What a comprehensive piece- Thanks. Here’s the dystopian part- will insurers toss tau-217 positives off coverage, as these patients will be more expensive going forward, sooner or later?
I have been following your work recently and up until this point formed the opinion that it is better to live optimally to prevent disease and let the chips fall where they may. But what you suggest today makes me lean more toward testing.
I know from taking the Huntington's Gene test that it can be very stressful and although my results were favorable, other family members were not so lucky. When I tested, I thought that course of action was a no-brainer. Most people who have not stared down that barrel do. But now I have a much different view. As the youngest in the family who tested positive said, "if I didn't test, I could have maintained 50% hope. Now I have none."
Here I am again staring down the barrel of another gun. (APOE4) But much later in life, and although some say different, I refuse to believe it is determinative. AND what I read here today is that beyond the lifestyle factors, we are close to interventions that will at least delay the onset.
I have taken some time to investigate clinics and research on Alzeheimers and I do not yet see affordable options. I do not want to merely be a piece of data and live the remainder of my life in limbo. I want to know that I will be involved in treatments. Please keep the ethical testing framework in mind as your write. Many (like I did with HD) will run out and test, but that too can prove detrimental to health. Also, I hope you will point out ways to identify good clinics that support the patient through the ups and downs of treatments.
Referring to the APOE 4 gene you said "I refuse to believe it is determinative", here is more information which supports what you say. My mother in law had Alzheimer's and when my husband learned there was no treatment being a scientist he started reading the research. To find a way to help his mother. That was 12 years ago and he achieved his goal. Her symptoms began interfering with her daily functioning around the age of 85. She lived to 97 and was not in end stage Alzheimer's. She was APOE 3/4. At the time we didn't know that the information my husband learned would also help me as I am APOE 3/4. I wrote a comment on Dr. Topel's write up about TAU. Through reading the research my husband learned aluminum is the primary cause of Alzheimer's and drinking mineral water which is rich in silica removes the aluminum from the body including the brain. I have reduced my aluminum (24 hour urine test) to that of a 20 year old. My husband applied Causal Inference and found all the biomarkers which have been identified for diagnosing Alzheimer's are caused by aluminum. If you would like to learn more check out my substack and my husband (Dennis N Crouse) and my website. I don't know if I can post links here.
Nancy and Laurie, This complex research on genetics, biomarkers etc. is conducted by doctors and other researchers who are not aware that most or all of the Alzheimer's disease, Parkinson's disease etc. patients they are studying have very low levels of circulating 25-hydroxyvitamin D compared to the 50 ng/mL (125 nmol/L) which the immune system needs to function properly. Most people have only a fraction of this, and the lower the level, the greater the risk of neurodegeneration, severe infectious disease, autoimmune disorders etc.
These messy details ("7000 blood plasma proteins") of what goes wrong with this or that form of neurodegeneration are all very interesting and are good for getting research funding, but they are analagous to detailed study of all the ways an internal combustion engine can fail when it runs out of oil, by researchers who do not recognise that the failures they are studying almost never occur with adequate oil supply. Imagine PhD theses into exactly how big-end bearing surfaces (steel in the crankshaft and lead on bearing shell) touch, melt and destroy the crankshaft when the researcher does not realise the cause is inadequate oil being pumped into the bearing.
Most people have only a fraction of the circulating (in the bloodstream) 25-hydroxyvitamin D they need for their immune systems to operate properly. Most doctors and immunologists are not aware of this and have never heard of the 25-hydroxyvitamin D to calcitriol (1,25-dihydroxyvitamin D) intracrine (inside an individual cell) signaling system which many types of immune cell rely upon in order to adapt their behaviour according to their changing circumstances.
This includes recommendations from New Jersey based Professor of Medicine Sunil Wimalawansa for how much vitamin D3 to supplement, on average per day, to safely attain at least 50 ng/mL circulating 25-hydroxyvitamin D without the need for blood tests or medical monitoring. For 70 kg (154 lb) body weight without obesity, 0.125 milligrams (5000 IU) a day is a good amount = a gram every 22 years.
Prof. Wimalawansa has been researching vitamin D since the mid 1990s. In these now three decades, the most common response he has found from physicians to this research is "Its too simple. How can it be true?". It is simple, compared to many aspects of biology, and especially compared to the mountain of Alzheimer's disease research which has now been done, in ignorance of the importance of proper 25-hydroxyvitamin D levels.
It is also true. I know a couple who got to their mid-90s without a trace of dementia, following a doctor recommending proper (as just described) vitamin D3 supplementation 25 years ago. I have a friend who died of multiple system atrophy when he was 66. He had never supplemented vitamin D3. Proper vitamin D3 supplementation two years before was insufficient to reverse the damage, which surely occurred over decades
I had never heard of multiple system atrophy. It is essentially the same as Parkinson's disease and dementia with Lewy bodies. All three involve misfolding (prion disease) of the same α-synuclein protein, which exists naturally in the brain. Ayers et al. 2022 https://www.pnas.org/doi/abs/10.1073/pnas.2113489119 established that the difference between these three diseases (really three somewhat different sets of symptoms) is that the self-catalysing (prion) misfolding pattern is different for each of these three, though the researchers could not discover exactly what the misfolding patterns were.
When I first searched Google Scholar for "multiple system atrophy" and "vitamin D" it took me two or so minutes to find Ogura et al. 2021 https://www.sciencedirect.com/science/article/pii/S2405650221000617 who report that healthy controls averaged 26.8 ng/mL circulating 25-hydroxyvitamin D (half of what the immune system needs to function properly) while Parkinson's disease patients averaged 13.4 ng/mL and multiple system atrophy patients 10.5 ng/mL. (Both p = 0.0001.)
Evatt et al. 2011: https://sci-hub.se/10.1001/archneurol.2011.30 found that the low 25-hydroxyvitamin D levels in PD etc. patients is not, or is perhaps only very slightly, due to reverse causation (the disease lowering the level).
So there you have it - low -25-hydroxyvitamin D is a very strong, and easily corrected, risk factor for PD etc. Alzheimer's disease and numerous other kinds of neurodegeneration follow the same pattern, as you can read from the research cited and discussed at: https://vitamindstopscovid.info/00-evi/#3.3.
However, most doctors and researchers are pathologically uninterested in such simple explanations, with equally simple and desirable interventions to greatly reduce the risk.
There's no further research required - however all these researchers are trying to get funding for their multi-year, pseudo-sophisticated, programs investigating all manner of biochemical intrigues which occur with aging in the absence of sufficient 25-hydroxyvitamin D to run the immune system properly.
You are correct, low levels of Vitamin D are found in people with Alzheimer's. The reason for this is enhanced aluminum levels inhibit the production of the active form of vitamin D from vitamin D2 in the kidney. My husband Dennis N Crouse discusses the research on Vitamin D in his book 'Prevent Alzheimer's Autism and Stroke......." page 86-93. I wish I could put photos of these pages here. Here are 2 of the references. 300. Robertson, J.A., et al.; Animal model of aluminum induced osteomalcia: role of chronic renal failure; Kidney Int.; 23:327-35 (1983) 301. Goodman, W.G., et al.; Parenteral aluminum administration in the dog II. Induction of osteomalcia and effect on vitamin D metabolism; Kidney Int.; 25:370-75 (1984)
Hi Laurie, Thanks for the links to these articles.
The first one: Robertson et al. 1982 "Animal model of aluminum-induced osteomalacia: Role of chronic renal failure" https://www.sciencedirect.com/science/article/pii/S0085253815328933 PDF at: https://sci-hub.se/https://doi.org/10.1038/ki.1983.23 doesn't mention anything about aluminium disrupting the hydroxylation of 25-hydroxyvitamin D in the kidneys to calcitriol (1,25-dihydroxyvitamin D). " Thus, although it is an attractive hypothesis to link aluminum-induced osteomalacia with altered vitamin D metabolism, firm evidence in support of such a hypothesis is lacking."
Likewise the second article: Goodman et al. 1984: "Parenteral aluminum administration in the dog II. Induction of osteomalcia and effect on vitamin D metabolism" https://www.sciencedirect.com/science/article/pii/S0085253815331355 . "The failure of dialysis osteomalacia to respond to treatment with 1,25(OH)2D3 [29—3 1] makes it unlikely that altered vitamin D metabolism induced by aluminum is responsible for the osteomalacia observed clinically."
The cause of lower than 50 ng/mL levels of circulating 25-hydroxyvitamin D is not related to aluminium. It is due to insufficient vitamin D3 production in the skin, from UV-B light converting 7-dehydrocholesterol to vitamin D3, in the absence of sufficient vitamin D3 (or, in principle, 25-hydroxyvitamin D calcifediol) supplementation. There's very little vitamin D3 (or calcifediol) in food, fortified or not. To get 50 ng/mL 25-hydroxyvitamin D primarily or solely via UV-B skin exposure, there would need to be extensive all year round high elevation sun exposure on ideally white skin, or artificial UV-B light exposure. Both would damage DNA to the extent of greatly increasing the risk of skin cancer, although the higher 25-hydroxyvitamin D level will reduce the risk of all cancers compared to the typical levels of most of the population today, in the range of 5 to 25 ng/mL, except perhaps in summer, for those who do not properly supplement vitamin D3.
I have not studied aluminium toxicity but it is a serious and perhaps pervasive problem. Dr Christopher Exley has his Subtack devoted to this problem, including the use of drinking water containing silica (silicon dioxide = quartz) to reduce the level of aluminium in the bloodstream: https://drchristopherexley.substack.com.
I am very familiar with Chris Exley's work. I can't represent here what my husband has written in his book about aluminum and Vitamin D (pages 86 to 93). There are more references in his book. IF you are interested in Preventing Alzheimer's you should check out my husband's writings. He applied Causal Inference and proved aluminum is the cause of all the biomarkers which have been identified for diagnosing Alzheimer's. His book is titled "Finding a Cause and Potential Cures for Alzheimer's". Thank your for reading my comment.
Reading the second article more closely, some sentences imply a potential causal link between aluminium in the kidneys and reduced production of calcitriol (in this article, "1,25(OH)2D3") in the kidneys: "The present observations suggest that not only might aluminum be responsible for impaired mineralization of bone but also raise the possibility that aluminum could be responsible for the reduced biosynthesis of l,25(OH)2D3 in these patients.". However, without me reading the whole article, I don't see how this is consistent with an earlier statement:
"Some degree of aluminum retention is common with renal insufficiency; this fact and the present observations also raise the possibility that the retention of aluminum could be a factor which contributes to the low serum levels of 1 ,25(OH)2D in patients with advanced renal failure. This possibility seems unlikely, however, in that markedly reduced serum levels of 1 ,25(OH)D are observed almost uniformly in patients with endstage renal failure, regardless of whether they exhibit clinical features which suggest aluminum accumulation."
Thank you Laurie for your thoughtful reply and information. I have heard about the association of aluminum to Alzheimers but have not investigated the research. I will do that.
Hi Eric. Thank you for presenting this information about blood tests for dementia, specifically p-tau217!
As a primary care doc, I'm on the front lines of patients trying to translate these sorts of findings into actual blood test orders, and the results that come back to us. I'm concerned about opening this sort of Pandora's box with them when there are no solid clinical guidance guardrails for us in primary care (or even neurology) yet.
I crunched some prompt engineering and clinical questions with Claude.AI this morning about this whole dilemma, and here is what I came up with. It seems valid to me. I can send you the actual references Claude cites (but I can't get these to work in comment fields).
If you have time please let us know what you think at the point of care.
# **p-Tau 217 in Primary Care: Current Evidence for Clinical Implementation**
## **Bottom Line Up Front**
p-Tau 217 shows excellent diagnostic accuracy and strong predictive value for dementia risk, but comprehensive primary care guidelines for risk counseling and serial monitoring remain limited. Unlike LDL monitoring for cardiovascular disease, p-Tau 217 is not yet ready for routine serial testing, though it provides valuable risk stratification through quartile-based analysis.
## **Current Clinical Implementation Status**
The most robust evidence comes from a major 2025 Nature Medicine study involving 1,767 participants across primary and secondary care settings, which found that plasma p-tau217 using the fully automated Lumipulse immunoassay achieved 85% accuracy in primary care, with positive predictive values of 82% and negative predictive values of 88%.
Mayo Clinic now offers clinical p-Tau 217 testing with specific cutoffs: sensitivity of 92% at the lower cutpoint (≤0.185 pg/mL) and specificity of 96% at the upper cutpoint (≥0.325 pg/mL) for detecting abnormal amyloid-PET. However, this assay should not be ordered for individuals younger than 50 years or in cognitively unimpaired individuals regardless of age, and should not be used to predict the development of dementia or other neurologic conditions.
A **two-cutoff approach** provides enhanced clinical utility: using cutoffs at <0.22 and >0.34 pg/mL achieved 92% accuracy in primary care, excluding 16% of patients with intermediate results who would need confirmatory testing. This approach reduces uncertainty and expensive follow-up testing while maintaining high accuracy.
## **Specific Risk Prediction Data**
The most comprehensive predictive data comes from a landmark Swedish community-based study that followed 2,148 dementia-free older adults for up to 16 years and provides robust predictive data by p-Tau 217 quartiles, revealing a dose-response relationship between baseline p-tau217 levels and higher hazards of all-cause and AD dementia.
**Ten-Year Dementia Prediction Performance:**
- P-tau217 demonstrated AUC of 81% for all-cause dementia and 77% for AD dementia
- Negative predictive values were consistently >90% for all biomarkers
- When p-tau217 was combined with NfL for all-cause dementia prediction, PPVs reached 43% while maintaining high NPVs of 95.1%
**AUC (Area Under the Curve)** represents the biomarker's ability to distinguish between people who will and won't develop dementia. An AUC of 0.81 (81%) means that if you randomly selected one person who will develop dementia and one who won't, the test would correctly identify the higher-risk person 81% of the time. Values above 0.8 indicate good predictive performance.
In cognitively unimpaired individuals, almost all individuals with abnormal p-tau217 values developed mild cognitive impairment within a 10-year follow-up. Additionally, in a community-based study from Shanghai, higher concentrations of plasma p-tau217 were correlated with higher decline speed of Mini-Mental State Examination scores and higher risk of incident dementia and AD.
## **Primary Care Implementation Challenges**
Only 20% of primary care providers reported high confidence in interpreting cognitive test results, and only 13% were highly confident in making specific neurocognitive disorder diagnoses. A quarter of all providers identified lack of familiarity with diagnostic criteria for neurocognitive disorder syndromes as a barrier to clinical practice.
The new 2024 Alzheimer's Association Clinical Practice Guidelines for primary care include obtaining information about the patient's risk profile for AD including age, family history of dementia, hypertension, and smoking, but don't yet provide specific protocols for counseling based on p-Tau 217 levels.
## **Why Serial Monitoring Isn't Like LDL**
Unlike LDL cholesterol monitoring, p-Tau 217 faces significant limitations for serial testing:
1. **No Treatment Goals**: p-Tau217 concentrations have not been established to correlate with disease severity, and the test has not been established for monitoring the effect of disease monitoring therapies.
2. **Biological Differences**: While LDL reflects modifiable cardiovascular risk and responds to statins, p-Tau 217 reflects irreversible brain pathology accumulation over decades.
3. **Technical Variability**: P-tau217 levels vary with time of day, which could have implications for their clinical value.
4. **Research-Only Recommendation**: Experts strongly advocate that people who do not yet have cognitive impairments should not be tested or diagnosed based on plasma p-tau217 in clinical settings, restricting use to research settings and specialized clinics where counseling is available.
## **Cost-Effectiveness and Access**
Using a two-cutoff approach with p-tau217, estimated savings of up to 60% compared with CSF testing alone and up to 81% compared with amyloid PET imaging alone were demonstrated, making it economically attractive for healthcare systems.
## **Current Clinical Recommendations**
**Appropriate Uses:**
- Diagnostic evaluation of patients ≥50 years with cognitive symptoms
- Risk stratification using quartile-based analysis
- Ruling out Alzheimer's pathology (excellent negative predictive value >90%)
- Reducing need for expensive CSF or PET testing
**Not Recommended For:**
- Screening asymptomatic individuals
- Serial monitoring of disease progression
- Precise individual risk percentages
- Predicting when dementia will develop
## **Future Directions**
While p-Tau 217 shows excellent promise, comprehensive primary care protocols for risk counseling and lifestyle modification based on biomarker levels are still being developed. The technology is ready for implementation, but clinical practice guidelines need further development to match the sophistication we see in cardiovascular risk management with LDL monitoring.
The evidence supports p-Tau 217 as excellent for **risk stratification** rather than precise risk prediction, with its greatest value being the ability to confidently rule out significant AD pathology when levels are low.
The easiest and best way of reducing the risk of Alzheimer's disease, Parkinson's disease, dementia with Lewy bodies etc. is to supplement vitamin D3 cholecalciferol sufficiently to attain the 50 ng/mL (125 nmol/L = 1 part in 20,000,000 by mass) level of 25-hydroxyvitamin D calcifediol (AKA "calcidiol") which the immune system needs to function properly.
This page begins with graphs from Quraishi et al. 2014 https://jamanetwork.com/journals/jamasurgery/articlepdf/1782085/soi130062.pdf which show that immune responses to bacterial pathogens which cause post-operative infections are highly effective (2.5% risk of hospital-acquired and 2.5% risk of surgical site infections) with pre-operative circulating 25-hydroxyvitamin D levels of 50 ng/mL or more, but are weak (25% risk of each type) with 20 ng/mL, which is a normal level for those who do not supplement vitamin D3 properly, and who have not recently had high levels of UV-B irradiation of ideally white skin.
Most doctors regard 20 ng/mL circulating 25-hydroxyvitamin D as sufficient for good health. However, this is only what the kidneys need to play their role in regulating calcium-phosphate-bone metabolism, which is the function of the three vitamin D compounds medical professionals all understand.
What they don't know is that many types of cell, including many or perhaps most types of immune cell, rely on a good supply of 25-hydroxyvitamin D as a raw material for their 25-hydroxyvitamin D to calcitriol (1,25-dihydroxyvitamin D) intracrine signaling systems. These are signaling systems entirely within a single cell and are crucial to the ability of many types of cell to alter their behavior in response to their changing circumstances. A related paracrine signaling system is used by some types of immune cell to signal to nearby immune cells, typically of different types.
25-hydroxyvitamin D to calcitriol intracrine signaling was first discovered in dendritic cells and macrophages by Martin Hewison and colleagues in the late 2000s. In 2021 a team of NIH researchers elucidated this signaling system (which they named, somewhat incorrectly, as "autocrine" signaling) in Th1 regulatory lymphocytes: Chauss et al. https://www.nature.com/articles/s41590-021-01080-3. I summarized this dense (and I think Nobel Prize worthy) cell biology article at: https://aminotheory.com/cv19/icu/#2021-Chauss.
Th1 cells from the lungs of hospitalized COVID-19 patients were found to be stuck in their pro-inflammatory startup program, despite detecting the condition (a high level of a complement protein) which should cause them to transition to their anti-inflammatory shutdown program. The researchers describe in great detail the 25-hydroxyvitamin D to calcitriol intracrine signaling system which achieves this transition. The failure or this system in these patients, who almost invariably have very low circulating 25-hydroxyvitamin D levels) was found to be due entirely or almost entirely, to these cells having inadequate supplies of 25-hydroxyvitamin D in their cytosols. (These people had severe disease due to poor initial responses to the virus and due to the pathologically sustained inflammatory response once the virus infected the alveoli. If they had had 50 ng/mL or more circulating 25-hydroxyvitamin D they would generally have had only mild COVID-19 symptoms.)
This and the related paracrine signaling systems enable cell-type specific conditions to be detected so that the cell responds in a cell-type specific manner. These signaling systems are unrelated to hormonal (endocrine) signaling, which the kidneys use (with calcitriol as the blood-borne signaling molecule, AKA hormone) to alter the behavior of cells in distant parts of the body.
Neither vitamin D3 cholecalciferol no 25-hydroxyvitamin D are hormones. They are not signaling molecules. The immune system does not use hormonal signaling. Medical professionals and many vitamin D researchers refer to all three compounds as if they were all "vitamin D". This is a mistake, as Reinhold Vieth pointed out in 2004: https://sci-hub.se/10.1016/j.jsbmb.2004.03.037.
25-hydroxyvitamin D, as measured in "vitamin D" blood tests, is made, primarily in the liver, by hydroxylating vitamin D3 cholecalciferol on the 25th carbon.
There's very little vitamin D (2 or the natural 3) in food, including food with is fortified with vitamin D3 or the less effective D2.
Vitamin D3 can be produced with UV-B irradiation of ideally white skin, but most people cannot attain this naturally (dark or black skin, living too far from the equator etc.) to get this all year round - and all such UV-B exposure damages DNA and so increases the risk of skin cancer.
Supplementation of vitamin D3, in quantities well above those needed for 20 ng/mL 125 nmol/L 25-hydroxyvitamin D is needed by almost everyone in order to be healthy. The exceptions are babies breast fed by 25-hydroxyvitamin D replete mothers (about equal amounts of vitamin D3 and 25-hydroxyvitamin D in the breast milk, with the latter being 4 times as effective as vitamin D3 in raising the baby's 25-hydroxyvitamin D) and those who get so much UV-B skin exposure, year after year, that they are at high risk of developing skin cancer.
In 2022, New Jersey based Professor of Medicine Sunil Wimalawansa published an article with proper vitamin D3 supplementation recommendations, as ratios of body weight, with higher ratios for those suffering from obesity, since obesity reduces the rate of hydroxylation in the liver to 25-hydroxyvitamin D and because the excess adipose tissue absorbs 25-hydroxyvitamin D and vitamin D3: https://vitamindstopscovid.info/00-evi/#obesity-deficit.
He later simplified these recommendations somewhat to average daily supplemental intakes of vitamin D3:
70 to 90 IU / kg body weight for those not suffering from obesity (BMI < 30).
100 to 130 IU / kg body weight for obesity I & II (BMI 30 to 39).
140 to 180 IU / kg body weight for obesity III (BMI > 39).
For 70 kg (154 lb) without obesity, this is about 0.125 milligrams (5000 IU) a day. This takes several months to attain the desired > 50 ng/mL circulating 25-hydroxyvitamin D. This is 8 or more times what most governments recommend. "5000 IU" sounds like a lot, but it is a gram every 22 years - and pharma grade vitamin D costs about USD$2.50 a gram ex-factory. This will safely attain at least 50 ng/mL 125 nmol/L 25-hydroxyvitamin D over several months, without the need for blood tests or medical monitoring. He recently repeated these recommendations in an article co-written by two other professors - one of medicine and the other of pediatrics:
Integrating Endocrine, Genomic, and Extra-Skeletal Benefits of Vitamin D into National and Regional Clinical Guidelines Sunil J. Wimalawansa, Scott T. Weiss and Bruce W. Hollis, Nutrients 2024-11-20 https://www.mdpi.com/2072-6643/16/22/3969.
For patients who are interested in understanding their polygenic risk for common conditions like heart disease or cancer—but whose physicians don’t currently use or offer this kind of testing—where do you recommend they turn? Are there trustworthy, clinically valid platforms you would feel comfortable pointing people toward, especially those trying to be proactive about their health?
My 83-year-old brother just succumbed to complications of Alzheimer's. I am one or two years away from when he started to show symptoms. My PCP refuses to order a p-tau217 test for me, and says that a referral to a neurologist would not be appropriate due to my age and condition. What you describe is utopian, not realistic in the current health care climate.
What a comprehensive piece- Thanks. Here’s the dystopian part- will insurers toss tau-217 positives off coverage, as these patients will be more expensive going forward, sooner or later?
I have been following your work recently and up until this point formed the opinion that it is better to live optimally to prevent disease and let the chips fall where they may. But what you suggest today makes me lean more toward testing.
I know from taking the Huntington's Gene test that it can be very stressful and although my results were favorable, other family members were not so lucky. When I tested, I thought that course of action was a no-brainer. Most people who have not stared down that barrel do. But now I have a much different view. As the youngest in the family who tested positive said, "if I didn't test, I could have maintained 50% hope. Now I have none."
Here I am again staring down the barrel of another gun. (APOE4) But much later in life, and although some say different, I refuse to believe it is determinative. AND what I read here today is that beyond the lifestyle factors, we are close to interventions that will at least delay the onset.
I have taken some time to investigate clinics and research on Alzeheimers and I do not yet see affordable options. I do not want to merely be a piece of data and live the remainder of my life in limbo. I want to know that I will be involved in treatments. Please keep the ethical testing framework in mind as your write. Many (like I did with HD) will run out and test, but that too can prove detrimental to health. Also, I hope you will point out ways to identify good clinics that support the patient through the ups and downs of treatments.
Referring to the APOE 4 gene you said "I refuse to believe it is determinative", here is more information which supports what you say. My mother in law had Alzheimer's and when my husband learned there was no treatment being a scientist he started reading the research. To find a way to help his mother. That was 12 years ago and he achieved his goal. Her symptoms began interfering with her daily functioning around the age of 85. She lived to 97 and was not in end stage Alzheimer's. She was APOE 3/4. At the time we didn't know that the information my husband learned would also help me as I am APOE 3/4. I wrote a comment on Dr. Topel's write up about TAU. Through reading the research my husband learned aluminum is the primary cause of Alzheimer's and drinking mineral water which is rich in silica removes the aluminum from the body including the brain. I have reduced my aluminum (24 hour urine test) to that of a 20 year old. My husband applied Causal Inference and found all the biomarkers which have been identified for diagnosing Alzheimer's are caused by aluminum. If you would like to learn more check out my substack and my husband (Dennis N Crouse) and my website. I don't know if I can post links here.
Nancy and Laurie, This complex research on genetics, biomarkers etc. is conducted by doctors and other researchers who are not aware that most or all of the Alzheimer's disease, Parkinson's disease etc. patients they are studying have very low levels of circulating 25-hydroxyvitamin D compared to the 50 ng/mL (125 nmol/L) which the immune system needs to function properly. Most people have only a fraction of this, and the lower the level, the greater the risk of neurodegeneration, severe infectious disease, autoimmune disorders etc.
These messy details ("7000 blood plasma proteins") of what goes wrong with this or that form of neurodegeneration are all very interesting and are good for getting research funding, but they are analagous to detailed study of all the ways an internal combustion engine can fail when it runs out of oil, by researchers who do not recognise that the failures they are studying almost never occur with adequate oil supply. Imagine PhD theses into exactly how big-end bearing surfaces (steel in the crankshaft and lead on bearing shell) touch, melt and destroy the crankshaft when the researcher does not realise the cause is inadequate oil being pumped into the bearing.
Most people have only a fraction of the circulating (in the bloodstream) 25-hydroxyvitamin D they need for their immune systems to operate properly. Most doctors and immunologists are not aware of this and have never heard of the 25-hydroxyvitamin D to calcitriol (1,25-dihydroxyvitamin D) intracrine (inside an individual cell) signaling system which many types of immune cell rely upon in order to adapt their behaviour according to their changing circumstances.
Please see my comment below, which is a mini vitamin D pep talk intended to encourage people to read the relevant research, much of which you can find cited and discussed at: https://vitamindstopscovid.info/00-evi/ - specifically regarding neurodegeneration: https://vitamindstopscovid.info/00-evi/#3.3.
This includes recommendations from New Jersey based Professor of Medicine Sunil Wimalawansa for how much vitamin D3 to supplement, on average per day, to safely attain at least 50 ng/mL circulating 25-hydroxyvitamin D without the need for blood tests or medical monitoring. For 70 kg (154 lb) body weight without obesity, 0.125 milligrams (5000 IU) a day is a good amount = a gram every 22 years.
Prof. Wimalawansa has been researching vitamin D since the mid 1990s. In these now three decades, the most common response he has found from physicians to this research is "Its too simple. How can it be true?". It is simple, compared to many aspects of biology, and especially compared to the mountain of Alzheimer's disease research which has now been done, in ignorance of the importance of proper 25-hydroxyvitamin D levels.
It is also true. I know a couple who got to their mid-90s without a trace of dementia, following a doctor recommending proper (as just described) vitamin D3 supplementation 25 years ago. I have a friend who died of multiple system atrophy when he was 66. He had never supplemented vitamin D3. Proper vitamin D3 supplementation two years before was insufficient to reverse the damage, which surely occurred over decades
I had never heard of multiple system atrophy. It is essentially the same as Parkinson's disease and dementia with Lewy bodies. All three involve misfolding (prion disease) of the same α-synuclein protein, which exists naturally in the brain. Ayers et al. 2022 https://www.pnas.org/doi/abs/10.1073/pnas.2113489119 established that the difference between these three diseases (really three somewhat different sets of symptoms) is that the self-catalysing (prion) misfolding pattern is different for each of these three, though the researchers could not discover exactly what the misfolding patterns were.
When I first searched Google Scholar for "multiple system atrophy" and "vitamin D" it took me two or so minutes to find Ogura et al. 2021 https://www.sciencedirect.com/science/article/pii/S2405650221000617 who report that healthy controls averaged 26.8 ng/mL circulating 25-hydroxyvitamin D (half of what the immune system needs to function properly) while Parkinson's disease patients averaged 13.4 ng/mL and multiple system atrophy patients 10.5 ng/mL. (Both p = 0.0001.)
Evatt et al. 2011: https://sci-hub.se/10.1001/archneurol.2011.30 found that the low 25-hydroxyvitamin D levels in PD etc. patients is not, or is perhaps only very slightly, due to reverse causation (the disease lowering the level).
So there you have it - low -25-hydroxyvitamin D is a very strong, and easily corrected, risk factor for PD etc. Alzheimer's disease and numerous other kinds of neurodegeneration follow the same pattern, as you can read from the research cited and discussed at: https://vitamindstopscovid.info/00-evi/#3.3.
However, most doctors and researchers are pathologically uninterested in such simple explanations, with equally simple and desirable interventions to greatly reduce the risk.
There's no further research required - however all these researchers are trying to get funding for their multi-year, pseudo-sophisticated, programs investigating all manner of biochemical intrigues which occur with aging in the absence of sufficient 25-hydroxyvitamin D to run the immune system properly.
You are correct, low levels of Vitamin D are found in people with Alzheimer's. The reason for this is enhanced aluminum levels inhibit the production of the active form of vitamin D from vitamin D2 in the kidney. My husband Dennis N Crouse discusses the research on Vitamin D in his book 'Prevent Alzheimer's Autism and Stroke......." page 86-93. I wish I could put photos of these pages here. Here are 2 of the references. 300. Robertson, J.A., et al.; Animal model of aluminum induced osteomalcia: role of chronic renal failure; Kidney Int.; 23:327-35 (1983) 301. Goodman, W.G., et al.; Parenteral aluminum administration in the dog II. Induction of osteomalcia and effect on vitamin D metabolism; Kidney Int.; 25:370-75 (1984)
Hi Laurie, Thanks for the links to these articles.
The first one: Robertson et al. 1982 "Animal model of aluminum-induced osteomalacia: Role of chronic renal failure" https://www.sciencedirect.com/science/article/pii/S0085253815328933 PDF at: https://sci-hub.se/https://doi.org/10.1038/ki.1983.23 doesn't mention anything about aluminium disrupting the hydroxylation of 25-hydroxyvitamin D in the kidneys to calcitriol (1,25-dihydroxyvitamin D). " Thus, although it is an attractive hypothesis to link aluminum-induced osteomalacia with altered vitamin D metabolism, firm evidence in support of such a hypothesis is lacking."
Likewise the second article: Goodman et al. 1984: "Parenteral aluminum administration in the dog II. Induction of osteomalcia and effect on vitamin D metabolism" https://www.sciencedirect.com/science/article/pii/S0085253815331355 . "The failure of dialysis osteomalacia to respond to treatment with 1,25(OH)2D3 [29—3 1] makes it unlikely that altered vitamin D metabolism induced by aluminum is responsible for the osteomalacia observed clinically."
The cause of lower than 50 ng/mL levels of circulating 25-hydroxyvitamin D is not related to aluminium. It is due to insufficient vitamin D3 production in the skin, from UV-B light converting 7-dehydrocholesterol to vitamin D3, in the absence of sufficient vitamin D3 (or, in principle, 25-hydroxyvitamin D calcifediol) supplementation. There's very little vitamin D3 (or calcifediol) in food, fortified or not. To get 50 ng/mL 25-hydroxyvitamin D primarily or solely via UV-B skin exposure, there would need to be extensive all year round high elevation sun exposure on ideally white skin, or artificial UV-B light exposure. Both would damage DNA to the extent of greatly increasing the risk of skin cancer, although the higher 25-hydroxyvitamin D level will reduce the risk of all cancers compared to the typical levels of most of the population today, in the range of 5 to 25 ng/mL, except perhaps in summer, for those who do not properly supplement vitamin D3.
I have not studied aluminium toxicity but it is a serious and perhaps pervasive problem. Dr Christopher Exley has his Subtack devoted to this problem, including the use of drinking water containing silica (silicon dioxide = quartz) to reduce the level of aluminium in the bloodstream: https://drchristopherexley.substack.com.
I am very familiar with Chris Exley's work. I can't represent here what my husband has written in his book about aluminum and Vitamin D (pages 86 to 93). There are more references in his book. IF you are interested in Preventing Alzheimer's you should check out my husband's writings. He applied Causal Inference and proved aluminum is the cause of all the biomarkers which have been identified for diagnosing Alzheimer's. His book is titled "Finding a Cause and Potential Cures for Alzheimer's". Thank your for reading my comment.
Reading the second article more closely, some sentences imply a potential causal link between aluminium in the kidneys and reduced production of calcitriol (in this article, "1,25(OH)2D3") in the kidneys: "The present observations suggest that not only might aluminum be responsible for impaired mineralization of bone but also raise the possibility that aluminum could be responsible for the reduced biosynthesis of l,25(OH)2D3 in these patients.". However, without me reading the whole article, I don't see how this is consistent with an earlier statement:
"Some degree of aluminum retention is common with renal insufficiency; this fact and the present observations also raise the possibility that the retention of aluminum could be a factor which contributes to the low serum levels of 1 ,25(OH)2D in patients with advanced renal failure. This possibility seems unlikely, however, in that markedly reduced serum levels of 1 ,25(OH)D are observed almost uniformly in patients with endstage renal failure, regardless of whether they exhibit clinical features which suggest aluminum accumulation."
Thank you Laurie for your thoughtful reply and information. I have heard about the association of aluminum to Alzheimers but have not investigated the research. I will do that.
You are welcome. Thank you for your interest. Here is a link to a video on substack I made on the relationship between ApoE 4 and aluminum. https://open.substack.com/pub/silicawater/p/apoe-4-reduce-your-risk-of-alzheimers?r=1be2vl&utm_medium=ios
Hi Eric. Thank you for presenting this information about blood tests for dementia, specifically p-tau217!
As a primary care doc, I'm on the front lines of patients trying to translate these sorts of findings into actual blood test orders, and the results that come back to us. I'm concerned about opening this sort of Pandora's box with them when there are no solid clinical guidance guardrails for us in primary care (or even neurology) yet.
I crunched some prompt engineering and clinical questions with Claude.AI this morning about this whole dilemma, and here is what I came up with. It seems valid to me. I can send you the actual references Claude cites (but I can't get these to work in comment fields).
If you have time please let us know what you think at the point of care.
# **p-Tau 217 in Primary Care: Current Evidence for Clinical Implementation**
## **Bottom Line Up Front**
p-Tau 217 shows excellent diagnostic accuracy and strong predictive value for dementia risk, but comprehensive primary care guidelines for risk counseling and serial monitoring remain limited. Unlike LDL monitoring for cardiovascular disease, p-Tau 217 is not yet ready for routine serial testing, though it provides valuable risk stratification through quartile-based analysis.
## **Current Clinical Implementation Status**
The most robust evidence comes from a major 2025 Nature Medicine study involving 1,767 participants across primary and secondary care settings, which found that plasma p-tau217 using the fully automated Lumipulse immunoassay achieved 85% accuracy in primary care, with positive predictive values of 82% and negative predictive values of 88%.
Mayo Clinic now offers clinical p-Tau 217 testing with specific cutoffs: sensitivity of 92% at the lower cutpoint (≤0.185 pg/mL) and specificity of 96% at the upper cutpoint (≥0.325 pg/mL) for detecting abnormal amyloid-PET. However, this assay should not be ordered for individuals younger than 50 years or in cognitively unimpaired individuals regardless of age, and should not be used to predict the development of dementia or other neurologic conditions.
A **two-cutoff approach** provides enhanced clinical utility: using cutoffs at <0.22 and >0.34 pg/mL achieved 92% accuracy in primary care, excluding 16% of patients with intermediate results who would need confirmatory testing. This approach reduces uncertainty and expensive follow-up testing while maintaining high accuracy.
## **Specific Risk Prediction Data**
The most comprehensive predictive data comes from a landmark Swedish community-based study that followed 2,148 dementia-free older adults for up to 16 years and provides robust predictive data by p-Tau 217 quartiles, revealing a dose-response relationship between baseline p-tau217 levels and higher hazards of all-cause and AD dementia.
**Ten-Year Dementia Prediction Performance:**
- P-tau217 demonstrated AUC of 81% for all-cause dementia and 77% for AD dementia
- Negative predictive values were consistently >90% for all biomarkers
- When p-tau217 was combined with NfL for all-cause dementia prediction, PPVs reached 43% while maintaining high NPVs of 95.1%
**AUC (Area Under the Curve)** represents the biomarker's ability to distinguish between people who will and won't develop dementia. An AUC of 0.81 (81%) means that if you randomly selected one person who will develop dementia and one who won't, the test would correctly identify the higher-risk person 81% of the time. Values above 0.8 indicate good predictive performance.
In cognitively unimpaired individuals, almost all individuals with abnormal p-tau217 values developed mild cognitive impairment within a 10-year follow-up. Additionally, in a community-based study from Shanghai, higher concentrations of plasma p-tau217 were correlated with higher decline speed of Mini-Mental State Examination scores and higher risk of incident dementia and AD.
## **Primary Care Implementation Challenges**
Only 20% of primary care providers reported high confidence in interpreting cognitive test results, and only 13% were highly confident in making specific neurocognitive disorder diagnoses. A quarter of all providers identified lack of familiarity with diagnostic criteria for neurocognitive disorder syndromes as a barrier to clinical practice.
The new 2024 Alzheimer's Association Clinical Practice Guidelines for primary care include obtaining information about the patient's risk profile for AD including age, family history of dementia, hypertension, and smoking, but don't yet provide specific protocols for counseling based on p-Tau 217 levels.
## **Why Serial Monitoring Isn't Like LDL**
Unlike LDL cholesterol monitoring, p-Tau 217 faces significant limitations for serial testing:
1. **No Treatment Goals**: p-Tau217 concentrations have not been established to correlate with disease severity, and the test has not been established for monitoring the effect of disease monitoring therapies.
2. **Biological Differences**: While LDL reflects modifiable cardiovascular risk and responds to statins, p-Tau 217 reflects irreversible brain pathology accumulation over decades.
3. **Technical Variability**: P-tau217 levels vary with time of day, which could have implications for their clinical value.
4. **Research-Only Recommendation**: Experts strongly advocate that people who do not yet have cognitive impairments should not be tested or diagnosed based on plasma p-tau217 in clinical settings, restricting use to research settings and specialized clinics where counseling is available.
## **Cost-Effectiveness and Access**
Using a two-cutoff approach with p-tau217, estimated savings of up to 60% compared with CSF testing alone and up to 81% compared with amyloid PET imaging alone were demonstrated, making it economically attractive for healthcare systems.
## **Current Clinical Recommendations**
**Appropriate Uses:**
- Diagnostic evaluation of patients ≥50 years with cognitive symptoms
- Risk stratification using quartile-based analysis
- Ruling out Alzheimer's pathology (excellent negative predictive value >90%)
- Reducing need for expensive CSF or PET testing
**Not Recommended For:**
- Screening asymptomatic individuals
- Serial monitoring of disease progression
- Precise individual risk percentages
- Predicting when dementia will develop
## **Future Directions**
While p-Tau 217 shows excellent promise, comprehensive primary care protocols for risk counseling and lifestyle modification based on biomarker levels are still being developed. The technology is ready for implementation, but clinical practice guidelines need further development to match the sophistication we see in cardiovascular risk management with LDL monitoring.
The evidence supports p-Tau 217 as excellent for **risk stratification** rather than precise risk prediction, with its greatest value being the ability to confidently rule out significant AD pathology when levels are low.
The easiest and best way of reducing the risk of Alzheimer's disease, Parkinson's disease, dementia with Lewy bodies etc. is to supplement vitamin D3 cholecalciferol sufficiently to attain the 50 ng/mL (125 nmol/L = 1 part in 20,000,000 by mass) level of 25-hydroxyvitamin D calcifediol (AKA "calcidiol") which the immune system needs to function properly.
Please see the research cited and discussed at: https://vitamindstopscovid.info/00-evi/. Regarding low circulating 25-hydroxyvitamin D being a huge risk factor for neurodegeneration: https://vitamindstopscovid.info/00-evi/#3.3 and (in pregnancy and early childhood) preeclampsia, pre-term birth, mental retardation, ADHD and autism: https://vitamindstopscovid.info/00-evi/#3.2.
This page begins with graphs from Quraishi et al. 2014 https://jamanetwork.com/journals/jamasurgery/articlepdf/1782085/soi130062.pdf which show that immune responses to bacterial pathogens which cause post-operative infections are highly effective (2.5% risk of hospital-acquired and 2.5% risk of surgical site infections) with pre-operative circulating 25-hydroxyvitamin D levels of 50 ng/mL or more, but are weak (25% risk of each type) with 20 ng/mL, which is a normal level for those who do not supplement vitamin D3 properly, and who have not recently had high levels of UV-B irradiation of ideally white skin.
Most doctors regard 20 ng/mL circulating 25-hydroxyvitamin D as sufficient for good health. However, this is only what the kidneys need to play their role in regulating calcium-phosphate-bone metabolism, which is the function of the three vitamin D compounds medical professionals all understand.
What they don't know is that many types of cell, including many or perhaps most types of immune cell, rely on a good supply of 25-hydroxyvitamin D as a raw material for their 25-hydroxyvitamin D to calcitriol (1,25-dihydroxyvitamin D) intracrine signaling systems. These are signaling systems entirely within a single cell and are crucial to the ability of many types of cell to alter their behavior in response to their changing circumstances. A related paracrine signaling system is used by some types of immune cell to signal to nearby immune cells, typically of different types.
25-hydroxyvitamin D to calcitriol intracrine signaling was first discovered in dendritic cells and macrophages by Martin Hewison and colleagues in the late 2000s. In 2021 a team of NIH researchers elucidated this signaling system (which they named, somewhat incorrectly, as "autocrine" signaling) in Th1 regulatory lymphocytes: Chauss et al. https://www.nature.com/articles/s41590-021-01080-3. I summarized this dense (and I think Nobel Prize worthy) cell biology article at: https://aminotheory.com/cv19/icu/#2021-Chauss.
Th1 cells from the lungs of hospitalized COVID-19 patients were found to be stuck in their pro-inflammatory startup program, despite detecting the condition (a high level of a complement protein) which should cause them to transition to their anti-inflammatory shutdown program. The researchers describe in great detail the 25-hydroxyvitamin D to calcitriol intracrine signaling system which achieves this transition. The failure or this system in these patients, who almost invariably have very low circulating 25-hydroxyvitamin D levels) was found to be due entirely or almost entirely, to these cells having inadequate supplies of 25-hydroxyvitamin D in their cytosols. (These people had severe disease due to poor initial responses to the virus and due to the pathologically sustained inflammatory response once the virus infected the alveoli. If they had had 50 ng/mL or more circulating 25-hydroxyvitamin D they would generally have had only mild COVID-19 symptoms.)
Since there are no peer reviewed articles which explain 25-hydroxyvitamin D to calcitriol intracrine signaling, I wrote a non peer reviewed tutorial: https://vitamindstopscovid.info/02-intracrine/. A simpler version is at: https://vitamindstopscovid.info/00-evi/#02-compounds.
This and the related paracrine signaling systems enable cell-type specific conditions to be detected so that the cell responds in a cell-type specific manner. These signaling systems are unrelated to hormonal (endocrine) signaling, which the kidneys use (with calcitriol as the blood-borne signaling molecule, AKA hormone) to alter the behavior of cells in distant parts of the body.
Neither vitamin D3 cholecalciferol no 25-hydroxyvitamin D are hormones. They are not signaling molecules. The immune system does not use hormonal signaling. Medical professionals and many vitamin D researchers refer to all three compounds as if they were all "vitamin D". This is a mistake, as Reinhold Vieth pointed out in 2004: https://sci-hub.se/10.1016/j.jsbmb.2004.03.037.
25-hydroxyvitamin D, as measured in "vitamin D" blood tests, is made, primarily in the liver, by hydroxylating vitamin D3 cholecalciferol on the 25th carbon.
There's very little vitamin D (2 or the natural 3) in food, including food with is fortified with vitamin D3 or the less effective D2.
Vitamin D3 can be produced with UV-B irradiation of ideally white skin, but most people cannot attain this naturally (dark or black skin, living too far from the equator etc.) to get this all year round - and all such UV-B exposure damages DNA and so increases the risk of skin cancer.
Supplementation of vitamin D3, in quantities well above those needed for 20 ng/mL 125 nmol/L 25-hydroxyvitamin D is needed by almost everyone in order to be healthy. The exceptions are babies breast fed by 25-hydroxyvitamin D replete mothers (about equal amounts of vitamin D3 and 25-hydroxyvitamin D in the breast milk, with the latter being 4 times as effective as vitamin D3 in raising the baby's 25-hydroxyvitamin D) and those who get so much UV-B skin exposure, year after year, that they are at high risk of developing skin cancer.
In 2022, New Jersey based Professor of Medicine Sunil Wimalawansa published an article with proper vitamin D3 supplementation recommendations, as ratios of body weight, with higher ratios for those suffering from obesity, since obesity reduces the rate of hydroxylation in the liver to 25-hydroxyvitamin D and because the excess adipose tissue absorbs 25-hydroxyvitamin D and vitamin D3: https://vitamindstopscovid.info/00-evi/#obesity-deficit.
He later simplified these recommendations somewhat to average daily supplemental intakes of vitamin D3:
70 to 90 IU / kg body weight for those not suffering from obesity (BMI < 30).
100 to 130 IU / kg body weight for obesity I & II (BMI 30 to 39).
140 to 180 IU / kg body weight for obesity III (BMI > 39).
For 70 kg (154 lb) without obesity, this is about 0.125 milligrams (5000 IU) a day. This takes several months to attain the desired > 50 ng/mL circulating 25-hydroxyvitamin D. This is 8 or more times what most governments recommend. "5000 IU" sounds like a lot, but it is a gram every 22 years - and pharma grade vitamin D costs about USD$2.50 a gram ex-factory. This will safely attain at least 50 ng/mL 125 nmol/L 25-hydroxyvitamin D over several months, without the need for blood tests or medical monitoring. He recently repeated these recommendations in an article co-written by two other professors - one of medicine and the other of pediatrics:
Integrating Endocrine, Genomic, and Extra-Skeletal Benefits of Vitamin D into National and Regional Clinical Guidelines Sunil J. Wimalawansa, Scott T. Weiss and Bruce W. Hollis, Nutrients 2024-11-20 https://www.mdpi.com/2072-6643/16/22/3969.
For patients who are interested in understanding their polygenic risk for common conditions like heart disease or cancer—but whose physicians don’t currently use or offer this kind of testing—where do you recommend they turn? Are there trustworthy, clinically valid platforms you would feel comfortable pointing people toward, especially those trying to be proactive about their health?
My 83-year-old brother just succumbed to complications of Alzheimer's. I am one or two years away from when he started to show symptoms. My PCP refuses to order a p-tau217 test for me, and says that a referral to a neurologist would not be appropriate due to my age and condition. What you describe is utopian, not realistic in the current health care climate.
Question regarding apoe 4/4 and statin use
With elevated LDL
p tau 217 0.44( limit in Canada 0.47 )
Thanks
Find your info very useful