Skin can look older or younger than a person’s chronological age, but visible appearance is only one layer of the story. A newer question is whether molecular ageing signals in skin can separate changes associated with internal ageing from those linked more strongly to accumulated environmental exposure.
In September 2026, researchers from King’s College London, Erasmus MC, Unilever and collaborating institutions reported two purpose-built DNA-methylation clocks: one aimed at endogenous skin ageing and one aimed at exogenous ageing, especially photoageing.
The study is particularly relevant to TwinPare because TwinsUK/MuTHER skin data were used for validation, sun-seeking analyses and gene-expression work. But the strongest exposed-versus-protected skin comparisons came from separate donor-matched datasets — an important distinction for interpreting what the twin data actually add.
Quick answer
Can skin be biologically “older” than your chronological age?
A skin epigenetic clock can estimate an age-related molecular pattern that differs from chronological age, and the new exogenous clock was sensitive to sun-linked differences. But that estimate is model-specific: it is not a single definitive biological age for your skin or your body.
Key takeaways
- The researchers trained separate Endogenous Skin and Exogenous Skin DNA-methylation clocks using more than 400 skin samples.
- The Exogenous Skin clock estimated significantly older ages in sun-exposed than donor-matched sun-protected epidermis and cultured fibroblasts in independent validation datasets.
- In 414 older female TwinsUK participants, the exogenous clock was tested in abdominal whole-skin methylation data; 242 had historical sun-seeking data and 364 had overlapping RNA-seq profiles.
- Exogenous age acceleration correlated modestly with sun-seeking behavior and with facial photodamage in a separate 44-person skin-swab sample.
- The endogenous clock tracked chronological age more accurately, reinforcing that “skin age” depends on what the clock was designed to measure.
- The clocks are subject to a patent held by Unilever, several authors had Unilever affiliations, and Unilever funded multiple component datasets and parts of the clock work.
One tissue, two kinds of ageing
Skin ageing is not driven by one process. Endogenous ageing reflects changes arising within the body and over chronological time. Exogenous ageing reflects accumulated external exposure, with ultraviolet radiation among the most important influences on exposed skin.
The 2026 study asked whether those two components could be separated molecularly. Instead of building one general skin clock, the researchers constructed an Endogenous Skin clock and an Exogenous Skin clock from different DNA-methylation patterns.
How the two clocks were built
The Endogenous Skin clock was trained using CpG methylation sites that changed with age in sun-protected skin but were not strongly different between exposed and protected sites. Its final model used 25 CpG sites.
The Exogenous Skin clock was built from CpG sites selected for age-related methylation changes in sun-exposed skin and used 24 CpG sites. The aim was not simply to predict a person’s birthday, but to make the model more sensitive to the molecular signature of accumulated external ageing.
Sun-exposed skin looked older to the exogenous clock
In independent donor-matched validation data, the Exogenous Skin clock produced significantly older age estimates for sun-exposed than sun-protected epidermis. A cultured fibroblast dataset from four younger and four older donors showed the same direction, with fibroblasts from protected sites estimated as younger than those from exposed sites.
This is an important within-donor comparison because it reduces some between-person differences. But it is not a randomized lifetime sun experiment, and it is not the TwinsUK co-twin comparison. Body site, tissue layer, cell culture and cumulative exposure can all influence the observed difference.
Where TwinsUK enters the story
The researchers also analysed 414 whole-skin DNA-methylation profiles from older female TwinsUK participants in the MuTHER project. These biopsies came from an abdominal area close to the umbilicus — a relatively sun-protected site.
Historical sun-seeking questionnaire data were available for 242 participants. Exogenous Skin age acceleration correlated with sun-seeking behavior, but the association was modest (rho about 0.18). This is useful validation, not a precise personal sun-damage calculator.
For 364 TwinsUK samples with overlapping RNA sequencing, 202 genes showed expression significantly associated with exogenous age acceleration after multiple-testing correction. Twin-pair relatedness was modeled as a random effect. That strengthens the molecular interpretation while still leaving open exactly which pathways are causal, consequential or correlated.
A separate facial-swab test linked the clock to visible photodamage
In 44 facial skin-swab samples from Rotterdam, higher Exogenous Skin age acceleration correlated with higher photodamage grading (rho 0.33). That result connects the molecular signal with a visible skin phenotype using a non-invasive sample.
The sample was small and the association was moderate. It is therefore encouraging validation rather than evidence that the clock can yet provide a clinically meaningful personal skin-age score.
The endogenous clock answered a different question
The Endogenous Skin clock and the established Skin and Blood clock tracked chronological age more closely than the Exogenous Skin clock. Their age-acceleration estimates were not significantly associated with the tested sun-exposure parameters or skin mRNA levels.
That contrast is the central result: a clock optimized for chronological or intrinsic ageing and a clock optimized for exogenous ageing do not necessarily rank people or samples the same way. “Biological age” is therefore not one universal number that every clock is measuring imperfectly.
What this does — and does not — say about sunlight
The study supports the idea that cumulative sun-linked exposure can leave DNA-methylation patterns that a purpose-built skin clock can detect. It does not tell us that sunlight adds an exact number of years to a person’s face, nor does it separate every environmental exposure from every internal ageing process.
The main sun-seeking measure was self-reported and had been collected roughly a decade before the skin biopsy. Several validation datasets were small, and further independent validation across age groups, ancestries, body sites, skin types and exposure histories is needed.
Why the funding, patent and affiliations matter
The authors report that the Endogenous and Exogenous Skin clocks are subject to a patent held by Unilever. David Gunn, Y. Jeremy Shen and Laura Ramos del Caño had Unilever affiliations, and Christopher Shore was supported by a BBSRC–Unilever iCASE PhD studentship.
Unilever also funded several component datasets, the facial-swab epigenetic study and support of the clock work. These disclosures do not invalidate the results, but they are relevant when interpreting a biomarker with potential clinical and cosmetic applications. Independent external replication will be particularly valuable.
The TwinPare boundary
TwinPare Research can use this study to explain how environmental exposure and intrinsic ageing can leave partly different molecular signatures, and why TwinsUK is valuable for relating molecular skin data to behavior and gene expression.
TwinPare Health & Fitness does not currently measure epigenetic skin age. A photo, wearable, sleep score or routine lifestyle log cannot be translated into one of these DNA-methylation clocks without the underlying molecular assay and validated model. The research is relevant to future scientific understanding, not a current TwinPare diagnostic or cosmetic-age feature.
Source notes
The source has been verified and editorially reviewed for this article. The limitations below show which level of conclusion the sources support.
- [shore-skin-clocks-2026] Epigenetic clocks can differentiate between exogenous and endogenous skin ageing Christopher J. Shore; Roy B. Simons; Laura Ramos del Caño; Ziyi Xiong; Luba Milena Pardo; Christina M. W. Eecen; Tamar Nijsten; Kerrin S. Small; Athina Vidaki; Manfred Kayser; Jordana T. Bell; Y. Jeremy Shen; David A. Gunn. Clinical Epigenetics, 2026. Evidence type: Development and multi-dataset validation of separate endogenous and exogenous skin DNA-methylation clocks, including TwinsUK/MuTHER skin methylation and transcriptomic data Limitation: Published 11 September 2026. The exogenous clock distinguished sun-exposed from donor-matched protected skin in independent datasets and was associated with sun-seeking, skin gene expression and facial photodamage. Several validation samples were small, associations were modest, further external validation is needed, and the clocks are subject to a Unilever-held patent. Multiple authors were Unilever employees or received Unilever-linked support. PubMed DOI
Editorial source review
This section shows how the article's key factual claims are linked to the source.
Phrasings that require caution
- Do not write that the sun adds an exact number of years to a person’s face.
- Do not describe the Exogenous Skin clock as a universal or clinically validated “true skin age”.
- Do not describe the TwinsUK analysis as an identical-twin sun-exposure experiment.
- Do not imply that sun-seeking caused the observed methylation or gene-expression changes.
- Keep the modest sun-seeking correlation and small facial-swab sample visible.
- Do not generalize the older-female TwinsUK skin sample to all ages, sexes, ancestries or skin types.
- Disclose the Unilever patent, affiliations and funding when presenting the study.
- Do not imply TwinPare can estimate epigenetic skin age from photos, wearables or ordinary app data.
| ID | Claim | Source support | Caution |
|---|