Predicting skin cancer risk from facial images with an explainable artificial intelligence (XAI) based approach: a proof-of-concept study.

Background: Efficient identification of individuals at high risk of skin cancer is crucial for implementing personalized screening strategies and subsequent care. While Artificial Intelligence holds promising potential for predictive analysis using image data, its application for skin cancer risk prediction utilizing facial images remains unexplored. We present a neural network-based explainable artificial intelligence (XAI) approach for skin cancer risk prediction based on 2D facial images and compare its efficacy to 18 established skin cancer risk factors using data from the Rotterdam Study. Methods: The study employed data from the Rotterdam population-based study in which both skin cancer risk factors and 2D facial images and the occurrence of skin cancer were collected from 2010 to 2018. We conducted a deep-learning survival analysis based on 2D facial images using our developed XAI approach. We subsequently compared these results with survival analysis based on skin cancer risk factors using cox proportional hazard regression. Findings: Among the 2810 participants (mean Age = 68.5 ± 9.3 years, average Follow-up = 5.0 years), 228 participants were diagnosed with skin cancer after photo acquisition. Our XAI approach achieved superior predictive accuracy based on 2D facial images (c-index = 0.72, 95% CI: 0.70-0.74), outperforming that of the known risk factors (c-index = 0.59, 95% CI 0.57-0.61). Interpretation: This proof-of-concept study underscores the high potential of harnessing facial images and a tailored XAI approach as an easily accessible alternative over known risk factors for identifying individuals at high risk of skin cancer. Funding: The Rotterdam Study is funded through unrestricted research grants from Erasmus Medical Center and Erasmus University, Rotterdam, Netherlands Organization for the Health Research and Development (ZonMw), the Research Institute for Diseases in the Elderly (RIDE), the Ministry of Education, Culture and Science, the Ministry for Health, Welfare and Sports, the European Commission (DG XII), and the Municipality of Rotterdam. G.V. Roshchupkin is supported by the ZonMw Veni grant (Veni, 549 1936320).

The Rotterdam Study. Design update and major findings between 2020 and 2024.

The Rotterdam Study is a population-based cohort study, started in 1990 in the district of Ommoord in the city of Rotterdam, the Netherlands, with the aim to describe the prevalence and incidence, unravel the etiology, and identify targets for prediction, prevention or intervention of multifactorial diseases in mid-life and elderly. The study currently includes 17,931 participants (overall response rate 65%), aged 40 years and over, who are examined in-person every 3 to 5 years in a dedicated research facility, and who are followed-up continuously through automated linkage with health care providers, both regionally and nationally. Research within the Rotterdam Study is carried out along two axes. First, research lines are oriented around diseases and clinical conditions, which are reflective of medical specializations. Second, cross-cutting research lines transverse these clinical demarcations allowing for inter- and multidisciplinary research. These research lines generally reflect subdomains within epidemiology. This paper describes recent methodological updates and main findings from each of these research lines. Also, future perspective for coming years highlighted.

Association between lifetime smoking and cutaneous squamous cell carcinoma: A 2-sample Mendelian randomization study.

Background/Purpose: Cutaneous squamous cell carcinoma (cSCC) is one of the most common malignancies worldwide. While several environmental risk factors for cSCC are well established, there is conflicting evidence on cigarette smoking (and its potential causal effect) and cSCC risk. Furthermore, it is unclear if these potential associations represent causal, modifiable risk factors for cSCC development. This study aims to assess the nature of the associations between cigarette smoking traits (smoking initiation, amount smoked, and lifetime smoking exposure) and cSCC risk using two-sample Mendelian randomization analyses. Methods: Genetic instruments, based on common genetic variants associated with cigarette smoking traits (P < 5 × 10-8), were derived from published genome-wide association studies (GWASs). For cSCC, we used GWAS summary statistics from the Kaiser Permanente GERA cohort (7701 cSCC cases and 60,167 controls; all non-Hispanic Whites). Results: We found modest evidence that genetically determined lifetime smoking was associated with cSCC (inverse-variance weighted method: OR[95% CI] = 1.47[1.09-1.98]; P = .012), suggesting it may be a causal risk factor for cSCC. We did not detect any evidence of association between genetically determined smoking initiation or amount smoked and cSCC risk. Conclusion: Study findings highlight the importance of smoking prevention and may support risk-stratified cSCC screening strategies based on carcinogen exposure and other genetic and clinical information.

Longitudinal Assessment of the Prevalence of Actinic Keratosis and Extensive Risk Factor Evaluation: An Update from the Rotterdam Study.

Population-based studies available to analyze the prevalence, risk factors, and longitudinal outlook of actinic keratoses (AKs) are limited. These features mentioned earlier were assessed using Rotterdam study participants aged ≥40 years who underwent a full-body skin examination by a dermatology-trained physician. ORs with 95% confidence intervals were calculated for the associations between risk factors and the presence of AK. Among 8,239 eligible participants, the prevalence of one or more AKs was 21.1% (95% confidence interval = 20.2-22.0) and was higher in men. Male sex, age, lighter hair and eye color, baldness, genetic risk score, and digital photoaging measures (digitally assessed pigmented spots, telangiectasias, and global facial wrinkling) had a positive association with AK. Cigarette smokers had reduced odds of having AK, with current smokers having the lowest risk. Among patients with two AK assessments, there was no difference in the presence of AK during follow-up between treated and untreated participants. In conclusion, genetic risk score and digital photoaging measures showed associations with increased lesion count. At the individual level, patients were most likely to decrease in AK severity group over time, possibly regardless of whether or not participants were treated.

Younger facial looks are associate with a lower likelihood of several age-related morbidities in the middle-aged to elderly.

BACKGROUND: Looking older for one's chronological age is associated with a higher mortality rate. Yet it remains unclear how perceived facial age relates to morbidity and the degree to which facial ageing reflects systemic ageing of the human body. OBJECTIVES: To investigate the association between ΔPA and age-related morbidities of different organ systems, where ΔPA represents the difference between perceived age (PA) and chronological age. METHODS: We performed a cross-sectional analysis on data from the Rotterdam Study, a population-based cohort study in the Netherlands. High-resolution facial photographs of 2679 men and women aged 51.5-87.8 years of European descent were used to assess PA. PA was estimated and scored in 5-year categories using these photographs by a panel of men and women who were blinded for chronological age and medical history. A linear mixed model was used to generate the mean PAs. The difference between the mean PA and chronological age was calculated (ΔPA), where a higher (positive) ΔPA means that the person looks younger for their age and a lower (negative) ΔPA that the person looks older. ΔPA was tested as a continuous variable for association with ageing-related morbidities including cardiovascular, pulmonary, ophthalmological, neurocognitive, renal, skeletal and auditory morbidities in separate regression analyses, adjusted for age and sex (model 1) and additionally for body mass index, smoking and sun exposure (model 2). RESULTS: We observed 5-year higher ΔPA (i.e. looking younger by 5 years for one's age) to be associated with less osteoporosis [odds ratio (OR) 0.76, 95% confidence interval (CI) 0.62-0.93], less chronic obstructive pulmonary disease (OR 0.85, 95% CI 0.77-0.95), less age-related hearing loss (model 2; B = -0.76, 95% CI -1.35 to -0.17) and fewer cataracts (OR 0.84, 95% CI 0.73-0.97), but with better global cognitive functioning (g-factor; model 2; B = 0.07, 95% CI 0.04-0.10). CONCLUSIONS: PA is associated with multiple morbidities and better cognitive function, suggesting that systemic ageing and cognitive ageing are, to an extent, externally visible in the human face.

Akkermansia muciniphila and Faecalibacterium prausnitzii in Immune-Related Diseases.

Probiotics and synbiotics are used to treat chronic illnesses due to their roles in immune system modulation and anti-inflammatory response. They have been shown to reduce inflammation in a number of immune-related disorders, including systemic lupus erythematosus (SLE), human immunodeficiency virus (HIV), and chronic inflammatory skin conditions such as psoriasis and atopic dermatitis (AD). Akkermansia muciniphila (A. muciniphila) and Faecalibacterium prausnitzii (F. prausnitzii) are two different types of bacteria that play a significant part in this function. It has been established that Akkermansia and Faecalibacterium are abundant in normal populations and have protective benefits on digestive health while also enhancing the immune system, metabolism, and gut barrier of the host. They have the potential to be a therapeutic target in diseases connected to the microbiota, such as immunological disorders and cancer immunotherapy. There has not been a review of the anti-inflammatory effects of Akkermansia and Faecalibacterium, particularly in immunological diseases. In this review, we highlight the most recent scientific findings regarding A. muciniphila and F. prausnitzii as two significant gut microbiota for microbiome alterations and seek to provide cutting-edge insight in terms of microbiome-targeted therapies as promising preventive and therapeutic tools in immune-related diseases and cancer immunotherapy.

NNT mediates redox-dependent pigmentation via a UVB- and MITF-independent mechanism.

Ultraviolet (UV) light and incompletely understood genetic and epigenetic variations determine skin color. Here we describe an UV- and microphthalmia-associated transcription factor (MITF)-independent mechanism of skin pigmentation. Targeting the mitochondrial redox-regulating enzyme nicotinamide nucleotide transhydrogenase (NNT) resulted in cellular redox changes that affect tyrosinase degradation. These changes regulate melanosome maturation and, consequently, eumelanin levels and pigmentation. Topical application of small-molecule inhibitors yielded skin darkening in human skin, and mice with decreased NNT function displayed increased pigmentation. Additionally, genetic modification of NNT in zebrafish alters melanocytic pigmentation. Analysis of four diverse human cohorts revealed significant associations of skin color, tanning, and sun protection use with various single-nucleotide polymorphisms within NNT. NNT levels were independent of UVB irradiation and redox modulation. Individuals with postinflammatory hyperpigmentation or lentigines displayed decreased skin NNT levels, suggesting an NNT-driven, redox-dependent pigmentation mechanism that can be targeted with NNT-modifying topical drugs for medical and cosmetic purposes.

Genome-wide meta-analysis identifies eight new susceptibility loci for cutaneous squamous cell carcinoma.

Cutaneous squamous cell carcinoma (SCC) is one of the most common cancers in the United States. Previous genome-wide association studies (GWAS) have identified 14 single nucleotide polymorphisms (SNPs) associated with cutaneous SCC. Here, we report the largest cutaneous SCC meta-analysis to date, representing six international cohorts and totaling 19,149 SCC cases and 680,049 controls. We discover eight novel loci associated with SCC, confirm all previously associated loci, and perform fine mapping of causal variants. The novel SNPs occur within skin-specific regulatory elements and implicate loci involved in cancer development, immune regulation, and keratinocyte differentiation in SCC susceptibility.

Association between Diet and Seborrheic Dermatitis: A Cross-Sectional Study.

Current treatments for seborrheic dermatitis provide only temporary relief. Therefore, identifying modifiable lifestyle factors may help reduce disease burden. The objective of this study was to determine whether specific dietary patterns or total antioxidant capacity are associated with seborrheic dermatitis. Participants of the Rotterdam Study with a skin examination and a food frequency questionnaire were included. Total antioxidant capacity was assessed on the basis of ferric reducing antioxidant potential of each food item. Dietary patterns were identified with principal component analysis (PCA). Multivariable logistic regression analyses were used to assess the association between total antioxidant capacity, dietary pattern-derived PCA factors, and seborrheic dermatitis adjusted for confounders. In total, 4,379 participants were included, of whom 636 (14.5%) had seborrheic dermatitis. The PCA identified vegetable, Western, fat-rich and fruit dietary patterns. The fruit pattern was associated with a 25% lower risk (quartile 1 vs. quartile 4: adjusted odds ratio = 0.76, 95% confidence interval = 0.58-0.97, P = 0.03), and the Western pattern with a 47% increased risk (quartile 1 vs. quartile 4: adjusted odds ratio = 1.47; 95% confidence interval = 0.98-2.20, P = 0.03), but only for females. Other factors were not associated with seborrheic dermatitis. In conclusion, a high fruit intake was associated with less seborrheic dermatitis, whereas high adherence to a "Western" dietary pattern in females was associated with more seborrheic dermatitis.

Prevalence and determinants for xerosis cutis in the middle-aged and elderly population: A cross-sectional study.

BACKGROUND: Determinants and the extent of dry skin in healthy middle-aged and elderly populations have not been well established. OBJECTIVE: We aimed to identify the prevalence and determinants for generalized dry skin (GDS) and localized dry skin (LDS) within a large prospective population-based cohort of middle-aged and elderly individuals of the Rotterdam Study. METHODS: Dry skin was physician-graded as none, localized, or generalized. For GDS and LDS, separate multivariable logistic regression analyses were performed to search for association with participant characteristics, lifestyle factors, environmental factors, several comorbidities, and drug exposure. RESULTS: Among the 5547 eligible participants, 60% had dry skin, of whom a fifth had GDS. Age, female sex, skin color, body mass index, outside temperature, eczema, and chemotherapy in the past were significant determinants for both GDS and LDS. Smoking, the use of statins and diuretics, poorer self-perceived health, and several dermatologic conditions increased the likelihood of having GDS only. Daily cream use was associated with less LDS. LIMITATIONS: Interobserver variability and residual confounding could have influenced our results. Because of our cross-sectional design, we could not infer causality. CONCLUSION: We identified factors significantly associated with dry skin in a general middle-aged and elderly population, with health parameters more strongly associated with GDS.

Association of BRCA2 K3326* With Small Cell Lung Cancer and Squamous Cell Cancer of the Skin.

Background: Most pathogenic mutations in the BRCA2 gene carry a high risk of hereditary breast and ovarian cancer (HBOC). However, a stop-gain mutation, K3326* (rs11571833), confers risk of lung cancer and cancers of the upper-aero-digestive tract but only a modest risk of breast or ovarian cancer. The Icelandic population provides an opportunity for comprehensive characterization of the cancer risk profiles of K3326* and HBOC mutations because a single mutation, BRCA2 999del5, is responsible for almost all BRCA2-related HBOC in the population. Methods: Genotype information on 43 641 cancer patients and 370 971 control subjects from Iceland, the Netherlands, and the United States was used to assess the cancer risk profiles of K3326* and BRCA2 999del5. BRCA2 expression was assessed using RNAseq data from blood (n = 2233), as well as 52 tissues reported in the GTEx database. Results: The cancer risks associated with K3326* are fundamentally different from those associated with 999del5. We report for the first time an association between K3326* and small cell lung cancer (odds ratio [OR] = 2.06, 95% confidence interval [CI] = 1.35 to 3.16) and squamous cell carcinoma of the skin (OR = 1.69, 95% CI = 1.26 to 2.26). Individuals homozygous for K3326* reach old age and have children. Unlike BRCA2 999del5, the K3326* allele does not affect the level of BRCA2 transcripts, and the allele is expressed to the same extent as the wild-type allele. Conclusions: K3326* associates primarily with cancers that have strong environmental genotoxic risk factors. Expression of the K3326* allele suggests that a variant protein may be made that retains the DNA repair capabilities important to hormone-responsive tissues but may be less efficient in responding to genotoxic stress.

Lifestyle and Physiological Factors Associated with Facial Wrinkling in Men and Women.

Facial wrinkling is one of the most notable signs of skin aging. Men and women show different wrinkling patterns yet the lifestyle and physiological factors underlying these sex-specific patterns are relatively unknown. Here, we investigated sex-specific determinants for facial wrinkles. Wrinkle area was quantified digitally using facial photographs of 3,831 northwestern Europeans (51-98 years, 58% female). Effect estimates from multivariable linear regressions are presented as the percentage difference in the mean value of wrinkle area per unit increase of a determinant (%Δ). Wrinkle area was higher in men (median 4.5%, interquartile range: 2.9-6.3) than in women (3.6%, interquartile range: 2.2-5.6). Age was the strongest determinant, and current smoking (men: 15.5%Δ; women: 30.9%Δ) and lower body mass index (men: 1.7%Δ; women: 1.8%Δ) were also statistically significantly associated with increased wrinkling. Pale skin color showed a protective effect (men: -21.0%Δ; women: -28.5%Δ) and, in men, sunburn tendency was associated with less wrinkling. In women, low educational levels and alcohol use were associated with more wrinkling, whereas female pattern hair loss and a higher free androgen index were associated with less wrinkling. In summary, we validated known and identified additional determinants for wrinkling. Skin aging-reducing strategies should incorporate the sex differences found in this study.

Genome-Wide Association Studies of Multiple Keratinocyte Cancers.

There is strong evidence for a role of environmental risk factors involved in susceptibility to develop multiple keratinocyte cancers (mKCs), but whether genes are also involved in mKCs susceptibility has not been thoroughly investigated. We investigated whether single nucleotide polymorphisms (SNPs) are associated with susceptibility for mKCs. A genome-wide association study (GWAS) of 1,666 cases with mKCs and 1,950 cases with single KC (sKCs; controls) from Harvard cohorts (the Nurses' Health Study [NHS], NHS II, and the Health Professionals Follow-Up Study) and the Framingham Heart Study was carried-out using over 8 million SNPs (stage-1). We sought to replicate the most significant statistical associations (p-value≤ 5.5x10-6) in an independent cohort of 574 mKCs and 872 sKCs from the Rotterdam Study. In the discovery stage, 40 SNPs with suggestive associations (p-value ≤5.5x10-6) were identified, with eight independent SNPs tagging all 40 SNPs. The most significant SNP was located at chromosome 9 (rs7468390; p-value = 3.92x10-7). In stage-2, none of these SNPs replicated and only two of them were associated with mKCs in the same direction in the combined meta-analysis. We tested the associations for 19 previously reported basal cell carcinoma-related SNPs (candidate gene association analysis), and found that rs1805007 (MC1R locus) was significantly associated with risk of mKCs (p-value = 2.80x10-4). Although the suggestive SNPs with susceptibility for mKCs were not replicated, we found that previously identified BCC variants may also be associated with mKC, which the most significant association (rs1805007) located at the MC1R gene.

IRF4, MC1R and TYR genes are risk factors for actinic keratosis independent of skin color.

Actinic keratosis (AK) is a pre-malignant skin disease, highly prevalent in elderly Europeans. This study investigates genetic susceptibility to AK with a genome-wide association study (GWAS). A full body skin examination was performed in 3194 elderly individuals from the Rotterdam Study (RS) of exclusive north-western European origin (aged 51-99 years, 45% male). Physicians graded the number of AK into four severity levels: none (76%), 1-3 (14%), 4-9 (6%) and ≥10 (5%), and skin color was quantified using a spectrophotometer on sun-unexposed skin. A GWAS for AK severity was conducted, where promising signals at IRF4 and MC1R (P < 4.2 × 10(-7)) were successfully replicated in an additional cohort of 623 RS individuals (IRF4, rs12203592, Pcombined = 6.5 × 10(-13) and MC1R, rs139810560, Pcombined = 4.1 × 10(-9)). Further, in an analysis of ten additional well-known human pigmentation genes, TYR also showed significant association with AK (rs1393350, P = 5.3 × 10(-4)) after correction for multiple testing. Interestingly, the strength and significance of above-mentioned associations retained largely the same level after skin color adjustment. Overall, our data strongly suggest that IRF4, MC1R and TYR genes likely have pleiotropic effects, a combination of pigmentation and oncogenic functions, resulting in an increased risk of AK.