The last time Doc Gumshoe surveyed the Alzheimer’s landscape, the most prominent feature was the FDA’s approval – misguided, I thought – of Biogen’s Aduhelm (aducanumab) and the consequences this would have for everybody’s Medicare B premium. Sure enough, on November 12, 2021, the Centers for Medicare and Medicaid Services (CMS) confirmed the prediction that the cost of Aduhelm would indeed be borne by all Medicare subscribers. They announced that in 2022 the standard monthly payment for Medicare B would increase to $170.10, up from $148.50 in 2021. The reason for this was that FDA-approved treatments have to be covered by Medicare, and the estimated cost of Aduhelm treatment was $56,000 per year, despite the marginal benefit that this drug could provide for patients with Alzheimer’s disease (AD).

But on March 17th of this year, the FDA narrowed the terms of its approval of Aduhelm, limiting its use to patients participating in clinical trials. This will reduce the costs to Medicare by a substantial amount. Xavier Becerra, HHS Secretary, said that he will soon announce his decision on whether to lower Medicare premiums based on the FDA decision and the consequent greatly reduced usage of Aduhelm. Overall, that would be good news for most Medicare recipients, and probably not such bad news for Alzheimer’s patients who might otherwise have been given the drug – not such bad news, because Aduhelm’s very small benefit is countered by the fact that about 40% of patients receiving Aduhelm in clinical trials experienced painful brain swelling.

Researchers in AD had hoped that the FDA’s approval of aduhelm would spur more research on BACE inhibitors. Aduhelm, in case you don’t remember the pesky details, works (to the extent that it does work) by reducing the accumulation of amyloid beta (Aβ), which is one of two substances generally thought to contribute to the cognitive decline in Alzheimer’s patients. BACE is beta (β) secretase, the enzyme that cuts apart amyloid precursor protein (APP), which results in Aβ accumulation in the synaptic space between neurons, and the inhibition of the transmittal of neurotransmitters across this space. Since the transmittal of neurotransmitters from one neuron to another essentially constitutes brain activity, Aβ accumulation impeding this activity has been considered to be, if not the, at least an essential cause of AD.

The mechanism of action of BACE inhibitors is certainly promising. If we can prevent the formation of amyloid beta, and if BACE inhibitors effectively accomplish this task, it would seem evident that BACE inhibitors would significantly alleviate AD symptoms. But BACE inhibitors are very large molecules, and they have great difficulty in passing through the blood-brain barrier in enough concentration to be at all effective. Several BACE inhibitors have been developed, but up to now they have disappointed all parties – pharmaceutical companies, clinicians, and patients. Patients in particular have been waiting for a drug – something! – that will meaningfully slow the progression of AD. Aduhelm does not appear to be it.

When Aduhelm/aducanumab was granted accelerated approval by the FDA back in 2021, it was something of a surprise. Not only had a panel of 10 FDA advisors unanimously voted against the approval, but just a couple of years before that, Biogen and Eisai had announced that they were ending the Phase 3 ENGAGE and EMERGE trials of aducanumab in early AD, because an early analysis had demonstrated that the neither of the two trials had any chance at all of meeting its primary endpoint. These endpoints were the slowing of cognitive decline as measured by a widely used test called the Clinical Dementia Rating Scale – Sum of Boxes.

The FDA’s approval of aducanumab was not based on a published study, but on an analysis presented to the FDA that subjects in the EMERGE study taking aducanumab experienced somewhat slower cognitive decline than placebo subjects. The difference attained statistical significance, but the difference between aducanumab and placebo subjects was quite small – a fraction of a point on an 18 point scale. Biogen performed the same analysis on the ENGAGE study, but on this analysis aducanumab demonstrated no benefit at all in terms of cognitive decline.

Now the analysis on which the FDA’s approval was based has been published. Commenting on the publication Dr David Knopman (Mayo Clinic, Rochester, MN) was quoted as follows:

“Fully 3 years after the futility stop and 9 months after the accelerated approval, it is a relief to have a published manuscript on the aducanumab Phase 3 trials…. But the report contains no information that was not otherwise available from the sponsor or the FDA proceedings. The venue – the Journal of Prevention of Alzheimer’s Disease – in which the report was published is a lower-tier journal that is not widely available, that clearly has a laissez faire attitude towards standards for reporting clinical trials. In addition, there is an appearance of a conflict of interest in that two of the authors of the report happen to be editors of this journal. That Biogen was forced to publish in a venue of low impact and low editorial standards is a revealing message about the unrealistically optimistic interpretation of results.”

Medical journals are commonly ranked according to their impact factor, which is a measure of the frequency with which the average paper in a journal has been cited in a year. The New England Journal of Medicine (NEJM) has the highest impact factor, currently 92.24. The Journal of Prevention of Alzheimer’s Disease has an impact factor of 4.5.

Critics of the aducanumab studies, and, indeed, of studies of drugs aiming to treat AD, or at least slow the progression, frequently make the point that these studies start too late in the progression of the disease. If Alzheimer’s is caused by deposition of amyloid beta or the spread of tau proteins, then what we want to do is prevent those toxic substances from accumulating in the first place. The subjects in the aducanumab studies already had been diagnosed with Alzheimer’s disease, although at relatively early stages. But the brain sludge was already present. What if they had been able to treat these subjects at the very beginning of the process that led to their increasing dementia? Is it not possible that a BACE inhibitor at lower doses could have kept their brains relatively free from Aβ in the first place and prevented any significant impact on their cognition?

That is certainly a possibility, but demonstrating that a BACE inhibitor (or indeed any other drug) if given to patients that were predicted to develop AD does actually prevent the development of the disease would be extremely difficult. It would require the identification of persons with a significant likelihood of developing AD and enrolling a sufficient cohort of such persons in a clinical trial. There are certainly factors that increase the likelihood that an individual will eventually develop AD, but, since enrollment in a clinical trial requires subjects to volunteer, getting a large enough cohort to be evaluated for the characteristic that predicts
Alzheimer’s would be a monumental task.

If there were an existing drug or treatment of any kind that would reliably prevent or even slow the progress of AD, getting people to volunteer to be tested for whatever characteristic predicts the onset of the disease would be easy. But, as we know, at present there is no such drug. Women willingly have mammograms and men PSA tests, because there are highly effective treatments that can forestall the development of breast and prostate cancer. But why would people voluntarily submit to be evaluated for the likelihood that they will develop Alzheimer’s if the best result they can hope for is that they will be in a clinical trial that tests whether an experimental drug will provide any benefit? And in the meantime, the proposed trial subject is burdened with the unpleasant knowledge that he or she is fated to be an Alzheimer’s victim.

I don’t want to say it’s a hopeless task, but progress on that front is likely to be slow. A possible scenario is that a treatment modality may be found to have some effect on slowing progression in some patients already diagnosed with Alzheimer’s. It is tested in a small cohort of volunteers not yet diagnosed with AD but likely to develop AD, based on early signs and symptoms and perhaps genetic characteristics. The treatment appears to prevent a significant percentage of these volunteers from progressing to frank AD. Based on this hopeful information, more volunteers are found for a larger clinical trial. Meanwhile, the developers of the treatment have tinkered further and come up with a modified and improved version of the treatment, so that the new clinical trial will be larger and (hopefully, at least) produce more definitive results. And eventually a drug or a treatment will be identified that really makes a significant difference in the treatment of AD. The key word here is “eventually.”

Update on the current Alzheimer’s disease picture

This is an update of the data that was in the most recent Doc Gumshoe piece on AD. The data come from the Alzheimer’s Association “2022 Alzheimer’s Disease Facts and Figures Special Report. More than Normal Aging: Understanding Mild Cognitive Impairment.”

• 6.5 million persons in the US are currently diagnosed with AD. That’s diagnosed with AD; an equal number are thought to have undiagnosed AD, perhaps in the early stages.
• This includes nearly a quarter of a million persons under 65 years of age.
• One in nine individuals 65 years old or older has Alzheimer’s dementia. That percentage increases with age; 35.7% of people 85 years old or older have AD.
• Almost two-thirds of Americans with AD – 3.6 million – are women. The chief reason for this large imbalance is that women’s life expectancy is about 5 years greater than men’s life expectancy, and it’s in those 5 years that the incidence of AD increases steeply.
• AD prevalence is expected to increase as our population ages. By 2030, 8.5 million Americans are predicted to have AD, and by 2050, the number is predicted to be 12.7 million.
• The prevalence of AD and other dementias is 19% in African Americans and 14% in Hispanic Americans compared with 10% in non-Hispanic whites. The reasons for this are not entirely clear. The disparity is thought to be in part due to higher prevalence of conditions and circumstances which contribute to AD and dementia in African Americans and Hispanics.
• However, some genetic factors are involved. Individuals with one copy of the e4 allele in their APOE gene have three times the risk of developing AD, while those with two copies of e4 have eight times the risk. About 4.5%% of African Americans have two copies of that allele, compared with 2.4% of European Americans
• According to the Alzheimer’s Association, the 2021 costs to the nation associated with AD will be about $355 billion. About half of this cost will be paid by Medicare, and another significant chunk by Medicaid. Unless there is a treatment breakthrough, total costs may reach $1.1 trillion by 2050.
• Annual Medicare costs per patient with AD are estimated to be $26,358. Out-of-pocket costs are $11,571.
• These costs don’t figure in the huge economic impact of AD on caregivers, who are often the immediate family members of the person with AD. In 2020, these caregivers provided about 15.3 billion hours of care, valued at nearly $257 billion.
• AD was the 6th leading cause of death in the US in 2019; with the onset of COVID-19, AD dropped to the 7th leading cause of death in 2020 and 2021.
• US annual mortality from AD increased from 17.6 per 100,000 population in 2000 to 27.0 per 100,000 in 2019. The lowest mortality was 17.5 per 100,000 in Alaska while the highest rate was 56.0 in South Dakota.
• Among people aged 70 or older, about 61% of those with AD will not make it to age 80. This is twice as high as the death rate in people without AD.

The death rates in the 5 other leading causes of death in the US – breast cancer, prostate cancer, HIV, and especially heart disease and stroke,– have declined since the year 2000, while the death rate from AD has increased by 145.2%. At present, there is no cure, and the most effective treatments do nothing to stop the progress of the disease. At best, they slow it a little. This, may I note, is not unlike some cancer treatments that have been deemed moderately successful.

So, in the grim context outlined above, can we point to developments that appear to sound at least a few positive notes? I will leave it to the readers of this piece to pass judgment on whether the following news items offer significant promise to the prospects of effective management alternatives for Alzheimer’s disease.

TNF Inhibitors May Reduce the Risk of Developing Alzheimer’s Disease

The follow-on clause to that positive-sounding phrase is “in a subset of patients who also have some forms of cardiac disease.” This information comes from the DREAM study, which followed over 45,000 rheumatoid arthritis (RA) patients ages 65 and older from 2007 to 2017. The study subjects were grouped into three cohorts, based on the drugs the subjects were taking to control their RA. The three types of drugs being tested were a janus kinase (JAK) inhibitor, an interleukin-6 (IL-6) inhibitor, and one or more TNF inhibitors. These were assessed against a comparator drug, which inhibits the activation of T-cells. All drugs in the trial are commonly used as treatments for rheumatoid arthritis. The chief outcome of the study was the onset of Alzheimer’s and related dementia.

There were no observed associations between the JAK inhibitor or the IL-6 inhibitor and AD or dementia, and also no observed association between the TNF inhibitor and dementia, except in a subset of patients with established cardiovascular disease. In these patients, researchers found a lower incidence of AD and dementia in two analyses. In one, AD/dementia incidence was 26% lower over a 6-month period. In another, based on diagnosis codes and symptomatic prescriptions, the incidence of Alzheimer’s and dementia was 55% lower in subjects taking the TNF inhibitor.

The investigators were quick to point out that the study had several limitations. Outcomes were small for the JAK and IL-6 inhibitors, partly owing to short follow up periods. Also, they noted that the pathogenesis of AD and dementia likely begin several years before clinical diagnosis; therefore, longer treatment and observation periods may be needed to draw firmer conclusions. A study co-author, Dr Madham Thambisetty of the NIH National Institute on Aging observed, “A one-size-fits-all approach to treating and preventing Alzheimer’s is unlikely to be as effective as identifying particular risk profiles of patients who may benefit from specific drugs to lower their risk of Alzheimer’s disease.” (Desai R, et al. Comparative Risk of Alzheimer Disease and Related Dementia Among Medicare Beneficiaries With Rheumatoid Arthritis Treated With Targeted Diseases-Modifying Antirheumatic Agents. JAMA Open. doi:10.1001/jamanetworkopen.2022.6567).

Tumor necrosis factor (TNF) and inflammation as factors in Alzheimer’s

A bit of background may be called for. Tumor necrosis factor (TNF) is a cytokine. Cytokines are an enormous group of small proteins that carry on signaling functions between our cells; they cannot actually enter our cells, but they make contact with cell exteriors and transmit information. TNF is an essential messenger in our immune system. Tumor necrosis factor alpha (TNFα), a member of the larger TNF family, is a highly active mediator of inflammation. That sounds like a bad thing, but we need to remember that inflammation is a necessary physiologic response to a number of potential harms. TNFα is involved in causing the death of harmful cells through the process called apoptosis. However, TNFα can also cause harm to important cells, such as those forming the surfaces of bones in our joints; thus TNFα is a factor in several auto-immune diseases such as rheumatoid arthritis (RA). The word “rheumatoid” points to the auto-immune aspect of RA; before we had any clue as to the precise pathology of RA, the presence of “rheum” – a bodily discharge – had been detected in some arthritic joints.

TNFα inhibitors work in different ways. Some, such as Humira (adalimumab) and Remicade (infliximab) are monoclonal antibodies, while others, such as Enbrel (etanercept) work by occupying the site on the TNFα molecule that would otherwise bind to the TNFα’s target. In either case, they prevent the TNFα molecules from inflicting harm on their target. What the TNFα inhibitors do, in two words, is suppress inflammation.

Why would suppressing inflammation have the effect of preventing or slowing AD? The obvious answer: inflammation must be part of the pathology of Alzheimer’s disease.

When researchers into AD looked at the brains of persons who had died with the disease, they found two anomalies. One was the presence of those amyloid beta (Aβ) plaques; the other was the presence of neurofibrillary tangles (NFTs). These tangles arise from the hyperphosphorylation of tau protein. Phosphorylation and de-phosphorylation of tau is part of normal brain function, allowing for communication via microtubules, but hyperphosphorylation (i.e., too much phosphorus) causes the disruption of a number of cellular processes in the brain.

These two anomalies – Aβ plaques and NFTs – were quite rationally assumed to be fundamental causes of Alzheimer’s disease. What caused these anomalies to emerge was not so clear.

A major obstacle to investigations into AD is that researchers had no data on what was going on in the brains of patients before AD symptoms appeared, and, indeed, very little data before AD’s final stages. All they had to go on in the first several decades of AD research was post-mortem brain studies, and, while those studies confirmed the presence of the Aβ plaques and neurofibrillary tangles, they revealed very little about the underlying pathology.

In the past ten years or so, it has become clear that chronic inflammation is present in brains of persons with Alzheimer’s, and a substantial body of research points to chronic inflammation as a factor that leads to both of those anatomical features of AD – the Aβ plaques and the neurofribillary tangles. Chronic inflammation needs to be distinguished from acute inflammation, which occurs as an immune defense against infections, toxic substances, and injury.

Chronic inflammation has been observed in the brains not only of persons with Alzheimer’s, but in the brains of those with Parkinson’s disease, traumatic brain injuries, amyotrophic lateral sclerosis (ALS), and multiple sclerosis. A sustained immune response leading to chronic inflammation has been recognized as a central feature of several neurodegenerative disorders.

Several cytokines have been identified as inflammatory triggers. These include several interleukins, such as IL-1β and IL-6. IL- 1β in particular is considered to be the “master regulator” of the brain inflammatory cascade, due to its role in regulating the expression of IL-6 and of tumor necrosis factor (TNF). In a ten-year study, high levels of IL-6 in middle-aged patients have been found to predict cognitive decline. Elevated IL-6 is present in the cerebrospinal fluid and serum of AD patients and is thought to be a significant contributor to the chronic inflammation found in the brains of AD patients.

TNFα is one of the most important proinflammatory cytokines in Alzheimer’s, allowing white blood cells and immune cells to flow into certain brain areas. TNFα binds to two receptors, and high levels of these specific receptors are found in the cerebrospinal fluid of patients diagnosed with mild cognitive impairment as long as six years before these patients progress to frank Alzheimer’s disease. TNFα may also increase the production of Aβ plaques by upregulating the generation of beta secretase, which, as mentioned earlier, is the enzyme that cleaves amyloid precursor protein (APP), and which therefore results in Aβ accumulation in the synaptic space between neurons, and the inhibition of the transmittal of neurotransmitters across this space.

A lot of the information in the section above came from a long and thorough review article by JW Kinney et al in Alzheimer’s & Dementia: Translational Research & Clinical Interventions 4 (2018) 575-590.

A conclusion that might be drawn from this complex web of potential causes and effects is that, while drugs like aducanumab have fundamental difficulties preventing the build-up of Aβ, drugs that inhibit the action of TNFα may have the same ultimate effect in preventing Aβ without the essential drawbacks of the BACE inhibitors. Which is to say, TNFα inhibitors may be able accomplish superior results. However, there are reservations.

I can readily see how you loyal readers would be somewhat perplexed at what all this might imply. If signs of inflammation can be detected in persons years before progression to frank AD, does that mean that they need to start taking drugs like TNFα inhibitors as soon as those signs are detected? TNFα inhibitors are potent and effective drugs, but in patients with rheumatoid arthritis they are not usually prescribed until other alternatives have been tried.

Let’s look at this another way. If Aβ and those neurofibrillary tangles are the gunk that clogs our brains and impedes the effective transmission of impulses between neurons, which after all is what constitutes mental activity, and if inflammation is the part of the process that causes that gunk to accumulate, then whatever we can do to prevent chronic inflammation will slow the accumulation of brain gunk and impede the progression of dementia and Alzheimer’s. We’re talking about chronic inflammation here – not the episodes of inflammation that occur as a result of a knock on the head or a bit too much booze.

A great deal has been written and promulgated about the kind of diet that controls inflammation. The following is (condensed) from the Harvard Health Letter:

Foods that cause inflammation: refined carbohydrates; fried foods; sodas and other sweetened beverages; red meat and processed meats; margarine, shortening, and lard.

Anti-inflammatory foods: tomatoes; olive oil; green leafy vegetables; nuts; fatty fish such as salmon, mackerel, tuna, and sardines; and fruits such as blueberries, strawberries, cherries, and oranges.

As you may have noticed, this looks a lot like the Mediterranean diet.

The point that needs to be made is that controlling inflammation is not only a good thing generally, but there seems to be a clear indication that it may be a safe and effective way of keeping our wits about us and avoiding the awful fate of Alzheimer’s disease.

Two pathways to Alzheimer’s disease and dementia

An aspect of Alzheimer’s that doesn’t get nearly as much attention as the cognitive decline and dementia that come with AD is its effect on our ability to do just about anything, including breathing and digesting food. AD as a cause of death is simply that as the brain slowly grinds to a halt, all the functions that the brain controls also stop working. That goes way beyond remembering where you left the car keys and remembering the name of your next-door neighbor’s daughter. The autonomic nervous system, which controls all our bodily functions, is also impeded by the accumulation of that combination of Aβ and neurofibrillary tangles.

But there’s also strong evidence of other factors that are predictive of cognitive decline. Early results from a study in 2,000 older Americans is showing that those who adhered to at least four of five specifically described behaviors had a 60% lower risk of developing signs of dementia than those who adhered to at most one of those behaviors. The behaviors were: physical activity, not smoking, light-to-moderate alcohol consumption, a high-quality diet, and participation in activities that required exercise of cognitive function. The trial, U.S. POINTER (U.S. Study to Protect Brain Health Through Lifestyle Intervention to Reduce Risk) is led by Klodian Dhana, MD, PhD, of Rush University School of Medicine in Chicago.

That 60% lower risk is a major risk reduction, and I’m repeating it so it sticks out!

Three of those behaviors, relating to smoking, alcohol consumption, and diet, are related to chronic inflammation. However, the fifth behavior, which amounts basically to keeping the brain active, would appear to have no connection with inflammation. Intuitively, exercising the brain would have the same positive effect on brain health that exercising the body has on physical health. And the evidence that this is the case was presented convincingly in Lancet a couple of years ago.

The authors of the Lancet study concluded that more than one-third of all cases of Alzheimer’s disease were due to what they called “Population Attributable Factors” (PAFs). The study was based in part on observation of about 10,000 adults in the UK. According to the authors, an important concept in slowing down dementia is a factor that they call “cognitive reserve.” This sounds fairly obvious – if you have more reserve of any faculty, including cognition, you can afford to spare more of that faculty than a person that has less of it to start out with.

The factors that they identified included poor education (i.e., no schooling beyond 11th grade), hearing loss, hypertension, obesity, smoking, depression, physical inactivity, social isolation, and diabetes. It’s highly likely that a number of those risk factors overlap; for example, high-school dropouts are far more likely to be smokers than are college graduates, and hypertension, obesity, physical inactivity, and diabetes certainly do tend to go together.

These PAFs are, at least in theory, preventable, or at least treatable. Of course, modification of factors such as poor education would have to start quite early in life, although the effects of a poor education – also at least in theory – are remediable, i.e., keep your brain busy and active even if you didn’t finish high school.

Other research tends to support the idea that there are AD/dementia-causing factors more associated with life-style and behavior than with physiology. I have frequently mentioned the famous nun’s study, which showed a clear relationship between the richness of the vocabulary that nuns had employed in the essays they wrote when they became nuns and the likelihood of surviving into old age with their wits intact. The nuns who had used a more limited vocabulary were significantly more likely to demonstrate dementia in their later years than the ones with a larger lexicon. And post-mortem examination the brains of some nuns who had been totally lucid in their later years found that they had remained totally sharp despite considerable deposits of those Aβ plaques and neurofibrillary tangles.

If we look at the whole Alzheimer’s/dementia picture from a distance, it would appear that there are separate pathways to AD and dementia. The growth and accumulation of the stuff in the brain that impedes all brain function and leads to Alzheimer’s disease and, eventually, death, is one pathway. The other pathway, fueled by lack of cognitive reserve, leads to dementia, although not necessarily to AD.

What that means is that management of Alzheimer’s needs to have two separate foci. One is preventing the accumulation of Aβ plaques and neurofibrillary tangles, by whatever means appear to work, regardless of the apparent effect of this mode of treatment on cognition. The other is preventing the progress of dementia, again by whatever means appear to work, regardless of the apparent effect of this mode of treatment on the accumulation of brain gunk.

Those two approaches to managing Alzheimer’s are not, strictly speaking, treatment strategies. Since at this moment we do not have any effective ways of reversing AD once it becomes manifest, adopting whatever measures exist to prevent or at least delay its onset looks like the best available option.

This is not to ignore the enormous efforts that are currently being made both to identify those individuals who are at the highest risk of developing Alzheimer’s and also to develop forms of treatment that can stop the progress of this devastating disease. Genuinely effective treatments might be on their way, but in the meantime, we should all do what we can to prevent Alzheimer’s from affecting our lives.

* * * * * * * *

This diatribe does not begin to touch on the numerous small and perhaps hopeful findings in recent studies, many of which have received little notice, in particular when a lot of the health-care media is focused on COVID-19. I’ll come back to that in the next installment. In the meantime, stay well, avoid whichever new strain pops up in the next few days, and keep the comments coming! Thank you all! Michael Jorrin (aka Doc Gumshoe)