Abstract
Background/Aim: Actinic keratosis (AK) is a common skin condition associated with cumulative sun exposure and advancing age. As a precursor to squamous cell carcinoma, it is clinically relevant as a target for prevention. While epidemiological studies have suggested various risk factors, causal inference is often limited by confounding. Mendelian randomization (MR), which uses genetic variants as proxies for exposures, can help address this. This review aimed to provide an overview of published MR studies that have examined AK either as an exposure or an outcome.
Materials and Methods: We conducted a systematic literature search in MEDLINE and Embase from inception to August 2025. Eligible studies were original human research employing MR to investigate associations involving AK and any health-related trait.
Results: Eight studies were included, all of which applied two-sample MR using summary-level genome-wide association data, primarily from European populations. The studies explored a variety of exposures, including metabolic traits, immune cell characteristics, dermatological conditions, and sun sensitivity. One study examined AK as an exposure to assess its association with skin cancers. Most analyses were exploratory, and no exposure–outcome pair was studied in more than one paper, limiting conclusions and precluding meta-analysis.
Conclusion: The body of MR research concerning AK is limited but methodologically diverse. Although the current evidence base is small, this review highlights how MR has been applied to AK and may inform directions for future investigation.
Introduction
Actinic keratosis (AK) commonly occurs in older individuals, particularly those with fair skin and a history of prolonged sun exposure (1). It presents as rough, scaly lesions on sun-exposed skin and is considered a precursor to squamous cell carcinoma, underscoring its clinical importance (1-3). While AK is primarily linked to ultraviolet radiation, the broader biological and clinical context in which AK occurs is not fully understood (1, 4). Traditional epidemiological studies have explored various environmental, genetic, and lifestyle associations with AK (1, 4). However, these studies often face challenges in establishing causality due to confounding factors and reverse causation.
In recent years, Mendelian randomization (MR) has emerged as a tool for exploring causal relationships between genetic proxies for exposures and health outcomes (5, 6). By using genetic variants as instrumental variables, MR analyses reduce bias from confounding (5, 6). This makes it a tool for identifying exposures that may play a causal role in disease development and for evaluating potential targets for further research. This review aimed to provide a comprehensive synthesis of MR studies that have examined any potential causal relationship involving AK.
Materials and Methods
Searches were conducted in MEDLINE and Embase from database inception to August 28, 2025. The searches combined AK-related and MR-related terms. Strategies are shown in Supplementary Table I. Additionally, reference lists of included papers were screened for any relevant studies not initially identified.
Mendelian randomization studies on actinic keratosis (AK).
To be included in the review, studies were required to use the MR analytical approach to investigate associations involving AK and any health-related outcome or trait in humans. Studies not employing MR methodology, without AK, or using non-human data were excluded. Additionally, review articles, editorials, commentaries, or conference abstracts lacking original data were excluded.
Studies clearly not using MR methodology or not focusing on AK were excluded in the initial screening phase. Full texts of potentially eligible studies were reviewed to confirm inclusion. Data were extracted using a standardized template and included study design, exposure and outcome definitions, data sources, and main results. Due to heterogeneity in exposures and outcomes across studies, no quantitative synthesis or meta-analysis was conducted. Instead, findings were summarized narratively, with an emphasis on the methodological characteristics and reported associations. The data work was conducted in duplicate by IH and FE.
Results
The searches identified 17 studies, of which 8 were duplicates. This resulted in 9 unique studies subjected to full-text review (7-15). One study was excluded as it did not include a MR study (15). Eight studies were included in the review (7-14). A summary of all included studies is provided in Table I. The studies were published between 2022 and 2025. In 7 studies, AK served as the outcome (7-13). One study treated genetic liability AK as the exposure (14). All included studies applied two-sample MR designs using genome-wide association study (GWAS) data, primarily from populations of European ancestry. The three main GWAS data sources used for AKs were FinnGen (16), UK Biobank (17), and the Genetic Epidemiology Research on Adult Health and Aging cohort (18). A complete overview of all data sources used across the included studies is in Supplementary Table II.
Two studies investigated systemic metabolic traits (8, 12). Chen et al. (8) found that genetic liability to type 1 and type 2 diabetes was associated with a reduced risk of developing actinic cheilitis, but the data actually cover AK. Shao et al. (12) analyzed 1,400 genetically predicted plasma metabolites and identified 155 associated with AK. The metabolite with the strongest effect was 5-hydroxy-2-methylpyridine sulfate, although the findings were exploratory and not replicated. In the context of immune and inflammatory factors, Yin et al. (13) assessed 731 immune cell characteristics and identified four that were inversely associated with AK, including HLA-DR expression on monocytes and dendritic cells, suggesting protective immune modulation. Two studies investigated the chronic inflammatory skin conditions atopic dermatitis and rosacea, respectively, and found no evidence of causal effects on the development of AK (10, 11). One study investigated the skin microbiome and found no association with AK (7). Ping et al. (9) found a positive association between genetically predicted self-reported tanning response to sun exposure and AK. This trait was based on participants’ reports in the UK Biobank on how their skin reacts to sunlight, with higher tanning ability (i.e., tanning brown rather than burning red) linked to increased AK risk. Finally, Qian et al. (14) investigated AK as the exposure and found that genetic liability to AK was associated with increased risk of three types of skin cancer: melanoma, basal cell carcinoma, and squamous cell carcinoma. An overview of reported associations is in Supplementary Table III.
Discussion
This review provides an overview of existing MR studies that investigate potential causal relationships involving AK. Eight studies were included. The studies applied GWAS data on AK from FinnGen, UK Biobank, and the Genetic Epidemiology Research on Adult Health and Aging cohort (16-18). Together, these represent genetic resources that may be leveraged in future MR investigations of AK.
MR relies on three core assumptions: first, the genetic variants used as instruments must be strongly associated with the exposure of interest; second, these variants must not be associated with confounders of the exposure-outcome relationship; and third, the genetic variants must influence the outcome solely through the exposure and not via alternative biological or behavioral pathways (5, 6, 19). Violations of any of these assumptions can bias causal inference and must be considered when interpreting MR findings (5, 6, 19).
One of the included studies, Ping et al. (9), may be particularly susceptible to violation of these assumptions. The study found that genetic liability to increased tanning response was associated with a higher risk of developing AK (9). An association that is not intuitively expected, given that AK primarily occurs in individuals with a low tanning response (4). A possible explanation is that individuals who tan easily may experience more cumulative sun exposure over time due to behavioral patterns, such as spending more time outdoors or using less sun protection. If this is the case, the exclusion restriction assumption may be violated, as the genetic instrument would affect the outcome not solely via the biological exposure (tanning response), but also through correlated behaviors. This demonstrates the importance of careful interpretation, especially when exposures are behaviorally modulated.
Conclusion
In conclusion, while only a few MR studies on AK have been published, this review summarizes current evidence and methodological approaches. With AK GWASs available, there is potential for further MR analyses to investigate causal relationships and guide future research.
Footnotes
Authors’ Contributions
IH and HH designed the study. IH and FE collected and analyzed data. IH drafted the manuscript, and all authors revised and approved the final version.
Supplementary Material
Available as: Heerfordt, Ida M.; Esmann, Frederik Viggo Lautrup; Horwitz, Henrik (2025), “Supplementary Material for Mendelian Randomization Studies on Actinic Keratosis”, Mendeley Data, V1, doi: 10.17632/m2v7xzx7gj.1.
Conflicts of Interest
None.
Funding
None.
Artificial Intelligence (AI) Disclosure
During the preparation of this manuscript, a large language model (ChatGPT-4.5, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors.
- Received August 9, 2025.
- Revision received August 30, 2025.
- Accepted September 2, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.