Protein damage could provide a valuable measure of how potential anti-ageing therapies perform, but turning complex biological changes into a single score raises important questions for drug developers.

3D molecular illustration of a folded protein structure showing alpha helices and beta sheets.

Longevity research has a measurement problem. An experiment can produce a convincing molecular signal without answering the question that matters for development: is this candidate worth taking further? Researchers need ways to compare results and choose the next experiment. A carefully defined score can help make that decision, provided its meaning remains attached to the biology behind it.

Ageing biology is too complex to be captured by one number. But a specific response associated with ageing can be summarised in a way that makes it easier to interpret. That distinction should guide how we develop, validate and communicate longevity metrics.

What should a score represent?

The hallmarks of ageing describe an interconnected set of processes, including loss of proteostasis, mitochondrial dysfunction, cellular senescence and impaired autophagy.1 They provide a map for research, rather than a universal scale on which every intervention can be ranked.

A score becomes useful when it addresses narrower questions. What changed after treatment? In which tissue? Under what challenge? Compared with which control? Without those definitions, a number can appear precise while leaving its biological meaning uncertain.

Consider two experimental candidates. One produces a strong effect in a selected assay; the other produces a smaller but more consistent effect. A development team must decide whether to prioritise magnitude, reproducibility or a mechanism aligned with the intended application. A summary metric can organise that discussion. The underlying results must remain available to resolve it.

Measuring ageing in drug discovery

Experimental readouts of ageing-associated processes include measurements of DNA damage, mitochondrial function, cellular senescence and protein integrity. The selection of appropriate endpoints depends on the proposed mechanism of action and the biological model under investigation.

Protein damage as a research readout

Protein damage offers one example of a measurable process relevant to ageing. Research has described the accumulation of oxidatively modified proteins and the involvement of protein maintenance, energy metabolism and other cellular pathways.2 This supports examining protein integrity as part of a broader experimental strategy.

An experimental score could summarise a treatment-associated change in a defined damage readout. Researchers would need to specify the model, challenge, comparator and calculation. If several measurements contribute, the weighting should reflect a stated biological rationale. The result should retain a specific interpretation: an effect on the measured process under the conditions tested.

That interpretation creates a practical starting point for research. It also sets a boundary. The result does not establish a person’s biological age or predict lifespan. Nor should an effect in a skin model be presented as proof of benefit in another organ or in a patient.

A published example co-authored with L’Oréal illustrates the approach.3 Human skin explants were pretreated for 48 hours with a topical formulation of a salicylic acid derivative, then challenged with particulate matter and UVA. Protein carbonylation was then quantified by fluorescent labelling in tissue sections. The challenge approximately doubled carbonylation in whole skin relative to unstressed explants and pretreatment brought it back close to baseline levels. The effect was significant in the epidermis and dermis but not in the stratum corneum.

Making the number useful for development

The strongest case for a score is its ability to improve a decision. Early development involves choosing among candidates before committing to more demanding studies. A common readout could help a team identify which candidates warrant investigation, especially when experiments use comparable conditions.

For drug developers, this is a broader methodological lesson rather than evidence that a skin assay can evaluate every longevity therapy. Any application to a therapeutic programme would need a model, exposure and endpoint suited to the disease and mechanism under investigation.

A promising result should generate a testable next question. Does the response persist at relevant concentrations? Is it reproducible across independent donors? Does it accompany a functional improvement? Is it consistent with the proposed mechanism? These questions turn an initial signal into a development strategy.

They also help interpret disappointing results. A candidate that performs poorly in one oxidative-damage assay may act through a different pathway. Selecting compounds solely by a single score could discard useful biology. The metric should inform prioritisation alongside mechanism and other evidence.

The evidence behind the ranking

Standardisation requires more than putting results on the same numerical scale. Researchers need to understand how measurements behave across runs, operators, donors and laboratories. They should also understand the normalisation method and how individual measurements are combined into a single score. For example, the stratum corneum, epidermis and dermis can respond differently to the same treatment, so researchers need to understand how measurements from each compartment are weighted. They should also establish whether reference comparisons were generated under equivalent conditions.

Uncertainty must be visible, too. If two scores differ only slightly, that difference may be too small to support a ranking. Replicate structure matters: repeated measurements of the same tissue do not provide the same evidence as independent biological samples. A useful report should make that distinction clear.

As a metric develops, independent replication and transparent validation should become priorities. The field needs evidence that using the score improves candidate selection, not simply evidence that a number can be calculated consistently. Ultimately, researchers should ask whether it adds information beyond the underlying assay results.

Communicating longevity without outrunning the data

A simple number is attractive because it travels easily beyond the laboratory. That makes precise language essential. A claim about reduced protein damage in a defined experimental model should remain a claim about that result. Assertions about tissue rejuvenation, disease prevention or longer life require evidence designed to answer those separate questions.

This discipline can strengthen the case for longevity research. Clear limits make results easier to evaluate and give developers a credible basis for further work. They also allow readers to distinguish an experimental lead from a demonstrated benefit.

Can ageing biology be reduced to a single score? A score can summarise a defined part of it and make that result more useful. Its scientific value depends on whether researchers can trace the number back to the experiment, understand its uncertainty and use it to choose a better next step. That is a worthwhile ambition for longevity measurement.

References

1. López-Otín C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: an expanding universe. Cell. 2023;186(2):243-278. doi:10.1016/j.cell.2022.11.001

2. Baraibar MA, Friguet B. Oxidative proteome modifications target specific cellular pathways during oxidative stress, cellular senescence and aging. Exp Gerontol. 2013;48(7):620-625. doi:10.1016/j.exger.2012.10.007

3. Gras Lavigne E, Cavagnino A, Steinschneider R, Breton L, Baraibar MA, Jäger S. Oxidative damage prevention in human skin and sensory neurons by a salicylic acid derivative. Free Radic Biol Med. 2022;181:98-104. doi:10.1016/j.freeradbiomed.2022.01.029