More evidence of telomere length as a biomarker of aging emerges from a study of “exceptionally old persons” (subjects >90 years old): Peripheral blood leukocytes from very old individuals show telomere lengths that are about what you’d expect based on extrapolation from the steady decline in length between youth and regular (as opposed to “exceptional”) old age. The very aged also show an enrichment in what the authors call “ultra-short telomeres”; I suspect this peak in the histogram can be rationalized by the fact that telomere lengths are bounded below by zero, such that steady linear decline in length over time would result in a pileup in the “undetectable” category.

Is telomere length important to the mechanism of aging, in addition to being a marker of advanced chronological age? The jury is still out on that, likely frustrated by the circumstantial nature of the evidence. We know that telomere length appears to be inversely proportional to life expectancy, but of course correlation ≠ causation, and lifespan and telomere length could be unrelated signifiers of the same underlying phenomenon. To complicate the issue, we are reminded in a recent review that all telomere ends are not equal:

In addition, recent studies in both normal and pathological contexts point to the existence of chromosome-specific mechanisms of telomere length regulation determining a telomere length profile, which is inherited and maintained throughout life.

…i.e., bulk telomere length might be less important than the lengths of particular telomeres, a forest hiding the most significant trees. More to the point, the telomere lengths of the various chromosomes might be regulated independently, making bulk telomere length something of a red herring.

In the cellular senescence field, the idea of “sentinel telomeres” is still alive and kicking, but hardly ascendant. The emerging consensus is that a specific group of chromosomes with the shortest telomeres in a given clonal lineage, rather than a single sentinel telomere, herald the onset of senescent cell cycle arrest. Furthermore, where senescence is concerned, it is likely that end-to-end joining between “signal-free ends” (probably the same as the “ultra-short telomeres” discussed above) are more relevant than bulk telomere shortening. I mention this not to argue one way or the other about the relative importance of individual vs. population lengths as related to telomeres as a biomarker of aging, but rather to give a cautionary example of a situation in which the easiest measurement to make (the distribution of lengths in a population) doesn’t appear to be the most relevant.