Biological Age

Well, it's mostly a supply and demand issue. Insurance companies driving up both demand and the costs to obnoxious, extravagant levels in the same way that student loans drove up tuition costs. On the supply side we have protectionism from government, licensing, regulations, control over the medical boards and school numbers limiting the supply of doctors, consolidating all the doctors into large institutions.

Here's the whole story right here if you can stand reading dry economics and history blog posts. This guy goes all the way back to the 19th century.

I'll probably get dinged for being off topic here but for those of you interested in getting panels done, this is what you're facing in the U.S.

My deductible will be met after I pay off this giant bill and I'll be ordering all kinds of panels and making them pay for the whole thing this year. Question is whether to do it before or after I do my first fast. Before would tell me whether all the supplements I've taken and diet changes over the past year have done squat as far as pushing me in a positive direction. The latter would be a different story.

Edited by Nate-2004, 09 May 2017 - 07:25 PM.

I am not challenging your general thesis that medical expenses have gotten way out of whack. Are we as consumers also part of the problem?

We can order lab tests on line at any time in some States at least. I have done it through LifeExtension multiple times. Around $300 for a very comprehensive panel.

Edited by Heisok, 09 May 2017 - 07:44 PM.

I'll keep these in mind for next year but now that I've had to pay off my whole deductible this year I might as well take advantage of all the panels I can get for that. I like that there are inexpensive means of getting around the system now. I do hope that AGI evolves with medical technology soon to replace or at least alleviate demand on undersupplied medical personnel.

I am not challenging your general thesis that medical expenses have gotten way out of whack. Are we as consumers also part of the problem?

Yes, sort of, not really. Insurance has enabled high artificial demand and consumption for sure, but much like student loans and venture capital, costs rise not only with the demand and restricted supply, but with "other people's money". People will always charge more when far more abundant resources are willing to pay.

Edited by Nate-2004, 09 May 2017 - 09:10 PM.

Q: You’ve been using InsideTracker for quite a while. How many times had you tested prior to this?

David: “Yeah, I test routinely. So it’s many times per year; three or four times. I have a long record of my blood work, going back to 2012, I believe — maybe even 2011. So I know what my body does and I’ve been able to actually see my body getting older over time, which was quite disturbing to me."

http://blog.insidetr...i-aging-success

Q: You’ve been using InsideTracker for quite a while. How many times had you tested prior to this?[/size]
David: “Yeah, I test routinely. So it’s many times per year; three or four times. I have a long record of my blood work, going back to 2012, I believe — maybe even 2011. So I know what my body does and I’ve been able to actually see my body getting older over time, which was quite disturbing to me."[/size]
http://blog.insidetr...i-aging-success[/size]

 

That's very disappointing: the InsideAge measure hasn't been publicly validated against anything, and may be nothing more than an arbitrary guess by the founders.
 
Meanwhile, an independent group tested the Aging.AI panel in a set of different populations, and found that it gives wildly different results in different populations:
 

Deep biomarkers of aging are population-dependent

... There is a substantial caveat to Putin et al. [1]’s approach that was not mentioned in their article. Their algorithm was developed based on clinical data from a single source covering Eastern Europe (90% Russia), and the applicability to data from other settings or to population subsets was not verified. There are a number of reasons to suspect that their algorithm would need to be adjusted for application in other settings: (1) Aging rates may differ across countries; (2) Genetic and environmental determinants of physiology may differ across countries/cultures, independent of aging; and (3) There may be specific biases in how clinical lab samples are taken and analyzed that differ substantially across health systems.  ...

We have access to similar data to that used by Putin et al. [1] for three major aging cohort studies, the Women’s Health and Aging Study I &II (WHAS) [8], the Baltimore Longitudinal Study on Aging (BLSA) [9,10], and Invecchiare in Chianti (InCHIANTI) [11], as well as publicly available cross-sectional data for a representative sample of the American population from the National Health and Nutrition Examination Survey (NHANES) [12]. For each study, we randomly chose 110 participants, stratified by age when necessary to achieve a broad age range, and input their values for the 10 basic biomarkers ... at www.aging.ai. Alpha-2-globulins were only present in InCHIANTI, so we left the field empty in the other data sets (the DNN is capable of treating missing data, though this reduces accuracy). In addition, we ran as many of the full 41 biomarkers as possible for a set of 10 individuals per study, chosen randomly by age stratum from among the 110 run with 10 biomarkers. The number of biomarkers available was: WHAS: 34 biomarkers out of 41, BLSA: 37, InCHIANTI: 38, and NHANES: 33.

We found that indeed the performance of the model was substantially diminished in all four of our data sets. In the original study, the 10-biomarker version of the DNN has a 10-year epsilon accuracy (i.e., percentage correct prediction within age±10 years) of 70% and R² = 0.63; across our datasets the mean epsilon accuracy was 38% and mean R² = 0.37, with maximum epsilon accuracy = 56% (InCHIANTI) and maximum R² = 0.59 (NHANES, Fig. 1). The 41-biomarker versions performed neither markedly better nor worse, with a mean age error (MAE) actually increasing by 0.45 (95%CI: [-2.2, 1.3]) across our 40 samples. The confidence intervals and consistency across data sets are sufficient to exclude the possibility that our core results are due to the use of the 10-biomarker rather than the 41-biomarker tool (Fig. 1).

Recent study showed the shape of your neck arteries may be as good as anything:

https://mipt.ru/engl...d_artery_health

I have heard Bill Andrews make the case that telomeres have a lot of value as well.  You need the comprehensive test though probablyl

I feel that tools such as aging.ai or similar, when put in an appropriate context using complex network analysis they might actually guide in (cautious) prediction of biological aging too and possibly help in a clinical setting. I would need also to research better on DNA methylation which I often hear from experts mentioned as a powerful aging biomarker.

Even if causality cannot be inferred, what the authors below refer here as "integrated albunemia" is an example of what can come out of the network analysis, apparently conserved across populations and related to frailty, one of the typical aging characteristics: e.g. there is connection between chronic inflammation (CRP), hand grip strength and physical decline in the old and very old as well albumin and sarcopenia, even if to a limited extent. We do need in depth analytics integrating all this and research for mechanistic explications of the underlying of "integrated biomarkers" implying both therapeutics research and in parallel to "damage-repair" approaches à-la-SENS.

"Many studies of aging examine biomarkers one at a time, but complex systems theory and network theory suggest that interpretations of individual markers may be context-dependent. Here, we attempted to detect underlying processes governing the levels of many biomarkers simultaneously by applying principal components analysis to 43 common clinical biomarkers measured longitudinally in 3694 humans from three longitudinal cohort studies on two continents (Women’s Health and Aging I & II, InCHIANTI, and the Baltimore Longitudinal Study on Aging). The first axis was associated with anemia, inflammation, and low levels of calcium and albumin. The axis structure was precisely reproduced in all three populations and in all demographic sub-populations (by sex, race, etc.); we call the process represented by the axis “integrated albunemia.” Integrated albunemia increases and accelerates with age in all populations, and predicts mortality and frailty – but not chronic disease – even after controlling for age. This suggests a role in the aging process, though causality is not yet clear. Integrated albunemia behaves more stably across populations than its component biomarkers, and thus appears to represent a higher-order physiological process emerging from the structure of underlying regulatory networks. If this is correct, detection of this process has substantial implications for physiological organization more generally." (bold mine)

Cohen AA, Milot E, Li Q, et al. Detection of a novel, integrative aging process suggests complex physiological integration. PLoS ONE. 2015;10(3):e0116489.

https://www.ncbi.nlm...les/PMC4356614/

Sousa AC, Zunzunegui MV, Li A, Phillips SP, Guralnik JM, Guerra RO. Association between C-reactive protein and physical performance in older populations: results from the International Mobility in Aging Study (IMIAS). Age Ageing. 2016;45(2):274-80.

https://www.ncbi.nlm...pubmed/26822196

"...There is a long history of attempts to determine biological age and quantify the tempo of the process of ageing. Typically, age determination utilizes one or another molecular facet of ageing, for example, the degree of the damage to cell's DNA [2]. Among more recently developed integrative biomarkers of aging is the GlycanAge index that profiles the structural details of sugar chains attached to the conserved N-glycosylation sites of three types of IgG molecules. This index reflects the level of systemic inflammation, predicts chronological age with standard deviation of 9.7 years, and is superior to age evaluation using telomere length [3]. Peripheral blood mononuclear cells (PBMCs) mRNAs-based “trans-criptome age” index predicts chronological age with mean absolute error of 7.8 years [4]. Even more precise PBMCs-based “epigenetic age” relies on quantitation of the methylation of three CpG sites located in ITGA2B, ASPA and PDE4C genes with standard deviation of less than 5 years [5]. An increase in the number of profiled CpG dinucleotides to 353 improves epigenetics-based age estimates by decreasing an error down to 2.9 years [6]. This technique is also capable of predicting mortality (p < 0.003) [7], but not the probability of major cardiovascular event [8]. It should be noted that all the techniques described above require specialized equipment and skilled laboratory personnel, thus, limiting their clinical applicability. On another end of the spectrum are age-predicting models not specifically connected to any particular mechanism of aging, for example, deep neural networks (DNNs) modules evaluating common blood biochemistry and cell count tests [9]. Though the accuracy of this model is quite high, the number of parameters in the model is also high. Since deep neural nets are, in a nutshell, “black boxes”, the dissection of these models into mechanistic insights into the process of ageing is impossible..." (bold mine)

Fedintsev A, Kashtanova D, Tkacheva O, et al. Markers of arterial health could serve as accurate non-invasive predictors of human biological and chronological age. Aging (Albany NY). 2017;9(4):1280-1292.

https://www.ncbi.nlm...les/PMC5425127/

I was at the meeting in Basel (Switzerland) where Insilico announced Young.Ai. I never tried it but it might develop into an AI system to track how you are doing with your interventions. I expect a blossoming of tools like this but what would really matter will be their clinical relevance, adoption, reproducibility, consistency across multiple platforms using the same set of curated data etc ....

In any case, an interesting space to watch:

 youngAI.PNG   297.16KB   0 downloads

https://young.ai/#/

... A good biological aging score would accurately predict mortality (and preferably morbidity) in basically healthy community-dwelling people, and would do so better than their chronological age. Ideally, it would track the accelerating increase in mortality rate across the lifespan from middle-age onward. They haven't yet even attempted to do any of this yet: again, they've just tried to construct a score that correlates well with chronological age....

The following is not likely what you rightly point out as a biological aging score predicting mortality but maybe a step in that direction:

"MortalityPredictors.org is a database of human biomarkers associated with all-cause mortality in humans"

http://mortalitypredictors.org/

I try using it as a biomarker relative importance comparator to mortality and as literature source. Any comment?

Trying to merge insight from the main and cross-ethnicity blood bio-markers from InSilico Medicine AI algorithms (e.g. albumin, ALP, ....) and the all causes mortality largest blobs in mortalitypredictors.org , there is indeed some recoupling as one would expect,  e.g. here with liver funtion tests as albumin and GGT (in the "composite" category), ALT enzyme (in the "blood" category) and ALP. 

I think much more work is needed though along these directions to find effective biomarkers of biological aging but think definitively machine learning and AI are helping finding a path.

 Top 5 cross ethnicity.PNG   25.34KB   0 downloads

 Top 5.PNG   294.2KB   0 downloads

Population specific biomarkers of human aging: a big data study using South Korean, Canadian and Eastern European patient populations

Putin E, Mamoshina P, Aliper A, et al. Deep biomarkers of human aging: Application of deep neural networks to biomarker development. Aging (Albany NY). 2016;8(5):1021-33.

ALT= Alanine transaminase

ALP= Alkaline phosphatase

GGT= Gamma-glutamyltransferase

....

 

FWIW, the only decent biological aging score of which I'm aware are the composite Frailty Indexes built up in different publications by Rockwood and Mitnitski (see here, here, here, here, and here). However, it's mostly been validated in elderly people, and may not be sensitive enough for significant effects of metabolically-based interventions; it's also a much more complicated construct than a simple blood test.

....

Thank you for your input Michael. I agree. I just wish to gather your knowledgeable advise on whether or not also methylation based aging clocks, which I understand provide an accurate representation of chronological age, suffer of the same problem i.e. lacking sensitivity to metabolically base interventions. I think this would mainly due to epigenetics information being relatively stable. In addition there are probably issues with practicality of epigenetics tests , where mortality validation of the Insilico Medicine algorithms would be welcome, and standardization. Also, as with the frailty tests, there is a clear direction of using composite markers.

Shipony Z, Mukamel Z, Cohen NM, et al. Dynamic and static maintenance of epigenetic memory in pluripotent and somatic cells. Nature. 2014;513(7516):115-9

Murabito JM, Zhao Q, Larson MG, et al. Measures of Biologic Age in a Community Sample Predict Mortality and Age-Related Disease: The Framingham Offspring Study. J Gerontol A Biol Sci Med Sci. 2017

Belsky DW, Moffitt TE, Cohen AA, et al. Eleven Telomere, Epigenetic Clock, and Biomarker-Composite Quantifications of Biological Aging: Do They Measure the Same Thing?. Am J Epidemiol. 2017

From the first paper (Population specific ..) I quoted above:

 validation.PNG   86.11KB   0 downloads

I think this is a good validation of the algorithm, as mentioned in my previous post and suggest the algorithms approaches the biological age. I understand the data tell that high (low) mortality rate is found for those people, across ethnicity, who are predicted older (younger) than the chronological by the algorithms.

"... A Cox proportional hazards regression model was used to relate survival time to the accelerated aging group (delta >5) and slowed aging group (delta <5). Patients predicted younger their chronological age has a lower mortality risk, while patients predicted older has a higher risk. Each row represents a hazard ratio and 95% confidence interval ..."

Bumping up this interesting thread. 

Michael was actively and critically following as usual but, as the OP, he seems has decided to stop. A pity because the topic is important.

Is there another thread?

I'm still following, I know lifespan.io crowdfunded a project for developing a way to measure aging but not sure where they are with it.

I'm still following, I know lifespan.io crowdfunded a project for developing a way to measure aging but not sure where they are with it.

Same for me. I am happy Nate you do follow this. If I recollect well you said this is THE thread in the Forums to discuss this topic and hopefully others will add contributions. I miss a bit Michael but guess he is taken on too many other fronts.

Interesting ...

Wei Gan, Xin-Le Liu, Ting Yu, Yuan-Gao Zou, Ting-Ting Li, Shuang Wang, Jin Deng, Lan-Lan Wang, Jian-Ping Cai. Urinary 8-oxo-7,8-dihydroguanosine as a Potential Biomarker of Aging. Frontiers in Aging Neuroscience, 2018; 10 DOI: 10.3389/fnagi.2018.00034

"...To date, most studies have dealt with urinary 8-oxodGsn, and a very limited number of studies have focused on 8-oxoGsn. This discrepancy in focus may be because 8-oxoGsn is an adduct from RNA, which does not pass on to the next generation. However, RNA plays an important role in protein translation, and any modification of its bases will directly lead to the formation of abnormal proteins. One important characteristic of aging is the accumulation of dysfunctional proteins. It is therefore reasonable to consider that 8-oxoGsn is a biomarker of aging.

Our study demonstrated that 8-oxoGsn is a better aging marker than 8-oxodGsn in two respects. First, the level of 8-oxoGsn was higher (approximately 2-fold) than 8-oxodGsn in age-matched counterparts. Second and more importantly, the levels of 8-oxoGsn correlated better with the rate of aging (Table 2 and Figure 1). The 8-oxoGsn content does not always correlate with chronological aging (Sprott, 2010) but instead reflects the actual physiological stage of aging.

In conclusion, a simple, rapid, sensitive and reliable methodology for the analysis of urinary excretion of 8-oxoGsn and 8-oxodGsn was developed. Based on this large population study, we concluded that 8-oxoGsn in urine may be a novel way to evaluate the aging process in adults...." (bold mime)

I just wonder if someone out there has been doing the small exercise to see if there is a trend in the age predicted by the Insilico's algorithm vs. the chronological age. I was curious to see if a trend could hint to a possible impact of our lifestyle, nutrition, supplementation etc. In my case the predicted age has been basically flat at around 32 yo, between 50 and 60 yo. I used Aging.Ai v. 3.0 where you enter 19 biomarkers plus weight and height (Insilico indicates a MAE of 5.9 years).

I just wonder if someone out there has been doing the small exercise to see if there is a trend in the age predicted by the Insilico's algorithm vs. the chronological age. I was curious to see if a trend could hint to a possible impact of our lifestyle, nutrition, supplementation etc. In my case the predicted age has been basically flat at around 32 yo, between 50 and 60 yo. I used Aging.Ai v. 3.0 where you enter 19 biomarkers plus weight and height (Insilico indicates a MAE of 5.9 years).

Test and bio-hack since 9 years due to various diagnoses (PAD, COPD, T2D, an old brain infarction I didn't even knew of..). Though there are large fluctuation between lab tests, in general most parameters have improved during these yeas.

If I for simplicity enter the average values for these 9 years, my predicted age was 39 years, while 4 1/2 years ago I actually was 46 - and despite having struggled with so many chronic diseases all these years.

In my opinion for this to work more accurately, much more bio-markers would have to be included. For example had consistent low fT3, which is associated with chronic disease and mortality.

 

Very informative! Indeed this is what I'm looking for "http://www.aging.ai/  ".

 

Strange, my age-range must be truly in bad shape.

 

Had a PAD diagnosed due to a 80% stenosis at my abdominal aorta bifurcation and was predicted a 30% change of dying within 5 years. That was 8 years ago, since which I did regular blood-tests and 2 years ago also reverted a government-certified 60% walking disability, mainly with consistent Linus Pauling's therapy and diet. Other conditions: prediabetes (controlled with diet), non-symptomatic COPD and countless infections: spondylodiscitis, schistosomiasis, amoebic hepatitis, 7 times malaria, psoriasis, rhinitis..

 

Entered the average values because of quite some fluctuations within 8 years, even though actually much improved during this time, still came out 1 year younger than I was 4 years ago. And they got my gender right too.

Just realized already took this test 1 year ago. Just since the last year seem to already have gained 5 years!
 

 

I just wonder if someone out there has been doing the small exercise to see if there is a trend in the age predicted by the Insilico's algorithm vs. the chronological age. I was curious to see if a trend could hint to a possible impact of our lifestyle, nutrition, supplementation etc. In my case the predicted age has been basically flat at around 32 yo, between 50 and 60 yo. I used Aging.Ai v. 3.0 where you enter 19 biomarkers plus weight and height (Insilico indicates a MAE of 5.9 years).

 

Test and bio-hack since 9 years due to various diagnoses (PAD, COPD, T2D, an old brain infarction I didn't even knew of..). Though there are large fluctuation between lab tests, in general most parameters have improved during these yeas.

 

If I for simplicity enter the average values for these 9 years, my predicted age was 39 years, while 4 1/2 years ago I actually was 46 - and despite having struggled with so many chronic diseases all these years.

 

In my opinion for this to work more accurately, much more bio-markers would have to be included. For example had consistent low fT3, which is associated with chronic disease and mortality.

 

The version 1.0 includes 41 biomarkers vs the 19 of the version 3.0.  fT3 is not included but there are others, e.g. related to liver function (e.g. ALT Alanine Transaminase) which also have a link to the mortality risk, e.g. see the blob size of ALT in http://mortalitypredictors.org/. As you mention fT3 I would add kidney function (as estimated by the eGFR) which has a strong link to age and mortality and Apo A-1, both have also large blob size in the mortality predictor. I also know Insilico is working on their own mortality predictor.

In a couple of points I tested the v1 predicts a higher age vs the v3 typically by ca. 30%. Would be curious to see if you got similar results.

 

 

 

Just realized already took this test 1 year ago. Just since the last year seem to already have gained 5 years!
 

 

I suggest you do not put too much trust in it. A trend, as I tried to extract but could not determine (at least with the version 3.0), would be more important than absolute values. Remember the algorithm itself has an absolute error of 5.5 to 6.2 years, depending on the versions.

In a couple of points I tested the v1 predicts a higher age vs the v3 typically by ca. 30%. Would be curious to see if you got similar results.

Oops, different versions I wasn't aware. Entered in v1 again and now ended up with the prediction of 66 years

 

Oops, different versions I wasn't aware. Entered in v1 again and now ended up with the prediction of 66 years

 

I got a similar result.  Only off by 20+ years.

Did additional measurements:

- v1.0 returns systematically higher predicted age values than v3.0 (by about 50%)

- v1.0 shows (contrary to v3.0) a pleasant trend downward at higher chronological ages, possibly hinting to some positive effects of lifestyle

However, interpreting all this, the Richard Feynman's words resound loudly: "The first principle is that you must not fool yourself and you are the easiest person to fool." ;-)

Has anyone tried the $65 epigenetic aging test from Osiris Green?

https://www.osirisgreen.com/

https://www.fightagi...-to-the-public/

I should have some results from them in a couple of weeks.

Several things, here.
 
<< SNIP >>

.

FWIW, the only decent biological aging score of which I'm aware are the composite Frailty Indexes built up in different publications by Rockwood and Mitnitski (see here, here, here, here, and here). However, it's mostly been validated in elderly people, and may not be sensitive enough for significant effects of metabolically-based interventions; it's also a much more complicated construct than a simple blood test.

<< SNIP >>

i sometimes wonder if Michael and I are living in the same universe...

Studies about Estimated Remaining Lifespan via techniques focusing on Epigenetic Methylation...

• 2011, Epigenetic Predictor of Age, Horvath
• 2013, DNA methylation age of human tissues and cell types, Horvath
• 2014, Biomarkers and ageing: The clock-watcher, Gibbs... (A summary of the early Horvath work.)

Studies about Estimated Remaining Lifespan via techniques focusing on Telomere Length...

•  

Studies about Estimated Remaining Lifespan via techniques focusing on both Epigenetic Methylation and Telomere Length...

• 2016, Telomere, epigenetic clock, and biomarker-composite quantifications of biological aging: Do they measure the same thing?
• 2016, The epigenetic clock and telomere length are independently associated with chronological age and mortality

AgeMeter... A new means for estimating age in the project/development stage. I like the name but don't know much it. Including it here because Longevity Science Movement (LSM) organizations are involved in promoting and/or developing it.

• 2017, New Lifespan.io Campaign Launch: AgeMeter!

 

Here's one not referenced above...

2017 - DNA methylation-based measures of biological age: meta-analysis predicting time to death, Horvath

Abstract

Edited by HighDesertWizard, 08 March 2018 - 07:24 AM.