complete “will place many older and multimorbid people at risk of overdiagnosis, which in turn could lead to initiation of treatments with limited benefit and high potential for harm in this population”. But this is explicitly not what the Working Group is proposing. Instead, the Working Group proposes only that clinicians use these tests in “determining eligibility for treatments targeting core disease pathology based on drug registration criteria” — that is, based on the criteria for which the AmyloSENS therapies are FDA-approved — namely, having actual Alzheimer’s dementia.
And as to the worry about the worry about the “high potential for harm” in “older and multimorbid people at risk of overdiagnosis,” the Working Group says:
We do not advocate initiating treatments targeting core AD pathology in all symptomatic persons with biologically confirmed AD without regard to clinical context. Rather we emphasize that treatment in symptomatic individuals with biologically proven AD should be based on clinical assessment of risk/benefit at the individual patient level.
And, finally, to the extent that the Times, the AGS, or anyone else worries about the risks associated with beta-amyloid-clearing antibodies, it is likely the case that these therapies will be safer when they are given to people earlier in the process than these therapies are currently licensed to treat. The major side-effect of these therapies is amyloid-related imaging abnormalities (ARIA), which are brain swellings or bleeding in the brain. While these side effects sound alarming and are sometimes life-threatening, they are common in people suffering with neurodegenerative aging of the Alzheimer’s type even without therapy, and in most cases pass with no symptoms or apparent effect if the dose is tapered down or withdrawn.
ARIA are thought to result from either injury during the movement of the antibody-amyloid complex across the brain’s protective barrier, or from some other disruption to the amyloid burden in the brain’s blood vessels. Thus, you might expect that the earlier you receive an amyloid-clearing AmyloSENS therapy, the less likely it will be that you would get ARIA, because there would be less beta-amyloid in the brain to clear and less baseline damage to the vessels. And indeed, while the definitive evidence will again come from the new Phase III trials, the additional analysis of the donanemab trials presented at the 2023 CTAD conference found that people with beta-amyloid burden at the beginning of the trial were at lower risk of the more common brain-swelling variety of ARIA.
Similarly, other risk factors for ARIA included poorly-controlled high blood pressure, having existing microbleeds in the brain, or evidence of old injuries there — all less likely to occur the earlier in the “disease” process and the younger the chronological age of the person receiving therapy.
Keeping Medicine and Social Policy In Their Own Lanes
The second group of rational concerns the Times piece raises are social and bioethical ones. One is the worry that people may be denied health or life insurance, or subject to employment discrimination, or required to disclose personal health information on the basis of such testing. These seem like reasonable concerns.
Instead of opposing people’s ability to get information about their own medical risks and the opportunity to access medicines to hold dementia at bay, people and interest groups concerned about these outcomes should be advocating for protecting people’s right to access insurance without discrimination and to keep their medical information private — and above all, to access damage-repair therapies should they prove effective for prevention, in order to let people neutralize their risk while keeping insurance companies’ risk pools in balance.
Another rational-sounding concern in the article is the psychological impact that a diagnosis of “Stage 1 Alzheimer’s disease” based on molecular damage biomarkers might have on people: “People with no memory problems who learn they are positive for abnormal levels of amyloid or tau proteins can suffer from depression, anxiety and thoughts of suicide, studies have found.”
But in fact, the studies reviewed in the paper that the story links under “studies have found” do not support this concern. Most of the reviewed papers are speculations by bioethicists or medical professionals about how people might react to being told they had beta-amyloid or other molecular damage in their brains, or surveys of the public asking them how they think they might react to such a disclosure. And despite the Times’s well-meaning concern, most people who participated in the survey-based studies said they would want to know if they tested positive. One such study with a randomized design found that substantial majorities of people would be willing to participate in a hypothetical study that involved biomarker testing whether they would be told the results or not, with slightly more people expressing willingness to join if they would be told (70%) as if not (61%).
In some of the survey-based studies, a minority of people say they wouldn’t want to get tested because they worry that they would find a positive finding upsetting. But people declining to be tested because they’re worried they about what they might learn is quite different from people actually suffering anxiety or depression in response to a positive finding. And of course, no one is advocating mandatory testing of people who don’t want to get tested, any more than anyone advocates for mandatory cancer screening or mandatory testing for blood pressure.
And when we turn to the four studies where people were actually scanned or spinal-tapped for signs of Alzheimer’s-driving molecular damage, the wave of psychological harms that the Times and its bioethicist sources imagine prove to be grossly overblown. In the largest study, scientists used brain scans to look at beta-amyloid burden and brain shrinkage in people over the age of 65 who were cognitively normal for their age and enrolled in a clinical trial testing the effects of exercise on such damage. More than one in four of the volunteers had such damage evident in their scan. After researchers told people whether or not they had Alzheimer’s-driving molecular damage in their brains, people with brain amyloid detected had no worse depression scores than did those without. And while people with a positive scan expressed a slight increase in anxiety scores upon first being told, their anxiety levels returned to baseline or were actually better than where they started after a six-month followup. The investigators concluded that informing people of their amyloid status “has a low risk of psychological harm.”
A caveat to that finding is that the volunteers in this trial were given counseling about the implications of the test before and after the researchers informed them of their status, which might have helped to allay any psychological distress. One might reasonably worry that outcomes could be less sunny if health systems began offering widespread testing in the community without such counseling.
Rather than grasping at this nuance as a reason to reject testing people who are cognitively normal for their age, those who are concerned for the psychological well-being of people who are tested should work to ensure that such people have similar access to counseling. 23andme has proven that this can be scaled using telemedicine as part of a modestly-priced low-barrier consumer test for the Alzheimer’s risk genetic variant APOE4; there is no obvious reason why it similar counseling could not be bundled with testing for Alzheimer’s-driving molecular damage. In fact, one of the other four studies simply provided test-takers with an informational brochure about the test; the volunteers said they found it useful, and again there was no evidence of psychological distress from receiving the results. (Conversely, the people who tested positive said the result motivated them to adopt more healthy lifestyle habits).
In a third study, people who were tested and their caregivers uniformly expressed relief at their result, whether it was positive or negative. And in one of the smallest studies, this one in people with MCI rather than people with normal cognition for their age, only two out of eight people who tested positive for beta-amyloid said they suffered emotional distress such as sadness and worry after learning the results. Counterbalancing such distress,
Patients reported that they experienced advantages after the disclosure, such as information about their health status, the possibility of making practical arrangements, medication, enjoying life more, and a positive impact on relationships.
The final thing to say about this concern about the psychological impacts of widespread testing is that to whatever extent people might experience sadness or anxiety upon learning that they have the molecular damage that leads to Alzheimer’s disease in their brains, it seems reasonable to assume that much of that distress would be due to the fact that most people believe that nothing can be done about an Alzheimer’s diagnosis — even the presymptomatic, damage-based “Alzheimer’s Stage 1” diagnosis that the Working Group is proposing.
This fatalism was excessive even at the time that the studies we’ve just reviewed were conducted; there is much less justification for it today, and it would be actively harmful if it were to interfere with the efforts of the Working Group to line up testing and treatment. At the time that the Times article and its cited review article were written, we already had AmyloSENS therapies that slow the slide into darkness for people in the early stages of actual dementia. And the Working Group proposal is an effort to ensure that people who have beta-amyloid or tau aggregates in their brains can initiate treatment with damage-repair therapies while their cognition is still intact, to keep as many people as possible from falling into dementia in the first place.
As we have just seen, the available evidence finds very little psychological harm to people who learn they have Stage 1 Alzheimer’s even in the absence of an effective therapy; how much less of a negative impact will it have when we have such therapies available?
This hand-wringing is reminiscent of some of the social concerns that people raise about greatly-extended lifespans arising from postponing age-related disease and debility with rejuvenation biotechnologies: a crisis in Social Security, cultural stagnation, gerontocracy, and so on. The way to prevent this parade of horribles is not to put roadblocks in the way of developing therapies to keep people healthy and prevent them from suffering Parkinson’s, and muscle wasting, and hip fractures, and multiple kinds of blindness. Instead, people concerned about these consequences should work to reform the legal and social structures that have been built up around the assumption that everyone will age, become disabled, and die on a predictable threescore-and-ten schedule.
Reports of Alzheimer’s Demise are Greatly Premature
The Times piece also notes that the proposal “comes amid a decades-long decline in the risk of dementia,” apparently suggesting that it doesn’t make sense to start rolling out an aggressive testing and treatment regimen when the disease at issue is in retreat. There are three problems with this line of argument.
First, while some reports do find that overall dementia is, in a sense, in decline, this is not true of neurodegenerative aging of the Alzheimer’s type specifically. The prevalence of dementia has fallen after adjusting for the age of the population, but this has been driven by our great progress in preventing and treating ASCVD (and specifically stroke), and thus beating back vascular dementia. Better control of pollution and infection, more education, and lifestyle change also likely played a role. But even after adjusting for the aging of the population, the incidence of Alzheimer’s specifically is either depressingly flat or possibly even on the rise in older people.
On top of that, there is evidence of an increase in new cases of early-onset Alzheimer’s, meaning the unusual cases that occur before the age of 65, according to a report from the insurance company Blue Cross Blue Shield. Walter Rocca, a neurology researcher at the Mayo Clinic, has reported the same thing. Because early-onset is uncommon, many of the studies that have looked at the trajectory of Alzheimer’s incidence over time (including the study we just cited above) have neglected to look at cases in people under the age of 65, and have therefore missed an important part of the story.
Clearly, it makes no sense to preempt testing and anticipated treatment of Alzheimer’s disease because of a decline in the incidence of non-Alzheimer’s kinds of dementia. Indeed, the fact that we are seeing more people fall prey to the disease at an age where no one is expecting it is another argument for testing people before they suffer substantial cognitive decline.
Second, while the rate at which older people are being diagnosed with Alzheimer’s at a given age is overall staying flat, the total number of older people is increasing due to our successes in combatting cardiovascular disease and other specific causes of death. As a result, the total number of people suffering with Alzheimer’s is unambiguously going up, as is the share of them in the overall population, and it will continue to do so for decades to come if nothing changes. This adds societal urgency to the call to test and (when we can) to treat, so as to prevent a tsunami of personal grief and burden on the economy and a healthcare system that is already showing strain. Indeed, projections from 2018 — before we even had proven AmyloSENS interventions against beta-amyloid — show the enormous potential of such therapies to hold the line against the coming wave.
And finally: even if the incidence of Alzheimer’s really was declining — well, so what? Would that population-level decline comfort the person who is starting to get lost at the mall and can’t recognize her son? Deaths from cardiovascular disease and cancer have been on the decline for decades, as has smoking with all its attendant harms; should we take that as a signal that any further efforts on these fronts are overkill?
We are on the brink of finally being able to do something about neurodegenerative aging of the Alzheimer’s type. The Times is calling on us to abandon the fight before even firing a shot.
Play Where the Puck is Going, Not Where it Is
In response to the Times’ inquiry, the Alzheimer Association’s Chief Science Officer said, “The purpose of this initiative is to advance the science of early detection and treatment. In order to prevent dementia, we need to detect and treat the disease before symptoms appear.” The evidence already shows that even in people who are already suffering from MCI or early dementia, clearing beta-amyloid sooner rather than later is safer and more effective than removing it later. And three early-intervention trials are already underway to prove out the clear implication that starting clearance before cognitive impairment gets a chance to set in will work even better, delaying the onset of dementia and buying time for additional rejuvenation biotechnologies to emerge and clear a wider range of aging damage in the brain. So policymakers and clinicians must be ready to seize the moment when these trials cross the finish line.
Policymakers should, at minimum, be setting up the human and technological infrastructure and reimbursement policies to accommodate a surge in newly-eligible aging people. As things stand, people with actual Alzheimer’s dementia are having a hard time accessing AmyloSENS therapies because no one was ready when the first such therapies succeeded. Pre-identifying people at risk and staging these resources now will save minds and lives when these therapies are licensed to treat them.
But it seems unlikely that the author or editor of the Times piece and some of its interview subjects are likely to embrace the Working Group proposal even after their more rational-sounding objections are addressed. They seem nonspecifically uncomfortable with the very idea of testing and treating people who have no obvious signs of looming dementia other than evidence of Alzheimer’s-driving molecular damage in their brains. University of North Carolina neurologist Dr. Andrea Bozoki is “troubled” by “taking a bunch of people who may never develop dementia or even cognitive impairment and you’re calling them [Alzheimer’s disease] Stage 1. That doesn’t seem to fit.” Dr. Madhav Thambisetty at the National Institute of Aging has what the Times characterizes as “serious bioethical concerns that come with testing healthy people for signs of Alzheimer’s.” Dr. Eric Widera objects to “redefining what it means to have Alzheimer’s” such that “You no longer need to have cognitive impairment to have this disease. You just need the positive blood test [for Alzheimer’s-driving molecular damage in the brain].”
But even within the limits of today’s metabolism-targeting medicine, clinicians routinely apply this same approach to preventing terrible outcomes from other diseases of aging. Just as not every person with beta-amyloid or aberrant tau in their brains will develop dementia within their lifetimes, not single person with chronically high blood sugar is fated to develop vision problems, or failing kidneys, or erectile dysfunction, or dementia, or any other specific complication of diabetes. Much like people with normal cognition for their age but with amyloid in their brains, people in the early stages of diabetes are typically still fully functional and feel “normal.” But as soon as doctors recognize that a person has chronically high blood sugar, they diagnose him or her with diabetes and begin giving them lifestyle counseling and medications to bring their blood sugar down. We intervene before they develop complications because we know that if they go without treatment, they are at high risk of developing one or two or all of them. By intervening while a person still seems healthy, we prevent these catastrophic outcomes from happening.
Similarly, we don’t wait until a person has congestive heart failure to diagnose him or her with hypertension and bring their blood pressure down. We don’t wait until a person has been rushed into the ER with a heart attack to diagnose them with hyperlipidemia and give them a statin. We don’t wait until we see a suspicious speck on a low-dose lung CT scan to counsel a person to quit smoking. Again: we make a diagnosis, and we act to prevent those catastrophic outcomes from happening. Why would we do any differently when a person has Alzheimer’s molecular aging damage in his or her brain?
In fact, the case for making diagnosis and treatment decisions based on biomarkers for beta-amyloid and aberrant tau is even stronger than it is for these various metabolic abnormalities. High blood sugar, high apoB-containing lipoproteins like LDL, and high blood pressure are only relative increases in aspects of normal metabolism: we would die if any of them fell to zero, and there is a relatively narrow and contested gray zone between the levels of these metabolic risk factors that are considered “normal” and the levels that are considered high-risk and necessitating some kind of intervention. We base thresholds for intervention on how much a given level of a metabolic actor increases one’s risk of “hard,” damage-based outcomes and on how risky it is to push them lower with available drugs.
Of necessity, all of this is indirect: we try to slow the accumulation of damage in our tissues by way of inhibiting otherwise health-sustaining metabolic processes that unintentionally inflict it. By contrast, the Working Group proposal — consistent with the SENS “damage-repair” strategy — is to test for stable molecular damage that should not be present in the brain, make a diagnosis on the basis of the extent of that damage, and (once the trials show it’s safe and effective) to remove the damage before it reaches a level that causes dementia.
So the Times’ objections are actually even more unreasonable than we’ve thus far suggested. Let us follow the analogies we’ve already reviewed one step further, to the point where actual structural damage is present in the aging body. LDL and other apoB-containing lipoproteins are the root metabolic driver of atherosclerotic cardiovascular disease (ASCVD), accelerated by additional insults such as high blood pressure, smoking, and insulin resistance. Today, clinicians help people with high apoB hold off ASCVD by lowering it via lifestyle and drugs. But it’s not the apoB itself that causes heart attacks, strokes, or chest pain on exertion (angina): it’s the structural damage to the arteries (ASCVD plaques) that does it.
Now, there are some diet tribes that want to insist that having high levels of apoB-containing lipoproteins is not a problem in the absence of other risk factors. (This is incorrect). They say that there’s nothing to worry about unless and until one has ASCVD detectable on a coronary calcium scan. This is poor reasoning. By the time a person has the kind of advanced plaque detectable on a coronary calcium scan, his or her arteries have already suffered substantial damage. Cyclarity Therapeutics and Repair Biotechnology’s Cholesterol Degrading Platform are working on therapies that would directly remove ASCVD damage even after it has formed, but we do not have any such therapies at this time. Therefore, our very best chance of averting future heart attacks and strokes with today’s medicine is to slow ASCVD development as much as possible before it reaches the calcified plaque stage. As of early 2024, all that can be done for a person who already has calcified plaques is to try to hold off the worst outcomes (heart attacks and strokes).
By analogy, the critics in the Times article are saying something even further off base: that a person should still not be diagnosed with cardiovascular disease (by analogy, Alzheimer’s Stage 1) even after he or she has calcified plaque in their arteries (beta-amyloid or aberrant tau), unless and until they also have their first heart attack (advanced cognitive decline like MCI or dementia)!
To Know and To Choose
Suppose there were a test, covered by your health insurance, that would tell you whether you had suffered molecular damage in your brain that would set you on the path toward developing Alzheimer’s in your lifetime. Wouldn’t you want to know about it?
The answer to this question seems obvious to me —and apparently to the great majority of Americans. Whereas the Times and most of the clinicians they interview object to testing people for beta-amyloid and aberrant tau when they don’t have any remarkable cognitive impairment, a landslide 85% of Americans in a recent poll said they would want to know if they had Alzheimer’s disease in its earliest stages.
And this poll was taken in 2020, before we had disease-modifying therapies for Alzheimer’s to offer anyone diagnosed with MCI or early Alzheimer’s dementia. That may explain some seemingly contradictory findings in the poll. For instance, people were less eager to know, not more, the more severe cognitive impairment became: only 54% would still want to know in the MCI stage, and even fewer would want to be told if they were in the early phase of Alzheimer’s dementia proper (31%) — and only 8% if they had “moderate” Alzheimer’s dementia. Similarly, only 40% of people polled said they would promptly make an appointment with their doctor if they began experiencing symptoms of MCI.
It seems reasonable to think that enthusiasm for testing would rise substantially at all stages of the disease if people knew that, were they to test positive, there was an FDA-approved therapy available that could keep their minds intact for many more years and potentially hold dementia it at bay for the rest of their lives. Indeed, among those who said they would not want to get an early diagnosis of Alzheimer’s, some of the top reasons included the fact that treatment options for Alzheimer’s are limited (31%), that there is no cure (28%), and that the tests might not be covered by insurance (13%). If upcoming trials validate the idea that starting on beta-amyloid-clearing AmyloSENS therapies while one’s mind is still intact can give a person additional years of life free of dementia, and if the Working Group proposal made testing widely available at low or no cost, many of these people would presumably line up to get scanned. Even those who gave the number one reason to decline testing (“A diagnosis of Alzheimer’s disease would be difficult for me to accept”) might feel differently if the consequences of the diagnosis were changed by the availability of a therapy that impacted the disease trajectory itself.
The Times piece is misguided in its biological conservatism, its medical fatalism, and its paternalism. The Working Group proposal is a good one and deserves support.
And in the meantime, cognitively-intact people who meet the age and other criteria can help unlock preventive rejuvenation biotechnologies for the brain by signing up for dementia prevention trials: the AHEAD Study, TRAILBLAZER-ALZ 3, or (if you are at high genetic risk) DIAN-TU Tau NexGen (which will test both the beta-amyloid-clearing antibody and an antibody targeting aberrant tau).
* An extremely small number of people (less than 1% of cases) suffer a kind of “pure” Alzheimer’s disease that occurs as a result of rare inherited mutations that cause their brains to produce so much beta-amyloid that they become demented when they are still only middle-aged by today’s standard, and their brains are therefore still in many ways structurally young.
