Accurate prognostication guides optimal clinical management in skin cancer. Merkel cell carcinoma (MCC) is the most aggressive form of skin cancer that often presents in advanced stages and is associated with poor survival rates. There are no personalized prognostic tools in use in MCC. We employed explainability analysis to reveal new insights into mortality risk factors for this highly aggressive cancer. We then combined deep learning feature selection with a modified XGBoost framework, to develop a web-based prognostic tool for MCC termed 'DeepMerkel'. DeepMerkel can make accurate personalised, time-dependent survival predictions for MCC from readily available clinical information. It demonstrated generalizability through high predictive performance in an international clinical cohort, out-performing current population-based prognostic staging systems. MCC and DeepMerkel provide the exemplar model of personalised machine learning prognostic tools in aggressive skin cancers.

9570 Background: Precision-based personalised biomarkers able to identify both low-risk and high-risk patient subpopulations with localised cutaneous melanoma are urgently needed to guide clinical follow up and treatment stratification. The combined immunohistochemical expression of AMBRA1 and Loricrin (AMBLor) in the epidermis overlying non-ulcerated AJCC stage I/II melanomas as prognostic biomarker able to accurately identify genuinely low-risk patient subpopulations (NPV >96%, clinical sensitivity >95%, Ewen et al Brit J Dermatol. 2024). To further identify distinct subsets of patients at high risk of metastasis, the present study aimed to develop a machine learning (ML) risk-prediction model combining AMBLor ‘at-risk’ status with six specific patient clinical and tumour pathological features. Methods: Using common and widely used ML models a Naïve Bayes and a Generalized Linear Model with adaBoost, ML algorithms were trained and tested using three geographically distinct retrospective-prospective cohorts of AMBLor at-risk non-ulcerated AJCC stage I/II melanomas from Australia, USA and Spain (n=552), with validation studies performed in a 4 th independent retrospective-prospective cohort of 120 AMBLor at-risk non-ulcerated localised melanomas derived from the UK. Results: Based on a training: test data split of 50:50, 20% of patients were defined as high-risk, with a 5-year recurrence-free survival (RFS) probability of 56% (Log-rank [Mantel-Cox) P < 0.0001, HR 6.88, 95% CI 3.03-15.63, clinical specificity 87.2%, PPV 44.4%). Further validation of the ML algorithms in the UK validation cohort identified 24% patients as high-risk, with a 5-year RFS of 56.3% (Log-rank [Mantel-Cox) P < 0.0001, HR 7.59, 95% CI 2.94-19.6, clinical specificity 82.1%, PPV 50%). Conclusions: Through the proven negative predictive power of AMBLor with the cumulative power of prognostic clinical and pathological features these data provide a novel and improved risk- prediction model to stratify patients with non-ulcerated localised melanomas at low or high risk of tumour recurrence thereby aiding optimal personalised patient management and treatment stratification.

Accurate prognostication is essential to guide clinical management in localised cutaneous melanoma (CM), the form of skin cancer with the highest mortality. While the tumour microenvironment (TME) plays a key role in disease progression, current staging systems rely on limited tumour features and exclude key clinicopathological prognostic features. Here we show that MelanoMAP, a multimodal AI model integrating TME-derived digital biomarkers and clinicopathological features from over 3,500 histology slides, improves prognostication of localised CM. MelanoMAP achieved a C-index of 0.82, a 24% improvement over traditional AJCC staging (0.66) and consistently outperformed clinicopathological-only models across six international patient cohorts. SHAP analysis identified TME-derived digital biomarkers, alongside traditional clinicopathological factors including age, mitotic count, and Breslow depth, were critical determinants of metastatic risk. MelanoMAP establishes a potential foundation for precision oncology in CM, demonstrating how AI-driven digital biomarkers can advance personalised prognostication and inform clinical-decision making.

Selective degradation of damaged mitochondria by autophagy (mitophagy) is proposed to play an important role in cellular homeostasis. However, the molecular mechanisms and the requirement of mitochondrial quality control by mitophagy for cellular physiology are poorly understood. Here, we demonstrated that primary human cells maintain highly active basal mitophagy initiated by mitochondrial superoxide signaling. Mitophagy was found to be mediated by PINK1/Parkin-dependent pathway involving p62 as a selective autophagy receptor (SAR). Importantly, this pathway was suppressed upon the induction of cellular senescence and in naturally aged cells, leading to a robust shutdown of mitophagy. Inhibition of mitophagy in proliferating cells was sufficient to trigger the senescence program, while reactivation of mitophagy was necessary for the anti-senescence effects of NAD precursors or rapamycin. Furthermore, reactivation of mitophagy by a p62-targeting small molecule rescued markers of cellular aging, which establishes mitochondrial quality control as a promising target for anti-aging interventions.

Chronic non-healing cutaneous wounds represent a major burden to patients and healthcare providers worldwide, emphasising the continued unmet need for credible and efficacious therapeutic approaches for wound healing. We have recently shown the potential for collagen peptides to promote proliferation and migration during cutaneous wound healing. In the present study, we demonstrate that the application of porcine-derived collagen peptides significantly increases keratinocyte and dermal fibroblast expression of integrin α2β1 and activation of an extracellular signal-related kinase (ERK)-focal adhesion kinase (FAK) signalling cascade during wound closure in vitro. SiRNA-mediated knockdown of integrin β1 impaired porcine-derived collagen peptide-induced wound closure and activation of ERK-FAK signalling in keratinocytes but did not impair ERK or FAK signalling in dermal fibroblasts, implying the activation of differing downstream signalling pathways. Studies in ex vivo human 3D skin equivalents subjected to punch biopsy-induced wounding confirmed the ability of porcine-derived collagen peptides to promote wound closure by enhancing re-epithelialisation. Collectively, these data highlight the translational and clinical potential for porcine-derived collagen peptides as a viable therapeutic approach to promote re-epithelialisation of superficial cutaneous wounds.

BACKGROUND:Cutaneous squamous cell carcinoma (cSCC) incidence continues to increase globally with, as of yet, an unmet need for reliable prognostic biomarkers to identify patients at increased risk of metastasis. The aim of the present study was to test the prognostic potential of the combined immunohistochemical expression of the autophagy regulatory biomarkers, AMBRA1 and SQSTM1, to identify high-risk patient subsets.METHODS:A retrospective cohort of 68 formalin-fixed paraffin-embedded primary cSCCs with known 5-year metastatic outcomes were subjected to automated immunohistochemical staining for AMBRA1 and SQSTM1. Digital images of stained slides were annotated to define four regions of interest: the normal and peritumoral epidermis, the tumor mass, and the tumor growth front. H-score analysis was used to semi-quantify AMBRA1 or SQSTM1 expression in each region of interest using Aperio ImageScope software, with receiver operator characteristics and Kaplan-Meier analysis used to assess prognostic potential.RESULTS:The combined loss of expression of AMBRA1 in the tumor growth front and SQSTM1 in the peritumoral epidermis identified patients with poorly differentiated cSCCs at risk of metastasis (*p < 0.05).CONCLUSIONS:Collectively, these proof of concept data suggest loss of the combined expression of AMBRA1 in the cSCC growth front and SQSTM1 in the peritumoral epidermis as a putative prognostic biomarker for poorly differentiated cSCC.

The epidermal expression of AMBRA1 and loricrin overlying primary melanomas has recently been validated as a prognostic biomarker able to identify genuinely low risk nonulcerated American Joint Committee on Cancer stage I/II tumours. We have previously shown that the loss of these biomarkers is due to secretion of transforming growth factor-β2 by the underlying tumour cells, and hypothesize that this mechanism is used by melanomas at risk of metastasis to facilitate premalignant dissemination. The aim of the present study was to evaluate the potential for the combined expression of AMBRA1 and loricrin to characterize a range of melanocytic naevi and melanoma in situ as benign or malignant. Formalin-fixed paraffin-embedded tissue sections were obtained derived from a representative selection of benign naevi and melanoma in situ from the cellular pathology archive of Newcastle upon Tyne Hospitals NHS Foundation Trust. These comprised 5 intradermal naevus, 16 dysplastic naevus (severe/moderate), 5 junctional naevus, 5 compound naevus, 6 common blue naevus, 5 cellular blue naevus, 2 combined naevus, 1 melanocytoma and 26 melanoma in situ. All sections were subjected to automated immunohistochemical analysis for AMBRA1 and loricrin expression. Staining patterns and expression were evaluated using consensus scoring by the study cellular pathologists (A.H., P.S.). All of the benign and dysplastic melanocytic naevi examined showed maintenance of both AMBRA1 and loricrin in the epidermis overlying the lesion, in line with their expression in normal skin. A maintained pattern was also seen in 24 of 26 cases of melanoma in situ, but two cases displayed focal losses of both AMBRA1 and loricrin. On review, both cases showed no evidence of melanoma in further levels and there was no clinical progression. The maintained pattern of AMBRA1 and loricrin was expected in the melanocytic naevi and is consistent with their benign nature. However, loss of expression of AMBRA1 and loricrin in two cases of melanoma in situ was unexpected, suggesting that some cases of melanoma in situ may contain melanocytes with malignant potential. Collectively these data warrant the further investigation of AMBRA1 and loricrin expression in a larger cohort to investigate their potential as biomarkers to distinguish between benign melanocytic lesions and those at risk of malignant progression. This study is funded by the Innovate UK Combined Investor Partnerships Award Round 11 (reference 10083008-1586) awarded to AMLo Biosciences.

Alternative 3-dimensional (3D) skin models that replicate in vivo human skin are required to investigate important events during wound healing, such as collective cell migration, epidermal layer formation, dermal substrate formation, re-epithelialisation and collagen production. In this study, a matched human 3D skin equivalent model (3D-SEM) was developed from human skin cells (fibroblast and keratinocytes), characterised using haematoxylin and eosin, immunofluorescence staining and microRNA profiling. The 3D-SEM was then functionally tested for its use in wound healing studies. Mesenchymal stem cells (MSCs) were isolated and characterised according to the criteria stipulated by the International Society for Cell Therapy. Cytokine and growth factor secretions were analysed by enzyme-linked immunosorbent assay. MSC-conditioned medium (MSC-CM) was then tested for wound healing capacity using the developed 3D-SEM at different timepoints i.e., at one, two and four weeks. The constructed 3D-SEM showed consistent development of skin-like structures composed of dermal layers and epidermal layers, with the ability to express epidermal differentiation markers and full stratification. They also showed prolonged longevity in culture media, retaining full differentiation and stratification within the four weeks. MicroRNA profiling revealed a strong correlation in microRNA expression between the developed 3D-SEM and the original native skin (p<0.001; R=0.64). Additionally, MSC-CM significantly enhanced migration, proliferation and differentiation of epidermal cells in the wounded models compared to control models at the different timepoints. In conclusion, in this study, the developed 3D-SEM mimicked native skin at the cellular and molecular levels, and clearly showed the important stages of skin regeneration during the healing process. MSC secretome contains growth factors that play a pivotal role in the healing process and could be used as a therapeutic option to accelerate skin healing.

BACKGROUND:Combined expression of the autophagy-regulatory protein AMBRA1 (activating molecule in Beclin1-regulated autophagy) and the terminal differentiation marker loricrin in the peritumoral epidermis of stage I melanomas can identify tumour subsets at low risk of -metastasis. OBJECTIVES:To validate the combined expression of peritumoral AMBRA1 and loricrin (AMBLor) as a prognostic biomarker able to identify both stage I and II melanomas at low risk of tumour recurrence. METHODS:Automated immunohistochemistry was used to analyse peritumoral AMBRA1 and loricrin expression in geographically distinct discovery (n = 540) and validation (n = 300) cohorts of nonulcerated American Joint Committee on Cancer (AJCC) stage I and II melanomas. AMBLor status was correlated with clinical outcomes in the discovery and validation cohorts separately and combined. RESULTS:Analysis of AMBLor in the discovery cohort revealed a recurrence-free survival (RFS) rate of 95.5% in the AMBLor low-risk group vs. 81.7% in the AMBLor at-risk group (multivariate log-rank, P < 0.001) and a negative predictive value (NPV) of 96.0%. In the validation cohort, AMBLor analysis revealed a RFS rate of 97.6% in the AMBLor low-risk group vs. 78.3% in the at-risk group (multivariate log-rank, P < 0.001) and a NPV of 97.6%. In a multivariate model considering AMBLor, Breslow thickness, age and sex, analysis of the combined discovery and validation cohorts showed that the estimated effect of AMBLor was statistically significant, with a hazard ratio of 3.469 (95% confidence interval 1.403-8.580, P = 0.007) and an overall NPV of 96.5%. CONCLUSIONS:These data provide further evidence validating AMBLor as a prognostic biomarker to identify nonulcerated AJCC stage I and II melanoma tumours at low risk of disease recurrence.

Intrinsic and acquired resistance to mitogen-activated protein kinase inhibitors (MAPKi) in melanoma remains a major therapeutic challenge. Here, we show that the clinical development of resistance to MAPKi is associated with reduced tumor expression of the melanoma suppressor Autophagy and Beclin 1 Regulator 1 (AMBRA1) and that lower expression levels of AMBRA1 predict a poor response to MAPKi treatment. Functional analyses show that loss of AMBRA1 induces phenotype switching and orchestrates an extracellular signal-regulated kinase (ERK)-independent resistance mechanism by activating focal adhesion kinase 1 (FAK1). In both in vitro and in vivo settings, melanomas with low AMBRA1 expression exhibit intrinsic resistance to MAPKi therapy but higher sensitivity to FAK1 inhibition. Finally, we show that the rapid development of resistance in initially MAPKi-sensitive melanomas can be attributed to preexisting subclones characterized by low AMBRA1 expression and that cotreatment with MAPKi and FAK1 inhibitors (FAKi) effectively prevents the development of resistance in these tumors. In summary, our findings underscore the value of AMBRA1 expression for predicting melanoma response to MAPKi and supporting the therapeutic efficacy of FAKi to overcome MAPKi-induced resistance.

Background: Loss of epidermal AMBRA1 and Loricrin (AMBLor) overlying AJCC stage I/II primary melanomas has recently been validated as a prognostic biomarker1. However, while loss of AMBLor identifies low risk tumour subsets, a need for a robust biomarker able to identify early-stage melanomas at risk of metastasis remains. To this aim, our pilot data shows loss of epidermal AMBRA1 overlying primary melanomas is also associated with loss of expression in the surrounding tumour endothelium, suggesting endothelial AMBRA1 (eAMBRA1) expression may define high-risk tumour subsets. To evaluate this potential, the aim of the present study was to evaluate the immunohistochemical expression of eAMBRA1 in a cohort of primary melanomas with loss of epidermal AMBLor and corelate expression levels with clinical outcome over 5 years. Methods: Automated immunohistochemical analysis of eAMBRA1 expression was evaluated in a cohort of FFPE tissue derived from 47 AJCC stage I/II melanomas (23 metastatic/24 non-metastatic) with loss of epidermal AMBLor. eAMBRA1 expression was determined by positive pixel count using Aperio ImageScope software and normalised to vessel area. Results: Data revealed an increased proportion of intra/peri-tumoural vessels with low eAMBRA1 expression (in which AMBRA1 expression was decreased by more than 55% compared to the average AMBRA1 vessel score throughout the tissue section) in melanomas that developed metastasis; reduced eAMBRA1 expression was observed in 21.5% of vessels in tumours that developed metastasis compared to 9.9% in non-metastatic tumours (p < 0.004). Grouping patients according to whether the percentage of intra/peri-tumoural vessels with reduced eAMBRA1 expression was lower than or at least 9.4% (determined by ROC analysis), Kaplan Meier analysis revealed a significantly reduced 5-year recurrence-free survival for those patients with melanomas with ≥ 9.4% vessels with reduced eAMBRA1 expression (logrank HR 18.32, 95% CI 8.1-41.6, p < 0.001). Conversely no difference in the proportion of vessels with low eAMBRA1 expression was observed between normal skin, primary melanomas with maintained epidermal AMBLor or tumours with loss of AMBLor that did not metastasise. Conclusions: Collectively, these data suggest the increased proportion of vessels in the melanoma microenvironment with reduced eAMBRA1 expression as a novel putative biomarker to identify and aid the stratification of early AJCC stage I/II melanomas at risk of metastasis.1.Ewen at al. Brit J Dermatol 2023 in press

Background: Merkel cell carcinoma (MCC) is a highly aggressive cutaneous malignancy for which accurate prognostication is necessary to support clinical management. Objective: (1) To determine which survival endpoint-disease-specific death (DSD) or overall survival (OS)-was better stratified by MCC American Joint Cancer Committee eighth edition staging. (2) To develop a multivariable model for enhanced MCC survival predictions. Methods: A retrospective analysis of 10,958 histologically confirmed MCC patients between January 2000 and December 2020 was performed. Patient and tumor features were analyzed from 2 cohorts: a US cohort and an external validation UK cohort. A multivariable Fine and Gray competing risk (FG) model was utilized to account for the competing risk. Results: DSD demonstrated greater discriminatory power as a survival endpoint when compared with OS. Multivariate FG analysis identified the most impactful features of DSD: truncal lesions (subdistribution hazard ratios [SHRs] = 1.96, P < .001), age >84 years (SHR = 1.79, P < .001), male sex (SHR = 1.34, P < .001), and marital status (SHR = 1.09, P < .001). A second FG model incorporating those impactful features enhanced survival predictions beyond current MCC staging criteria alone in both the US (C-index 0.75 vs 0.64, P < .001) and external validation UK cohort (C-index 0.77). Conclusions: MCC staging can stratify DSD better than OS. Additional patient and tumor featuresenhanced prognostication beyond current staging systems.

BACKGROUNDCombined loss of the autophagy-regulatory protein AMBRA1 and the terminal differentiation marker loricrin in the peritumoural epidermis of stage I melanomas can identify tumour subsets at low risk of metastasis.OBJECTIVESThe aim of the present study was to validate the combined loss of peritumoural AMBRA1 and loricrin (AMBLor) as a prognostic biomarker able to identify both stage I and II melanomas at low risk of tumour recurrence.METHODSAutomated Immunohistochemistry was used to analyse peritumoural AMBRA1 and loricrin expression in geographically distinct discovery (n = 540) and validation (n = 300) cohorts of non-ulcerated AJCC stage I and II melanomas. AMBLor status was correlated with clinical outcomes in the discovery and validation cohorts separately and combined.RESULTSAnalysis of AMBLor in the discovery cohort revealed a recurrence-free survival (RFS) rate of 95.5% in the AMBLor low risk group compared to 81.7% in the AMBLor at-risk group (multivariate log-rank, P < 0.001), and a negative predictive value (NPV) of 96%. In the validation cohort, AMBLor analysis revealed a RFS rate of 97.6% in the AMBLor low risk group compared to 78.3% in the at-risk group (multivariate log-rank, P < 0.001) and a NPV of 97.6%. In a multivariate model considering AMBLor, Breslow thickness, age and sex, analysis of the combined discovery and validation cohorts showed that the estimated effect of AMBLor was statistically significant with a hazard ratio of 3.469 (95% confidence interval 1.403-8.580, P = 0.007), with an overall NPV of 96.5%.CONCLUSIONSThese data provide further evidence validating AMBLor as a prognostic biomarker to identify non-ulcerated AJCC stage I and II melanoma tumours at low risk of disease recurrence.

legend to supplementary figure 1