The DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index; none of the 3 sources its claims cite is checked for support yet.
A model of aging as accumulated damage matches observed mortality patterns and predicts the life-extending effects of prospective interventions
Open accessAbstract
The relative insensitivity of lifespan to environmental factors constitutes compelling evidence that the physiological decline associated with aging derives primarily from the accumulation of intrinsic molecular and cellular side-effects of metabolism. Here we model that accumulation starting from a biologically based interpretation of the way in which those side-effects interact. We first validate this model by showing that it very accurately reproduces the distribution of ages at death seen in typical populations that are well protected from age-independent causes of death. We then exploit the mechanistic basis of this model to explore the impact on lifespans of interventions that combat aging, with an emphasis on interventions that repair (rather than merely retard) the direct molecular or cellular consequences of metabolism and thus prevent them from accumulating to pathogenic levels. Our results strengthen the case that an indefinite extension of healthy and total life expectancy can be achieved by a plausible rate of progress in the development of such therapies, once a threshold level of efficacy of those therapies has been reached.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
The authors present a biologically motivated simulation of aging as accumulating molecular and cellular damage, fitted to US 1999 mortality data, and use it to model therapies that repair damage. They conclude that if therapies halve each category of aging damage roughly every 40 years or faster, “longevity escape velocity” is quantitatively plausible.
Key findings
- The simulation reproduced the distribution of ages at death in US 1999 mortality data with goodness-of-fit exceeding r² = 0.99, better than the Gompertz formula, especially at ages where most deaths occur.
- Simulated damage accumulated during old age at roughly twice the rate seen during youth; the slowest-aging simulated individuals had a damage level at age 85 that the fastest agers reached by age 70.
- In a schedule where therapies arrive every 6 years and there are 7 damage categories, each category’s damage and new accumulation are halved every 42 years; cohorts first treated at age 80 gained only a few years, while first treatment at age 70 gave a 10% chance of reaching negligible age-related mortality, and the cohort 10 years younger had almost a 50% chance.
- The authors conclude that if each category of aging damage is halved every ∼40 years or faster, beneficiaries can experience progressive damage decline and indefinite lifespans; a harmonic slowdown (1/2, 1/3, 1/4…) may still work, but exponential deceleration (1/2, 1/4, 1/8…) fails.
Methods
- A simulation program modeled each person with multiple damage categories, each containing multiple mechanisms; each mechanism contributed additively to its category via a basal constant rate plus a feedback rate proportional to existing damage.
- Fitting parameters were adjusted to match observed United States mortality data from 1999 (Wilmoth 1997); after fitting, the model simulated schedules of “cures” that set a mechanism’s accumulated damage and future accumulation rate to zero.
- The intervention analysis used 100,000 individuals, 7 damage categories, therapies arriving every 6 years, and a schedule eliminating 50% of remaining damage in each category per therapy; survival projections assumed no age-independent death.
Limitations
- The paper states that the simulation has many parameters, which could invite criticism, but argues all parameters have biological reasons and that goodness-of-fit is insensitive to many parameter values.
- The paper states that survival projections assume no age-independent death and that the timeframe for initial fundamental breakthroughs is factored out by defining time zero as the first breakthrough.
- Evident from the text: the intervention predictions are model outputs under assumed cure schedules, not validated against real rejuvenation therapies; the model also focuses on therapies that remove damage rather than prevent its accumulation.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 of 3 cited sources checked · 0 supported · 0 flagged
- Findingown data
The rate of accumulation of damage is dominated by the feedback component, i.e. by how much damage is already present, rather than the basal (constant) component.
“The explanation is to be found in the parameters that we found fit real-life mortality data: the rate of accumulation of damage is dominated by the feedback component, i.e. by how much damage is already present, rather than the basal (constant) component.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Findingown data
The model very accurately reproduces the distribution of ages at death seen in typical populations well protected from age-independent causes of death.
“We first validate this model by showing that it very accurately reproduces the distribution of ages at death seen in typical populations that are well protected from age-independent causes of death.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Findingown data
Inter-individual variation in rate of simulated aging is plausible, with the slowest-aging simulated individuals having a damage level at age 85 that the fastest agers attain by age 70.
“The inter-individual variation in rate of simulated aging is also plausible, with the slowestaging simulated individuals having a damage level at age 85 that the fastest agers attain by age 70.”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Findingreasoning
Damage for which there is as yet no therapy has approximately a constant doubling time.
“Thus, damage for which there is as yet no therapy has approximately a constant doubling time.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Findingown data
The simulation achieved a goodness-of-fit to observed United States mortality data from 1999 exceeding r^2 = 0.99.
“The goodness-of-fit of our simulation to the data exceeds r 2 =0.99.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Findingreasoning
Longevity escape velocity cannot be achieved by a schedule of cures that slows down too fast in terms of the proportion of each category of damage eliminated per unit time.
“We have also established that longevity escape velocity cannot be achieved by a schedule of cures that slows down too fast (in terms of the proportion of each category of damage eliminated per unit time).”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Findingown data
Simulated damage accumulates during old age at roughly twice the rate at which it accumulates during youth.
“As shown in Fig. 3, simulated damage accumulates during old age at roughly twice the rate at which it accumulates during youth.”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Findingreasoning
In the simulated cure schedule, category-maximum damage accumulation begins to be affected even after the first intervention because each category's level influences the rates of accumulation of other categories.
“Categorymaximum damage accumulation begins to be affected even after the first intervention (whose timing is identified by the circles), because of the influence of each category's level on the rates of accumulation of other categories.”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Findingown data
A periodically applied repair therapy that removes half the damage benefits people already in middle age much more than a halving of the default rate of aging.
“A halving (blue line) of the default rate of aging (red line), applied to people already in middle age, will benefit them much less than a periodically applied therapy that removes half the damage (pink line).”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Findingown data
The goodness-of-fit of the simulation to the source data is extremely insensitive to the values of many parameters.
“It is also worth noting that the goodness-of-fit of our simulation to the source data turns out to be extremely insensitive to the values of many of the parameters (data not shown).”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Findingreasoning
With improvements in efficacy, repeated repair therapy can indefinitely avoid a pathogenic level of damage even though the therapy is never perfect.
“But if, by the second time the repair therapy is applied, it has been improved to the point where it can fix half the damage that it originally could not fix, and the third time it can fix half the rest, and so on, the result (Fig. 1, orange line) is an indefinite avoidance of a pathogenic level of damage, even though the therapy is never perfect.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Findingnone given
The older a cohort is when therapies start to arrive, the sooner they return to their pre-treatment level of damage.
“Notably, however, the older a cohort is when therapies start to arrive, the sooner they return to their pre-treatment level of damage.”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 79.9961, normSD 0.1851, skew -1.1368, kurtosis 5.0180, and r-squared 0.9913.
“//Sim 79.9961 0.1851 -1.1368 5.0180 0.9913”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 81.7828, normSD 0.1645, skew -1.1589, kurtosis 5.4316, and r-squared 0.9807.
“//Sim 81.7828 0.1645 -1.1589 5.4316 0.9807”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 80.0860, normSD 0.1766, skew -1.0961, kurtosis 5.1927, and r-squared 0.9819.
“//Sim 80.0860 0.1766 -1.0961 5.1927 0.9819”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 81.2735, normSD 0.1696, skew -1.0879, kurtosis 5.0940, and r-squared 0.9839.
“//Sim 81.2735 0.1696 -1.0879 5.0940 0.9839”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 16000, the model reproduced the ideal mortality distribution with mean 80.9397, normSD 0.1684, skew -1.0840, kurtosis 5.3031, and r-squared 0.9893.
“//Sim 80.9397 0.1684 -1.0840 5.3031 0.9893”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 80.7430, normSD 0.1759, skew -1.0144, kurtosis 4.9862, and r-squared 0.9844.
“//Sim 80.7430 0.1759 -1.0144 4.9862 0.9844”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 81.0839, normSD 0.1699, skew -1.0875, kurtosis 5.0973, and r-squared 0.9870.
“//Sim 81.0839 0.1699 -1.0875 5.0973 0.9870”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 16000, the model reproduced the ideal mortality distribution with mean 80.3307, normSD 0.1755, skew -1.1611, kurtosis 5.2429, and r-squared 0.9913.
“//Sim 80.3307 0.1755 -1.1611 5.2429 0.9913”
Quote in the paperSubstantial claim - Findingown data
In a simulated population of 8000, the model reproduced the ideal mortality distribution with mean 81.1780, normSD 0.1665, skew -1.0472, kurtosis 5.3162, and r-squared 0.9793.
“//Sim 81.1780 0.1665 -1.0472 5.3162 0.9793”
Quote in the paperSubstantial claim - Findingown data
The reflected normal distribution provides a good fit to the mortality data.
“/* normal reflected -good fit* /”
Quote in the paperSubstantial claim - Interpretationreasoning
Studies of functional decline and mortality are necessarily limited to analysis of the later stages of life and cannot illuminate the lifelong accumulation of initially latent molecular and cellular changes.
“it is necessarily limited to analysis of the later stages of life, when functional decline is present; it cannot illuminate the lifelong accumulation of initially latent molecular and cellular changes that is the necessary precursor of that decline.”
Quote in the paperSubstantial claim - Interpretationreasoning
An indefinite extension of healthy and total life expectancy can be achieved by a plausible rate of progress in development of repair therapies, once a threshold efficacy is reached.
“Our results strengthen the case that an indefinite extension of healthy and total life expec-tancy can be achieved by a plausible rate of progress in the development of such therapies, once a threshold level of efficacy of those therapies has been reached.”
Quote in the paperSubstantial claim- Rests on an assumption: The physiological decline associated with aging derives primarily from the accumulation of intrinsic molecular and cellular side-effects of metabolism.
- Rests on an assumption: The timeframe for fundamental breakthroughs is very hard to predict, but once made, cumulative refinement by individually modest advances proceeds at a rather predictable rate so long as public enthusiasm for progress remains strong.
- Premisecitation
The proportion of people reaching age N who died before reaching age N+1 is approximated by an exponential function of N across the range of ages at which most people died.
“the proportion of people reaching age N who died before reaching age N+1 was approximated with remarkable precision by an exponential function of N across the range of ages at which most people died (Gompertz 1825).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Gompertz 1825
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Premisereasoning
The rate of accumulation of damage tends to accelerate as the overall level of damage increases because metabolism's homeostatic task becomes harder.
“Thirdly, metabolism includes many processes that limit the rate of accumulation of damage; thus, that rate will tend to accelerate as the overall level of damage (of all types) increases and makes metabolism's homeostatic task harder, even before it begins to fail in that task (i.e.functional decline emerges).”
Quote in the paperSubstantial claim - Premisecitation
Gavrilov and Gavrilova's reliability-theory model can approximate reasonably accurately the mortality patterns in real populations.
“Analyses using reliability theory have shown that their model can approximate reasonably accurately the mortality patterns in real populations.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Gavrilov andGavrilova (1991, 2004)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Interpretationreasoning
Indefinite lifespans can be achieved even without the total elimination of all age-related damage.
“Indefinite lifespans can therefore be achieved, even without the (probably unrealistic) total elimination of all age-related damage.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Interpretationreasoning
The concept of a longevity escape velocity is a quantitatively realistic scenario once therapies that eliminate important contributors to aging begin to emerge.
“Our main finding in this study is that the concept of a”
Quote in the paperSubstantial claim0 of 1 sources checked- cites de Grey 2004 source
A check is scheduled.
- Rests on an assumption: The physiological decline associated with aging derives primarily from the accumulation of intrinsic molecular and cellular side-effects of metabolism.
- Rests on an assumption: The timeframe for fundamental breakthroughs is very hard to predict, but once made, cumulative refinement by individually modest advances proceeds at a rather predictable rate so long as public enthusiasm for progress remains strong.
- Interpretationreasoning
A retardation therapy can achieve indefinite avoidance of pathogenic damage only if the improvement in its efficacy is far more rapid or begun at a much earlier age.
“This can only be achieved with a retardation therapy if the improvement in its efficacy is far more rapid or begun at a much earlier age-and, of course, the beneficiaries of such a therapy grow steadily closer to frailty all the time (Fig. 1, green line).”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- Interpretationreasoning
The Gavrilov-Gavrilova model is unrealistic because it depicts the organism as made of components that are individually non-aging and not replaced over time.
“it is unrealistic in the important sense that it depicts the organism as made of components that are (1) individually non-aging and (2) not replaced over time.”
Quote in the paperSubstantial claim - Interpretationnone given
There is an equivalence between wholly eliminating one minor contributor to aging and partly eliminating a major one.
“It should be stressed that this is a reasonable representation of such therapies across the spectrum of real-life precursors of aging, because there is an equivalence between wholly eliminating one minor contributor to aging and partly eliminating a major one.”
Quote in the paperSubstantial claim - Interpretationreasoning
The model predicts that if progress in eliminating successive types of damage continues at a sufficient pace, the amount of damage present in the body can be kept low, with a negligible rate of age-related death per year, irrespective of a person's age.
“Our model predicts that, if progress in eliminating successive types of damage continues at a sufficient pace (which, we stress, is a pace very typical of past technologies-prominent examples include powered flight, computers and the combatting of infectious diseases), the amount of damage present in the body can be kept low, with a negligible rate of age-related death per year, irrespective of a”
Quote in the paperSubstantial claim- Rests on an assumption: The physiological decline associated with aging derives primarily from the accumulation of intrinsic molecular and cellular side-effects of metabolism.
- Rests on an assumption: The timeframe for fundamental breakthroughs is very hard to predict, but once made, cumulative refinement by individually modest advances proceeds at a rather predictable rate so long as public enthusiasm for progress remains strong.
- Premisenone given
The Gompertz relationship holds with great accuracy across all human and animal populations that exhibit aging and do not suffer appreciable mortality from causes unrelated to aging.
“This relationship has since been found to hold, with great accuracy, across all human and animal populations that (1) exhibit aging at all and (2) do not suffer appreciable mortality from causes unrelated to aging.”
Quote in the paperNot a substantial claim
- It may state two separate claims as one. 12
- Probably not a substantial claim. 12
- The quote only partly states this claim. 2
It may state two separate claims as one.
Findingown dataThe model's fit to contemporary human mortality data is better than the Gompertz formula, including at the ages at which most deaths occur.
“Figure 2a and b illustrate our first major finding: our model can generate a mortality pattern that fits contemporary human mortality data extremely well, better even than the Gompertz formula.”
- Also stated in: § Fitting human mortality data · boosted
“it can be clearly seen that the superiority does not derive simply from a better fit at early ages, but at the ages at which most deaths occur.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
- Also stated in: § Fitting human mortality data · boosted
It may state two separate claims as one.
Findingown dataCures can slow down to roughly a harmonic rate of progress (remaining amount cured every 40 years is 1/2, 1/3, 1/4... rather than always 1/2); only an exponential deceleration (e.g. 1/2, 1/4, 1/8...) fails.
“However, our experiments indicate (data not shown) that this still allows cures to slow down somewhat, to roughly a harmonic rate of progress (in which the remaining amount cured every 40 years is 1/2, 1/3, 1/4... rather than always 1/2): only an exponential deceleration in progress (e.g.1/2, 1/4, 1/8...) fails.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
It may state two separate claims as one.
FindingreasoningIt takes more than 36 years from the arrival of the first therapy before the DMAX of any cohort falls back to its pre-therapy level, because until then at least one category of damage has yet to be addressed at all.
“As one would expect, it takes more than 36 years from the arrival of the first therapy before the DMAX of any cohort falls back to its pre-therapy level, since until then at least one category of damage has yet to be addressed at all.”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
It may state two separate claims as one.
FindingreasoningIf the impact of each category of aging mechanism is halved every ∼40 years or faster, beneficiaries of the first round of such therapies will experience a progressive decline in their damage level over the long term, and thus a reduction to negligible levels of their risk of death from age-related causes.
“Specifically, we show that, if the impact of each category of aging mechanism is halved every ∼40 years or faster, beneficiaries of the first round of such therapies will experience a progressive decline in their damage level over the long term, and thus a reduction to negligible levels of their risk of death from age-related causes.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
It may state two separate claims as one.
FindingreasoningAs few as 10 years of head start (first therapies at age 70) gives a 10% chance of surviving long enough that the chance of dying from aging at any age becomes infinitesimal, and the cohort 10 years younger has almost a 50% chance of achieving indefinite youth.
“As few as 10 years of”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: In simulations of survival, there is no age-independent death.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
It may state two separate claims as one.
FindingreasoningFor individuals who first receive therapies when already 80 years old, the survival curve differs only slightly from a contemporary curve, with only a small minority obtaining a few additional years of life.
“The survival curve for individuals who first receive therapies when already 80 years old differs only slightly from a contemporary curve (black line in Fig. 5), with only a small minority of the cohort obtaining a few additional years of life.”
Quote in the paperSubstantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
- The authors state this as an assumption: In simulations of survival, there is no age-independent death.
- The authors state this as an assumption: Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
Findingnone givenThe normal reflected model with zero SD gives an OK fit to the Gompertz model.
“/* normal reflected, zero SD -OK fit to Gompertz* /”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
Findingnone givenThe lognormal, big basal model gives a mediocre fit with bad skew.
“/* lognorm, big basal -mediocre fit (bad skew)* /”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one. The quote only partly states this claim.
Findingown dataThe simulation with the normal reflected model yields mean 79.4792, normSD 0.1741, skew -0.8564, kurtosis 4.2376, ar^2 0.9690, and gr^2 0.9850.
“//Sim 79.4792 0.1741 -0.8564 4.2376 0.9690 0.9850”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
Findingown dataThe optimised values of the fitting parameters were virtually identical whatever goodness-of-fit measure was chosen.
“our simulation achieved so high a goodness-of-fit that it was unnecessary to choose between these alternatives (i.e. the optimised values of the fitting parameters were virtually identical whatever measure was chosen).”
Quote in the paperNot a substantial claim- The authors state this as an assumption: Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
- The authors state this as an assumption: Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
- The authors state this as an assumption: The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
Probably not a substantial claim.
Findingown dataThe simulation was fitted to United States mortality data for 1999.
“Fitting of the simulation to United States mortality data for 1999.”
Quote in the paperSubstantial claimThe quote only partly states this claim.
Findingown dataWith the normal reflected distribution and the listed parameters, the simulation model produces a good fit to the mortality data.
“/* normal reflected -good fit*/”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it.
Findingown dataThe absolute-normal, big basal model gives a lousy fit.
“/* absnorm, big basal -lousy* /”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
Findingown dataThe Gompertz model yields mean 80.0171, normSD 0.1722, skew -1.0700, kurtosis 4.8656, ar^2 0.9876, and gr^2 1.0000.
“//Gomper 80.0171 0.1722 -1.0700 4.8656 0.9876 1.0000”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
Findingnone givenThe high basal rate model is poor with or without the reflected normal distribution, and impossible with the lognormal distribution.
“/* High basal; poor with or without refl norm, impossible with lognorm* /”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
Findingnone givenThe actuarial model yields mean 78.7283, normSD 0.1977, skew -1.8408, kurtosis 8.4616, ar^2 1.0000, and gr^2 0.9876.
“//Actuar 78.7283 0.1977 -1.8408 8.4616 1.0000 0.9876”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
Premisenone givenTypes of damage can be considered in clusters whose members contribute additively to impair essentially the same aspects of metabolism by the same mechanisms.
“Second, and conversely, types of damage can be considered in clusters, in which the members of a cluster contribute additively to impair essentially the same set of aspects of metabolism by the same mechanisms”
Quote in the paperNot a substantial claimIt may state two separate claims as one.
PremisereasoningDeath is not a deterministic outcome of progression to a given level of damage; damage level dictates near-future risk of death rather than causing instant death.
“Fourth, death is not a deterministic outcome of the progression to a given level of damage, because death generally results from the body's inability to cope with a stochastic, often external, challenge such as an infection or an injury: thus, one's level of damage dictates one's risk of death in the near future rather than causing instant death.”
Quote in the paperSubstantial claimProbably not a substantial claim. The paper’s conclusions may not depend on it. The paper may give no reason why it holds.
Premisenone givenEach type of damage affects different aspects of metabolism, such that a single type could kill the organism once it rose to a sufficient level even if all other types of damage were absent.
“First, each type of damage affects different aspects of metabolism to different degrees, such that in principle a single type could kill the organism once it rose to a sufficient level even if all other types of damage were totally absent.”
Quote in the paperNot a substantial claimProbably not a substantial claim. The paper may give no reason why it holds.
Premisenone givenMetabolism is a highly robust homeostatic network that maintains the organism in a functional state in the face of transient perturbations.
“Metabolism (here defined in its broad sense, as the entire set of biochemical and cellular processes that maintain organismal function) is a highly robust homeostatic network that maintains the organism in a functional state in the face of phenomena that transiently perturb that state-infections, meals, exertion, etc.”
Quote in the paperNot a substantial claim
The paper relies on this without support.
The physiological decline associated with aging derives primarily from the accumulation of intrinsic molecular and cellular side-effects of metabolism.
“The relative insensitivity of lifespan to environmental factors constitutes compelling evidence that the physiological decline associated with aging derives primarily from the accumulation of intrinsic molecular and cellular side-effects of metabolism.”
The authors state this as an assumption.
Aging damage can be represented as multiple categories, each with multiple mechanisms that contribute additively to the damage level of its category.
“Each category has multiple mechanisms, and each mechanism contributes additively to the damage level of its category.”
The authors state this as an assumption.
Each mechanism's damage rate has a basal component adding a constant amount each year and a feedback component increasing damage in proportion to existing damage.
“Each mechanism has two contributors to its rate: a basal component that adds a constant amount of damage each year, and a feedback component that increases damage in proportion to the person's existing damage (the average of its category damages), creating an accelerating increase in damage.”
The authors state this as an assumption.
The probability of death in the coming year given a particular damage level follows an exponential curve with value 0 at damage level 0 and 1 at damage level 1.
“One additional fitting parameter determines the probability of death in the coming year given a particular damage level.The curve is exponential and chosen to have a value of 0 at damage level 0 and 1 at damage level 1”
The authors state this as an assumption.
In simulations of survival, there is no age-independent death.
“Figure 5 shows the consequences of these trajectories of damage level for the probability of survival to a given age assuming no age-independent death.”
The paper relies on this without support.
The timeframe for fundamental breakthroughs is very hard to predict, but once made, cumulative refinement by individually modest advances proceeds at a rather predictable rate so long as public enthusiasm for progress remains strong.
“Specifically, the timeframes for fundamental breakthroughs are very hard to predict, but once those breakthroughs have been made, their cumulative refinement by individually modest advances proceeds at a rather predictable rate so long as public enthusiasm for progress remains strong.”
The authors state this as an assumption.
Curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.
“In simulations in which aging mechanisms are cured, curing a mechanism sets both its accumulated damage and damage accumulation rate to zero.”
The authors state this as an assumption.
There is a frailty threshold of loss of reserve, beyond which intervention becomes much harder as pathologies are emerging and exacerbating their own rate of progress.
“It is presumed that there is a “frailty threshold” of loss of reserve, beyond which intervention becomes much harder as pathologies are emerging and exacerbating their own rate of progress.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
The implemented death rule is not the stated exponential curve
MajorThe methods specify an exponential death-probability curve that reaches certainty at damage level 1. The appended code instead draws a fatal threshold from a uniform or other fitted distribution and tests whether damage exceeds it; that rule is not the specified curve and does not generally make death certain at damage 1.
“The curve is exponential and chosen to have a value of 0 at damage level 0 and 1 at damage level 1”
“if (CONST_FATAL) { fatalLevel = U.Next()*FATAL_LEVEL; } else { fatalLevel = dist(FATAL_LEVEL, FATAL_LEVEL_SD, FATAL_LEVEL_LOGNORM); } if (maxDamage > fatalLevel && !WAIT_CHALLENGE) { dead = 1; }”
Person-level feedback damage is specified as both a sum and an average
MinorThe methods table defines PD(t), the damage used for feedback, as the sum of category damages. The appended implementation divides that sum by the number of categories before using it for subsequent updates.
“PD(t)=[SUM c=1 to N_CAT] D_C c (t)”
“damage = sumDamage/NUM_CATEGORIES;”
Citations
Cites
Gompertz, B. (1825) · Philosophical Transactions of the Royal Society of London · doi:10.1098/rstl.1825.0026 · Cited by 3,816
Aging, natural death, and the compression of morbidity
Fries, J. F. (1980) · New England Journal of Medicine · doi:10.1056/nejm198007173030304 · Cited by 3,356
The biology of life span: A quantitative approach
Gavrilov, L. A., & Gavrilova, N. S. (1992) · Free Radical Biology and Medicine · doi:10.1016/0891-5849(92)90121-v · Cited by 415
Manton, K. G., Gu, X., & Lamb, V. L. (2006) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0608483103 · Cited by 273
Limits to deficit accumulation in elderly people
Rockwood, K. J., & Mitnitski, A. (2006) · Mechanisms of Ageing and Development · doi:10.1016/j.mad.2006.01.002 · Cited by 237
The biology of life span—A quantitative approach
Dean, W. (1992) · Experimental Gerontology · doi:10.1016/0531-5565(92)90049-6 · Cited by 195
Time to Talk SENS: Critiquing the Immutability of Human Aging
de Grey, A. D. N. J., Ames, B. N., Andersen, J. K., Bartke, A., Campisi, J., Heward, C. B., McCarter, R., & Stock, G. B. (2002) · Annals of the New York Academy of Sciences · doi:10.1111/j.1749-6632.2002.tb02115.x · Cited by 188
Escape Velocity: Why the Prospect of Extreme Human Life Extension Matters Now
de Grey, A. D. N. J. (2004) · PLoS Biology · doi:10.1371/journal.pbio.0020187 · Cited by 97
The Reliability‐Engineering Approach to the Problem of Biological Aging
Gavrilov, L. A., & Gavrilova, N. S. (2004) · Annals of the New York Academy of Sciences · doi:10.1196/annals.1297.094 · Cited by 41
How individual age-associated changes may influence human morbidity and mortality patterns
Ukraintseva, S. V., & Yashin, A. I. (2001) · Mechanisms of Ageing and Development · doi:10.1016/s0047-6374(01)00277-9 · Cited by 37
A strategy for postponing aging indefinitely.
de Grey, A. D. N. J. (2005) · PubMed · Cited by 19
Flagged first, then by citations
Cited by
Sorted by citations
Sarcopenia: Aging-Related Loss of Muscle Mass and Function
Larsson, L., Degens, H., Li, M., Salviati, L., Lee, Y. I., Thompson, W., Kirkland, J. L., & Sandri, M. (2019) · Physiological Reviews · doi:10.1152/physrev.00061.2017 · Cited by 1,797
The Demographic and Biomedical Case for Late-Life Interventions in Aging
Rae, M. J., Butler, R. N., Campisi, J., de Grey, A. D. N. J., Finch, C. E., Gough, M. S., Martin, G. M., Vijg, J., Perrott, K. M., & Logan, B. J. (2010) · Science Translational Medicine · doi:10.1126/scitranslmed.3000822 · Cited by 127
Is 60 the New 50? Examining Changes in Biological Age Over the Past Two Decades
Levine, M. E., & Crimmins, E. M. (2018) · Demography · doi:10.1007/s13524-017-0644-5 · Cited by 83
Avanesov, A. S., Ma, S., Pierce, K. A., Yim, S. H., Lee, B. C., Clish, C. B., & Gladyshev, V. N. (2014) · eLife · doi:10.7554/elife.02077 · Cited by 74
Toward a Unified Theory of Aging and Regeneration
West, M. D., Sternberg, H., Labat, I., Janus, J., Chapman, K. B., Malik, N. N., de Grey, A. D., & Larocca, D. (2019) · Regenerative Medicine · doi:10.2217/rme-2019-0062 · Cited by 48
Strategies for Engineered Negligible Senescence
Zealley, B., & de Grey, A. D. N. J. (2012) · Gerontology · doi:10.1159/000342197 · Cited by 42
Aging in psoriasis vulgaris: female patients are epigenetically older than healthy controls
Borský, P., Chmelařová, M., Fiala, Z., Hamáková, K., Palička, V., Krejsek, J., Andrýs, C., Kremláček, J., Řeháček, V., Beránek, M., et al. (2021) · Immunity & Ageing · doi:10.1186/s12979-021-00220-5 · Cited by 19
The singularity and the Methuselarity: similarities and differences.
de Grey, A. D. N. J. (2009) · PubMed · Cited by 8
Scoping Mortality Research: (Report of the Mortality Research Steering Group)
Macdonald, C. (2009) · British Actuarial Journal · doi:10.1017/s1357321700005444 · Cited by 6
WILT: Necessity, Feasibility, Affordability
de Grey, A. D. N. J. (2010) · Springer eBooks · doi:10.1007/978-90-481-3999-6_22 · Cited by 4
Strategies for Engineered Negligible Senescence
de Grey, A. D. N. J., & Rae, M. J. (2021) · Springer eBooks · doi:10.1007/978-3-030-22009-9_387
Strategies for Engineered Negligible Senescence
de Grey, A. D. N. J., & Rae, M. J. (2021) · Springer eBooks · doi:10.1007/978-3-319-69892-2_387-2
Strategies for Engineered Negligible Senescence
de Grey, A. D. N. J., & Rae, M. J. (2019) · Springer eBooks · doi:10.1007/978-3-319-69892-2_387-1
Borský, P., Chmelařová, M., Fiala, Z., Hamáková, K., Palicka, V., Krejsek, J., Andrýs, C., Kremláček, J., Řeháček, V., Beránek, M., et al. (2021) · Figshare · doi:10.6084/m9.figshare.14157565
Metadata
- Authors
- Phoenix, C., & de Grey, A. D. N. J.
- Affiliations
- Methuselah Foundation (US)
- Corresponding author
- de Grey, Aubrey D. N. J.
- Venue
- AGE
- Published
- 2007-09-18
- Volume
- 29
- Issue
- 4
- Pages
- 133-189
- ISSN
- 0161-9152 (ISSN-L)
- 1574-4647
- PMC
- PMC2267031
- OpenAlex
- W2031504628
- Semantic Scholar
- 31f321b4a33c2d1455ff4c417c5ffbcfa67767a5
- Cited by
- 14
- References
- 11
- Citations by year
- 2021: 4 · 2019: 2 · 2018: 2 · 2014: 1 · 2012: 1
- FWCI
- 0.34
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- www.springer.com/tdm (tdm)
- Locations
- AGE (publishedVersion)
- Europe PMC (PubMed Central) (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Aging › Genetics, Aging, and Longevity in Model Organisms
- Social Sciences › Demography › Insurance, Mortality, Demography, Risk Management
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Cell Biology › Telomeres, Telomerase, and Senescence
- Keywords
- aging, mortality patterns, age-at-death distribution, life expectancy extension, anti-aging interventions, healthy life expectancy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-11-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-11-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
