Average Telomere Length is a Terrible Measure of Aging

Here I'll point out one of numerous studies providing evidence to illustrate that telomere length isn't all that useful as a biomarker of aging. Telomeres cap chromosomal DNA, a length of repeated sequences that shortens every time a cell divides. This forms a part of the limiting mechanism that stops ordinary somatic cells from dividing indefinitely. Stem cells and cancer cells maintain their ability to divide by periodically extending their telomeres via various mechanisms. At present telomere length is usually measured from a blood sample, taking the average of lengths in immune cells. This will reflect some combination of cell division rates, cell replacement rates, and the immune status of the individual.

Statistically, considered over large populations, average telomere length in immune cells trends downward with aging, indicating that cell populations are dividing more frequently, or not receiving as great an influx of new cells with long telomeres as they were in youth, or both. The latter is a part of the well-known decline of stem cell activity with age. Unfortunately telomere length is nonetheless a terrible measure of biological age on a practical, individual basis, as the correlation just isn't that good, and measures can vary greatly over time for reasons that have little to do with aging, such as ill health due to infectious disease. This all ties in with the idea that the age-related statistical trend towards diminishing telomere length in cell populations is a reflection of the effects of damage on other processes, as well as the influence of numerous other environmental circumstances, and not a cause of aging in and of itself.

Advances in technology allow scientists to measure intricate details about the human body that greatly enhance understanding of health, disease and aging. Yet, when it comes to predicting death, more rudimentary measures - like a person's age or a person's ability to climb stairs or walk a short distance - are much more powerful predictors of survival than certain biomarkers. Using data from the United States, Costa Rica and Taiwan, researchers compared a broad set of predictors of death - like age, smoking habits and mobility - with the length of telomeres, DNA sequences that generally shrink with age.

Decades ago, researchers discovered that telomeres - which are protective caps on the ends of our chromosomes - act as a 'molecular clock' in human cells. Every time cells divide, telomeres shorten until they become critically short and signal the cell to stop dividing. Telomere length is typically measured in white blood cells (leukocytes), and shorter leukocyte telomeres have been associated with disease, aging and death. For these reasons, there has been great interest in the ability of this biomarker to predict mortality.

After evaluating data, the research team found that using telomere length to predict a human's death was only marginally better than a "coin toss." Chronological age was, by far, the single best predictor of death in all three countries. "Scientific evidence on telomere length has been sensationalized and, in some cases, exaggerated by the media and by companies that have capitalized on the research to market products that may promise more than they can deliver. This is what fueled our research. We wanted to determine whether telomere length could predict mortality better than other well-established predictors of survival, most of which are less invasive and much less costly to measure."

The researchers note some potential limitations of the findings. People who are critically ill might exhibit changes in the distribution of different types of leukocytes that makes their telomeres appear longer. In this study, telomere length is measured in leukocytes, which is common across most research. But some types of leukocytes tend to have longer telomeres than others. "Telomere length tends to be longer in the type of leukocyte that becomes more dominant when a person is ill. Therefore, a sick person might appear to have 'longer' telomere length, but that is deceptive. In fact, these critically ill individuals may be much more likely to die in the short-term despite the appearance of 'longer' telomeres."

It also is plausible that telomere length is a better predictor of long-term mortality, compared to short-term survival, since it reflects the gradual process of cellular aging. "Alternatively, telomere length might be a predictor of mortality only for certain groups of patients, such as those with cancer. An interesting possibility is that telomere length might not be a good predictor of mortality, but it could be a good predictor of healthy aging. Increasing evidence demonstrates that shorter telomeres are associated with cardiovascular disease, but additional research is needed to clarify the association between telomere length and other diseases of aging such as cancer."

Link: http://wws.princeton.edu/news-and-events/news/item/age-and-mobility-predict-death-better-one%E2%80%99s-%E2%80%98molecular-clock%E2%80%99


"It also is plausible that telomere length is a better predictor of long-term mortality, compared to short-term survival, since it reflects the gradual process of cellular aging."

What do they mean when they say "reflects the gradual process of cellular aging?" Are they saying that telomere attrition is, in a very general sense, a measure of the gradual decline in gene network integrity, but otherwise a poor measure of the person's actual age, because it doesn't account for the various other drivers of aging?

Posted by: Eric at April 7th, 2016 11:01 AM

@Eric: Telomere length is affected by all sorts of things that are not considered aging per se. The distinction between primary and secondary aging is perhaps a useful one. Primary is the things that happen anyway, and secondary is avoidable or avoidable to some degree, such as smoking, being overweight, and so on. The mechanisms overlap and interact. There is also the question of what to say about states of age-related illness that can be impactful, but may be cured or at least managed on a timescale of a few months or years: there is up and down, and that tends to reflect in average telomere length also.

I'm not sold on the idea that if you can abstract out shorter term things then telomere length measures become useful in a practical, medical sense. I think DNA methylation measures of biological age are already showing themselve to be much better at everything that people were suggesting telomere length measures would be useful for. If a correlation isn't strong, then a correlation isn't strong, and that pretty much rules out use in personalized medicine.

Posted by: Reason at April 7th, 2016 9:28 PM

Hey there,

I believe this study in lost out in the field and is seriously underestimating what has been found about telomeres.

''After evaluating data, the research team found that using telomere length to predict a human's death was only marginally better than a "coin toss."''

lol. What ?? a joke much...well, okay maybe yes if we use the old LD50 for the coin toss odds, you have 50% chance of being right or 50% chance of being wrong; basically 1 chance out of 2 (heads 50% or tails 50% = 50/100%). Which, in fact is pretty good when you think about it; when many markers and therapies can't even reach 50% chance of prediction power; some have barely 15% odds outcome and are far fetched-reaching on results.

If Telomeres have 50% correlation, than that is very good. Chronological aging is very accurate too. The reason for that is because the large mass of humans aged about the same speed (for the healthy people) biologically; as such, don' deviate much on the biological-to-chronological aging ratio/speed difference.
Centenarians who reach 100 and more; are simply ''using up'' more of their telomeres and post-poned replicative senescence or stress-induced senescence (the inflammatory-pathological kind, creating diseases that kill people and cut their ''aging reservoir clock'' left (replicative senescence) to a halt).

One study, precisely on a Super*-Centenerian woman (115 years old), disproves this study here. That supercentenarian study proved beyond a doubt that humans who go to the extreme (110 years old or more) have used up their telomeres beyond what is compatible with function. Her blood leukocyte telomeres were measured and were in the 2000 to 3000 base pairs; which is the very limit that activates the second final crisi point (M2) of replicative senescence. If cells obviate that M2 point, they become immortal, rogue and cancerous like. Some may be non-cancerous; that is extremely rare as the healthy cell telomeric system does not allow such illegit/aberration (p53 and p16, p21 are in fulll force and cyclins are abrogated). The shortest telomeres of these supercentenarians prove that once it goes very very low, the telomeres Overall/on average are *all* short telomeres (in the leukocytes at least); this creates an overall 'uncapped state' of telomeres (activating telomere DDR signal and senescence genes) that the telomeric shelterin proteins, that capp telomeres for chromosome stability, try to mitigate (POT1, TRF1, TRF2) but it becomes harder and harder; as the amount of short telomeres rises; creating overaly instability and can create telomere fusion, SCE, and recombination (the chromosomal/genomic DNA danger of short 'uncapped telomeres').

The fact that that woman had ultra-short telomeres and was 115 years old, demonstrate that telomeres are still one of the best predictors; same goes for DNA methylation. In fact DNA methylation is 100% tied to telomeres, and vice versa. Hyper methylated telomeres/sub-telomeres/centromeres are long and strong. Hiumans, with aging, show loss of global methylome methyl count (5-methylcytosine loss with aging).
The confusion of these papers is the blurred line between people with changes of telomeres during pathological bouts and why it differs from healthy people who aging normally. Yet, as told, that super centenarian woman reached 115, she was Healthy - enough, to reach that, as such her intrinsic aging was 'regular' and she had good antioxidative protection (high Redox reduced power in blood leukocytes and RBC) in her blood/plasma/serum to protect organs from aging effect. She olny 'used up' 'more' of her 'telomere reservoirs/clocks', whereas other sick people die before that without ever 'spending up' the entire telomeric reservoir.

Telomeres are still it..

Posted by: CANanonymity at April 8th, 2016 12:29 PM

It is fairly obvious to me that a variety of biomarkers should be used. First and foremost in my view, DNA Methylation ala Steve Hovarth though telomeres are also a useful indicator depending on which method you use. Q-TRAP assay is better than the more traditional methods but again taken in isolation telomeres do not give a complete picture. They are an indication of cellular "fatigue" if you will and they most certainly do correlate with various pathology so they are not an entirely invalid measurement.

Posted by: Steve Hill at April 9th, 2016 2:58 AM

Post a comment; thoughtful, considered opinions are valued. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.