For most of the history of human civilization, humanity expanded at an astonishing pace: faster than exponential, nearly hyperbolic. This trend was famously described in a 1960 paper by Heinz von Foerster and colleagues, who extrapolated global population data to predict a so-called “Doomsday”, a demographic singularity in which human numbers would become infinite by 2026, assuming growth continued unchecked [1].
However, something unexpected happened: birth rates began to fall even as lifespans continued to rise. This shift wasn’t directed by any authority; it was a spontaneous outcome of billions of individual choices. It was a striking example of emergent behavior in a complex system. In doing so, humanity collectively dodged what had once been seen as its greatest existential threat: overpopulation.
Yet, the success of this transition has brought a new crisis: as we live longer, we don’t necessarily live better. The burden of chronic disease rises sharply with age, driving healthcare costs higher and threatening to destabilize pension systems worldwide. The same demographic shift that saved us from overpopulation now demands a new kind of solution; we must either decouple aging from disease or stop aging altogether.
This imperative has fueled the rise of longevity biotechnology, a modern crusade to unlock the secrets of extended life. Billions of dollars have poured into startups, research labs, and bold promises of reversing aging. We have made progress: we know that human life, and especially the lives of lab animals, can be stretched impressively.
Caloric restriction (CR), cutting food intake while maintaining nutrition, remains the gold standard, consistently extending lifespan in mice, worms, and even monkeys by slowing metabolism and reducing cellular wear. It’s a trick that nature has been hinting at for eons. Here’s the rub: despite all our high-tech tools, no cutting-edge intervention, whether cellular reprogramming with Yamanaka factors or advanced drug cocktails, has outperformed rapamycin or caloric restriction in animal models, whether tested alone or in combination. As Matt Kaeberlein has emphasized, nothing yet beats the effect of simply eating less.
This stagnation echoes a tale from 800 years ago, when Genghis Khan, the conqueror of empires, turned his gaze to conquering death itself. Around 1222, as he ravaged all of the taxable-at-the-time world, the aging Khan summoned Qiu Chuji, a Taoist monk famed for his wisdom on longevity. Genghis, nearing 60 and feeling the weight of his relentless campaigns, demanded the secret to eternal life: an elixir to defy time.
After a grueling two-year trek to meet the Khan, Qiu offered no potion, no magic. Instead, he counseled moderation: curb your excesses, avoid overindulgence in wine and war, live simply. It was caloric restriction and a balanced life in all but name, pragmatic advice rooted in observation, not mysticism. Genghis, perhaps disappointed, still honored the monk but died just five years later in 1227. Even then, with all his power and the wisdom of the age, nothing better than moderation emerged.
Ironically, it is traditional pharma that has edged closer to practical anti-aging interventions. Drugs like Ozempic, which were originally developed for diabetes and obesity, have shown real, though modest, mortality benefits across a growing list of conditions. These effects are meaningful, but they still fall short of fundamentally altering the aging process.
Meanwhile, the longevity field projects a contradictory message. On one hand, it claims we are close to developing a drug against aging; on the other, it acknowledges that we still lack a shared understanding of what aging actually is. We are like early aviators tinkering with wings and engines, achieving powered flight through trial and error. Drugs that mimic the effects of caloric restriction, like rapamycin and metformin, are our first creaky airplanes: promising, but still crude.
The ambition to truly defeat aging is not just about building better airplanes; it’s about realizing that no airplane, no matter how refined, can reach the Moon. To get there, humanity needed rockets, which are based on entirely different principles. Similarly, halting aging will demand not just incremental improvements, but a deep, principled mastery of the fundamental mechanics that drive the aging process.
As with the case of flying machines, nature provides numerous examples of evolutionarily advanced creatures, including some mammals, that exhibit little to no signs of aging and live up to 10 times longer than expected for animals of their size: an effect size far greater than that achieved by CR. This phenomenon, known as negligible senescence, is increasingly recognized as a distinct regime of aging. Notably, recent studies by the Calico team, using increasingly large animal cohorts of naked mole rates, have shown the absence of the acceleration of mortality, a defining feature of aging in humans.
We must begin thinking more deeply to make sense of these observations. In a 2007 paper in PLOS Genetics, Leonard Hayflick proposed entropy, the universal force driving systems toward disorder, as the fundamental cause of aging. Building on this idea and using modern molecular-level data, our team developed a comprehensive, quantitative theory of aging that integrates both dynamic (reversible) and entropic (irreversible) components of the aging process.
In a series of peer-reviewed studies, published in Nature Communications [2] and Aging Biology [3], we demonstrated that human aging is governed by a dual mechanism: a slow, linear accumulation of entropy that expands the footprint in physiological configuration space and leads to irreversible information loss, and, alongside it, dynamic fluctuations—short-term, reversible stress responses—that gradually destabilize the system and drive the onset of chronic diseases with age.
This combined framework not only confirms Hayflick’s hypothesis but extends it, providing a coherent and, crucially, quantitative explanation for how different biological systems age over time and eventually fail. Critically, it allowed us to classify organisms into two distinct aging regimes: relatively short-lived (“unstable”) species, like mice, whose aging is dominated by dynamic instabilities, and longer-lived (“stable”) species, like humans, where aging is driven primarily by the slow, irreversible accumulation of entropic damage. A key insight from this model is that interventions targeting only the dynamic components of aging, such as senescence or inflammation, have little to no effect on the underlying entropic damage, which is consistent with the expectations of macroscopic irreversibility as dictated by the second law of thermodynamics.
These findings are not merely of academic interest; they form the foundation for a new, quantitative theory of aging: one that explains mortality trends, biomarker divergence, and why long-lived species age differently from short-lived ones. At the heart of this theory lies a simple but powerful principle: aging in humans is driven by the accumulation of microscopic molecular insults—each individually benign and reversible, but collectively irreversible—that gradually erode physiological resilience. As this process unfolds, the organism becomes increasingly fragile, until even small fluctuations (biological noise) can push it past critical thresholds, triggering disease or death.
These findings and the underlying theory not only explain much of what is known about aging biology, but also generate important new predictions. In our 2021 Nature Communications publication, we provided one of the first direct measurements of the maximum human lifespan using longitudinal physiological data. By analyzing changes in biological markers, such as blood composition and physical activity patterns, we demonstrated that the variance and recovery time of physiological signals diverge near a critical age, between 120 and 150 years. This divergence signals a fundamental loss of resilience, indicating a hard upper limit on human lifespan. Our results suggest that maximum lifespan is not merely a statistical artifact of demographic data, but an objective, measurable, and potentially modifiable feature of human physiology.
Understanding the physics and biology behind this upper bound on lifespan is critical for evaluating the potential of longevity interventions. Our theory identifies three primary levers for intervention, which we classify into three distinct levels based on their potential effect size [4].
- Level-1 therapies target key molecular hallmarks of aging. These include CR mimetics, cellular rejuvenation therapies, senolytic therapies, telomere activators, and most other areas of research currently in the drug development pipeline. These work well in short-lived organisms where aging is unstable and markers of aging are tightly coupled. Level-1 therapies hold significant promise for addressing individual age-related diseases, including those with the largest market potential. Of those, diabetes alone reduces human lifespan by up to 8 years (depending on the age at diagnosis). This is why we expect that drugs aimed at improving metabolic health are expected to deliver the greatest benefits in this category.
- We predict the emergence of a new class of drugs, Level-2 therapies, designed to reduce physiological noise: the random fluctuations that destabilize health as organisms approach the limits of resilience. By damping this noise, these therapies could decouple aging from the onset of diseases, effectively “squaring” the survival curve. In practical terms, Level-2 interventions could add 30-40 years of healthy life by bridging the gap between today’s average lifespan (70-80 years, depending on the country) and the maximum natural lifespan of 120-150 years. However, they would not significantly extend the maximum lifespan itself.
- Level-3 therapies would aim to halt—or, as some hope, reverse—the accumulation of entropic damage itself. These therapies would not merely extend life; they would arrest functional decline and push maximum lifespan beyond the current 120-150 year limit. Because entropic damage accumulates slowly, future experiments and clinical trials will require the development of actionable biomarkers to track it. Targeting or controlling this damage will be challenging and will likely demand novel technologies, such as advanced organ replacement and new animal models of aging. Nevertheless, our theoretical framework provides a solid analytical foundation and brings these ambitious goals within conceptual reach.
Without a clear theoretical understanding of the aging process, drug development efforts often fall into the trap of focusing narrowly on specific disease indications. This is why most researchers, investors, and entrepreneurs in longevity biotechnology are currently centered on Level-1 therapies. These interventions may delay the onset or progression of individual diseases and modestly improve healthspan. However, they will not alter the fundamental dynamics of aging or extend the maximum lifespan.
The theoretical picture sends a dire warning. Level-1 biology addresses diseases in humans but has only a modest effect on lifespan. Level-2 interventions may further reduce the incidence of diseases and mortality. However, neither Level-1 nor even Level-2 therapies alone can alter the rate of functional decline. Aging is not merely the sum of diseases. Even in the absence of illness, humans grow increasingly fragile over time. A 90-year-old free of disease remains a diminished, less resilient version of their younger self.
The reason is that irreversible damage accumulates over time, leading to the progressive and likely irreversible decline of key functional indicators such as IQ, VO₂max, kidney function, and others that collectively define physiological resilience and quality of life. While squaring the curve, extending healthspan toward maximum lifespan, could significantly prolong life compared to current averages, it would not intercept the aging process itself. Without directly addressing underlying decline, it risks becoming a path to prolonged disability rather than true rejuvenation.
Only a combination of Level-2 therapies that decouple aging from disease—and, even more critically, Level-3 therapies that target or reverse damage accumulation—will extend lifespan and preserve function, opening the path to negligible senescence in our species.
The era of low-hanging fruit is coming to an end. Rather than chasing isolated hallmarks of aging or targeting individual diseases, we must now approach aging as a system-level, entropy-driven process. My scientific aspirations are firmly at Level-3, but my instincts tell me that Level-2 therapies, those that reduce biological noise, can be discovered and developed into real medicines much sooner with today’s technology.
This is the current focus of our research and development at Gero. We are investigating the biology of physiological noise using longer-lived mammals, such as dogs, as model organisms. Whether it is us or another research team that ultimately cracks the code of Level-2 or Level-3 interventions, true success will come only through a comprehensive understanding of the aging process and by raising the bar for what we expect from interventions. Without that foundation, no amount of billions spent will carry us much farther than the same old wisdom Genghis Khan got almost exactly 800 years ago.
Literature
[1] Von Foerster, H., Mora, P. M., & Amiot, L. W. (1960). Doomsday: Friday, 13 November, AD 2026: At this date human population will approach infinity if it grows as it has grown in the last two millenia. Science, 132(3436), 1291-1295.
[2] Pyrkov, T. V., Avchaciov, K., Tarkhov, A. E., Menshikov, L. I., Gudkov, A. V., & Fedichev, P. O. (2021). Longitudinal analysis of blood markers reveals progressive loss of resilience and predicts human lifespan limit. Nature communications, 12(1), 2765.
[3] Tarkhov, A. E., Denisov, K. A., & Fedichev, P. O. (2024). Aging clocks, entropy, and the challenge of age reversal. AgingBio, 2, e20240031.
[4] Denisov, K. A., Gruber, J., & Fedichev, P. O. (2024). Discovery of Thermodynamic Control Variables that Independently Regulate Healthspan and Maximum Lifespan. bioRxiv, 2024-12.
