New Evidence that Long Telomeres cause Cancer, and Why I Think It’s Wrong

Since 2003, I’ve been saying that long telomeres are a path to long life.  The opposing view says that nature allows our telomeres to shorten to protect us against cancer.  Up until this spring, there has been little evidentiary support for the cancer theory.  Now, a major new study uses genetics to argue that longer telomeres increase risk of cancer as much as five-fold.  The study contains many statistical checks, but I’m going out on a limb to say I think the experts have made a conceptual error.

Up until now, epidemiological studies in humans and lab studies in animals have shown consistently that shorter telomeres increase risks for all the diseases of old age.  People’s telomere length tends to decline with age, but among people of the same age, those with shorter telomeres tend to die sooner.

The new study finds a very different conclusion: that shorter telomere length leads to much lower risk of cancer, while longer telomere length leads to slightly lower risk of heart disease.  Put these two together, and you predict pretty much the same life expectancy for people with long telomeres and short telomeres.

The new studies are based on genetics and account for telomere length only indirectly.  Nevertheless, it is claimed, they are more reliable than the old studies (based on direct observation) because they are able to eliminate a statistical anomaly that (they claim) is super-important.  

I believe the new study is actually less reliable, and that we should believe the more direct studies like the ones I have reported here in the past.  My reasons are that

  • The previous studies are straightforward, direct correlations.  Methodology in the new study relies on very small differences in telomere length, tiny differences that are lost in the noise and very difficult to detect.
  • The new studies require very large implicit extrapolation that is not necessary in the old studies.  The 50 to 1 extrapolation is very speculative, and it magnifies the noise along with the signal.
  • It is likely what these new studies are seeing are actually direct effects of genetics on cancer risk.  Even very small (direct) effects of genotype on cancer would appear in their methodology as though they were huge (indirect) effects of telomere length.  This is what I believe is happening, and why I don’t trust their results.

I may be wrong about this.  I’m questioning seasoned experts in the field based on my general knowledge of statistics.

Two years ago, I reported on a Danish study linking short telomeres to higher mortality, especially heart disease (CV).  I took this as clear proof that telomere length was not just a marker of aging but a cause.  The implication is that you can live longer by adopting lifestyles and taking supplements that extend your telomeres.

The core of my argument, based on the the Danish study, was this:

  • Impact of telomere length on mortality, raw data:   3.38 (meaning that the 10% of people with the shortest telomeres were dying at a rate 3.38 as high as the 10% with the longest telomeres)
  • Same calculation, corrected for age:  1.54
  • Same calculation, corrected for age and all other hazard variables:  1.40

Conclusion: This demonstrates that age is the biggest factor in mortality, and telomere length is second, with a strong effect, independent of age.  All the health variables together have only minor effect compared to age and telomere length.

The Danish study did a multivariate analysis, also called ANOVA.  This is a statistical technique designed to separate out the factors that contribute to an outcome (in this case, mortality) and assign percentages of causality.  What their analysis revealed was that the strongest cause of increased mortality is age itself, and that telomere length comes second.  Everything else, from smoking to depression to a history of infections, is much less important than age and telomere length.  I interpret this to say that short telomeres are probably a direct cause of increased disease risk.

A popular theory is that the association of short telomeres with higher mortality is only incidental.  Stresses, infections, smoking, etc. cause both shorter telomeres and higher mortality.  But these are separate pathways.  It is not the shorter telomeres that are causing higher mortality, but short telomeres happen to be associated with higher mortality because both are caused by various stressors in a person’s past.  If this is true, you can’t improve your odds of living longer just by extending your telomeres.

But I believe that the Danish study disproves this theory.  If the stressor theory were correct, then the Danish analysis would have found that the relationship between stressors and mortality was stronger than the relationship between telomere length and mortality.  In fact, they found the opposite.


The New Genetic Study

The result reported by the new study is that longer telomeres creates a very much higher risk of several common cancers.  On the other hand, longer LTL (=leucocyte telomere length) protects against heart disease.  The protective effect for heart disease is much smaller, but many more people die of heart disease than of these particular cancers.  The result is a wash.  Longer LTL is neither a net benefit to health nor is it a net risk.  People with longer and shorter LTLs have similar overall mortality risk, about the same life expectancy.


Disease Odds Ratio
Glioma 5.27
Ovarian cancer 4.35
Lung cancer 3.19
Neuroblastoma 2.98
Bladder 2.19
melanoma 1.87
Testicular 1.76
Kidney 1.55
Endometrial 1.31
Basal cell skin 1.22
Breast Cancer 1.06
Heart disease 0.78

“Odds ratio” refers to a person’s probability of contracting the corresponding disease.  For example, the first line means that people whose telomere length is one standard deviation (1 sigma) longer than average have a risk of glioma 5 times greater than people who have average telomere length.

This result gains credibility because it is exactly what the theory would predict.  Nature has optimized LTL by compromising between two risks.  If the average LTL for our species were longer, then we’d get more cancer.  If it were shorter, we’d get more heart disease.  The reason there is so much variation among the population, people with much longer and much shorter telomere length, is that it doesn’t matter very much.

So here is agreement between experiment and theory, a tidy situation that scientists like to see.  What is more, there is a widely-held belief that the methodology of the new study is more reliable than studies in the past that are more direct and simpler.  Nevertheless, I’m about to offer my opinion that the previous studies were right, the theory is wrong, and, in fact the design of the new study is seriously flawed.

This was the latest and far the largest in a series of GWAS studies going back four years [ref, ref, ref].  GWAS stands for Genome-Wide Association Study.  The idea is to work around life experience variables that might create a correlation without a causal connection.  In the present case, the target is to detect any causal relationship between leucocyte telomere length (LTL) and various diseases, while filtering out associations between LTL and disease risk that might be incidental, as described above.  The researchers looked for small genetic differences (called SNPs) that are linked to telomere length.  These vary from one individual to the next, and they persist through a lifetime.  The next step is to compare numbers of people with a particular SNP variant among those who have the disease and those who don’t have the disease.  Are people who have the SNP associated with longer telomeres more or less likely to develop the disease?  From the answer to this question, they infer a causal relationship, not between SNP and the disease but between telomere length and the disease.

Observational studies look for a direct relationship between LTL and disease.  GWAS studies look for an indirect relationship between SNP and LTL, SNP and disease.  The indirect study is widely considered to be a more reliable indicator of causal connection than the direct study.  Why?

“Mendelian randomization studies are less susceptible to confounding in comparison to observational studies…Given the random distribution of genotypes in the general population with respect to lifestyle and other environmental factors, as well as the fixed nature of germline genotypes, these results should be less susceptible to confounding and reverse causation than those generated by observational studies.”

The reasoning is that people have their genomes for their entire lives, independent of how they live, what they do, what they are exposed to.  By working with the genome, the statisticians can be sure to eliminate the standard objection that (for example):

  • Stress directly decreases LTL
  • Stress directly increases risk of disease
  • Therefore, short LTL will appear to be linked with disease, even though short LTL doesn’t cause disease.


Problems with GWAS studies

But the GWAS methodology also introduces new problems of its own.  The main problem is that the statistical sensitivity of the study is seriously reduced.  This is because the relationship between SNP and LTL is very weak.  All sixteen SNPs together constitute a very small factor among many larger ones that create difference in LTL between one person and the next.

“The selected SNPs correspond to 10 independent genomic regions that collectively account for 2% to 3% of the variance in leukocyte telomere length”

And of course, very few people have all 16 SNPs going in the same direction.  The study is forced to work with people who have, for example 10 positive SNPs out of 16 compared to others who may have 5 positive SNPs out of 16.

Their LTL is really quite close together.  To compensate for this, the statisticians divide by a small number to extrapolate outwards.  For example, the difference between typical people in the study is about 1/20 sigma*.  And the difference between risk of glioma (brain cancer) for these people is only about 0.08** .  But the difference is reported as “what would have been the risk of brain cancer if the difference had been not 1/20th but one full sigma.  They extrapolate exponentially, so the conclusion comes out quite startling: They claim that people with 1 sigma of extra LTL have 5 times greater chance of getting brain cancer.

What they find: people with 0.05 sigma extra LTL have 1.08 times the risk of getting brain cancer.

What they report: people with 1 sigma extra LTL would have (by extrapolation) 5 times the risk of getting brain cancer.

They conclude that there is a large effect of telomere length on cancer, but they do this by the following reasoning:

  • There is a small effect of these genetic variations on telomere length.
  • There is a small effect of these genetic variations on cancer risk.
  • Dividing the small by the small, they conclude: if the mechanism for these genetic variations affecting cancer is mediated by their effect on telomere length, then the effect of telomere length on cancer must be quite large.

I’m sorry to belabor this, but it’s important, and it’s hidden in the methodology.  People who do these studies know that an odds ratio (OR) of 1.08 means nothing.  The state of the art in epidemiology is rarely able to attach meaning to odds ratio that is close to 1.  It is lost in the nosie.  But an OR of 5 is something easy to see.  It stands out from the noise and is easy to detect.

The description of the methodology in this study hides the fact that they are working with ORs less than 1.08 and extrapolating exponentially outward to make the ORs look very large and significant.


What I think is really going on

The study finds a large and consistent result that demands some explanation.  I’m claiming that the explanation they offer (in terms of telomere length) is wrong.  So why do I think they get the results that they did?

A few of the sixteen SNPs that are considered in the study correspond to slight variations on the form of the telomerase molecule.  I’m guessing that these mutated forms of telomerase cause an increased risk of cancer.  The increased risk doesn’t have to be large.  As in my example above, the increased risk for brain cancer would have to be just 8%, and the increased risk for lung cancer (more important because it is more common) only 6%.  Because of the extrapolation by an exponent of 20 that is implicit in their methodology, these small effects would be reported as though they were odds ratios of 5 (for brain cancer) and 3 (for lung cancer).

Another possibility is that one or more of the SNPs happen to be more common in a segment of the population that is prone to cancer, for whatever reason.  It may be that a particular SNP is more common in an ethnic group that has high smoking rates, or that is prone to melanoma because of lighter skin, or has a diet and lifestyle that leads to a slightly greater risk of cancer.  For example, it is known that people of African extraction have SNPs associated with longer telomere length, and they also have higher risks for many cancers, including lung and [ref].  (Africans have lower risk of glioma, so the correlation goes in the wrong direction for this particular example.)  At the risk of beating a dead horse, I emphasize again that even a small increased risk would be magnified by the extrapolation that is implicit in the methodology of the GWAS, and appear very large and scary when misinterpreted as an effect of telomere length.

GWAS is also referred to as “Mendelian randomization studies” because they depend very much on the assumption that different SNPs are randomly distributed in the population.  Of course, this assumption is not literally satisfied.  How significant is the deviation from random distribution?  I will be investigating this question, and I’ll let you know what I find.


The Bottom Line

There is a sharp conflict between the new GWAS results [Haycock, 2017] and the observational results [Rode, 2015] reported two years ago.  They can’t both be right.  If the GWAS results are as Haycock claims, there would have been glaring increases in cancer risk that Rode could not have missed.  If Rode is correct, then the methodology of Haycock must be flawed.

The reasoning in GWAS studies depends on a huge extrapolation.  I am saying it is more likely that the effect of genetic variations on cancer risk is direct, not (as per Haycock’s assumption) mediated by telomere length.  It could be that a very small direct effect of one of these SNPs is reported as though it were a large indirect effect, working via telomere length.

For now, I’m sticking with my previous counsel: Lengthening telomeres is a viable strategy for improving health and longevity.  If you take supplements that promote telomerase, you are not adding to your cancer risk.  Because of the large net benefit, lengthening of telomeres should be a major target for medical research.

But as I said at the outset, I am criticizing the new study from the outside, and it is quite possible that I have misunderstood the methodology.  I have sided with the direct observational studies and I have been skeptical of the GWAS studies, but it may be that the consensus in the field is correct, and that GWAS studies really are more reliable indicators of causality.

I intend to get to the bottom of this, and will report my findings in future columns.


* Sigma is a standard deviation of telomere length in the population at large. If you know what that means, that’s great; if you don’t it doesn’t matter to the logic of what I’m saying.

** Disease risk is typically reported as an odds ratio.  In this case, 0.08 would mean that, in their raw data, people in the study with the longer LTLs had a risk of 1.08 times as great as people with shorter LTLs.  You get to 1.08 not by adding 1 but by raising e to the power 0.08.

289 thoughts on “New Evidence that Long Telomeres cause Cancer, and Why I Think It’s Wrong

    • Hi Charles,
      You should very carefully consider that short lived mice have LONG telemeres and long lived humans have SHORT telemeres.
      It is very reasonable to believe that humans could have evolved whatever telemere length was best for their survival. Mice have long telemeres because that probable works best for repair and maintenance and their survival.
      This new study provides very compelling evidence that short telemeres help protect against cancer and is very likely the reason long-lived humans needed short telemeres.

      Furthermore, I have seen studies suggesting that telemeres follow the Goldilocks rule. Stem cells have telemerase and they try to maintain telemeres at best length, not too long or too short

      The idea of taking substances which length telemeres without very good evidence that such intervention is associated with increased all cause survival and not associated with increase risk of cancer death seems extremely foolhardy. Note odds ratio of 5.27 for gliomas, a very deadly brain cancer.

      As for me, I’m sticking with my short human telemeres and not going for long mice-like telemeres. Substance which increase telemere length will not be a part of my anti-aging formula any time soon. This new study reported by Josh is an excellent study and should be a red flag against untested telemere treatments.

      • On this rare occasion I’ll have to disagree with you Alan. It isn’t long telomeres that make mice short lived and short telomeres that make humans long lived. It is difference in body size. The bigger you are the greater the chance one of your many cells will go rogue and kill you with cancer. It would be risky to let all our cells express telomerase at will like mice cells do. But we aren’t trying to give all somatic cells the ability to extend telomeres indefinitely. We only want to reset them back to youthful levels in a one off treatment. No reason this would raise cancer risk. Will this treatment alone defeat aging altogether, I doubt it, but it would be a very good first step.

        • Hi Mark, Thanks for your comment. I was not trying to make causation argument that humans had long lifespan because they had short telemeres. That as you properly pointed out would be nonsense. A recent paper made point that zebra fish have short telemeres and so good model for study as regards humans and that mice bad model as mice have long telemeres. There is no causation argument regarding lifespan.
          The point is humans are unusual in having short telemeres.
          So question; why did humans evolve short telemeres. According to evolutionary theory, short telemeres must have provided humans an advantage in survival so trait for short telemeres was favored. The question is what was benefit of short telemeres. My point was that humans having a long life span needed extra protection against cancer compared to mice, which have short lifespan and breed very fast.
          Josh presented excellent recent 2017 paper showing increased cancer risk with long telemeres.
          In the universe I live in, medicine and law, there is very strong emphasis on determining what is reliable evidence. A well reasoned papers in a scientific journal, with lots of data could be considered reliable evidence. Opinions expressed in blogs would almost never be considered reliable. Blogs raise issue for consideration. Occasionally we get really great references like your recent reference on Mitochondria, which was of great value.
          I would advise readers to not take any opinions about medicine or science expressed by anyone on blogs as reliable information. It is the references which have real value as regards reliable evidence.

          • There are so many things that seem important in aging. TOR and mitochondrial, telomeres and inflammation. Someday we’ll have the model that links them all satisfactorily. For now I’d say it’s a good bet telomeres are a limiting factor in human lifespan (but not in mice), one of the downsides of doing lifespan studies in mice. I agree zebrafish should be persuaded to get more evidence for the importance of telomeres in aging.

  1. Josh, ALL,
    We know that our germline (eggs, sperm) are immortal due to plenty of telomerase enzyme to maintain full-length telomeres (or so the theory goes); thus the ends of the DNA in the cells never reach the Hayflick limit.
    But there are other (perhaps much more accurate) ways to measure our aging – namely DNA methylation (via methylation counts??).
    Do you know if our germline accumulates DNA methylation?
    IF DNA methylation is a more accurate measurement of actual chronological age and we also know that telomerase does not grant an organism immortality – if we then go back to the germline, perhaps it is not the telomerase maintaining the telomere length at all that grants immortality but somehow it is able to avoid accumulating methylation …
    Not sure if this makes sense/is a possibility …

  2. I have never been a fan of the whole telomere scene. Because for me telomere shortening with lower telomerase is a symptom of aging and not a cause. I have said this before in my comment here. If we are able to reverse aging telomere length would get back to normal due to regular telomerase replenishment.

  3. Re: “The reasoning is that people have their genomes for their entire lives . . .”. Certainly, that sounds like conventional wisdom. But is that even true? I’ve heard Craig Venter say they “can estimate your age just from the genetic code”.

  4. My contention is that Nature is not content to try to kill us by heart attack or cancer, but will try until She succeeds. In their amazing study, “Alterations of the systemic environment are the primary cause of impaired B and T lymphopoiesis in telomere-dysfunctional mice. Song, Z et al published in “Blood” 2010 – there are an entirely different constellation of aging diseases, and mortality by different means between mice that were telomere limited (They were telomerase -/- and had been bred for several generations such that their telomeres were short enough at birth that by old age, telomere attrition was apparent in cells of the tert -/- mutants. In the case of the short-telomere mice their diseases of aging were much more like human diseases of aging. Asked to guess on the basis of first principles, my guess would be that long telomeres give some stem and progenitor cells a longer replicative lifespan and hence the ability to accumulate more mutations, increasing the risk of cancer – while a short telomere would result in more stem and progenitor cells going into replicative senescence once their telomeres reach critical length (and start sending signals of their distress to the hypothalamus which responds accordingly by lowering it secretion of GnRH – as we learn later). Of course that’s the obvious and common sense view – usually wrong – but it would help explain the purpose of telomere attrition in aging.

  5. Pardon me, I left off the end of the first over-long sentence – which was just that normal long telomere mice suffered and died from different causes than telomerase negative mice with short telomeres at birth. And I forgot to emphasize the vital fact that replicative senescence in stem and progenitor cells means the conversion to senescent cells of the sort that produces SASP factors that lead to inflammation and heart disease.

    • Harold, wouldn’t the exponential runaway zombie apocalypse of senescent cell creation (which you argue for in your papers and I do agree) imply that most of them are somatic? (and I’d suggest also implies their bulk is to be found on fast renewed tissue, such as intestines leading to leakage, microbial migration and the ensuing immune response that not only increase the usual suspect of inflammation but lead to the diminished numbers of the most susceptible, and probably the healthiest, of the microbiota but, I digress)

      Stem/progenitor senescence is already the endgame for nature.

  6. Josh,
    The relationship to telomere length in ALL those studies was mortality, not aging.

    It is very difficult to see how stopping of cell division due to telomere shortening cound determine aging of neurons, heart or muscle tissue, the three tissues most important for aging and whose cells almost do NOT divide throughout life.

    I know you know this, my main argument against the telomere theory of aging. I just wanted to remember it.

    And, although age increases mortality rate in humans and rodents, Aging (the species-specific endogenous process) is NOT Mortality. Mortality is used as a surrogate of aging. This happens because gerontology continues to lack
    (unfortunately) a reliable marker of aging, and people wants to perform an publish studies nevertheless.

    In fact, ONLY if you die young, like Marilin Monroe, you can avoid aging. (Showing unsuccesfullness of gerontology for the moment)

  7. ALAN GREEN said:
    “Mice have long telomeres because that probably works best for repair and manteinance and their survival”.

    I fully agree with this. And that is an important general concept. Similarly, SHORT-lived animal species also have MORE (instead of LESS) endogenously determined antioxidant levels and MORE repair of DNA damage (e.g., BER) from endogenous origin. (compared to long-lived animal species). Because this is necessary for their survival due to their high rates of mitROS generation. But survival (inverse of mortality) is not aging.

    We should concentrate on parameters like low mitROSp (LOCAL damage just AT the site of ROS generation) or autophagy which can be causal effectors (executors) of the aging program instead of on telomere length, antioxidant levels, or DNA repair, which can increase survival rate but cannot decrease aging rate. (In the case of antioxidants because they can not be located between the mitROS generator and its aging target because those two are so near that they are likely in CONTACT). And in the case of DNA repair it does not make sense to endogenously generate a lot of damage to DNA and, afterwards, try to eliminate the damage to produce a longer-living species. It is much more efficient just to avoid damage by lowering the mitROSp on the first place). Same for antioxidants. It is not efficient to generate a lot of ROS and then try to intercept them before they reach the target. It is much better (100% efficiency) to produce few ROS per unit time on the first place, and that is in fact what all longlived animals do (both longlived species and DR animals).

    And of course most important is to concentrate on research on the genetic aging program that determines longevity (both between species and in DRs) through variations in the intensity of mitROSp, autophagy and other mechanisms of aging yet to be discovered.

    • Very good point, humans might not need long telomeres if we produce less damage to them (in comparison to mice for example). Still that’s not enough evidence to say they don’t contribute to aging in humans. You cite long lived non dividing cells, but the heart is dependent on the circulatory system, which most certainly does age via fast proliferating endothelial cells. Ditto with neurons and glial cells, the latter divide quickly and could become exhausted by proliferation over a lifetime.

      As far as I am concerned it is not just what the most fundamental drivers of aging are, but what the most accessible intervention points are. Yes ultimately figuring out how genetics codes for lifespan will enable us to make longer lived humans, but we need some interventions that will give us a much needed boost now. Whether that is enhancing mitophagy, slowing cell division and senescent cell conversion with TOR inhibition, or telomere elongation I don’t much care.

    • Gustavo, what is your take on TAs like MitoQ an SKQ1? Are they likely to show any protective mechanism in mitigating MitoROSp, or does MitoROSp occur is such close proximity of the aging target that damage simply can’t be prevented?

      • Some Mitrosp can be even in contact with the target explaining the lack of effect of antioxidants.

        However damage generation can be decreased. We already know that mitROSp goes down with DR, with PR, with MetR, and with rapamycin as well.

        The same occurs for ox.-derived mtDNA modification and for its fragments within nuclear DNA (this investigated in the case of rapamycin)

        • Thanks Gustavo. Kind of ironic, when you consider, how protein powders, protein bars etc. are being promoted almost everywhere nowadays…

  8. It is the CHANGES in telomere length that alters gene expression. It’s well known that people on a plant based diet , who exercise everyday maintain telomere length in old age. This is likely due to lower level of systemic inflammation.

  9. Do you want to live to 300 and beyond? If yes you need to lenghten telomeres somehow. Because the idea of having enoug stem cells to repopulate the body with fresh young acting cells is preposterous. You can activate telomerase for short bursts of time so you dont have to have the telomerase gene on the whole time (whih would help randomly mutated cells become a cancer threat, because you made it easier for them to replicate indefinitely).

  10. Since my telomeres are quite long, I ‘ m wondering if I should start eating poorly, stop exercising, and introduce as much inflammatory stress as possible into my life to shorten those suckers and reduce my cancer risk! On the other hand, the Ashkenazi Jews have a very low overall mortality rate with very long telomeres and are long lived.
    It was always my impression Josh that it is the percent of very short telomeres that matters most, also called the critically short telomeres, and it is these that are elongated the most through supplements and healthful lifestyle. Is that not true? And the mean telomere length isn’t nearly as important as having a low percent of the very short ones, and this study didn’t seem to look at that.

  11. This is a nice article Josh. Thank you for writing about telomeres and telomerase again. I still think extending telomeres in a controlled way is a viable and realistic intervention point against aging, but I understand that the consensus is as I used to believe, that it is dangerous and inferior to the (much more complicated) procedure of replacing stem cells. I agree with you that the risk increase to 1.08 is probably just noise, or due to an unrelated SNP effect, and can’t be extrapolated out to much larger theoretical differences in blood cells telomere lengths.

    Perhaps a better apprpoach would be to look a a selection of different mammals? As I pointed out about to Alan above telomere lengths in mammals is mainly a function of size; evolution has managed to work out how to make both squirrels (long telomeres) and beavers (short telomeres) live to about 30 years. Or perhaps the length of telomeres is a red herring and it is preserving the length that matters. Definitely more complicated than simply telomere length. One other thing, why does no study ever measure anything but the telomeres lengths in the blood? Surely we need a range of tissues over a period of time to get and accurate picture?

    • In regard to cancer risk and telomere length, new paper noted in UPMC news letter states, “Telomere length predicts cancer risk, according to large epidemiological study”, 4/3/2017.

      “Longer than expected telomeres…are associated with increased cancer risk, from study of 28,000 Chinese in Singapore.
      Longest telomere group had 33% higher odds of any cancer.
      5 cancers in particular:
      Lung: 66%
      Breast: 39%
      Prostate: 37%
      Colorectal: 37%
      Pancreatic: HIGHEST 260%
      3 cancer showed “U shaped curve” and highest risk in groups with longest telomeres and shortest telomeres:
      Stomach, bladder and leukemia.

      A completely different study showed high risk of type 2 diabetes and short telomeres in American Indians.

      In April 2017 American Association for Cancer research (AACR) meeting Wash, D.C.
      So looks like long telomere length is a double edge sword which increases cancer risk but protects against other diseases.

      Also as Dr Paul Rivas pointed, noted other studies showed group Ashkenazi Jews had marked longevity and longer than expected telomeres.

      • I added a link to the U of Pittsburgh press release from which Alan was citing. I have written to the author (Yuan Jian-Min) of that study, but he declined to release his ms pending acceptance for publication.

      • We just don’t yet know enough about this. It reminds me Alan of the cholesterol issue. At first it seemed that an elevated total cholesterol was a sure heart attack, but then we realized that it was much more complicated, and there was ldl and Hdl, and even subdivisions of those, and people with normal totals had heart attacks and elevated totals did not. We then tried to artificially raise hdl’s and they had also had heart attacks. It’s complex.
        We are only looking at a mean length in one organ system, the immune system, and that says nothing about all of our other organs, none of which age at the exact same rate.It’s possible that a lengthening of telomeres in the immune system is the body’s response to a genetic cancer predisposition, or that a subset of people have many long telomeres and also many very short ones leading to an overall genetic instability. My point is that right now our knowledge of telomeres and their importance is just not at all clear.

          • Elizabeth Blackburn , the Nobel prize winner for telomerase here at Hopkins has shown that healthy lifestyle interventions like diet and exercise will lengthen the very short and dangerous telomeres, and that is how the mean increases in size . It’s all about the percent of critically short ones that leads to DNA instability and cancer risk.It’s not the long ones. Young people have long mean lengths, a low percent of shorts, and rare cancer.

  12. Fasting, Caloric Restriction, Protein Restriction … all of these downregulate mTOR.
    Rapamycin downregulates mTOR, metformin and aspirin upregulate AMPK to downregulate mTOR.
    Professor Gustavo Barja discusses much regarding Protein and Methionine restriction while Dr.Alan Green discusses much regarding the Koschei Formula (Rapamycin, Metformin, Aspirin, etc.) based on Blagosklonny’s work.
    What ties all of these together?? They all promote Autophagy.
    How can you guarantee autophagy?? Obviously fasting, but add in plenty of wheat germ!!
    Thanks to posters here on Josh’s blog I have been introduced to the fact that Wheat germ (highest food source of spermidine) UPREGULATES AUTOPHAGY OUTSIDE OF mTOR!!! (not to mention is loaded with folate, magnesium, potassium and fibre).
    Vitamin D3 and melatonin (both hormones) which have positive effects on all-cause mortality also promote autophagy.
    From Professor Vince Giuliano’s website blog:
    “There is a wide variety of genetic manipulations, pharmacologic manipulations, and nutrient manipulations that have been shown to alter lifespan in model organisms. These include caloric restriction, “loss of function” mutations, “gene knock out” models, phytochemicals, and drugs that down regulate aging pathways (mTOR, insulin/IGF-1, etc.). It also includes “gain of function mutations”, transgenic models, phytochemicals, and drugs that up regulate longevity promoting pathways (AMPK, FOXO, Klotho, etc.). At first glance, all these interventions may seem to be unrelated, suggesting that aging is a multifactorial problem with no common denominator to longevity. On further examination, however, there is a common denominator to all of these interventions – autophagy”
    “Autophagy now appears to be a downstream event following insulin/IGF-1 pathway down-regulation, mTOR inhibition, Klotho activation, AMPK activation, Sirtuin dependent protein deacetylation, and histone acetyl transferase inhibition. Autophagy explains in part, the beneficial effects of caloric restriction, caffeine, green tea, rapamycin, resveratrol, metformin, spermidine, lithium, exercise, hypoxia, Torin-1, trehalose, and a host of other natural and synthetic compounds.
    There is much stronger evidence of a link between autophagy activation and longevity than there is with any other longevity interventions such as exogenous anti-oxidant supplementation, endogenous anti-oxidant up regulation, micronutrient replacement, hormone replacement, anti-inflammatory therapy, telomerase activation, or stem cell therapy.”

    • Hi Clinton,
      Thanks for comment about wheat germ.
      Recently came across wheat germ in paper in which treating fronto-temporal lobar dementia by increasing Autophagy. Spermidine worked as well as rapamycin to increases autophagy. Also spermidine is different pathway, does not increase autophagy by decreasing mTOR; but seems to have direct pathway to increase autophagy.
      Best source of spermidine, as you correctly pointed out is Wheat Germ.

      • Thanks Dr.Green,
        I am curious as to ‘how much’ wheat germ people generally should aim for. I see reference of 2 tablespoons per day from Dr.Liverman’s metabolic health blog (fantastic blog regarding aging – he is a propronent daily 12+hr fast, weekly 24hr fast, wheat germ, sulforaphane, Hight Intensity Interval Training and resistance training)
        I have personally been consuming approximately 3 tablespoons in 2 ‘smoothies’ that I make (total of 6tbspn = say minimum 30grams per day).
        Wiki reports that there is 243mg of spermidine per kg of wheat germ. That would bring my 6 tablespoons in at a measly? 7+mg per day of spermidine.

        • Hi Clinton
          I have few bags wheat germ in Frig, but didn’t know dose.
          Your plan sounds great.

          For me, Leave the fasting, take the Wheat germ.

          • Hi Dr Green,

            I have just came across this paper about wheat germ agglutinin:
            “Antinutritive effects of wheat-germ agglutinin and other N-acetylglucosamine-specific lectins”
            Not sure if this should raise some concerns about consuming wheat germ but it looks like it.

          • Dr Green,
            Do you have any comments regarding the two posts above concerning wheat germ consumption?
            Thank you,

    • I agree, Clinton, that DR, PR, MetR, and Rapamycin can increase autophagy, but please note that it is well known that they all also decrease mitROSp. We have a lot of experimental evidence of such decrease in mitROSp from our lab during many years. I agree that these TWO responses are much more important for the increase in longevity that antioxidants, telomeres etcetera.

      My present model is that all those longevity increasing interventions (DR…..) in mammals signal the aging program lying in the cell nucleus, which in turn responds by: 1-lowering the rate of mitROS production, 2-increasing autophagy, and 3-other longevity effectors to be discovered. Therefore the cell produces less endogenous
      oxidative damage and increases the elimination of garbage. All this makes a lot of sense especially due to the well known relationship between lipofuscin and oxidative macromolecular damage.

      Concerning wheat germ I am not aware if it could also lower mitROSp in addition to increasing autophagy. It would be interesting to know as well as the right dose and possible unwanted effects in humans.

      • Hi Gustavo,

        Thank you for the link to your paper. This is very interesting but I need some time to digest it.

        A simple question come to mind about membrane fatty acid unsaturation:

        “The low delta-5 and delta-6 desaturase activities (which are rate limiting enzymes in the n-3 and n-6 fatty acid synthesis pathways) of long-lived animals will decrease
        the conversion of the less unsaturated 18:2n-6 and
        18:3n-3 diet-derived precursors to the highly unsaturated
        20:4n-6 and 22:6n-3 products. Thus, 18:2n-6 and 18:3n-3 would accumulate and 20:4n-6 and 22:6n-3 will diminish, which is just the general kind of fatty acid profile observed in long-lived animal species. ”

        Would it be feasible to create mutant mice with low delta-5 and delta-6 desaturase activities? If yes, and if they live longer, this would validate the relevance of membrane fatty composition to aging independently of other factors right?

        • Aldebaran,
          1) Yes, it is a big article. But there is a specific section dedicated to the mtDNA fragment issue. And please note that transfer of the fragments from the pericentromeric region, wehere they are present in large amounts, to the rest of the chromosome’s length seems to be performed by TEs. All this sounds to me like an organized intracellular mito-nuclear mechanism to promote aging finally by damaging the nuclear genome and chromosomes.

          2) knockouts of delta-5 and delta-6 desaturase genes has been performed in mice but it is lethal during development. To mimic what Nature allways does when evolving from short- to long-lived species what is needed is to DECREASE especially 22:6n-3 but not to zero levels. Some of it is needed.

          Shmookler-Reis (ortography’s name, I am not sure by memory) et al. did mutants lacking these desaturases in C. elegans nematodes (much simpler organisms) and they were succesful in increasing maximum longevity.
          We and others decreased the desaturases with atenolol and a similar beta-blocker in mice and found less tissues DBI and many molecular and functional benefits, but max. longevity (although high, 3.9 years!) was not increased compared to controls, in our case at least in part due to a side negative effect of the drug on the heart working through variation in blood pressure.

          • Hi Gustavo,

            Thanks! Very helpful comments. I was not aware that mtDNA could be inserted into nDNA. I am finishing chapter 8 where all this is detailed. Definitely like your paper. Very informative and well written.

    • Great comment. Dr. Frank Madeo is a an Austrian molecular biologist and an expert in autophagy and has discovered very high blood levels of spermidine in people over 90. He also has found that resveratrol and pterostilbene have highly additive effects with spermidine and so require much lower doses of each to be effective. His list of best food sources of spermidine are cooked soybeans, peas, pears and mushrooms, in that order.

    • Hi John,

      Thanks for the link.
      If the genome modifications over time are actively driving the aging process (not only through telomere changes but also as mentioned in paper through somatic mutations, retrotranspositions, and so on) it is hard to imagine a quick fix.

      Interestingly, it seems that Everon Biosciences is currently trying anti-retroviral drugs on mice to make them live longer (link mentioned by John previously), so they think genetic mobile elements are actively driving aging.

      Now what happen exactly when an old cell with some accumulated genetic modifications is reprogrammed into a stem cell? Does all its genetic modifications are magically reversed? Does only a part of it is reversed? Does it simply cannot be reprogrammed after reaching a threshold of genetic damage?

      We should be now in position to answer this sort of questions given the rapid progress in sequencing. From the paper: “The capacity to sequence deeply, and to sequence at the level of the single cell, is providing a new lens to understanding the diversity of genomes within an individual”.

      Where does mitoROS fit into this picture? Are they somehow driving a significant part of the nuclear genetic modifications (and epigenetic as well)?

      How about epigenetic changes? Are they driving genetic changes, are they just a consequence of them, or are they independant?

      Sorry, many questions on my side that have certainly been asked (and maybe answered) before but that I am unable to find a clear answer to.

      • Thank you, aldebaran. As you probably noticed, the authors are part of the Human Longevity, Inc. (HLI) team. I feel the same way as you; this paper tells us a quick fix is unlikely. But I heard co-author Greg Venter say they are making “8 to 10 new discoveries every day” in a video from 2016. It doesn’t look like they are publishing 8 to 10 papers per day, lol! I also heard him — or maybe it was Brad A. Perkins — say that they could help Calico. I got the impression he was inviting Calico to come to them for help and they would make a deal. So I am hopeful a lot of enormously useful data is forthcoming from this research!

        • Thanks John, looking forward to more discoveries from HLI and Calico in the coming years. At least, they won’t be too limited by budget, which was a main issue for research in this domain.

      • Aldebaran.
        Concerning your question on mitROSp and chromosome and genomic modifications during aging, the answer is positive. We and others have shown that mtDNA fragments accumulate during aging inserted inside the nucleus in the nuclear DNA, and this is reversed by rapamycin in parallel to decreases in mitROSp, in rodents.

        And those fragments accelerate aging in yeast.

        They enter the nDNA through the pericentromeric chromatin where they are heavily concentrated (a location which will facilitate their induction of many kinds of cromosome abnormalities including whole chromosome loss). From the pericentromeres , possibly helped by TEs, they seem to distribute along the full length of all the chromosomes causing diverse types of DNA damage and genomic modicications during aging. Therefore they can explain at least in part the observations of Crayg Venter et al.

        All this mtDNA FRAGMENTS-nDNA issue is summarized on my ROS (journal) 2017 review (which I cited on my previous comment to this blog).

        • Thanks for your comments Gustavo.
          I have already started to read your paper with lot of interest! The best part for me so far:

          “Substantially decreasing the aging rate in mammals including humans will be relatively easy once the underlying basic mechanisms controlling longevity at physiological, cellular, and molecular levels are clarified.”

          The optimism of so many experts in the field is really telling me that we are on the right track!


  13. A new study published on 8/22/17 out of UC Berkeley looks at the role of telomerase in immortalizing cancer cells. There is a two step process, the first of which involves a mutation in the TERT promoter which turns on telomerase a slight amount , but not enough to immortalize the cancer cells, that is where the second step must occur as described by lead author Hockemeyer, ” we have evidence that the second step has to happen,and that the second step is initiated by or is occurring at a time where telomeres ARE CRITICALLY SHORT and when telomeres can be dysfunctional and drive genetic instability.”

    • it also depends on the cell type and the culture medium wheter hTERT transfection is enough for immortalization. a second usual checkpoint is p16ink4a, which also seems strongly associated with senescence

      • But the evidence so far is that the very short telomeres are the culprits in that final step to malignancy, and not the long ones. Do you agree?

        • This makes a great deal of sense when you consider that one cancer cell must jump many hurdles to escape its bounds and replicate out of control. These hurdles are the mutations needed to overcome growth arrest, contact inhibition, telomerase blockade, etc. And to have a chance of getting all these mutations you need chromosomal instability. And what is the easiest way of getting chromosomal instability? Critically short telomeres! The other part of the equation is the immune system, which declines as you grow old as the immune cells replacement rate is slowed (as with all proliferative cells), making it more likely a potentially malignant cell will survive long enough to accumulate the necessary mutations. So this is a second reason to have youthful length telomeres. And this finally is a point everyone seems to miss: young people get less cancer and have longer telomeres. So clearly the optimal length for human telomeres is the length we have as young people.

          • I totally agree. If you look at the four speculative groups based on telomere distribution that I proposed on 8/31, then it makes total sense that young people would be in group D, and that should be the target for telomerase therapy. Ashkenazi jews also in that group.

          • Hi Mark,
            In science, facts trump theories.
            The presented papers show long telemeres increase cancer risk adjusted for age.
            When the facts contradict the theory, the theory is wrong.

          • Hi Alan, well the paper that sparked this blog post showed certain SNPs with 2-3% contribution to telomere length have some correlation to cancer rates, if you accept the methodology, which in my opinion is rather weak extrapolating out a signal from the noise. And this isn’t even measuring telomere lengths. Other papers have, but they give confusing and sometimes contradictory results, which is unsuprising given the problems in measuring telomere lengths in only blood cells, and the discussed issues in measuring average lengths vs criticallly short ones.

            What I think we should all be able to agree on is that there is an optimal telomere length for humans. And that optimal length is the length of a young person’s telomeres. Do you agree? If so a telomerase therapy is still useful, as long as caution is used in its application.

            More broadly speaking I think we have to up the proliferation rates of mitotic cells if we are to be truly youthful again. Repressing MTOR is great for keeping those quiescent cells from senescing, but this won’t make your skin look like it did when you were 25!

          • Mark,
            I agree. But that is true only for skin and other mitotic tissues. But not for the most relevant ones for aging, the postmitotic ones, made up of neurons, cardiomyocytes anf skeletal muscle cells.
            Therefore, the old person you mention with his telomeres lengthened back to the lenght when he was young perhaps wll LOOK younger, as you say, but it would still have the muscles, heart and brain of an old person. A situation rather in the line of those present day olds using cosmetics and plastic surgery……

          • Hi Mark,
            In regard to your comment about 25 year old skin: Old skin looks fairly normal microscopically and looks OK grossly if stretch it. Problem with old skin is wrinkles, which is due to too much skin. That is why operation to reduce wrinkles is remove excess skin and stretch the old skin.
            Wrinkles due to excess senescent cells. After 20 months on rapamycin my impression is my skin looks very good on back of hands with no wrinkles.
            As regards telemeres: Telemeres get shorter with age so short telemeres are by definition a diseases of aging. Short telemeres associated with increased mortality and marked increase in specific diseases like diabetes.
            The issue is do short telemeres play a role in promoting cancer. Certainly cancer incidence greatly increases with age. The problem is data showing increased risk cancer with longer telemeres. That means hard to argue short telemeres cause cancer. To me this suggests that development of cancer has nothing to do with telemere length. However, once cancer has developed as regards mutations which enable cell to be cancer; the cancer cell is hijacking normal length telemeres to help them divide. I don’t see long telemeres as causing cancer in any way; but rather cancer might find it easier in cells with long or normal length telemeres. Bottom line, as studies showing increase cancer risk with longer telemeres; don’t like theory that short telemeres cause cancer.
            Anyway Mark, very much appreciate all your comments.

          • Yes Gustavo, I am on the same page as you – I am suggesting that the body turns down MTOR etc during aging to preserve post mitotic cells, but this has a detrimental effect on replacement rates of mitotic cells. Rapamycin does the same thing only more so. Evolution does the same thing between species only even more so – metabolism of a human is much slower than for mice, and in addition evolution has produced more stable MtDNA.

            We have some hints of how to manipulate these pathways for slightly longer healthy lives now, but I agree, we’ll need to intervene much more thoroughly to remain youthful indefinitely.

          • Ok Mark, but then you agree that humans have much lower turnover of Mitotic cells than mice and nevertheless they live 30 fold longer. Not?
            Therefore a high turnover of mitotic cells is not important for longevity.

          • I think max lifespan is probably a function of post mitotic cell function, but that we won’t be youthful without returning mitotic cells to a youthful (human) replacement rate.

          • I understand your point Mark.
            But the turnover of mitotic cells of an OLD mouse (although slower than that of a young mouse) is still HIGHER than that of a YOUNG human. And nevertheless the old mouse is degenerated and the young human is not.

          • In reply Mark’s comment 9/3 5:33 about old skin.
            John just posted cosmetic surgery paper about treatment skin old age and last line was “excess skin” causes wrinkles. Judith Campisi, 1998, “Role of cellular senescence in skin aging” before anybody knew about rapamycin and said old skin looked normal histologically, just too many senescent cells.
            Unfortunately, I didn’t know to take photograph 2 years ago of back of hands or could demonstrate rapamycin clears senescent cells and removes wrinkles.

          • True but then again no one here is saying mice and humans die of the exact same things. As you have already pointed out mice have superior antioxidant defences and longer telomeres than humans, and in fact all their cells are capable of continued replacement, so they don’t really have post mitotic cells as such. Yet they perish much sooner than humans. I wonder if the usefulness of mice as a study species for aging is coming to an end now?

          • If function on age-related diseases, will see that most common age-related diseases are diseases of hyperfuction, frequently hyperfunction of senescent cells. Most classic example being osteoporosis being caused by hyperfunction of osteoclasts dissolving bone. That is reason why rapamycin ameliorates almost all age related diseases as rapamycin stops hyperfunction caused by elevated activity of mTOR. Now if talking about after age 95 or 100, that could be an entirely different story. Most people never make it to 95, so most people dying of aging due to hyperfunction. A little more focus on actual age-related diseases, like atherosclerosis and will see at every molecular action taking place, there is hyperfunction being driven by mTOR and being inhibited by rapamycin.

          • Mark,
            I was not aware that mice do not have any postmitotic cells at all, even in brain or heart? If you could mention me any citations about this important point it would be nice on your part.

          • What I am driving at — can we slow down or stop the changes in bone structure described in the “Facial Aging is More Than Skin Deep“ article via anti-aging medicine, and therefore indirectly help the skin look more youthful that way, particularly in the face. Of course, I’m not recommending anyone get implants, lol!

        • Comment above was for Gustavo.

          Interesting Alan. If skin aging is mainly due to senescence, then i wonder if this is due to replicative senescence or other damage such as ROS? I expect MTOR inhibition would only help with this very gradually. But do let me know if you continue to see improvements.

          Incidentally restoring telomeres has been shown to restore skin (and the underlying layers) to youthful elasticity and function.

          • Sorry Gustavo, mice do stop growing so they obviously do have post mitotic cells- what I meant was all the cells in their body can regenerate as they can express telomerase. It isn’t like human neurons or heart cells, which can’t be replaced. My point was that mice and humans are different and we probably can’t solve human aging just by looking at mice.

          • Mark, thre have reports that some strains of mice, line MRL, have extraordinary regenerative capacities, but the regenerative capacities of other strains like C57BL/6 are poor…

            Sorry it is time to go to sleep here.

          • Hi Mark,
            Regard to skin:
            I was only speaking about surface epithelium which looks wrinkled due to too many cells.
            Old people also have sensitivity to cold due to defects in subcutaneous layer with loss of thickness and quality. This could probably be improved with better maintenance from fibroblasts. So makes sense that longer telemeres could improve subcutaneous tissue. Notice we are talking about 2 different diseases of old skin.

        • Thanks Gustavo. I don’t disagree at all, I was just making the point that mitotic cells are important too. What I think is happening with aging is that the body is trying to balance the health of mitotic and post mitotic cells. Because as you say the neurons and heart muscle etc is so important to protect from senescence and mitochondrial dysfunction, the body is slowing everything down. Hormone levels, replacement rates of proteins, it all goes down to extend the lives of these all important post mitotic cells. But this means the cells that need replacement regualrly have to last longer too. It’s a double edged sword, we must address both. We could boost things like growth hormone, and maybe look and feel younger, but this would be detrimental to those post mitotic cells. Similarly we can slow things down further with intermittent MTOR inhibition, but then skin and blood cells may suffer. This is why we will need telomerase therapy or in Vivo stem cell replacement as well as MTOR inhibition and perhaps other protocols to keep mitochondria healthy.

          • Mark, just a further related comment.

            For me the most important is not DR, rejuvenation therapies, etc., which, at best, increase longevity by “only” 1.5 fold.
            The important thing is the interspecies variation. For example, what happens during evolution from an animal mouse-like (longevity 4 years) to humans (longevity 120yrs)? (30 fold difference in longevity).

            In relation to this interspecies difference, do you know if replacement of the mitotic cells (e.g. skin) is slower in the human than in the mouse? I think it should be, as with many other characteristics, necessarily, at least in part because the weight-specific metabolic rate is also 10 times slower in the human than in the mouse.

          • I think it’s clear that there are multiple factors and pathways involved in the aging process, many of which are presently unknown. The question is, from a practical standpoint, are there things that we can do right now to make a dent in the process and maybe live to 120 and beyond, at which point we’ll know alot more than we do now and there will be yet more interventions.
            At present there are several things that we can do which I believe do more good than harm.
            1. partial inhibition of mTor with weekly rapamycin use
            2. increase autophagy and reduce insulin/IGF with fasting
            3.increase AMPK with berberine
            4. angiotensin inhibition with meds or grape leaf extract
            5.GSK 3 inhibition with low dose lithium
            6.Nic riboside for optimal NAD levels and mitochondrial health along with resveratrol and pterostilbene.
            7.Elongate the critically short telomeres with lifestyle and supplements.
            8. Exercise, sleep well , reduce stress, limit sugars and refined carbs, don’t smoke. Drink coffee and get sun exposure.
            I’m sure there are many more in addition to the above which are practical for the here and now.

          • Thaks Paul, nice 8 practical points.

            Just a comment concerning points 1 & 2:

            Rapamycin has side effects and only 1/4 of the effect on mouse longevity compared to DR. Then, if you already do DR (point 2, fasting), why to take rapamycin simultaneously? With DR you already have mTOR inhibition without the need to suffer the rapamycin other negative side effects.

          • Thanks Gustavo. You make a good point.
            I somehow forgot about low dose naltrexone . Probably nothing is more important than anti-inflammation and a healthy immune system . By transiently blocking both opioid and toll receptors you get a remarkable degree of beneficial immune modulation, rejuvenation and function as well as a potent anti-inflammatory effect. My colleagues and I are starting a trial study on it this month, but so far the results are astounding on depression and auto- immune conditions of all kinds.Even cancer.
            Check it out!

          • In response to Gustavo,
            Rapamycin increased life span in mice when started in middle age by 23% male and 26% females in 2014 study. In 2016 study reported by Lamming intermittent rapamycin, once every 5 days, increased lifespan in mice. Only female mice were used. Lamming reported no side-effects with intermittent use.
            CR has only increased lifespan when started in young mice and not when begun in middle age mice.
            As regards side effects: Daily rapamycin is used to reduce mTOR1 and mTOR2 to produce toxic effect of knocking out immune system to allow transplants. Nobody would use daily rapamycin as an anti-aging drug as clearly not good for health.
            Weekly rapamycin is used as an anti-aging drug.
            It is extremely problematic when people take they side effects of daily rapamycin and attribute them to weekly rapamycin as if they have a scintilla of evidence to support their claim.
            In only study in humans, the rapalog Everolimus when given in dose of 5 mg once a week in 12/24/2014 Mannick study improved immune function in elderly and did not cause significant side effects.
            I challenge Gustavo to cite a single study in humans in which weekly rapamycin caused significant side effects. If he can not cite a single study showing significant side effects with weekly rapamycin, I would request that he state that the side effects with weekly rapamycin are totally unknown to him and he was only talking about side effects with daily rapamycin. On the other hand, I have used weekly rapamycin for 20 weeks and I am very aware of effects of weekly rapamycin.

          • Thank you Alan.

            The same as you say, the only published study of rapamycin in humans I am aware of is the Mannick study that you mention.

            I was referring to side effects of rapa in mice which have been reported repeatedly. But still, in spite of those side effects, I believe that the global effect of rapa is Positive necessarily because it increases significantly both mean and maximum longevity.

            I was just wondering about overlapping of rapa with DR at the level of mTOR inhibition since DR also lowers mTOR downstream activity.

            Another relevant point is that the number of studies on the DR effect on rodent longevity is huge, whereas the number of studies of rapa on rodent longevity is tiny (more especially compared to the DR ones). Therefore I think that we shoud wait for much more rapa studies coming before concluding that rapa works from middle age whereas DR works only when started early in life. Specially since this is not what one would expect if rapa only changes a part of the effects of DR (and DR can increase mlsp up to 40%, whereas rapa only 20 or tiny 10%).

          • I don’t know the figures on replacement rates for mouse Vs human cells, but they are going to be proportional to metabolism, so mice cells should have a much faster turnover.

            Regarding rapamycin, I think they have got a mouse up to 60% life extension now with intermittent treatment starting in middle age.


            This was a large dose and caused problems in female mice, who get higher serum levels for the same dose, so perhaps this is about the limit of what we can do with rapamycin, but the fact you can start in middle ages and get these kinds of benefits, whereas DR needs to be lifelong to get the full benefits, shows that DR is not necessarily better than rapamycin. In the future I expect there will be better and more selective MTOR inhibitors.

            I used to take berberine but I think that daily treatment combined with weekly rapamycin treatment runs the body down too much. I prefer a weekly or bi weekly mitochondrial housecleaning day using nicotinamide and d-ribose.

          • Actually we’re on the same page as I’ve noticed a similar effect with berberine and had to reduce it, also I can only tolerate NR twice a week which is probably adequate as you say.

          • Take care with that conclusion from Kaeberlein’s study. Other highly respected rapa USA specilaists warned me last year that the real increase was not 60% in that study. That figure came out mathematically because the authors referred to (if I remember well) the controls which had already high age. The real increase was much less and in the line with R. Miller’s et al. previous studies.

          • Mark, you agree that:
            “Human turnover of mitotic cells is one tenth that of equivalent mice mitotic cells”.
            Ok, then, what mother Nature does durinc evolution of increasing longevity is to get DOWN BOTH postmitotic (and their mitochondria) and mitotic cells. Not to get down the first but UP the second as you suggested for humans (intraspecifically).

            But my feeling is that, if we want to get much further than “merely” 1.5 (or 1.6-i ntermitent rapa) fold life extension, we have to copy what Nature did so many million times during evolution.

            Without much longer than 1.5 fold life extension, the goal to stay “forever” young (ALL the years that one lives) will be impossible.

          • In response to Gustavo regarding goal of staying young forever,

            As a physician in anti-aging medical practice, the primary goal is to reduce age-related disease in 8th and 9th decade of life and improve quality of life that is reduced by age-related conditions.
            Perhaps, the goal of science fiction is to stay young forever; but I do not think that is goal of most physicians in anti-aging medicine.

          • But (ironically?) I think the only way to avoid those illnesses of old age (desired, ok, by physicians) is to Eliminate aging (which is the same as to be “forever young”).

            Given the huge advances of molecular gerontology during the last 2 decades I do not consider that “science fiction” but an adequate future objetive.

            Whether we must wait many decades or not to achieve it is impossible to predict now. Seems far, I agree, but once the basic endogenous origin of the aging rate of each species is clearly identified and its mechanisms well dissected, that major goal will be at much closer hand. This I believe sincerely as a biologist.

          • Wow its getting tricky to read these posts in the right order!

            I am very interested in your idea Alan, that skin aging and wrinkles are primarily caused by too much skin, and senescent skin cells. If this is the case then continuing rapamycin therapy should improve this situation without other interventions. But I suspect skin turnover needs to be increased as well. You might be able to achieve this through some kind of exfoliant or some such. But I suspect telomerase woukd be more effective by far. We shall see.

            I very much appreciate the practical approach of what we can do now from Alan and Paul in particular. But science fiction or not my aim is very much to restore my body to a fully youthful state and stay that way indefinitely.

          • I couldn’t agree more but one thing we don’t know is the synergistic and additive effects of all of the measures which we now have at our disposal. But youthful immortality does feel like a pipe dream at this time, but I hope not!

          • Yes Gustavo, I am on the same page as you – I am suggesting that the body turns down MTOR etc during aging to preserve post mitotic cells, but this has a detrimental effect on replacement rates of mitotic cells. Rapamycin does the same thing only more so. Evolution does the same thing between species only even more so – metabolism of a human is much slower than for mice, and in addition evolution has produced more stable MtDNA.

            We have some hints of how to manipulate these pathways for slightly longer healthy lives now, but I agree, we’ll need to intervene much more thoroughly to remain youthful indefinitely.

          • Sorry, my post above, linking to “Facial Aging is More Than Skin Deep” was meant to be nested under this discussion of skin and youthful appearance. I doubt there’s a way to make aging skin look more youthful without dealing with subcutaneous fat and bone underneath, which changes with age.

          • Fascinating discussion!
            Thanks all of you.

            What Alan said about hyperfunction with aging make me think about this recent paper:

            “Nucleolar expansion and elevated protein translation in premature aging”


            This is primarily about HGPS cells but they also claim the following:

            “We found a significant and direct correlation between aging and nucleolar size in healthy individuals. ”

            So too much ribosomal RNA generation with age resulting in too much protein synthesis. Would it be an effect of hyperfunction?

            They conclude:
            “While modulation of the mTOR pathway by rapamycin and analogs has shown promise in extending lifespan and reversing cellular phenotypes of HGPS, enthusiasm has been tempered by the possibility of side effects arising from prolonged treatment. Our findings imply that direct inhibition of rRNA production, perhaps using drugs that are currently in the trial pipeline for other diseases could be an attractive target for treatment of HGPS and possibly for extension of human lifespan”

            Inhibiting rRNA looks scary to me but maybe with a very selective drug and careful dosage?

            By the way, I had also a question for Paul about point 8. What is the rationale for getting sun exposure? I assume this is not just about vitamin D.

          • Hi Aldebaran,
            Very interesting paper.
            At end of paper they mention rapamycin and too many side effects, ref 65. So looked up ref 65 and 2009 paper by Brian Kennedy, Buck institute. Paper talks about all these great anti-aging benefits of inhibition mTOR with rapamycin and then throw in line about too many side effects.
            Of course, I have been on weekly side rapamycin for 20 months and no side effects; so I’m always wondering, why can’t anybody else figure out that weekly rapamycin is safe.
            My guess was all about money. (as in, if I could figure out weekly rapamycin is safe, everybody else can’t be that stupid)
            Got my answer when Brian Kennedy from Buck Institute combined with Silicon valley money to form company with name like Mount Tam to market new rapalog knock-off.
            The point is the people around rapamycin always knew that intermittent rapamycin was safe. The problem was no money in generic rapamycin. In order to make big money from rapamycin needed to come up with knock-off and then use their knock-off rapalog to then show their knock-off was safe.
            Maybe you have better explanation; but best I can come up with is failure to have any interest in studies to show weekly rapamycin is safe is more about making money on a knock-off than any other reason.
            As regards mTOR hyperfunction, it acts partly through making proteins which goes through messenger RNA, ribosomes, nucleolus. These proteins are what drives inflammatory reactions and other hyperfunctions which drive age related diseases.

          • Great explanation Alan and I’m sure it’s right on the mark. It is very annoying though when these ” scientists” very matter of factly state that long term use of rapa has many “known” side effects without backing that up. They may not even know that it can be used weekly as a safe alternative.
            I was on TA65 for 3 years and it was the same thing. The “experts” were always stating as “fact” that it caused cancer, despite all of the evidence to the contrary, and many of them had a financial bias. Too bad.
            Full disclosure: I too have been on 2mg of weekly rapamycin but only for the past month. No problems.

          • Hi Aldebaran
            Sun exposure has many important benefits going way beyond vit D. Of course the production of D is extremely vital to health and the D produced by the skin from sunlight is more protective and bioavailable than from supplements alone. A recent Swedish study over many years showed that little sun exposure doubled the mortality rate compared with those who get frequent exposure. It was the same risk as SMOKING.
            The sun is more than just Vitamin D
            1. RE- sets the circadian rhythm
            2. Raises nitric oxide to reduce blood pressure and heart attack risk
            3. Lowers cancer risk, even increases survival rates of melanoma patients.
            4. Increases serotonin and dopamine levels
            5. Lowers the risk of schizophrenia and other mental disorders.
            6. Improves immune function
            7. The infra red spectrum protects against neurodegeneration
            8. Enhances wound healing

            Dr. Michael Holick is professor of medicine and biophysics at Boston Univ. Med. Center and got a PHD researching Vitamin D. He’s the expert on this and a big believer in sun exposure.

          • Regarding overlap between Rapamycin and CR, I would love to see a life span study in mice combining both to answer the question of possible cumulative effect. And another one combining MetR and CR together as well. However, I am not aware of any such study. The only related experiment that I have found is this one where they conclude that Rapa is not a true CR mimetic:


          • Aldebaran. Combining MetR and CR?
            But, when you do standard CR you decrease all dietary components (except vitamins and minerals), and that includes proteins and their constituent aminoacids including methionine. Therefore, standard CR studies already include MetR.

          • Hi Alan,
            Thanks for you comments. This is sad but your explanation about the too much publicized side effects of Rapamycin makes perfect sense to me.

          • It does not surprise me that the lack of money to be made from rapamycin has lead to certain academics losing their intellectual independence, shall we say!

            Full disclosure from me: I have been on weekly rapamycin treatment for 6 months, with many good effects and only the occasional mouth ulcer to let me know I should slightly reduce the dose or leave it an extra day before my next dose.

            And I apologize if this is rather glib, but on the subject of money trumping science, I am also rather suspicious when companies that have spent huge amounts on proprietary tech developing many stem cell lines say a comparatively simple, less invasive and potentially cheap telomerase therapy is unsafe.

          • Hi Paul,

            Thanks for your detailed explanation about the benefits of sun exposure. Doctors in California advise strongly against it because of melanoma risks. However, knowing now the multiple benefits, I’ll continue to enjoy moderate sun exposure without any concern.

            Hi Gustavo,

            Sorry, I should have been more clear about combining MetR and DR. My understanding is that 40% DR restrict MET by only 40% and I was wondering if combining 40%DR with 80% Met restriction would further improve life span. This is because I read from your paper that MetR80 was more efficient than DR40 to lower DBI:

            “At variance with mitROSp, which is low both in long-lived species and in DR rodents, a low DBI occurs in long-lived species (see the previous section) and in 80% methionine restricted (MetR) rats, but not in 40% MetR (see Section 6) or 40% DR rats.”

          • Hi Aldebaran,
            I understand your point. Both 40% and 80% MetR lower mitROSp, but only 80% lowers DBI. Furthermore, the increase in longevity in rodents has always been studied at 80% MetR (never investigated at 40% MetR).

            On the other hand, manipulations that decrease DBI have demonstrated increases in longevity in C. elegans but not in mice (2 independenr studies).

            But of course 80% MetR could be combined with (40%) DR.

            In any case, before experimenting with yourself be careful with the absolute le vel of methionine ingestion. While MetR is clearly beneficial (inclyding less of some cancers) compared to higher levels, methionine DEFICIENCY has bee show capable to induce cancer. (again U shape dose-response effects?).

          • Hi Gustavo,

            Thanks for your comment and you advice!
            My question about MetR + DR was more theoretical than practical. I was not thinking of experimenting strict MetR on myself or anyone else. I think it would be just too complicated to implement strict MetR in humans. Social life (dinner, restaurant) would be too difficult and we would need to be closely monitored by a doctor because of the possible dangerous side effects that you have mentioned.

            On the other hand, I think that very moderate MetR by privileging food with low ratio of Met (ex: almonds) could be a safe and easy option, but with only minor benefits.

          • I just want to correct my previous statement that Doctors in California advise strongly against sun exposure. First, it is an over generalization from the few cases that I am aware of. Second, it is more correct to say that they focus on the negative sides (melanoma risk) and don’t mention the positives ones (all the points explained nicely by Paul). In a way, I can understand. If I have a melanoma, it is easy to blame the sun exposure. But if I have a heart attack, I cannot blame the lack of sun exposure to be responsible even though sun exposure decreases heart attack risks.

            My point is that with only the negative sides, I cannot make an informed decision. That is why it is very useful to have both sides (thanks again Paul).

          • Well said Aldeberan and I totally understand since I argue this with patients all the time. But consider this
            It is true that a history of a severe burn ( blisters) at a young age is a risk factor for melanoma, it is also true that UVR is a risk for skin cancers in general, though more so in ozone depleted areas.
            BUT people in higher latitudes with little sun have a higher mortality rate from cancers of the ovary, pancreas, prostate, colon and more.
            High sun exposure increases the survival rates of people WITH melanomas
            Occupational exposure to sun results in a LOWER risk of skin cancers
            Even with frequent and heavy tanning bed usage the risk of melanoma goes from .2% to only .3%
            Most skin cancers occur on parts of the body covered by clothing and lastly:
            There is no data that I am aware of linking brief periods of non-burning sun exposure on a daily basis with melanoma.
            As an aside, in those cultures where topless bathing is common, women have a much lower breast cancer risk and incidence!
            As with everything do the benefit/risk analysis and draw your own conclusions. Dermatologists are great, but they focus entirely on one organ , the skin.

          • Hi Paul, thanks for fascinating and very little know discussion about sunlight.
            As a footnote: You mentioned sunlight activates Nitric oxide;
            In 19th century, before any antibiotics, the one effective treatment for Tuberculosis was sunlight, preferable at high altitude where UV radiation was greater. Recently this mystery has been unraveled (Curing TB with sunlight, Wells, 2006). The sunlight makes Vit D, and downstream this makes Nitric oxide, which activates pathway which makes cathelicidin, which acts as antibiotic and kills Mycobacterium.
            Sunlight also plays key role in ameliorating two immune type diseases of unknown etiology, Psoriasis and Multiple sclerosis. Pathways still unknown.
            Vitamin D and melatonin just the start of long list of remarkable chemicals in body activated by sunlight.
            I spend over 600 hours in sunlight this summer and never knew all the free benefits I was getting.

          • Wow Alan that’s amazing! So often in medicine we see that things work long before we understand why they work.

          • Hi Paul,

            Thanks for pointing out all this facts! Very informative. I will definitely continue to enjoy moderate (non burning) sun exposure when I have a chance.

          • So far in these discussions we’ve wandered from telomeres to rapamycin to youthful skin to sun exposure. This is a great site

  14. well, I havent read the study you referred to but I am interested in it. I just added what I read earlier that sometimes cells need to overcome the p16Ink4a checkpoint to achive imortality. overcoming this checkpoint goes relatively easily with positive selection favouring the cells escaping this limit. overcoming this checkpoint also made the cells take up premalignant properties such as loss of contact inhibition.
    maybe extra short telomere cells are favoured in overcoming the p16ink4a checkpoint?

    • I was thinking that perhaps it would explain the results of the study that Josh refenced, maybe that sub population have a somewhat longer mean telomere length but also a higher percentage of very short ones leading to malignant transformations, especially of the aggressive cancers cited

        • Yeah and it would explain the contradictory findings of the danish study where the very long telomere group had the lowest mortality rates, probably also the lowest percent of very shorts as well

        • It’s all very confusing but this may explain the discrepancies.If we divide the population into 4 groups based on telomere distribution we’d have:
          Group A : Small population group. Overall short LTL’s .
          Higher general all cause mortality rates.
          Group B: Large population group.
          Average LTL. Not a large percent of long or very short telomeres. Average overall mortality rates.
          Group C: Small population
          Long average LTL.
          Combination of long telomeres but also a high percent of critically short telomeres. Lower heart disease risk from long telomeres but a higher cancer risk from the very short ones. This is the group that skews the data and makes it appear as if long telomeres cause cancer.
          Group D: Small population group with long LTL’s and a very low percent of critically short telomeres leading to an overall protective effect. This group includes very young people, those with very healthful lifestyles, Ashkenazi Jews, and the upper 10 % of the Danish study. This also validates Josh’s thinking that telomerase therapy can reduce mortality and prolong life by placing more people into this group and out of all of the other groups.
          Just speculating but it may fit the data.

          • Hi Paul,
            Thanks for trying to clarify discrepancies in data with very good explanation.
            It would be very helpful to have clear idea about telemere length and impact on health.
            Some very common things effect telemere length: For example:
            cigarette smoking: shorten
            caffeine: shorten
            coffee: lengthen.
            Interesting part about coffee is the opposite effect of coffee and caffeine.
            So if very common things like cigarettes, caffeine and coffee affect telemere length, probably long list of things that effect telemere length.

            As regards coffee, I had always been very impressed about beneficial effects of coffee in large number of different studies. Perhaps effect on telemere length is explanation.

          • Makes sense. Do you have a link to the Ashkenazi study? Thanks.

            Oh, and Alan, any ideas on how coffee lengthens telomeres? Caffeine just flat-out breaks DNA, so that’s amazing.

          • the study was presented by the Albert Einstein School of Medicine at the Proceedings of the National Academy of Sciences in November of 2009. They tested 86 very old people with an average age of 97 and 193 of their offspring and found high levels of telomerase and very long telomeres in both groups. The other interesting thing is that their lifestyles are anything but healthy.

          • Very interesting coffee studies.
            Aldebaran provided reference for paper I cited.
            In regard to Bill Walker question, how does coffee length telemeres:
            Coffee is witches brew of very many chemicals. People drink coffee for the caffeine which is harmful. However, there is also apparently also a magic bullet in coffee that lengthens telemeres.
            The other studies were showing a link with genetically determined telemere length and all cause mortality.
            The coffee studies show the effect of treatment in very large groups over many years.
            The coffee studies convinced me.
            My conclusion is shortening of telemeres is an age-related diseases that probably affects repair and maintenance and preventing shortening of telemeres improves survival. Coffee (but not caffeine) is not best documented substance to increases length telemeres or prevent shortening with age.
            (I hate coffee, so sucks for me.)

          • New study on coffee, Dr. Navarro et al presented on a 18 year cohort study of almost 20,000 people at the European Society of Cardiology meeting in Barcelona on 8/29/17. Showed a dramatic 65% reduction in all cause mortality with 4 cups coffee/day ( caff or decaf didn’t matter) and this jumped another 30% with 6 cups a day. This just adds to the growing data supporting coffee intake to reduce mortality rates. So I’m up to 6 cups now and feel like Beevus and Butthead, ” I am Cornholio”!

          • “Showed a dramatic 65% reduction in all cause mortality with 4 cups coffee/day ( caff or decaf didn’t matter) and this jumped another 30% with 6 cups a day.”

            This looks impressive!
            I used to drink lots of coffee but cannot anymore due to stomach issues. I do love coffee though so knowing the benefits on mortality, I wish I could drink more again.

            I suspect that overweight people should be the ones with the most reduction in mortality from coffee consumption.

            This seems to the case in mice:


          • In the 2014 study cited about mice by Rustenbeck they turn out to be totally clueless. First, they think effect related to caffeine and forget coffee is very many chemicals. Second, never understand the protective effect related to telemeres and not glucose metabolism. 3 years later totally different understanding of what happening.

          • Hi Aldebaran,
            Fascinating paper and intriguing question.
            Coffee, the miracle brew, appears to have two independent effects regarding ageing. One is increasing length telemeres and now other is decreasing mTOR.
            Increasing length telemeres seems totally different than mTOR effect.

            So very good question.

          • That was a surprising effect Aldebaran. The phenolic compound found in coffee, chlorogenic acid, was found to reduce ROS levels in brain cells even more than resveratrol. It goes on and on with coffee.
            See Neurochem Res 2016 Aug (8) 2075-2085

          • I am a big believer in coffee. For me, it has been the only psychotropic drug that has helped me with my major depressive disorder — over time, possibly better than electroconvulsive therapy. (I’m not joking.)

            Meanwhile, the mTOR pathway is all the rage in depression research, except researchers are taking about stimulating it, not inhibiting it. For example:

          • that’s amazing. I’m starting to wonder if a combination of coffee, TMS, and low dose naltrexone might be a great anti-depressant combination without any of the myriad side-effects of antidepressants and without increasing TOR

          • Response to John,

            As regards depression, mTOR effect is all a matter of dose.

            Low dose studies with mice, the dose that resulted in life span extension cause mice to have less anxiety and improved mood. mTOR increases brain chemicals which are typically targeted by anti-depressants.
            So I think quite possible mTOR lowering effect helped your depression. Your comment very interesting.

          • Perhaps I should explain — brutal treatment-resistant major depression and chronic depersonalization/derealization caused me to be housebound with anxiety for about ten years, starting in 1995, when I was 39 years old. I started drinking coffee during the sixth year, and never stopped drinking it. I started to come out of the worst aspects of the depression during the tenth year. I read about an epidemiological study showing lower suicide rates among nurses. So I came up with the theory that coffee helped me come out of the worst aspects of the depression, very slowly, through several years. The only other explanation is natural remission, because I had stopped trying treatments by 2001.

            But I still suffer from fatigue, with no clear medical explanation. I believe it is a residual symptom of major depression. And I still have depersonalization, without the associated anxiety. (I never believed depersonalization is secondary to anxiety, and now I have proof for myself.)

            I lost a lot of years due to depression. Now I need anti-aging treatments to get my lost years back!

          • Great article John. The evidence is clear. Ignore the bean at great peril!
            I propose now that coffee drinking may have a more potent effect on health than exercise.

          • Hi Paul,
            Other interesting thing about paper is how went from coffee to rapamycin to Mount Tam Biotechnologies.
            They don’t have an approved drug; but they have a ton of silicon valley money invested in company. They have a rapamycin knock-off.
            Can see by the way Mount Tam discussed the huge difference that money makes in press coverage.

          • So here’s how you do it to get around the compounding problem.
            Ask him for the standard 50mg dose tabs. You dissolve them ( easily) in 50ml of water. It totally dissolves and is stable refrigerated for one month. Get a syringe and start with 1.5 ml per day for the first month, AM or PM, and then go up to 3mg’s for next 2-3 months. May then go to the max dose of 4.5mg but wouldn’t exceed that dose. If he wishes to contact me [email protected].
            You can always offer to sign a consent form.

  15. As an aside have you had any followup with James Green regarding success with his astragalus regime, which you previously wrote an article on? Regards, Stephen

  16. Greenland sharks live over 400 years. Bowhead whales live over 230 years (with ~500 times more cells than humans). I know the whales have similar telomere lengths to us (and cows), because I did the TRFs 😉

    …so the answer to cancer or aging isn’t in average starting somatic-cell telomere lengths. Maybe stem cell telomerase activity, for aging? And cancer is dependent on the cancer control systems, e.g. the 20 copies of P53 in elephants.

    Not stopping my TA-65 based on this paper 😉

  17. Josh, thanks for doing such a great series. I read through the Danish study and your interpretation, and it seems to me that there is a fairly significant weakness in using it to support the idea that the direct impact of Telomere length is anywhere close to as large as is implied by the study or studies with similar design.

    The issue I see is that the factors the study corrects for seem to be very incomplete, and if other significant factors not considered are also driving poor health outcomes, then that could reduce or eliminate the impact that appears to be due to Telomeres. (You use the phase “all the health variables” however I think it would be more accurate to say “all the considered health variables”.) For example, it is common for people (based on the Kraft analysis for example) who are fairly active, have healthy BMI, etc to still have chronically elevated insulin and in many cases elevated blood sugar, both of which are highly correllated ( and likely causative) of bad health outcomes, however they don’t appear to be corrected for in the study. Another example is lack of frequent brief exposure to intense sunlight which manifests as low vitamin D status, but also has other impacts such as reduced nitric oxide production and increased blood pressure. Beyond these there is a long tail of other things like stress, sleep quality, etc. that would also have an impact. So I think the best conclusion is that whatever impact that is left over to be directly caused by Telomere length is at least likely to be much smaller than that stated in the study, if there is actually any impact left over at all.

  18. Maybe you’re right. But it’s only important to correct for confounders if we have reason to believe they might correlate (positively or negatively) with telomere length. For example, if low vitamin D could cause both shorter telomeres (how?) and also higher cancer mortality (through a different mechanism than telomere length), then the failure to include vit D levels in the analysis would be problematic.

  19. I sent this article to Phillip Haycock, first author of the study, and he was kind enough to send the following extensive response:

    Hi Josh. Thank you for your comments on our paper ( and for the opportunity to respond. Below I provide some responses to the issues you raise.

    1. Observational studies of directly measured telomere length are more reliable

    Observational studies, due to their non-experimental nature – are highly susceptible to confounding and reverse causation (e.g. whereby disease causes changes in telomere length and not vice versa). It’s not too much of an exaggeration to say that in observational studies almost everything is correlated with everything, making judgements about causality basically impossible. For example, observational studies tend to find that telomere length is associated with everything-under-the-sun (from meditation to stroke). Such wide-spread observational associations are likely to reflect wide-spread confounding. In contrast, genetic variants do not generally correlate with classic environmental and lifestyle factors (predicted in theory by Mendel’s laws and observed in practice). Germline genetic variants are also fixed. This means gene-trait associations in population studies are less susceptible to confounding and reverse causation. There are many high profile examples of randomized controlled trials – designed on basis of conventional observational studies – that failed due to flawed observational findings (e.g. vitamin E supplements and HDL cholesterol and heart disease; selenium supplementation and prostate cancer). There is a strong sense in our field that traditional observational epidemiology has failed – at least from an aetiological point of view (risk prediction doesn’t require knowledge on causality). It’s hard not to feel cynical over each new finding reported in the popular press of how coffee/alcohol/Mediterranean diet, etc, protect or harm your health. And I think the public are growing increasingly wary of the often contradictory findings being reported by conventional observational studies. And the public are right to be doubtful since these findings are almost always due to confounding and not causality. It is in this context that the technique of Mendelian randomization was developed. It is actually an application of instrumental variables analysis, originally developed in the field of econometrics. Mendelian randomization does however have limitations – but these must be contrasted with those of conventional observational studies. Mendelian randomization studies make the following key assumptions: 1. the genetic variants are associated with the exposure of interest (e.g. telomere length; by convention we refer to the intermediate/aetiological trait as the exposure); 2) the genetic variants are not associated with confounders; and 3) the genetic variants are associated with disease exclusively through their effect on the exposure (e.g. telomere length). These assumptions are not provable but it is possible to falsify them through various types of sensitivity analyses. You rightly highlighted that our results could be due to violation of assumption 3 (more on that below).

    2. Observational studies of directly measured telomere length provide opposite conclusion.

    Our findings are generally in strong agreement with prospectively designed observational studies (where telomere length is measured before cancer diagnosis). The apparent conflict you cite is almost entirely due to the retrospective studies, where telomere length is measured after cancer diagnosis, and which generally do find that shorter telomeres increase cancer risk. In my opinion this is due to reverse causation bias in the retrospective studies, whereby shorter telomeres arise as a result of cancer and not vice versa. You can find references to the studies in the discussion section of our paper.

    One of the observational studies you cite as being in strong conflict is the Danish study. I believe you are referring to the Copenhagen City Heart Study and the Copenhagen General Population Study. These are well-designed population-based studies that have examined the prospective association between telomere length and various diseases, including cancer. They have also conducted Mendelian randomization studies of telomere length and cancer. Contrary to your claim, our findings are in strong agreement with the findings from these studies. The studies are large with samples sizes ranging from 47,000 to 96,000. The main studies are:
    “…Short telomere length is…not [associated] with cancer risk”
    “…genetically determined short telomeres were associated with low cancer mortality…”

    Our findings are also in agreement with other Mendelian randomization studies of telomere length and cancer (although many of these do overlap with our study so aren’t strictly independent).

    Overall, there is actually quite good epidemiological evidence (from observational and Mendelian randomization studies) that longer telomeres increase risk of cancer. An important caveat of all this is that we generally don’t have good evidence on the shape of the association (many studies assume linearity) or evidence on the nature of the association in population subgroups. We do know that individuals with dyskeratosis congenita, a disease caused by germline loss-of-function mutations in the telomerase component genes TERC and TERT have chronically short telomeres and are at increased risk of some cancers, presumably due to increased susceptibility to genome instability and chromosomal end-to-end fusions. It is therefore likely that in some circumstances – .e.g extreme end of the distribution – that shorter telomeres do increase risk of cancer. Our results should therefore be interpreted as reflecting the average association at the population level and may not be generalizable to the extreme ends of the telomere length distribution.

    3. Results are due to direct effect of SNPs on cancer

    This possibility would be a violation of assumption 3 above – that the SNPs affect cancer exclusively via their effect on telomere length. Horizontal pleiotropy is a well known genetic phenomenon that could induce such direct associations between SNPs and cancer that bypass telomere length. This is the most important potential limitation of Mendelian randomization studies. We observed some evidence for this in our results and we admit in the discussion that we cannot entirely exclude this possibility. However, in our sensitivity analyses (MR-Egger and weighted median, see supplement), which make some allowance for horizontal pleiotropy, our results remained similar for most cancers (although again we admit in the discussion that analyses to detect horizontal pleiotropy were also underpowered). Results are also less likely to be due to horizontal pleiotropy when similar results are observed across independent genomic loci, i.e. when telomere length raising alleles are associated with higher risk of cancer at independent gene regions (each independent gene acts as an independent natural experiment). Our main results are based on averaging across the 10 independent genomic regions but for glioma, lung adenocarcinoma and coronary heart disease, similar results were seen across independent genomic regions. As noted above, our results are also generally consistent with prospective observational studies, which are not susceptible to this type of bias.

    Methodological details on the sensitivity analyses can be found in the supplement.


  20. Phillip Haycock’s comments, continued

    4. Our study relies on detecting very small effects that are lost in the noise and difficult to detect.

    Yes, the SNPs have a relatively small effect, which limits statistical power. This is reflected in the wide confidence intervals for many of the diseases. Where confidence intervals are wide I wouldn’t take the effect sizes too seriously (e.g. Glioma). In your blog post above you include effect sizes for various disease but appropriate interpretation requires consideration of the uncertainty in the estimates, which is quite considerable for some of the disease. Due to the small effect of the SNPs on telomere length very large sample sizes are required to detect valid associations. Our primary analyses were therefore based on disease studies where power was expected to be modest-to-strong. We excluded diseases where a priori powered was expected to be low. The latter studies were typically either too small in size or had too few SNPs available (power depends on both sample size and number of SNPs).

    We consider this issue in more detail in the supplement of our paper.

    5. The new studies require very large implicit extrapolation that is not necessary in the old studies. The 50 to 1 extrapolation is very speculative, and it magnifies the noise along with the signal.

    We standardise the results to reflect a 1-SD (standard deviation) change in telomere length and therefore you are correct that we are extrapolating beyond the observed effect sizes of the SNPs. However, the extrapolation you describe is more like a 7 to 1 than 50 to 1 extrapolation because the average effect size of the SNPs is 0.13 SD units per copy of the telomere length raising allele. The standardisation step is desirable as it makes it easier to compare results across different cancers and diseases as well as with findings from independent studies – e.g. observational studies of directly measured telomere length. A 1-SD unit of comparison is very commonly used in epidemiology. This has the disadvantage of extrapolating the effect beyond the observed effect of the SNPs on telomere length. However, if the shape of the association between telomere length and disease is linear these effect sizes should be valid. The degree of the bias will depend on the degree of deviation from linearity over the telomere length distribution. However, even if the effect sizes are biased (and we are therefore uncertain about the precise magnitude of the effect), the general conclusions about the overall direction of association (i.e. longer telomeres increase cancer risk) will be valid.

    • great response and much appreciated for its detail and very comprehensive explanation. I continue to stand-by my proposed solution of 4 distinct groups where group D , with very long telomeres and low percent of critically shorts are protected against all causes including cancer, this group is at the extreme end. At the other extreme is group A, at risk for all diseases, and group C with long telomeres but also a high percent of critically shorts , low cardiac risk and higher cancer risk. They’re just not looking at the right variable.But in the future it will be critical for any telomerase intervention to aim to correct the critically short telomeres and not just a haphazard overall lengthening.
      Thanks Josh for providing that. Much appreciated.

    • Kudos for Josh for obtaining such a fantastic discussion of telemeres from “Dr Telemere” himself.
      I think fact that Josh could obtain such discussion from a leading scientist shows that this blog is respected as place where people try to have serious discussion.

  21. Phillip Haycock’s comments, continued

    6. GWAS not same as MR

    You write about GWAS and Mendelian randomization as if they are the same techniques. They are actually quite distinct methodologies and analytical approaches. In GWAS we measure the association between genetic variants and human traits across 100s of thousands to millions of loci across the genome (focus is gene-trait association). Mendelian randomization is the use of genetic variants as instrumental variables to appraise causality in hypothesized exposure-disease associations (i.e. the focus is the exposure-disease association). Instrumental variables don’t have to be genetic and the technique was originally developed in the field of econometrics. Mendelian randomization exploits the properties of genetic variation in the population: the tendency of genetic variants to be randomly distributed in the general population with respective to lifestyle and “environmental” factors. Genetic variants are also fixed. This makes gene-trait associations in the population less susceptible to confounding and reverse causation – a major limitation of traditional observational studies.

    7. Study relies on people with all 10 TL raising variants
    This is incorrect. All subjects contribute to the results regardless of their genotypes. The results will reflect the average number of telomere length raising alleles.

    8. Another possibility is that one or more of the SNPs happen to be more common in a segment of the population that is prone to cancer, for whatever reason.

    The problem you are referring to is known as confounding by population stratification – the tendency of cases and controls to have slightly different genetic ancestries and which can introduce confounding into genetic association studies. This issue is taken very seriously by genome-wide association studies. We did our best to take this into account. For example, our analyses were either adjusted for principal component scores of genome-wide genetic variation or we found little evidence for population stratification in diagnostic plots (these are standard techniques in the field). However, I agree and acknowledge in the paper that we can’t entirely rule out this possibility. More details in the discussion section of the paper.

    9. GWAS is also referred to as “Mendelian randomization studies” because they depend very much on the assumption that different SNPs are randomly distributed in the population. Of course, this assumption is not literally satisfied. How significant is the deviation from random distribution? I will be investigating this question, and I’ll let you know what I find.

    As noted above, Mendelian randomization and GWAS are quite distinct techniques. With regards to the randomization assumption, we expect it to hold in general ‘unselected’ populations (but probably not selected populations, e.g. disease—only studies). The randomization stems from Mendel’s laws of inheritance – the laws of independent assortment and segregation – and also the fixed nature of germline genotypes. As well as predicted from theory, this also tends to be observed in practice. The randomization assumption is likely to be violated to at least some small degree for many of the diseases in our study. This could introduce some bias or confounding into the analysis. However, our study is still likely to be more robust to confounding in comparison to most conventional observational studies. If deviations from randomization were widespread we would expect to see many more associations with telomere length but for the majority of the diseases and risk factors (see supplement) in our study we only saw weak evidence for associations. In contrast, observational studies tend to find that telomere length is associated with everything-under-the-sun (from meditation to stroke). Such wide-spread observational associations are likely to reflect wide-spread confounding.

    10. There is/no overall net benefit
    We don’t know what the net benefits are at the population level and can’t infer that from our study. This requires detailed statistical modelling of absolute as well as relative risks. For example, although the relative risk for glioma for longer telomeres was about 5, the absolute risk is very low, around 0.4 per 100,000 per year in the US. This implies that the additional cases of glioma caused by longer telomeres is 5*0.4-0.4 = 1.6 per 100,000 (still very low). In contrast the relative risk for CHD is about 0.8 but the absolute risk of CHD is around 250 per 100,000 per year. The additional cases of CHD prevented by having longer telomeres is therefore around 250-0.8*250 = 50 per 100,000 per year (compared to 1.6 cases of glioma caused by having longer telomeres per 100,000 per year).

    Our study assumes a linear shape of association between telomere length and cancer. However, there is evidence that the shape of the association may be J shaped, such that individuals with extremely short telomeres may also be at increased risk of cancer. Our results should therefore be interpreted as reflecting the population-level average effect. This also means that we are unable to infer the nature of the association in subgroups of the population, such as those with very long or very short telomeres.

    11. Evolutionary tradeoff
    There’s an interesting literature about potential evolutionary tradeoffs in cancer and vascular disease risk and the impact of body size and telomere length. Cancer incidence doesn’t seem to increase with increasing mammal body size (about same rates in mice and humans, known as Peto’s paradox). See this interesting review on “Telomere Length and the Cancer–Atherosclerosis Trade-Off”:

  22. Seems like a reasonable response from Phillip Haycock. My reading from his paper and his explanation above is that he is using a large number of people and statistical analysis to pull out a small effect that would otherwise be hidden in the noise. As he admits under point 10., making rare cancers slightly more likely still leaves them rare, in comparison to any factor that modifies a big killer like heart disease.

    The evolutionary trade off he mentions is intriguing as it agrees with my comment at the top of the thread regarding telomere length being inversely correlated with body size: larger mammals have shorter telomeres, probably because the chances of cancer rise with more cells to potentially become cancerous.

    None of this means we should be happy about our telomeres shortening with age. As he admits several times, his analysis does not say anything about very long or very short telomeres, and he acknowledges there is some evidence the latter can cause cancer, (as well the other diseases short telomeres are known to contribute to).

    • Hi Mark,
      In terms of evolutionary trade-off atherosclerotic heart disease, (number 1 killer) vs cancer:
      One should consider that for humans 100,000 years ago, the normal life style was that of the ultra long distance runner and atherosclerotic heart disease was a very rare condition.
      Cancer was the number one risk that had to be protected against by large long-lived animals. Very interesting that mentioned that of 25,000 known genomes,15,000 related to cancer, growth issues.
      So humans from evolutionary viewpoint had good reason to limit telemerase and telemere length.
      However, It may well be that if a person is a couch potato (anybody getting less than 25 hours moderate physical exercise a week) then short telemeres become very big problem.

    • I think the author did a great job defending his methodology. His study is valid, just not informative. If you believe as I do that very long telomeres are protective and conversely very shorts are risky, then this study neither confirms nor contradicts. It certainly doesn’t tell us that very long telomeres cause cancer. In fact, it basically tells us nothing.

      • “It certainly doesn’t tell us that very long telomeres cause cancer.”

        Insofar as there are linear, J or U-shaped associations between telomere length and cancer, then very long telomeres will definitely cause cancer. The contrary would require a ‘n’ shaped association but there is little evidence for this in observational studies. The evidence in observational studies generally points to linear, J or U shaped associations.

        “His study is valid, just not informative…[the study] basically tells us nothing…”

        Unsurprisingly I disagree that the study is uninformative. At the population level, the net effect of longer telomeres is to increase cancer risk and to reduce risk of coronary heart disease. This tradeoff has to be carefully considered in any applications based on telomere length, eg, as a tool for risk prediction or as an intervention target for disease prevention.

        For example, the results of our study are informative for what would theoretically happen if you were to intervene in the general population – i.e. give telomere length raising medications / interventions to generally healthy people. If we extend the thought experiment in my previous post about absolute and relative risks to all the cancers in my study (not just glioma), then I estimate the following net effects of shifting average telomere length in the general population by 1 standard deviation:

        1. 53 fewer CHD events and 250 additional cancer cases per year per 100,000 people in US population
        2. Extrapolating the results to the entire US population, this translates into: 111,441 fewer cases of cardiovascular
        disease and 210,892 additional cases of cancer

        I think this means that if you were to develop interventions for increasing telomere length these should only be targeted to specific groups of people – e.g. people already at high risk of cardiovascular disease or people with extremely short telomere length.

        Also, many companies offer telomere length testing services under the claim that shorter telomeres are a general indicator of poorer health status and older biological age and that such information can be used to motivate healthy lifestyle choices in individuals. Someone who is told they are “biologically young” i.e. have relatively longer telomeres, might think they are not at increased risk of cancer and don’t need to modify their lifestyle. Our results indicate that this would be mistaken and that, in unselected/otherwise healthy people, the advice should always be to adopt healthy lifestyles. Lifestyle advice should not be directed on basis of telomere length tests. However, it’s possible that such testing might be useful in certain circumstances (e.g. high risk of CVD or extremely shorter telomeres, etc) but we don’t have good evidence for that yet.

        These are the assumptions for the above thought experiment on net benefits/risks of intervening on telomere length in the general population:
        – telomere length is normally distributed
        – longer telomeres are defined as 1 SD above the mean
        – incidence of cancer in people with average telomere length = incidence per 100,000 in general US population
        – 1.38 = the odds ratio for cancer per 1-SD increase in TL. (This is the average odds ratio for cancer in our study, weighted by cancer incidence in the US population).
        -322,762,018 = the US population size in 2016
        -1,685,210 = number of expected new cancer diagnoses in US in 2016
        -522 per 100,000 = incidence of cancer per year in the US (assume this is the incidence of cancer in people with average telomere length)
        -1.38*522 = 720 per 100,000 = incidence of cancer per year in the US sub-population with longer telomeres
        -198 per 100,000 =absolute risk difference (720-522) due to longer telomeres
        -106,511,465 Americans have long telomeres (0.33*322,762,018, where long means 1SD above the mean)
        -198/100,000 * 106,511,465 = 210,892 additional cancer cases per year in US in people with longer telomeres

        • The absolute risk difference is 198 per 100,000 in terms of cancer in those people who have an above average mean telomere length. I am saying that that relatively small group may also have a higher percentage of critically short telomeres putting them at risk. This wasn’t measured and so we don’t know that from your study, but it certainly could be the case since we do know that critically short telomeres lead to DNA instability and are also involved in the final step of malignant transformations of cancer into an immortal state. If shortening of telomeres is nature’s way of protecting us from cancer then it’s doing a lousy job of it since it’s now the number one killer in the U.S., and predominantly in old people with very short telomeres. And of course , the very young with very long telomeres and a low percentage of very shorts, rarely get cancer.
          But I do agree that just because you get a lab value showing a long mean telomere length that you should stop exercising and start smoking!
          As an aside , it’s very nice of you to take time to respond us on this website, and I’m probably wrong since I usually am.

          • Yes seconded, it’s very nice of Phillip to respond here.

            I am personally still not convinced that longer telomeres is a bad thing, at least within the normal range (for a human). Only a small error in this study would extrapolate out an exaggerated risk. None of Blasco et als work with AAV HTERT in mice has shown an increase in cancer incidence; in fact there have been remarkable benefits to health and lifespan. Yes perhaps she didn’t use enough mice to show up this effect, but that just shows how small the effect must be, or that this applies to humans only, which throws doubt that this effect is mediated by telomere length at all.

            Anyway I am sure the telomere aging vs cancer debate will rage on for a while yet.

          • Phillip stated in his response that he wouldn’t like to see someone get a ” long telomeres”result back from a lab and conclude that they can stop their healthy lifestyles. But if you follow his findings to their logical conclusion then you should indeed stop any telomere lengthening lifestyle ( exercise, meditation, good sleep and diet, even smoking) since all of these have been shown to activate telomerase and elongate telomeres, thereby increasing cancer risk. There are many disconnects

        • Dear Prof Haycock,
          Thank you for detailed explanation and great scientific work. It is very remarkable to have one of leading scientist take the time to explain things. Greatly appreciated.

          • My opinion is Josh did a fabulous job in getting the person who appears to be the world’s leading expert on telemeres to have very detailed discussion on this board. For me. it was extremely informative. There is huge amount of junk science on internet about telemeres and in particular scams to have people spend $100 to get their telemeres measures based on bogus information of the significance of the measurement. Today we have very authoritative expert provide real science. Real science is always better than junk science, even when may not like the facts. The critical fact was that if everybody in US had telemeres lengthened 1 standard deviation, there would be 100,000 less cases of heart attack and 200,000 more cases of cancer. However, he did not say that if you had short telemeres by 1 or 2 standard deviations, that you might not benefit from longer telemeres.
            Note that at about 3 years old, in aging studies, all the mice are dying of cancer. Even when rapamycin extends lifespan, both the controls and treated mice almost all die from cancer. Humans on other hand, live to 8th, 9th and 10th decade until cancer starts increasing dramatically.
            What Philip Haycock was saying is not a good idea to willy nilly intervene and extend human telemeres.

  23. There is a very interesting new paper on the connection between the epigenetic clock and telomere length. Steve Horvath at al. find that SNPs associated with longer telomeres.

    So increased hTERT expression seems to hasten the epigenetic clock. Accelerated epigenetic aging is a good predictor for mortality. So this seems to corrobate the Haycock paper also sheding light on the causal mechanism.

    “TERT alleles associated with a longer LTL (indicative of younger biological age) were robustly associated with increased IEAA (indicative of older biological age) (P ~ 1.e-11)”
    Also they found that cultured cells expressing hTERT aged faster epigenetically if the cells were doubling.
    I think this is very significant finding even from a philosophic or evolutionary perspective.

      • The solution to this is actually pretty obvious and it stuns me that this isn’t elucidated in the discussion portion of this paper.

        If you have genetically shorter telomeres your proliferating cells will actually be replaced more often, so at a given instant in time your epigenetic marks will appear to be younger than someone whose cells are replaced less often (by stem cells whose epigenetic marks all start at age 0).

        [Aside: How this might correlate to a lower mortality (independent of the risk of short telomeres also being a risk factor) is not understood, but perhaps faster replacement means more cells will be in ‘tip top’ shape at all times?]

        Telomerase immortal cells obviously don’t ever need to be replaced so continue to clock up more epigenetic marks. But they don’t if they aren’t allowed to replicate (end of page 10). This suggests that epigenetic marks are acquired only during mitosis.

        [Aside2: this doesn’t make sense in light of the fact that non-TERT cells stop acquiring epigenetic marks beyond a certain point even though they continue to proliferate. My guess is that this is because their proliferation in slowing as they approach senescence, but I could be wrong.]

        Of interest to Alan is that epigenetic aging rate is correlated with age of puberty (in girls) and inversely correlated with age of menopause, strongly supporting the role of MTOR on aging.

        • I doubt there are stem cells involved, these are experiments in a dish with a mono layer of cells. Those cells are all differentiated and just keep dividing until they reach each other then they stop. Then they are scratched up, diluted and planted to a new empty dish, where they start doubling until confluence again.
          Adult SCs also dont have 0 epigenetic age.
          Only IPSCs and ESCs have 0 epigenetic age.
          Remember DNAm age is by far the best age predictor in vivo (much superior to telomere length).
          Postmitotic cells do inhibit epigenetic aging. The DNAm age is universal across tissues regardless of replacement rate.
          But this is a dish environment and here I believe they are experimenting with accelerated epigenetic aging. They create an environment where cells receive stimuli because of loss of contact or serum, or growth factor. So they behave like they are still in an embryo. Thats why the accelerated epigentic aging, which in the embryo constitutes the development program.

          • Actually I think this kind of in vivo accelerated aging is something where I would expect some real breakthroughs on aging.
            I believe the epigenetic changes are the true drivers of aging, because they irreversibly degrade the quality of the chromatin. Also I believe these epigenetic changes are the embryonic developmental program and aging is a continuation of the developmental program.

          • Good point, my explanation would only apply in vivo – so can’t explain the effects in a petri dish!

            Given that earlier menarche and menopause equals greater epigenetic aging, this may be some sort of MTOR growth signal. That would affect post mitotic cells too. Is the epigenetic age uniform across tissues?

          • Hi Mark,
            Since discussion turned to epigentic age markers and you mentioned you were taking rapamycin for 6 months, thought this paper would be of interest:
            Epigenetic aging signatures in mice livers are slowed by dwarfism, caloric restriction and rapamycin treatment. Wang…Ideker, 2017.
            Paper confirms claim that rapamycin slows aging.
            I think if rapamycin was being promoted by a company and a CEO that could make big money from rapamycin; raamycin would be biggest news story; but since rapamycin is a generic prescription drug and no way anybody can make money from rapamycin, almost zero interest.

          • Gabor. I tend to agree that these in vitro experiments are very useful.

            I guess the question that applies to epigenetic methylation is the same one that has been asked re: telomeres in the past – are they a consequence or cause of aging? Or a better question: can we reset them to reset aging? In cells the answer is we can, but then the same is true of telomeres and we have yet to adequately demonstrate that it reverses aging in humans.

        • A group of molecular biologists made me aware that telomerase activated through various interventions does indeed act preferentially on very short telomeres and one result is that the relative proportion of short telomeres is increased, and this increase causes the average length to go down in many people. Since most technologies used in studies measure only average telomere length this results in the grouping together of people who have very different telomere length distributions.
          Very similar to drawing conclusions on cardiovascular risk based on Total cholesterol levels without knowing the subgroup details. The devil is indeed in the details.

          • I guess this also reveals a potential weakness in telomere elongation therapy – if telomerase targets the shortest telomeres, repeated treatment would converge all telomeres on some average as the short ones were elongated but the long ones continued to shorten.

        • I still think this is a good explanation of why telomere length and epigenetic age sometimes point in the opposite direction (in Vivo).

          The in vitro work still confuses me. Why does the epigenetic count stop going up when the immortalised cells are quiescent, yet GaborB tells me post mitotic cells age epigenetically in the same way as proliferating ones? My best guess is contact is inhibiting the MTOR signal, so epigenetic count is slowing right down.

          • And I forgot to say, that is a fascinating paper Alan on slowing down of epigenetic clock by dwarfism, CAR and rapamycin. Shows MTOR signal is absolutely fundamental, and though Horvath suggests his measure is not affected by lifestyle i think he is wrong.

          • Hi Mark,
            A couple of other ideas on those same epigenetic DNA methylation lines:
            Methylation of DNA part of doomsday clock.
            Purpose is to prevent cancer.
            Each time transcription takes place also have methylation.
            Genome only gets limited number of transcriptions as when get too crowded with methy groups, blocks transcription. Just enough methyl groups can be added for expected lifespan.
            DNA methylation correlates with chronological age as number of transcriptions and methylations directly related to actual age.
            Cancer needs mechanism for demethylation to become immortalized.
            (Note: This actually feeds into Josh’s idea that organism plans its own mortality. Here the idea is organism evolves anti-cancer plan, and consequence if gives up potential immortality; however evolution just wants organism to live long enough to reproduce and expects to die from predators 99% of time before ever reaches mortality limits of anti-cancer scheme. So net result same but motivation different.)
            This explains how rapamycin slows aging in worms. Worms do not get cancer like mice or other age related diseases. Rapamycin slows down transcription in worms, this slows down epigenetic methylation and prolongs lifespan. Also explains why CR and rapamycin only slow down aging; but doesn’t change overall scheme.
            Remember Philip Haycock stated 15,000 out of 25,000 known genomes related to cancer and growth. This suggests cancer was huge concern and justifies epigenetic DNA methylation and limiting telemere length and availability telemerase as both part of anti-cancer mechanism.
            Actually, anti-cancer works extremely well in humans as most live to 80s before cancer becomes very significant risk.
            Just some rough ideas.

          • Some great ideas Alan. The only issue i can think of is if methylation eventually stops transcription, and immortalised (non cancer) TERT cells continue to acquire methylation as per Horvath work, then why do they not stop proliferating?

            Perhaps the damage done to cells from methylation can be divided and hence only mounts up in arrested cells?

            I think you are right though, in that evolution is far more concerned with eliminating cancer in young people than it is in the health of old people. Hence controls on telomere length, etc.

          • Hi John,
            That was one fantastic study. As pathologist, I’m sure if the mummies had cancer, they could be identified, so must accept their conclusion that cancer was rare in ancient Egyptians.
            They make point that increase in cancer due to either diet, life style or environmental factors like exposure to carcinogens.
            In my website, I cite studies of Okinowans who followed lifestyle of @ 8% caloric restriction and very high level physical activity as they were farmers. they were all quite lean. They had about 20% overall cancer rate of Americans. This shows that diet and life style (physical activity) can make huge contribution to cancer incidence. Add in a single activity as smoking and have cause of 10% deaths world wide. Then there may be many other things in our environment which we don’t know about which might promote cancer. Think of all the cancer from just one job, ship building and asbestos. So that was one great study.
            I don’t want to get political; but would be nice if our government believed that science is a real thing and not a Chinese conspiracy.

          • Thank you, Dr. Green! The thing is, if cancer was very rare throughout most of the evolutionary history of humans, why would we have the “evolutionary trade off” process that people allude to on this blog? In other words, why would cancer be the explanation for epigenetic DNA methylation and limiting of telemere length?

          • I think that the explanation is that cancer prevention is not the reason for telomere shortening, especially since I read six studies just today that strongly and robustly link short telomeres to both cancer incidence and mortality

          • Hi John,
            Go back 1 billion years. Assume living things are getting bigger and more complex and cancer emerges as public enemy Number 1. So over next billion years anticancer defenses are developed by evolution. Any trait that decreases risk of cancer increases in population as more animals with that trait reach maturity and breed.
            Now consider term pleiotropy. That is single gene that has two or more apparently unrelated effects.
            Any trait that promotes aging will almost always to turn out be caused by a pleiotropic gene. The first initial effect is good and promotes life and is selected; but the late effect sucks and promotes aging.
            Telemere length is a pleiotrophic trait. Long lived animal like primitive humans buy short telemeres. Short telemeres are are hot. “Buy short telemeres and never worry about cancer.” Back in the day, most humans end up as cat food and never get old. Flash forward to 2017, everybody wants to live to 100. Now people start to say that short telemeres really suck; but you didn’t get cancer at 10 years old.

          • Hi Paul,

            Very interesting that you were able to backup your thoughts with 6 new studies. I hope you can prove me wrong about evolutionary trade-off cancer telomere length. Do you have any idea of what could be an evolutionary advantage of short telomeres if this is not cancer prevention?

          • Hi aldebaran,
            In evolutionary discussion about short telemeres to prevent against risk of cancer need to stress:
            ONLY TALKING ABOUT CHILDHOOD AND ADOLESCENT CANCERS. Evolution has no interest in cancers in old people.
            So looked at telemere length and childhood cancer.
            As regards neuroblastoma, a common deadly childhood cancer, there was increased risk in children and even greater risk in adolescence with long telemeres.
            Note that the children with “long” telemeres at increased risk are really very short telemeres compared to mice. So what would be the risk to children and adolescence of neuroblastoma if they had long mice length telemeres.
            Answer: we don’t know as they all died 100,000 years ago and never passed on their genes for long mouse-like telemeres.
            Also note: The evolutionary argument is an argument for older persons to want treatment to increase telemere length.
            The reasoning is, I got these crappy short telemeres to protect against childhood cancers like neuroblastoma; but now way past that stage, so now need longer telemeres.
            Also note that in old people regardless of actual risk of birth of cancer which is really unknown; Paul showed that old people with short telemeres at much greater risk of dying from cancer. My interpretation being, those with short telemeres can’t fight cancer as well as those with longer telemeres,

          • The problem here being that most children have long telomeres since for the most part telomere shortening is a matter of attrition over time, and also that neuroblastoma is more common, yet still quite rare , even in children. So in this case it’s like saying that children drink milk and children get neuroblastoma therefore milk must cause…..

          • Hi Alan,

            Thanks for your explanations. This make perfect sense to me. Short telomeres protect young people from cancer but increase mortality in old people and even probably cancer risks in old peolple. So this is a clear evolutionary trade-off.

            However, I was under the impression that Paul do not think that long telomeres increase cancer risk at any age, even in young (Paul please correct me if I am misinterpreting your thought). In that case, I don’t see the evolutionary trade-off cancer telomere length.

          • This is a very interesting study. Implies human cancer defences more than adequate in ancient times even for those old enough to get heart disease.

            Totally agree with Alan’s interpretation of young vs old cancer risk. Larger animals are born with shorter telomeres for the simple reason they have more cells.

            Still think we need more evidence and studies to tease apart the influence on cancer of overly long or overly short telomeres but I expect the mechanism is as follows:

            In the young: An oncogene is activated in a somatic cell without telomerase. The cell has until senescence to acquire the telomerase mutation. So the longer the telomeres the more chances it gets to do this.

            In the old: Conversely when telomeres get very short mutation rate rises considerably, so having lots of cells with short telomeres makes cancer much more likely. In additon short telomeres in immune cells make that line of defence weaker too.

          • Perhaps my perspective is simplistic, because I am not knowledgeable about all the biology, etc. But I just took it to mean the evolutionary trade off theory is wrong. My basic thinking: Cancerous is triggered by environmental factors that were rare before the Industrial Age. We wouldn’t evolve with a mechanism to defend against something that wasn’t a threat in the first place.

            The basic (simplistic?) idea: if there was no malaria, there would be no sickle cell mutation.

            Dr. Green, what is your evidence that cancer existed one billion years ago? Fossils?

          • Hi John,
            In response to last part of question, cancer and billion years.
            Philip Haycock noted that of 25,000 genomes, 15,000 were related to cancer and growth. This shows the extreme interconnection between growth and cancer. Evolutionary theories are critical for understanding cancer development and aging as most aging genes are really antagonistic pleiotropy genes.These are genes which are advantageous during youth but contribute to aging phenotypes. There is large group of fascinating papers about evolution and cancer. One such paper is “How cancer shapes evolution and how evolution shapes cancer. “(Casas-Selves)
            The paper which describes my own view is best described in article in “The Daily Galaxy”, Evolutionary Origins of Cancer– “A 1.6 billion year-old accident waiting to Happen” (Paul Davies)
            The idea is that cancer is like the child’s toy, Jack in the Box. Cancer is Jack and waiting to jump out any time unlock the box. The body keeps tight locks on Jack. (my analogy not Paul Davies). This is very different from standard idea that each cancer forms de novo from lots of mutations. Here the mutations are things that break the locks keeping Jack in the box.
            The idea was the first multicellular life was cancer. Better organized life evolved 600 million years ago at the dawn of the Cambian period from cancer. Cancer is an ancient program pre-loaded into genomes of all cells.
            These genes are need as embryos use these original cancer genes in first few weeks of life when ball of undifferentiated very rapidly growing cells. Cancer is like running evolutionary clock backward a billion years at very fast speed to Precambian multicellular life. That is why the cancer program is so tough; it is not something new that evolves overnight from a few random mutations.
            At any rate, looking at aging through the perspective of evolution and pleiotrophy genes is productive way to view aging.

          • Hi Alan
            A truly fascinating theory and it ties some things together. Cancer is there from the beginning but Jack is kept in the box due to a very healthy immune system. Over time mTor, a great example of antagonistic pleiotropy, deteriorates immune system functioning , and Jack is let loose, and ultimately takes over.

          • Yes Paul,
            It is amazing original theory; but suddenly things makes sense. Having looked at thousands of cancers, they all seem to follow the same script. They become more undifferentiated and finally they all look identical so eventually can’t tell if started as squamous type cancer or adenocarcinoma or even sarcoma and most of all, they don’t look like anything normal; but they do resemble extremely early embryonic tissue in first week. So big question is if everybody developing their own random mutations to make cancer, why do they all look exactly the same. So easy to believe this is a program that we all have for growth that predated life as we know it, precambian histology. Note the guys dreaming up this theory are from world of theoretical physics, kind of people who dream up stuff like black holes.

    • Short telomeres correlated with heart failure. Big surprise given how inactive cardiomiocytes are thought to be.

      Search for medicalxpress telomere length heart failure

      And perhaps less surprising, in muscular dystrophy:

      Search for medicalxpress shortened telomeres linked dysfunction duchenne

      • I think Mark that the telomere story is a long one and that we are only on chapter one. Until we delve deeper into the the details and carefully look at the percent distributions of very long, long, short, and very critically shorts in each participant of a well controlled study, we will continue to get conflicting and contradictory studies. But that is often how it is at the start.
        It’s been over thirty years since it was announced that high cholesterol causes heart disease and now, in many ways, it’s more confusing than ever, With HDL, LDL, dense and fluffy particles, endothelial function and stiffness, and much much more ( not to mention chronic inflammation).
        But it’s way too early in the game to draw any far reaching conclusions from any one study on telomeres in my opinion. Though it has made for a great and lively discussion.

      • Hi Mark,

        But couldn’t that be because Short telomeres correlate with cellular senescence which could be the real driver of heart failure?

        For example:
        “Markers of cellular senescence. Telomere shortening as a marker of cellular senescence”

        On the other hand when you activate telomerase, you are probably modifying the distribution of telomere length since it preferably acts on critically short ones, and this could actually reduce the proportion of cells that reach replicative senescence. So it seems to me that telomerase activation might indeed reduce heart failure risks.

        just my thought,

        • That’s a very good study that you referenced aldebaran. So I don’t see how we can reconcile the Haycock study with the Willeit study (Jama, June 2010) which showed a very clear association of short telomeres with both cancer incidence and mortality? Perhaps long ones are only protective when very long, and short ones risky only when very short.

          • I think the Haycock results are just referring to the small but real genetic disadvantages of having overly long telomeres on cancer incidence. It’s an evolutionary measure based on a very small increase in cancer, but enough to effect changes over time because it removed some people from the gene pool. That’s why humans have telomeres shorter than smaller animals. It is based on the length you are born with.

            All the health implications associated with aging we are discussing (including the separate age based rise in cancer incidence) are a different kettle of fish and based on the consequences of being a species with shorter telomeres that get shorter over time.

            A telomerase gene therapy is still a valid intervention against aging as it targets the shortest telomeres. You could make it even safer by preferentially targeting certain tissues that lose telomeres faster, or maybe even give people some extra cancer defence whilst you are at it. We’ll have to do that eventually anyway if we are to live 100s years!

          • Mark
            You may be right about the risk , but the Willeit study was a prospective , very well designed study following 1000 people over a ten and then a 15 year period and was published in JAMA 2010 Jul7;304:69-75. The results:
            “Short telomere length at baseline was associated with incident cancer independently of standard cancer risk factors . 95% confidence interval. Compared with participants in the longest telomere length group, the multivariable HR for incident cancer was 2.15 in the middle length group and 3.11 in the shortest length group. Incidence rates were 5.1 per 1000 person years in the longest telomere length group, 14.2 in the middle length group, and 22.5 in the shortest length group. The association equally applied to men and women and emerged as robust under a variety of circumstances. Furthermore , short telomere length was associated with cancer mortality and individual cancer subtypes with a high fatality rate.”
            It’s interesting that the Haycock extrapolation revealed the exact opposite. All of those cancers at risk from large telomeres were ones with high fatality rates.

          • Hi Paul,
            Key question is how to increase telemere length.
            In this regard looked at TA-65 which you mentioned. To me, TA-65 looks like very expensive junk. All studies done by people with conflict of interest in that have close financial connections to TA company.
            Judith Campesi, an independent expert, said something like money corrupts and dismissed research.
            There is also law suit to stop making unsupported claims. Now TA company appears to be just saying “health supplement” to avoid FDA scrutiny for false claims.
            Myself, highly suspicious of company that sets up company and makes claims without any independent evidence, aside from people who work for company.
            This is very treacherous area, perfect for false claims, due to lack of FDA involvement and reliability of FDA approved prescription drugs.
            I did look at TA-65 Maria Blasco paper and had 2 problems with paper, (1) No statement of absence conflict of interest, (2) Not convinced any mouse study applicable to humans as mice and humans very different as regards telemerase in tissues.

            I hate the taste of coffee; but found a supplement in pill form which is green coffee. Any opinion about green coffee supplement.

            Any other ideas about how to increase length telemeres aside from, “healthy life style”.

          • Hi Alan
            You make several good points as usual.
            Regarding TA 65 , it arrived with a big splash and miracles were expected of it, so I think it is suffering from a case of overly high expectations. I don’t take it myself.
            The green coffee bean extract supplement was being touted as weight loss miracle of sorts and didn’t pass muster. However, even though I’m skeptical of any supplement trying to replicate the benefits of a coffee bean which literally has over 100 polyphenols all interacting with each other, surprisingly this study, ” The potential effects of chlorogenic acid…….. Tajik N, 8 April 2017, European J of Nutrition”, did demonstrate improved glucose and lipid metabolism, and anti- inflammatory activity as well.
            As for your final point about those ever maddening telomeres, Elizabeth Blackburn won the Nobel Prize for her discovery of telomerase , and she recently wrote a very readible book on telomeres and their meaning. She quotes studies that show that:
            1.People who do moderate exercise for 45 mins , 3 times a week, have telomeres as long as marathon runners. They get even longer if you mix up the types of exercise that you do.
            2. Omega 3’s, whole foods, and vitamin D all lengthen them.
            3.Married with children , longer, but victims of domestic violence , shorter.
            4.She does claim that telomerse activity activation may lead to an increase in brain, lung and melanoma risk.
            So she strongly advocates for telomerase activation through life style, but not through “artificial ” means ( TA 65).
            Personally still not sure that that is logically coherent.

          • Hi Paul,
            Just ordered Elizabeth Blackburn book, 2017 and $20.
            I think her comment makes complete sense.
            She is saying the telemere length has U shaped curve, not straight curve for beneficial effect. She is looking for Goldilocks effect.
            She doesn’t recommend artificial means as nobody has any idea if their effect would be too small or too big.

          • I wonder Paul, could the cancer types be different? Also we’d need to compare the ages of the participants, was one study with older people and one over the whole range? We’ll get to the bottom of this eventually!

          • The ages were 40-79 years old. I quoted you the results of the 10 year study, they have since done a 15 year follow-up.
            “We extended the follow up from 10 to 15 years,increased the number of events, and added data from a second TL measurement because a single measurement may underestimate the true association.
            Incidence rates of cancer in the longest, middle , and shortest were 5.9, 16.9, and 22.8 respectively.
            The 15 year follow-up corroborates our previous findings that short telomeres are associated with cancer incidence and cancer mortality. An RDR of.59 and the srtonger associations for usual TL underscore the importance of telomere dynamics in carcinogenesis and the need for multiple measurements of TL in the characterization of individual cancer risk. The variability of TL over time is similar to previous evaluations of telomere dynamics and may reflect the cumulative effect of environmental and behavioral exposures, varying telomerase activity, and stress induced repopulation of peripheral blood by recently dividing hematopoietic bone marrow cells.”
            It’s maddening.

          • Hi Paul,
            That is a really great study, “Telemere length and risk incident cancer and cancer mortality.” 2010.
            As a pathologist, I had a totally different interpretation of their data. Note they followed 787 participants “free of cancer” for 10 years and then patients self reported cancer. 92 developed cancer.
            My interpretation: At start of study ALL 92 cancers were already present. The cancers were small and not clinically apparent.
            Over the next 10 years, the cancers grew and became diagnosed and reported.
            The participants with long telemeres handled their cancers. Many of their cancers never developed beyond very early stage and were never reported.
            The participants with short telemeres really sucked at keeping cancers under control. The had much greater chance to have cancer become clinically apparent and to die from their cancer.
            This is a study about progression of cancer and death from cancer and shows persons with short telemeres can not control and contain cancer. It is not a study about “birth” of a cancer.
            The participants came from town of Bruneck, South Tyrol, Italy. Maybe I would believe they were cancer free at start of study if had total body PET scan at start of study and then again at 1 year. If no cancer at 1 year PET scan, then cancer free at start of study.

          • Hi Alan
            As a pathologist you know more than I, But it stretches the imagination some to think that All 92 cancer cases were present at the beginning, does it not? I mean maybe a few were but still not enough to invalidate the results. But I did review the Singapore study results and it is pretty convincing that this is a U shaped curve with both very short and also the very long being most at risk, especially for the very nasty pancreatic cancer. I wonder if the link could be diabetes since we know that that group has a much higher pancreatic cancer risk and maybe metformin could be very helpful there.

          • Hi Paul,
            Based upon excellent study you cited, we both agree that older person with short telemerers has increased risk of death from cancer. To the patient who died from cancer, my guess is they probably don’t really care if you are correct that short telemeres increased their risk of getting cancer or I am correct that short telemeres resulted in weaker defense to control cancer.
            However, my general impression is that from day 1, (the birth of a single cancer cell) to death of the person is generally much longer than 10 years. I will leave this question in your court: After birth of a single cancer cell, what is chance that cancer cell will actually kill the patient in 10 years and what is chance that cancer cell will remain clinically silent and unknown and undetected without an autopsy or PET scan ?
            – Alan

            Alan –
            You’re undoubtedly right about “more than 10 years”. A corollary is that many cancers start up, and very few end up being symptomatic, a smaller percentage still being fatal. Studies of cadavers show that a high percentage have prostate cancers and thyroid cancers that never became a problem in the person’s lifetime. Why do most tumors lie dormant? Why do a few tumors become big problems? I think telomeres are only a small part of the answer.
            – Josh

          • Hi Alan
            Of course it all depends on cell type and the state of differentiation of those cells does it not? I think you would enjoy talking with a fellow pathologist , Dr. Shamsuddin at University of Maryland, a pioneer and expert on IP 6 – inositol , and he also developed a simple and inexpensive screening test for various cancers which are employed widely in China. It detects a sugar molecule , Gal- GalNac, released by cancer at its very earliest stages.
            His work on IP 6 is also impressive as it seems to actually convert moderate and poorly differentiated cells back to a well- differentiated state, functioning normally and without any actual killing of cells involved. He has a website and an excellent book on this.
            I suspect that would really enjoy his work.

          • Thanks Paul,

            Receive “The Telomere Effect, 2017, by Elizabeth Blackburn. Just started looking at it, but seems excellent. Thanks for tip.

          • Hi All

            I thought I’d chime in quickly with my opinion. Regarding the contradictory findings from Willeit et al, which we cite in our study, this is just 1 of 26 studies that evaluated the prospective association of telomere length with risk of overall or site-specific cancer (you can find the references in the discussion section of my paper). Of the 26 studies, only the Willeit study is difficult to square with our results. For example, the Copenhagen prospective observational studies, which are much larger than the study by Willeit et al (samples sizes ranging from 47,000 to 96,000), report results consistent with our findings:

            “…Short telomere length is…not [associated] with cancer risk”

            “…genetically determined short telomeres were associated with low cancer mortality…”

          • Personally, I am very impressed by the work accomplished by Haycock although I don’t understand statistics well enough to have a clear understanding of all its implications.

            The discussion initiated by Josh and contributed by others here is very helpful in that respect. This site is an incredible source of information!

            The evolutionary trade off to adjust telomere length in order to reduce cancer incidence at young age at the price of reducing maximal longevity (and maybe at the price of increasing cancer rate at old age) makes a lot of sense to me.

            So, I would tend to think that having long telomeres at birth would confer to people a higher probability of reaching an old age, but a higher probability also to die from cancer earlier in their life. However, I am not aware of any study to support this claim.

            I imagine that it might not be simple to correlate maximal longevity in humans with their telomere length at birth, since for example, I am not sure how we could measure the telomere length at birth of centenarians. However, I am wondering if we could not do that indirectly by looking at the distribution of SNP in centenarians knowing the relation between SNP and telomere length at birth, by using a similar approach than Haycock.

            However, I am probably missing something, since otherwise such a study would have already been made.

          • You may be right and as Mark said, one day we’ll get to the bottom of it. In the meantime all of this reminds me of an old telomere joke:

            Doctor: Well I have bad news and worse news. Which do you want first?
            Patient: In that case give me the bad news .
            Doctor: It seems that you have very long telomeres and only have 24 hours to live.
            Patient: OMG !!! What could be worse than that?!!
            Doctor: I meant to tell you yesterday.

          • Funny!

            I definitely hope I am wrong and that the mainstream view that telomere length is a double edge sword is incorrect.

            But my point is that we see many studies trying to find correlations with mortality rates and healthspan. This is very useful to guide us toward a healthy life style but this is of little relevance to understand the aging process.

            In contrast, it seems to me that there are very few studies trying to find correlations of something with maximal life span. However, these are the most informative studies to understand the aging process. So, I am wondering if this is because there is a lack of interest (or funding) for such studies or if it is because we don’t have a good methodology to find correlations with maximal life span in humans or long life species.

          • To my surprise, I was not able to find easily the maximal human life span from, say the cohort of humans born in 1900.

            The definition I want to use is the mean life span of the most long-lived 10% of a given cohort. Apparently, this is not a value typically published for humans (or it is published under another name).

            But surely, we can compute this data from the social security tables. For example, the age at which 90% of male american born in 1900 have died is about 80 years and a rough approximation of maximal life span of male american born in 1900 would be about 87 years.

            Now, is there a practical way to correlate the maximal life span in human with the telomere length at birth? I really don’t know.

          • If indeed people with short telomeres cannot fight as efficiently against cancer as people with long telomeres, it would be a very interesting discovery. However, it seems to me hat there is a confounding factor here.

            Paul showed us that people with healthy life style (who exercice, who have family, …) have longer telomeres than others. So couldn’t it be that that the reason why people with long telomeres can fight better cancer is because they have healthy life style?

            In order to find out it might be useful to differentiate people with genetically long telomeres (long telomeres at birth) from people with acquired long telomeres (from healthy life style).

            My understanding is that the Haycock study shows that people with genetically long telomeres have increased mortality rate from cancer. So this suggests to me that it might be rather the healthy life style that make people fight better with cancer (longer telomeres being just a consequence of life style). Does it make sense? Am I missing something?

          • You may well be right Aldebaran. The long telomeres may have nothing to do with the protection, but I wonder why it happens ? If the Singapore study that Alan alluded to is correct, then the risk is indeed U shaped, and it’s very interesting that the very short and the very long group are at risk for the same cancers, bladder, stomach, and leukemia, with pancreatic being the sole exception in the very long group. All of those cancers are relatively rare and are strongly associated with smoking. They are also very aggressive types.
            It could be, therefore, that if you have very long telomeres that you have a small absolute increase for several cancer types,ditto for very small telomeres, but this group is also at greater risk for all other diseases such as stroke and heart attack.
            It may be useful after all to get it measured so that you can be very vigilant to not smoke, keep your weight and blood sugars under control, exercise, etc., if you are at either extreme.
            And now that I think about it, cancer preventing measures such as rapamycin, IP 6, curcumin, metformin, and low dose naltrexone , should be highly considered also, since they are all cancer preventative.
            Screening measures as well.
            So maybe telomere testing to gauge risk isn’t a bad idea after all.

          • Hi Paul,

            Very interesting explanations about cancer prevention and telomere length. Always great to have your perspective on these complicated subjects.

          • The Singapore study is truly driving me crazy, and I can’t reference it because it hasn’t yet been published, but it was presented at an Oncology meeting last April. Can Anyone tell me how it’s possible for very short telomeres and very long ones to cause virtually the Same cancers ( bladder, stomach, and blood) at the almost the same incidences?? And all of those cancers are relatively rare anyway.
            Of course they’re all just relative risks, so the absolute risks remain very very low.
            I am going to send the lead author Wang an email and ask him for detailed telomere distributions for both of the extreme groups before I go absolutely insane thinking about this.It’s likely that either very long telomeres or the very short ones are giving these results, but not both.
            About low dose naltrexone ( LDN), it’s absolutely an amazing drug for being so poorly known and utilized. At high doses, > 25mg’s, the drug floods the opioid receptors and is good for heroin addicts, but not much else. At very low doses ( 1 to 4 mg’s) it’s a totally different animal and only blocks the receptor sites for several hours, leading to an outpouring of natural endorphins. The natural endorphins not only elevate mood, but are the most potent immune-modulators in the body. So if your immune system is weak, it makes it stronger, but it will weaken it some if it’s overactive. This makes it very effective for not only depression but for virtually ALL auto-immune diseases, of which there are many. See
            Furthermore, LDN very strongly inhibits cancer cell proliferation. So when the receptors are transiently blocked, the body responds by producing opioid growth factor, which leads to more opioid receptors being created on cancer cells, the endogenous endorphins attach to these new receptors and stop cancer in it’s track. So it can be used to both treat and prevent cancers of virtually all types.
            And there’s more. LDN also blocks Toll receptors which are responsible for releasing many pro inflammatory cytokines, So that LDN is also very anti- inflammatory as well. But it has none of these effects at high dose and actually is tumor promoting. The only side effect is weight loss since it inhibits hypothalamic related cravings. My partner, a Harvard trained internist , is so excited he actually smiled the other day, a first for him ( just kidding Daniel).

          • Paul,

            You have mentioned several times the benefits of low dose naltrexone and I was wondering if you could recommend a site where we could learn more about that?


          • Hi Paul,

            Thanks for providing information and link for LDN. This is very interesting and intriguing! So many different diseases potentially impacted (cancer, Crohn, MS, depression).

            It would be good to see more trials (I have found only one promising trial for Crohn in 2007). Problem is no company is probably interested since already FDA approved for large doses and so it would be hard to make money out of it (same story as with weekly rapamycin).

            Hopefully, the NIH could fund some trials. Are you yourself involved in running a LDN trial? Looks very promising.

          • We are running a pilot trial which is prospective but based on patient reporting, so observational. We considered a placebo control group, but we’re so certain of its efficacy that it didn’t seem to us to be ethical. I will be attending the LDN symposium on Sept. 22-24

          • What I’d really like to know is the dose and frequency for cancer prevention, but no one knows. We need Alan to figure it out like he did for rapamycin.

          • Hi Paul,

            Looking forward to get updated after the completion of your LDN study. Understand perfectly the ethical concern if it works evidently well but I imagine it is harder to be credible without a placebo group? Very hard dilemma I can imagine. Good luck!

        • Hi Mark,
          I looked at references in this Russian paper you are quoting as gospel regarding telemere shortening and senescence. References I checked are 1961, 1971, 1973, 1987, 1989, 2004 in references I checked regarding these claims.
          In my opinion anti-aging theory regarding cause of senescent cells starts in 2006, with Blagosklonny first paper about role of mTOR in aging and production of senescent cells. Not buying any references from 1961. Not buying any theory that doesn’t include mTOR’s role in aging.

          • I think someone else posted that paper Alan, but I know what you mean – senescence had been studied prior to Blagosklonny, but until it was tied to MTOR it’s fundamental link to aging was not yet established. We can now see a hazy outline of an aging process involving MTOR, mtROS, epigenetic aging, telomeres and inflammation starting to come into focus thanks to such pioneers.

          • Hi Mark,
            Thanks Mark, Sorry if my comment about Russian paper sounded rude.
            You actually explained in your note exactly what I meant.

  24. Mark,

    up until the end of 2016 there were only 3 ways to reset the epigenetic clock:
    1. conception
    2. somatic nuclear transfer (a.k.a. “cloning”)
    3. IPSC generation
    all three methods also reset the differentiation state and the age state of the cell. So the result was always the entry point of a whole new organism.
    Because epigenetic modifications of the DNA not only determine age but cell type as well.

    At the end of 2016 there was a successful experiment of in vivo resetting the epigenetic age state only. Dedifferentiation did not happen. Josh wrote about it extensively:

  25. Hi Mark,

    on the epigenetic clock in postmitotic cells:
    according to Figure 2 in the original Horvath paper
    the aging clock works in heart and brain tissue too.

    The in vitro experiments are very interesting yet there is not much interest in understanding the accelerated DNAm aging.
    I am startled that although IPSC reprogramming has been known for 10 years, there is still not much knowneledge about how IPSC reprogramming actually work. Belomnte and others just recently demonstrated that rejuvenation comes first followed by dedifferentiation. These are so easy repeatable and relatively experiments, yet no one seems to go for them.

    • Okay, well maybe the sheer ubiquity and reliability of epigenetic aging markers is simply a spread of stochastic events leading to methylation or demethylation of DNA. It could just be the chemical reactions going on as part of normal cellular operations are effecting these changes over time. Resetting telomeres is resetting enough gene expression to continue replication. Any problems caused by the un-reset remainder only show up in post mitotic tissues because they are not cycling. This explains the in vitro results with the immortalised cells.

      I agree some.basic work with ISPCs is needed to elucidate these fundamental mechanisms.

      • It is not known what drives the epigentic clock, but it has profound effects on the cell. There are two processes ongoing with age:
        demethylation of methylated (locked) chromatin. This maybe the result of oxidative damage for example. Note that the cell can repair all kind of oxidative damage except for DNA methylation because there exists no template.
        Demethylation allows transcription of DNA that the cell didnt want. For example transposable elements are set free. These are basically genome integrated rietroviruses. You dont want them transcribed and then be inserted randomly into your DNA. I think demethylation of TEs can greatly contribute to cancer burden related to age.
        The other process is more directed. That involves methylation (locking) of sequences which were beforehand accessible. What I read about t this process is that it preferentially methylates bivalent chromatin, that is DNA which is conditionally locked. DNA methylation makes them permanently locked. PRC2 binding domains are locked this way. Those domains are responsible for the development program and stem cell plasticity.
        So the epigenetic clock has an explicit negative effect on the health of your cells.
        For me there is no doubt: DNAm changes = development program = aging.
        Walks like a duck, quacks like a duck swims like a duck.

        • Thanks for that GaborB, it’s very interesting. I shall certainly read up more into this process.

          For me though it all links back to MTOR and mtROS. The lower those two are the slower the DNAm advances. We know that based on the paper Alan posted up thread about dwarf mice, CR treatment and rapa doses in mice affecting the rate of epigenetic aging. Josh’s site won’t let me post the link right now.

          Also a paper posted by me previously in one of Josh’s mitos articles, on how MTOR rate and mtROS together account for most of the difference in max lifespan in mammals

          Again, can’t post it, search for pubmed 18442324

          All other things like DNA instability are downstream of the damage done by basic metabolism leading to an inability by cells to express their DNA properly IMO. I could be wrong but the sheer consistency of epigenetic aging across diverse tissues leads me to this tentative conclusion.

          • hi Mark. if it was only up to basic metabolism, all life had been extinct after a few hundred doublings. in the izipusa paper they have shown that that ROS production fell back after epigenetic rejuvemation.

          • I am just saying a plausible mechanism is cell metabolism to epigenetic changes to gene expression changes to aging and death.

            I accept your argument that resetting the epigenetic changes could reverse aging. Life in the past has done this via meiosis, that is until Yamananka and Belmonte came along. It will be interesting to see if this can be made to work safely in humans, though given the cancer concerns currently holding up the relatively simple HTERT gene therapy, I am not holding my breath.

          • hi Mark, I think hTERT therapy is misguided because telomere restoration is downstream from epigenetic rejuvenation.
            looking at the success of CAR-T therapy fasltly approaching mainstream makes me optimistic that in vitro rejuvenated autologous stem cell therapy might be the way to go. though it wont be cheap though.
            2 things are needed:
            Izipusa has to show that the normal aged mice treated with OSKM can achieve actual life span extension
            successful stem cell engraftment into various tissues in vivo has to be proven

          • That’s not entirely true. Telomere restoration resets some important epigenetic markers (read up on the telomere position effect over long distances), and the paper you posted proves that this allows continued proliferation. So yes it’s not complete restoration but we know from the work in mice done by Blasco and Jesus etc., that it has a youthening, life extending effect.

            The more complete epigenetic restoration you are proposing is undoubtedly worth pursuing, but I see that as further off than an almost ready made HTERT therapy.

  26. Josh, all,
    Glutathione levels help maintain telomere length during replication iirc. But I am concerned about possible lower glutathione levels as a vegan due to less sulpher-containing amino acids. Is this the reason for NAC supplementation- to maintain glutathione and thus help maintain telomeres?

      • Mark,
        I know vegans that avoid grains will have lower cysteine and methionine- and that cysteine (and glycine) are 2 of the most important amino acids for the body to make glutathione. I have also read that glutathione levels being high helps maintain telomere length during replicating- it was from an Indian paper called “nutritional destiny and lifestyles influence telomeric bio molecules”. I know that Josh takes NAC and was wondering if the reason was 2-fold; for glutathione AND telomeres.

    • Thanks Ole,
      This is great news to me as less than 1 week ago I have finally switched to a vegan (without grains) diet similar to Josh (plus some D3, B-complex, fish oil, etc.).
      I posted above about spermidine but realized I can get plenty of spermidine by eating green peas, corn, chick peas, and even broccoli and cauliflower… I am so personally grateful for this blog and all of the people posting here. Now eating lots of vegetables, nuts and some fruit.

  27. Thanks Phillip
    I’m left with a couple of thoughts here.
    1. It would be nice to know not just the mean telomere lengths, but also the distributions in terms of percent of very shorts, very longs , etc. Those details could be revealing and helpful.
    2. The Singapore study suggests a U curve of risk , and your study also allows for such a possibility, where both very short and very long telomere lengths led to cancer risk, and to the identical rare cancers ( stomach, bladder, pancreas, blood). Again it would be nice to know the specific distributions i.e. , does the very short group also have a high percentage of very longs, and conversely does the very long group have a high percentage of very short telomeres.
    3. Finally, it may be clinically relevant for people to have telomere testing so that they can be extra vigilant to do all of the cancer preventive measures possible if they fall into one of those groups.
    Of course they also shouldn’t panic. The Singapore study showed a 200% increased risk of pancreatic cancer, But if we use the poisson distribution square root method of rare events then the absolute risk of 12/ 100,000 is + or – (3), and that would increase to 24/100,000 + or – (5), which only may take us from a risk of 15 to 19/100,000. Don’t jump off the bridge!

    • I just wonder Paul what the mechanism could be for the same cancers being more likely with either very short or very long telomeres.

      Having done some more reading on the subject the only way I know of for cancer cells to activate telomerase is to get short telomeres first, because this primes the TERT gene that would otherwise be inhibited by the close proximity of the looping telomere itself. So how could long telomeres cause this? I don’t see how. Can very long telomeres cause genomic instability in some other way? Something to research perhaps…

      It could be longer telomeres mean you keep mitotic cells for longer so they have longer to accumulate other oncogenic mutations, so when they finally get to short telomere stage and potentially activate telomerase they are more likely to be serious. But this explanation only works if oncogenes make it more likely to activate the TERT gene. Otherwise once the cancer has telomerase it can accumulate all the mutations it needs anyway. My best guess for a mutation that makes TERT activation more likely is something like p21, which is known to inhibit telomerase.

      Or it could be an indirect effect, like having very long telomeres might also mean you have short ones too, as you have suggested (but why would this be?). Or it could be an effect of the TERT protein that has nothing to do with telomeres, like a growth factor effect leading to more cancer? But then that should lead to other diseases of aging too.

      It’s an interesting mystery and a great deal of credit will go to he or she who solves it!

      • I couldn’t agree more with the entirety of your comment. Consider this example from my point of view now as a clinical physician. Let’s use my father as the example. He had his telomeres checked at HMS at 92 years old and had a very!! long mean telomere length with very long ones as well as part of the distribution. So if he were a patient I would have to advise him ( Haycock) that he has a significantly increased relative risk of cancer. BUT, he also had a very very low percentage of very short ones so now I would advise him ( Wlleit study) that his cancer risk is quite low.
        But in steps Wang ( Singapore ) study which shows a U-shaped risk, so I now inform him that his risk is high AND low. You see the practical dilemma here I’m sure.
        This is why the individual distributions are essential, but not performed, in these large population studies.

      • Let’s suppose that an individual has a large percent of very short telomeres in what is basically their immune system, since that’s what we keep measuring, and this leads to the last stage of malignant transformations. In response to this, the surrounding cells cells activate telomerase, which elongates the long telomeres already present to very long ones and also elongates the very short ones, all of this in an attempt to jack up immune defense and subsequent elimination of the cancer cells. Depending on the “snapshot” measurement , this person could be found to have very short telomeres, or very long ones , and both would appear to have a cancer association.

        • I agree that for useful information we need a complete distribution of telomere lengths, short, average and long. We could define short and long both absolutely (below or above a certain kb) and relatively (compared to age matched controls and also to your own average length telomeres). These measurements would need to be repeated several times over some set period of time (not just a snapshot as you say) to take account of variations due to replacement of somatic cells (more on this later).

          I would expect very long telomeres to correspond to either somatic cells that have just been replaced, or stem cells (which we would expect to form a small and fixed percentage). A larger than average number of long telomeres would imply a regular turnover of somatic cells and a healthy stem cell pool – but we need to use caution here because these readings could mislead if you are just recovering from some kind of stress or illness that lead to a larger than usual replacement of somatic cells. That kind of event would cause a depression on telomere lengths that would then recover, possibly overshooting your normal length. So that is why we need repeated measurements over time.

          Very short telomeres would correspond to cells approaching senescence (either through replicative senescence or damage). A larger number of these would mean some stressor is outpacing the ability of the stem cells to supply replacement somatic cells in good time.

          If we take the example of your father Paul, and took telomere lengths of his immune system, we would expect longer telomeres than average because his high level of physical activity will force a higher replacement rate and therefore a longer average telomere length and smaller number of short telomeres (so long as he wasn’t still recovering from a hard session or illness for example, when you would expect a temporary dip in telomere lengths).

          The mechanism for telomere length I’ve been discussing here is simply down to the replacement rate of the cells being measured (in this case the immune system). The Haycock work is talking about the length you have genetically and that you will maintain, all other things being equal. There is evidence that longer is not always better here. But I don’t think an active person could ever accidentally put themselves in this range if they weren’t already in it, as forcing their cells to replace faster would just restore their telomeres to lengths they were born with (at best, in practice it will be less as stem cells can’t fully compensate for their own telomere loss), not magically give them ever longer telomeres than they are genetically predisposed to have. The same would apply to some sort of pharmacological or gene telomerase therapy, because it is the shortest telomeres that get boosted, not the longest.

          The Singapore work I can’t comment on because I haven’t seen it, but what I have described is certainly consistent with a U shaped benefit of telomere lengths.

          I understand your difficulty in advising patients Paul. What if you measure the telomeres of someone who have much longer telomeres than average; how do you advise them? Do you recommend shorter telomeres? I think the answer is no because there is not a lot we can do about genetically long telomeres, the risks are small and age related telomere loss will attenuate this anyway. But short telomeres are also a cancer risk as well and this rises with age, so we need to make sure the number of short telomeres is kept to an absolute minimum. That is what I would advise your patients.

          So concluding this overly long post: ignore long telomeres, concentrate on reducing the number of short telomeres.

  28. I had a bit of a brainwave about how we might resolve this problem of long telomeres causing cancer.

    Correlation is not causation.

    Could it be that one or more of the SNPs surveyed by Haycock in their GWAS is the same or near to one of the mutations used by cancer to activate telomerase activity?

    If this is true then it would mean long telomeres achieved from lifestyle would not necessarily lead to the same risks as long telomeres caused by the culprit SNPs.

    It might also explain why some cancers are affected more than others, because cancers in different tissues use different pathways to achieve immortality; for example cancers in less proliferative tissues tend to have mutations in the HTERT promoter regions, but cancers in the fastest growing tissues are not thought to need to do this.

  29. Look at animals with negligible senescence: albatross, Blanding’s turtle. IIRC they maintain telomere length, yet they don’t have high cancer rates. Bowhead Whales do use somatic cell telomere shortening ( I know that for a fact, I did the TRFs, see link to old paper below), but they must lengthen telomeres in the stem cells.

    Anyone know results from Greenland Sharks? Quahogs? (I actually did some low-temp cell culture, but it’s probably not a common thing… I doubt that Carolina supplies quahog serum). BTW, don’t get too optimistic about “telomere measurements”… whether you measure with TRF or q-PCR, you aren’t measuring the lengths in the cells that died.

    Oh, and here’s a first… I’m going to go off-topic AWAY from telomerase 😉

    What’s the latest on nicotinamide riboside? Can we really trick cells just by forcing NAD+ higher? Li Ka-Shing obviously thinks so, but he’s just a billionaire, what does he know?

    • Hi Bill,
      An anti-aging drug does 3 things, (1) increases life span in very many animals including mammals, (2) prevents and ameliorates a large number of age-related diseases, (3) prevents accumulation of senescent cells.
      Forming a company and hyping product on internet doesn’t make it an anti-aging drug.
      Any study done by persons with financial interest with the company is worthless.
      Comments by billionaires are generally worthless as regards truthfulness.
      Cell-culture studies show ZERO about how works in real animals. A good scientist can show just about anything in cell culture experiments.

      Thanks for a large number of really great post.

      • Thanks Alan. I would be happy with a drug that did any one of the things you propose 😉

        As a cell culture guy, I would say that cell culture can tell you a few things… if you use normal cell feeder layers in your cancer experiments, physiological levels of O2 and nutrients, proper screening for mycoplasma, …. (any research guys are laughing now, 99.99% of cell culture experiments are with aneuploid cells, 21% O2, high sugar, etc. etc. etc. 😉

        There have been papers suggesting a life extension benefit (in rodents, at least) from NAD+ increase for a long time. There are two human Phase 2 trials in progress, which will report soon and give us some data on effects on markers. What I was hoping for here was some theory ideas… how would forcing the NAD+ level up affect the aging clocks?

        • Are you referring to this study regarding NAD Bill?
          “NAMPT-Mediated NAD biosynthesis as the Internal Timing Mechanism”
          B Poljsak. Rejuvenation Res 2017 Sept 08
          Are you also familiar with the intracellular role of NAD in regards to PCSK9 inhibition and prevention of cardiovascular disease?

          • Hi Paul,
            I looked at Nampt study you quote by Poljsak. I have absolutely no idea what that has to do with NR.
            Let me start by stipulating that NAD and NADH are two of most important molecules in living things, but what is the connection with NR.
            I would think NR pills are inactivated in digestive system and never get into blood stream. If get into blood stream, get inactivated in first pass through the liver. I would expect than NR has zero biological activity. I would expect that NR never enters cell and never changes NAD/NADH metabolism.

            Unless somebody proves NR is doing something in vivo after taken by mouth etc etc etc, then NR is just junk.

          • Hi Paul,

            My Bad.
            Found human trials with Niagen in which showed raised NAD+ levels in humans. So that changes everything.
            Niagen now looks like a the real deal as has bioavailability.

          • Hi Paul, Alan, Mark,

            I am a bit on the sideline about NR. I have read that it does raise the NAD+/NADH ratio but some people think it is just transient and after a while the ratio returns to normal. Any thought about that?

            Also, I am not aware on any study that shows normal mice have increase life span with NR.

            Do you think short term study on humans showing increased NAD+ is enough to recommend Niagen?

          • Hi Aldebaran,

            I just did a complete 180 degree turn on NR and ordered NR on line from amazon. $35 for 60, 250 mg.
            Maybe somebody else can give opinion if need 250 or 500mg.

            I was very impressed with studies showing raised NADH in humans.

            I always thought very important to raise NADH, I just didn’t believe NR really achieved these results until saw results from human trials.

          • The ratio is all important . It’s crazy how much NAD levels drop over a lifespan. In males from 8 ng to 1ng by 77 years old, females go from 4 to 1.5. I like NR because it goes to NMN and then straight to NAD. I don’t like nicotinamide because there is evidence that it’s accumulation is aging. I’ve been experimenting with NR for over the past year and I find that I feel best taking it once or twice a week max.

          • Hi Alan,

            Thanks for your inputs. I might do the same as you. Ideally, I would like to see life extension on normal mice before trying but the only experiment I am aware of is at the ITP and will take a while to complete. So maybe does not hurt to start now.

            The brand I would tend to prefer is the one from Chromadex (TRU Niagen: 135$ for 3 months) because I trust the quality. Maybe not that important though.

          • No Paul, I hadn’t seen the study in Rejuvenation Res, I was referring to the mouse study in Science.

            One of the Phase 2 studies has already “reported”, but only to a closed conference. I’m hoping to see some data on cardio markers this year.

            There’s this paper in Trends in Mol Med, but it’s paywalled (thanks for nothing, NIH system):


          • I’ve been taking Jarrow Formulas’ Niagen since March. I felt this press release from ChromaDex itself means it is the real thing:

            Jarrow Formulas Plans US Retail Launch of a Nicotinamide Riboside Product Featuring ChromaDex’s NIAGEN®

            But I might switch to Tru Niagen, because the dosage per capsule might be more convenient (250 mg. per capsule instead of 100 mg. per tablet.)

        • Speaking of Nicotinamide Riboside (Niagen), I watched this video the other day, an interview with the researcher who discovered it:

          Charles Brenner, PhD discusses Nicotinamide Riboside

          I think it’s been proven that Nicotinamide Riboside raises NAD+ levels. But whether raising NAD+ levels does anything beneficial in humans is not proven yet. It’s still a surrogate marker, I take it. So I hope the studies continue. I hope the folks selling the supplements don’t get spoiled by sales.

          • The NAD+/NADH ratio is important in mitochondria Alan, and more NAD+ causes mitos to fission into smaller pieces and undergo mitophagy. So this is definitely beneficial for humans. How much remains to be seen.

            I believe NAD+ levels also affect DNA repair, which might also be important.

            Incidentally I am not convinced you need niagen (NR). I think B3 plus D-ribose is probably better as I suspect this is what the body breaks niagen down into first anyway. I also wouldn’t recommend constant mitophagy. More like something you’d want to do every now and then.

          • Yes the pathway may be better for NR, but I am saying in the body NR becomes N+R, which is then used for raising NAD+. So if this is true, no advantage to taking expensive NR.

    • “Supplement suppliers (Elysium) recommend taking it daily”.

      I’m sure they do 😉

      Seriously, Elysium isn’t a very serious organization… they haven’t even paid Chromadex for supplying their NR! The companies are still fighting in court, IIRC. The only semi-trustworthy person involved in NR is Li Ka-Shing…. and he’s not a biologist, he’s an 89-year-old billionaire who (IMO, don’t sue me bro) is using NR and wants some more guinea pigs to join him. (Bill Andrews needs to get Li on board with telomerase activation).

      I guess the NAD+ results show that NR is affecting the cells… but what the long-term effects on the self-destruct programs are is the question.

  30. Oh, and there have been plenty of good studies by people with financial interests. The trick is whether other people can replicate them 😉

  31. This is interesting from Maegawa, Nature Communications 14 Sep 2017 article # 539. ” Caloric Restriction delays age-related methylation drift.”
    ” Previous research has shown that DNA methylation tends to drift with age. However it was not previously known whether there was a connection between this drift and lifespan. Team studied blood samples from mice, monkeys, and humans at different ages. Mice, few months to 3 years, monkeys , few months to 30 years, and humans between zero to 86 yrs. The analysis revealed gains and losses of DNA methylation at certain locations in the genome. The more methylated a genomic site was the less the genes were expressed. Further DNA analyses revealed an inverse correlation between methylation drift and lifespan. The more and the quicker epigenetic change occurred, the shorter the lifespan of each species.
    They then restricted calorie intake by 40% in mice at 3.4 months old and 30% in monkeys between 7-14 years old. They did this over a long period of time. In both species the effects were dramatic. Monkey’s blood methylation age seemed 7 years younger than their chronological age. They propose that epigenetic drift is a determinant of lifespan in mammals.
    The impacts of calorie resriction on lifespan have been known for decades, but thanks to modern techniques, we are able to show for the first time a striking slowing down of epigenetic drift as lifespan increases.” ” The findings may have major implications for age-related diseases.”

    • Hi Paul,

      Thanks. Very interesting study!

      Interestingly, they also noted the accelerated drift with chronic inflammation:

      “Together with previous findings showing that chronic inflammation (which shortens lifespan) accelerates methylation drift, our data suggest that epigenetic drift is an excellent biomarker of lifespan”.

        • Hi Paul,

          Yes it seems so! Indeed you’ve just reminded us yesterday how important it is to control chronic inflammation so when I saw the link between chronic inflammation and methylation drift in the paper, it just popped up in my mind.

          On the other hand, two compounds I am aware of that should help chronic inflammation (curcumin and fish oil) failed to increase mice life span (ITP:fish oil, and Spindler:curcumin).

          This suggests to me that we don’t have yet a magic bullet to control chronic inflammation (such as we have to control mTOR with rapamycin).

          Another possibility is that controlling inflammation only affects health span, not really the aging rate but this is not what the epigenetic drift is suggesting.

        • Hi Paul,
          This is comment on your long-running discussion that the important thing is not average length telomeres; but very short telomeres.
          I see telomere length, like everything else we measure, as having a standard distribution.
          To me, telomeres are either long enough to do their job or too short to do their job.
          The only telomere length that I think matters, is telomeres too short to do their job.
          According to Blagosklonny theory of aging, a key aspect is production of senescent cells. The theory is mTOR drives resting cells to grow and divide, if the cell gets stopped in division because problem with chromosomes, then cell becomes senescent cell, which acts as major troublemaker.
          The number one chromosome problem stopping cell division being allowed to proceed, I suspect is telomeres which are too short.
          In this way the very short telomeres can be acting directly with elevated mTOR in producing senescent cells and not producing more healthy cells.
          In regard to short telomeres and cancer; one specific cancer noted by Elizabeth Blackburn in her book that was increased with short telomeres was skin cancer (not melanoma). Note that squamous cell skin cancer was greatly increased in transplant patients who had impaired immune system function. So increased skin cancer in patients with short telomeres could reflect impaired immune function rather that actual production of more squamous cell cancer at chromosome-mutation level.

      • Hi Aldebaran,
        This discussion of chronic inflammation, epigenetic drift and aging is remarkable confirmation of Blagosklonny hyperfunction theory of aging. Chronic inflammation is hundreds of different specific actions, the sum total of which looks like chronic inflammation and this is a general manifestation of hyperfunction. Elevated mTOR is driving hyperfunction; methylation correlates with whatever ” function” and “hyperfunction” is on a molecular basis, and this becomes a biomarker of lifespan. All pointing to need to use rapamycin to reduce mTOR to reduce chronic inflammation.

        • Hi Alan,

          Very interesting comments about hyperfunction and mTOR! I am reading Blagosklonny paper 2006 that you’ve recommended and learning a lot.

          So mTOR reduction is certainly one way to control chronic inflammation (Rapamycin, CR) but it does not control all manifestations of inflammation right? For example Paul’s father still suffered chronic inflammation in late life despite aggressive CR. So this suggest to me that mTOR is not the only target to reduce chronic inflammation. Do you agree?

          • Hi Aldebaron.
            I see if differently.Paul’s father is extremely interesting case that Paul shared with us.
            In my opinion, at around 60, Paul’s father began suffering from elevated mTOR and the manifestation was onset of mild depression. He then discovered that a life style including very marked CR ameliorated depression and he felt good. This life style both reduced mTOR and helped maintain long telomeres. Then sometime in his 80s he developed an occult cancer. The occult cancer caused a thing called paraneoplastic syndrome which changed his immune system and he developed severe psoriasis, an immune disorder of unknown specific etiology. At this point his life dramatically changes, he had severe distress and his life style probably changed and could no longer keep mTOR low. Elevated mTOR was reflected in new onset of severe depression. Many of the bad things happening in his 90s could be reflecting a major increase in mTOR level. Note that with severe psoriasis and a developing cancer, his body would be under great stress, all of which would increase mTOR and shorten telomeres.

          • Hi Alan,

            Greatly appreciated comments as always! So in your opinion mTOR inhibition is the most efficient way to fight chronic inflammation.

            I have a somewhat related question. A friend of mine suffers from chronic calcific tendonitis. Do you think weekly rapamycin could be of any help?


          • Hi Alan,

            From your comments, it looks like that you think CR is not always enough to control TOR. Interestingly, Blagosklonny seems to think that as well. In his 2017 paper he says:

            “The effects of rapamycin and calorie-restriction are not identical and may be additive [248, 249]”.

            So it could make sense to combine moderate CR with weekly rapamycin (and metformin) to get larger effect.

          • Hi Aldebaran,
            The world’s oldest living man is Israel Kristal, age 113.
            In August 1944, at age 40, he was sent to Auschwitz concentration camp. He was rescued by the Russians in January 1945. He weighed 80 pounds. He survived starvation and hard labor.
            He did not need any rapamycin to get the maximum anti-mTOR effect.

            I prefer rapamycin combined with a smidge of caloric restriction.

          • Hi Alan,

            Got your point! Couldn’t agree more. Interesting example of Israel Kristal. He must have an incredible resistance to stress to have survived such conditions

  32. Hi Paul.
    Comment in regard to Blackburn and Hayflick limit:
    In 1961 Leonard Hayflick demonstrated in CELL CULTURE that human fetal cells will divide between 40-60 times and then become senescent. In 1974 Burnet coined name “Hayflick limit”.
    On Christmas day 1984, Blackburn showed that telomeres would grow in size. She had discovered telomerase. Life requires getting around the Hayflick limit.
    Since then, for the last 30 years, anybody talking about the Hayflick limit has no idea what they are talking about.
    Life is all about telomerase. The 1961 demonstration by Hayflick was a demonstration that cell culture is not life.
    It is true that telomeres shorten with age and it is true that short telomeres predict mortality.
    Length of telomeres is all about telomerase. Blackburn states typical trajectory of life of human telomers is 10,000 Newborn, 7,500 35 year old and 4,800 at 65.
    But you state your father had @ 9,000 base pairs near age 90.
    Regarding telomere length Blackburn states:
    “The wonderful news is that our research,…has shown you can step in and take control of how short or long–how robust they are.”

    On page 65 Blackburn states, “In the rest of book will hear us talking about how you can increase telomerase and protect your telomeres.”

    The Hayflick limit would make life impossible. Nothing can live with 40-60 divisions for stem cells. So the Hayflick limit is a totally bogus concept.
    All that matters is having enough telomerase.
    Fig 10 on page 51 shows very clearly that telomere length predicts: all cause mortality, other causes, cardiovascular disease and even cancer.

    So it is true that telemere length is of extreme importance and short telemere cells can’t divide and become senescent if try to divide; but this has nothing to do with Hayflick limit which is purely a cell culture mistake about how life works.

    • Hi Alan

      The way I see it we have this complicated system of…. shortening telomeres – senescence – clearance – stem cell replacement, because just having enough telomerase in every cell would lead to too much cancer. But unfortunately we know a lot can go wrong in our complicated system: senescent cells can accumulate faster than they are cleared, stem cells can go senescent too or become locked by high inflammation, and finally stem cells too can run out of telomerase. Only germline cells have enough to survive indefinitely.

      Life isn’t impossible with a limited number of divisions of stem cells because the somatic cells they supply have to use all of their divisions before a stem cell even has to do 1 division, whereby 1 copy remains a stem cell and the other becomes a somatic cell with its own 40-60 divisions.

      Note that low MTOR also means less division so telomeres last longer, so less senescent cells at any given time waiting for clearance and replacement. This also means less senescence in non dividing cells. This means less inflammation and ROS so less telomere attrition that isn’t due to replication.

      This all means we can get by for potentially over a 100 years with limited telomerase, provided nothing else kills us first.

      • Hi Mark,
        My point when came up with Hayflick limit 1970s nobody knew telomerase existed, so nobody knew any cells including stem cells or germline stem cells had telomerase.
        Quick math check: 2 x 40 power is 1 trillion.
        Number cells in human body 37 trillion.
        Therefore, life not possible without at least some cells having telomerase. Thank you Elizabeth Blackburn for making life possible.

        • I’ve done some calculations and I think you are right Alan. In a very, very optimistic scenario the small intestine lining, which turns over in about 3 days would run out of stem cells in 13 years.

          So either stem cells can elongate their telomeres, or somatic cells can become stem cells in some scenarios (we know this happens in a bad way in cardiovascular disease when inflamed epithelium cells become osteoblasts in calcify arteries).

  33. A couple things regarding Nicotinamide Riboside N(R) to note.
    First of all I’m an engineer that didn’t take biology even in high school however;
    1-NAD+ levels seem very correlated with circadian rythm so it seems that taking N(R) earlier in the day is more in tune with this that later (see paper “Circadian Control of the NAD+ Salvage Pathway by CLOCK-SIRT1” Nakahata, Sahar,
    2-Regarding dosages question above I’ve seen one regimen posted by Aubrey de Grey’s research assistant, Michael Rae (for whom I follow and have much respect for). He states in his personal regimen that he takes 250mg N(R) at 6:30am w breakfast and 125mg and 12:30 AND as noted above he does not take EVERY day rather 6 days per week and takes Friday ‘off’
    His Regimen on Longecity:

    • hi Clinton
      I have self experimented with all doses of the stuff and I find that dose impossible to tolerate. Anxiety and irritability.
      That could be just me though.

      • Paul,
        Thank you for that information – I’m curious what dose you use now if any?
        I was planning to go with just 125mg with breakfast 6days/week.

        • Clinton
          I’ve settled on a low dose of 125mg’s twice a week on average. I’m more energized at this dose without the stimulant effect. I also find that I feel really good when I combine it with 200mg’s of pine bark extract ( which by the way has had the most profound effect on the quality of my life than any other supplement)

  34. Hi Clinton,

    Thanks for your input and posting the link to Micheal Rae’ s regimen.
    We have two things in common. We are both engineers with limited background in biology (at least for me) and have both lots of respect in Micheal Rae’s opinion.

    The link you have posted contain lots of useful practical information.
    The only thing that looks weird to me is that Micheal mention that rapamycin and acarbose are both very promising as life extending drugs but does not seem to take either. Maybe I am missing something.

    • Aldebaran,
      I wouldn’t be surprised if some items in this regimen aren’t 100% accurate as to what Michael Rae is doing (even though it notes being updated per Aug 22 2017). I’m skeptical because he didn’t update his ‘diet’ to being vegan but still states it as lacto-ovo vegetarian which I thought he’d changed (to vegan).
      I really respect his perspective on things; I personally have been in addition to a very solid diet been taking the LEF 2 per day multi-vitamin which according to Michael is absolutely not helpful (and quite likely harmful), although a decent B-complex ought to be acceptable – not sure if anyone here would comment on basic vitamins that they take? I’ve been reluctant to scrap the MVM since it also contains some zinc and selenium that I think is good to take.
      He also takes phosphatidylcholine which Vince Giuliano scrapped due to it’s upregulation of NfKb and possible choline effects on cholesterol and cvd (if you read VG’s blog and his update on that supplement).
      Best Regards,

      • Hi Clinton,

        Thanks for your comments.
        I definitely like the depth and critical thinking that Michael put in his analysis. However, I am very far from being as serious as he is about diet and supplements. I try to avoid most of the bad stuff (refined sugar, excessive red meat, trans fat…), try to eat some of the supposed good stuff that I like (olive oil, salad, vegetable), and maintain a BMI in the low range of normal. That’s about it. In term of supplement, I am just starting with NR and thinking about other stuff such as Glucosamine. I am definitely interested to try rapamycin in the near future. There are lots of other interesting stuff discussed on this site that I might try at some point. I will take a look at Vince Giuliano’s blog.


Leave a Reply

Your email address will not be published. Required fields are marked *