Ahead of Longevity Investors Conference, Steve Horvath discusses how to separate genuine aging science from persuasive biological storytelling.
The Longevity Investors Conference (LIC) returns to Gstaad, Switzerland, next month, once again bringing together investors, founders, clinicians and researchers to examine the scientific advances and commercial opportunities shaping the rapidly evolving longevity sector. As the field matures beyond early promise, the conference has increasingly become a forum where capital and geroscience meet – a place where ambitious ideas are expected to withstand equally ambitious scrutiny.
A highlight of this year’s program is a masterclass entitled How to Tell the Difference – Spotting Real Aging Science, led by Dr Steve Horvath, Professor of Medicine in the Division of Geriatrics at the David Geffen School of Medicine at UCLA, alongside Dr Collin Ewald and Professor Evelyne Bischof. Few researchers have had a greater influence on modern geroscience than Horvath, whose pioneering work on epigenetic aging clocks transformed biological age from a theoretical concept into a measurable feature of human biology.
Longevity.Technology: Longevity science is entering an era in which the limiting factor is no longer generating rejuvenation claims, but distinguishing meaningful biological change from persuasive biological storytelling. The sector now has clocks, biomarkers and molecular readouts capable of detecting shifts that would once have been invisible; what remains rather less obliging is the question of whether those shifts preserve tissue function, alter disease trajectories or ultimately matter to the person whose biology is being so meticulously quantified. This is precisely why gatherings such as the Longevity Investors Conference matter – bringing capital into the same room as geroscientists does more than facilitate investment; at its best, it creates the intellectual friction required to test whether an elegant mechanism can survive translation from model to medicine, and whether an enticing data point deserves to become a therapeutic thesis.
Few people are better placed to impose some discipline on that conversation than Steve Horvath, whose work transformed biological age from an intriguing abstraction into something we could begin to measure – and who has remained gratifyingly clear-eyed about what those measurements can and cannot yet tell us. There is, admittedly, something rather reassuring about having one of the architects of the field gently confiscate its crayons; real aging science will not be identified by the boldness of the claim, but by the quality of the evidence connecting molecular change to meaningful function. For investors attempting to separate signal from an increasingly well-funded quantity of noise, that distinction is not academic – it is the difference between backing a biomarker that moved and backing biology that matters. Ahead of the conference, Longevity.Technology spoke with Horvath about biomarkers, causality and the limits of partial cellular reprogramming.
Function first, fanfare later
For investors evaluating early-stage longevity therapeutics, biomarkers have become indispensable. But Horvath argues that the first question should never be whether an aging clock moved, but whether biology did.
“There is no single number but if I had to reduce it to one rule it would be this,” he says. “Ask whether the treated animal or person actually functions better, in an old but otherwise healthy tissue and ask whether the biomarker change is carrying that improvement rather than simply sitting next to it.”
That distinction, he argues, is fundamental rather than academic. “Medicine has paid for this lesson before. A marker that correlates with an outcome has not yet earned the right to stand in for it.”
Horvath also urges investors to look beyond headline biomarker improvements and ask whether findings remain consistent across different aging clocks. If a result appears robust regardless of which clock is used, confidence naturally increases; if it disappears when measured differently, that too is informative. He also points to pan-mammalian clocks, which measure conserved DNA methylation sites across species – putting mouse and human on the same scale, so that a preclinical result and a clinical one are actually comparable.
When the clock behaves too well
The same principle underpins what Horvath describes as the “biomarker paradox”. If an intervention alters an epigenetic clock yet fails to preserve function or improve long-term outcomes, the most likely explanation, he says, “is the least flattering one.”
“The marker was tracking a consequence of aging rather than a driver of it, so the intervention moved the readout without moving the underlying biology.”
Before reaching that conclusion, however, he cautions against overlooking more mundane explanations – measuring the wrong tissue, following patients for too short a period, studies that are simply underpowered or changes in cell composition masquerading as biological rejuvenation.
“If all of that is excluded and the split still holds then yes, this is a genuine finding and worth publishing rather than quietly dropping,” he says. “It is also the reason I keep saying a clock is a screening and prioritisation tool and not a verdict.”
Three rungs, and most stand on the first
Consumer longevity diagnostics are proliferating, while the biology of aging itself remains an active area of scientific debate. For Horvath, the answer is not to constrain exploratory science but to be much clearer about what companies are actually claiming.
“I would not draw the boundary around the science,” he says. “I would draw it around the claim.”
Exploration, he argues, is the lifeblood of scientific progress. The problem begins when preliminary measurements are marketed as established clinical benefits.
Instead, Horvath proposes a simple hierarchy. First, does a test reliably measure what it claims to measure? Second, does that measurement predict something meaningful in a defined population? And finally, does changing that measurement actually change a person’s future health?
“Most consumer longevity products can support the first, sometimes the second, and essentially never the third,” he says. “That is fine provided they say so plainly rather than letting the marketing imply the third.”
Far from slowing innovation, he believes scientific honesty ultimately strengthens the field.
“A field is stifled when it loses credibility and the serious capital leaves,” he says. “Asking companies to be explicit about which rung they are standing on is the cheapest insurance we have against that.”
The long climb to causation
Epigenetic clocks have transformed aging research, but they have also raised one of the field’s biggest unanswered questions. Do methylation changes actively drive aging, or do they simply record its passage?
“Further off than the headlines suggest,” Horvath says when asked how close we are to truly causal aging clocks. “Mainly because the word ‘causal’ is doing several different jobs at once.”
He points to Mendelian randomization studies as one useful source of evidence, while noting their limitations. Genetic predispositions present from birth are not necessarily equivalent to the methylation changes accumulated later in life.
The more convincing experiment, he argues, remains largely undone.
“The stronger version asks whether changing the methylation at a site changes the cell’s or the tissue’s behaviour. That is the question that would settle the matter, but it has only been attempted at a handful of loci and rarely in aged tissue with a functional readout.”
Rather than viewing causality as a binary concept, Horvath describes a ladder of evidence – beginning with population-level correlations, progressing through longitudinal human data and genetic evidence before finally reaching direct manipulation and clinical trials demonstrating both clock changes and meaningful patient outcomes.
“The field is on the lower rungs,” he says. “Climbing the rest is probably a matter of years rather than months.”
Don’t ask the clock to carry the argument
Partial cellular reprogramming has become one of longevity biotechnology’s most closely watched frontiers, but Horvath believes the discussion extends beyond any single therapeutic approach.
“I would widen the question slightly because it applies to more than reprogramming,” he says.
Current aging clocks perform particularly well in interventions that produce large, biologically coherent effects, including caloric restriction, disruption of growth hormone signaling and heterochronic parabiosis in mice. They become less reliable, however, when expected effects are small, measurements are made in cultured cells or changes in cell composition complicate interpretation.
“Several therapeutic approaches, including reprogramming-based ones, sit in more than one of those categories at once,” he explains. “That is a statement about measurement difficulty and not about whether the biology works.”
Rather than expecting increasingly sophisticated clocks to answer every question, Horvath argues for combining molecular measurements with complementary approaches.
“The adaptation is less about a cleverer clock and more about not asking the clock to carry the whole argument,” he says. “Measure at single cell resolution, or estimate and adjust for the cell mixture so composition and intrinsic change can be told apart. Report the uncertainty rather than a single number. And always pair the molecular readout with a functional one in aged tissue.”
Ultimately, he says, it is the convergence of evidence that matters.
“If those two agree then you have something. If they disagree, you have a question worth answering, and the honest move is to say so.”
The 2026 Longevity Investors Conference takes place from 14–17 September at Le Grand Bellevue in Gstaad, Switzerland. For more information or to apply, please CLICK HERE.
