Search Results (107)

Background: Melanoma is one of the most aggressive forms of skin cancer, making early and accurate diagnosis essential for improving patient outcomes. Methods: In this work, we propose an Evolutionary Gabor-based Melanoma Descriptor (Evo-GMD), a lightweight and interpretable approach designed under the principles of Frugal AI. The method integrates multi-scale Gabor filtering with Differential Evolution to automatically learn discriminative texture patterns using a reduced set of parameters. The proposed approach was evaluated on the PH2 dataset, achieving competitive performance (accuracy above 95%) while maintaining low computational complexity and full interpretability. To further assess its robustness, complementary experiments were conducted on the ISIC 2017 dataset, which presents higher variability, class imbalance, and heterogeneous lesion characteristics. Results: The results reveal that multiple methods—including handcrafted descriptors, convolutional neural networks, and transfer learning models—exhibit significant performance degradation or converge to trivial solutions under these conditions. This behavior highlights that increasing model complexity does not necessarily improve classification performance when data constraints are present. Conclusions: Overall, the findings demonstrate that the proposed method provides a robust and efficient alternative for melanoma classification in low-resource scenarios, where data availability, computational capacity, and interpretability are critical factors. Full article

Cutaneous melanoma is an aggressive form of cancer that accounts for approximately 80% of skin cancer-associated mortality. Treatment strategies for melanoma have been rapidly evolving over the past two decades. Clinical outcomes for advanced melanoma patients have improved with the use of immune checkpoint inhibitors and targeted therapies, including combination therapies. In this review, we primarily focus on the evolution of novel therapies for advanced melanoma patients, existing and emerging melanoma biomarkers, and the most up-to-date treatment recommendations. Furthermore, we summarise ongoing studies, clinical trials, and emerging therapeutic strategies. In this review, we will walk you through how cutaneous melanoma treatment strategies have evolved over the past 50 years. We will provide background on how those strategies were developed, how they function, their pros and cons, and their performance in terms of clinical benefits. Lastly, we will provide insights into potential therapies and research areas. In summary, improving the efficacy of immune checkpoint blockade remains the most promising approach for improving cutaneous melanoma management. However, strategies such as CAR-T/NK, PROTAC, or extracellular vesicle-based therapies are also worth further investigation. Full article

Despite major advances in systemic therapy, surgery still plays a valuable role in the management of metastatic melanoma. This review summarizes the historical evolution of surgical management, outlines the traditional clinical risk factors, and examines biochemical, molecular, and mutational factors that impact melanoma tumor biology. Emerging tools such as predictive biomarker assays, gene expression profiling, and circulating tumor DNA are discussed in the context of patient selection. Finally, we consider contemporary indications for surgery, the management of oligoprogression, sequence of treatment, and optimal timing of resection while highlighting how operative intervention integrates with modern melanoma care. Full article

Many cancer-associated deaths result from metastases rather than primary tumors. Growing evidence suggests that DNA methylation alterations are crucial for inducing a plastic phenotype that allows cancer cells to adapt to the metastatic microenvironment. Brain metastases of melanoma (MBM) and glioblastoma (GB) share a neuroectodermal origin and the brain as tissue of residence, but their epigenetic regulation is poorly understood. Aiming at elucidating shared and tumor-distinct features, we analyzed the methylation of MGMT regulatory elements. We focused on MGMT because MGMT promoter methylation is used as a predictive marker for temozolomide response in GB, but its role in MBM has been discussed controversially. By targeting 12 CpG dinucleotides (CpGs) in the promoter, 68 CpGs in intergenic enhancers, and 31 CpGs in intragenic enhancers, we identified shared features, including an L-shaped relationship between promoter methylation and MGMT protein expression and an inverse L-shaped relationship between intragenic enhancer methylation and MGMT protein expression. GB exhibited higher methylation, particularly in promoter and intergenic enhancers, and stronger associations between methylation and overall survival than MBM. These results highlight both conserved and tumor-specific MGMT regulation, reflecting the complexity of epigenetic control in brain malignancies and emphasizing divergent evolution between MBM and GB. Full article

Background: Chromosome 3p (chr3p) is frequently deleted in multiple cancers, indicating the presence of shared tumor suppressors. In aggressive uveal melanomas (UVM), this deletion often co-occurs with chr8q amplification (8q+), suggesting strong selection pressure during UVM evolution. Methods: To understand the pattern of genomic alterations mediated by chr3p deletion, we have developed an algorithm for detecting isochromosomes in 10,632 TCGA cancer patients. We further perform integrative genomics analysis to investigate how chr3p deletion could affect subsequent cancer genome evolution and synthetic lethality in UVM. Results: Analysis of genomic alterations in 33 different cancer types implicates the deletion or deleterious mutations of SET-domain-containing 2 (SETD2) at chr3p21 in significantly facilitating the formation of isochromosomes, thereby promoting genomic instability conducive to rapid cancer genome evolution. Fracturing of dicentric isochromosomes during cell division is pervasive and follows the dynamic fragmentation pattern of solids under impulse. In the most aggressive UVM subtype, chr3 deletion includes MITF, a master regulator of melanocyte survival and differentiation, and co-occurs with 8q+. We demonstrate that MITF is a master transcriptional regulator of GNAQ/GNA11 and associated synthetic-lethal genes in UVM. MITF maintains MAPK and calcium homeostasis in UVM, and its hemizygous deletion is thus accidental, likely creating an early crisis during oncogenesis. We further show that MITF, MYC, and GNAQ/GNA11 form coupled regulatory feedback loops in the melanocyte lineage, and MITF deletion in UVM creates acute dependency on MYC-mediated rescue via 8q+. The discovered feedback loops predict both overall and relapse-free patient survival within the most aggressive UVM subtype, explain sensitivity to therapeutic gene perturbations, and inform effective combinatorial therapies. Conclusions: SETD2 deletion potentiates isochromosome formation across diverse cancers. Combinatorial targeting of MITF together with a previously identified synthetic lethal gene may benefit UVM patients harboring both chr3 deletion and 8q+. Full article

Melanocytic tumorigenesis is thought to occur through stepwise genomic evolution from normal skin to nevi and, ultimately, melanoma. To investigate this progression, we performed targeted deep sequencing of a 46-gene panel in matched healthy skin, nevus, and melanoma samples from 15 patients, including 14 complete tissue trios. Mutation burden increased progressively across tissues, with median mutation counts rising from benign skin to nevi and showing the highest levels in melanoma, consistent with cumulative somatic alterations. Canonical MAPK pathway mutations were common: BRAF V600E and NRAS Q61 variants were detected in many nevi and melanomas and were shared between lesions in 8 of 15 patients, providing direct evidence of clonal continuity. Variant allele frequencies for driver and nonsynonymous mutations were higher than those of passenger and synonymous mutations, reflecting selective expansion of functionally relevant clones. UV-signature substitutions were abundant, particularly among synonymous variants, suggesting background mutagenesis without clonal advantage. Melanoma-private mutations in genes such as ARID1A, ARID2, PIK3CA, and CDKN2A indicated additional late events contributing to malignant progression. Overall, this study supports a model in which many melanomas evolve from pre-existing nevi through sequential acquisition and clonal amplification of somatic mutations, while also revealing heterogeneous evolutionary trajectories. Full article

Temozolomide (TMZ) remains foundational in the management of adult-type diffuse gliomas in general, and glioblastoma specifically. However, its efficacy harbors an evolutionary trade-off. TMZ drives its cytotoxicity through generating O⁶-methylguanine lesions, especially active in MGMT-silenced, mismatch repair (MMR)-proficient tumors. By selecting for acquired MMR-deficient subclones, often via MSH6 inactivation, this process escalates into a hypermutator phenotype, generating thousands of de novo alterations. This is a hallmark of the mutational signature known as SBS11, characterized by C>T transitions, which is associated with TMZ treatment. The hypermutator phenotype drives heterogeneity, therapeutic resistance, spatial diversification, and distant recurrence. Despite harboring a mutational burden comparable to melanoma and lung cancer, TMZ-induced hypermutation does not sensitize gliomas to immune checkpoint blockade. This resistance reflects the profoundly immunosuppressive brain microenvironment, impaired antigen presentation, marked transcriptional plasticity, and perhaps also the frequent use of corticosteroids. Emerging strategies aim to exploit vulnerabilities created by TMZ-mediated genomic instability, including PARP, ATR, WEE1, and AURKA inhibition; alternative alkylators; metabolic rewiring; and G-quadruplex stabilization. Notably, the real-time detection of evolving mutational signatures via CSF-based liquid biopsies may enable adaptive therapy before radiographic progression. By reframing TMZ as a potent evolutionary agent rather than a conventional chemotherapy, this review synthesizes recent mechanistic insights and translational opportunities to guide a next-generation, evolution-informed treatment paradigm for glioma. Full article

Background: Despite advances in diagnosis and treatment, the psychosocial impact of skin cancer remains relatively underexplored. The aim of this study was to explore the emotional experiences of people diagnosed with melanoma and cutaneous squamous cell carcinoma (SCC) and their evolution over time. Methods: A qualitative study was conducted using semi-structured interviews. Purposive sampling was employed, taking into account gender, age, tumour type and tumour stage. Interviews were audio-recorded, transcribed verbatim and analysed thematically. Results: Thirty-six patients were recruited (18 with melanoma and 18 with SCC). Overall, 61.1% were men and the mean age was 63.8 years (SD 10.8). The analysis revealed three main themes: feelings, relationships and life changes. Patients in both groups experienced a range of emotions following diagnosis, such as anxiety, relief and fear of recurrence. Family dynamics played a crucial role in patients’ experiences, acting as both a source of support and a cause of concern. The diagnosis prompted changes in everyday life, affecting work, daily activities and patients’ sense of identity. Conclusions: The diagnosis of melanoma or SCC has a profound emotional and existential impact on patients. Personalised care is crucial to address their evolving concerns and information needs. Further research is needed on the long-term impact of skin cancer and the benefits of psycho-oncological support. Incorporating patients’ perspectives into current skin cancer guidelines should be considered. Full article

Melanoma cells in the brain may use similar mechanisms for adapting to injury and/or disease (that is, through continued reallocation of energy, matter, and information) as other cell types do to create an environment in which cancer cells can grow and sustain themselves within the confines of the brain. These adaptable mechanisms include the ability to reactivate dormant neural crest-derived migration and communication pathways. Unlike some other types of cancers that invade neural tissue as a simple invasion, melanomas are capable of achieving limited molecular, metabolic, and electrical similarity to the neural circuitry of the brain. Melanomas achieve this limited similarity through both vascular co-optation and mimicking synaptic functions, as well as through their engagement of redox-coupled metabolic pathways and feedback-regulated signal transduction pathways. The result is the creation of a metastable tumor–host system, where the relationship between tumor and host is defined by the interaction of stabilizing and destabilizing forces; forces that define the degree of coherence, vulnerability, and persistence of the tumor–host system. In this review, we integrate molecular, electrophysiological, and anatomical data to develop a single unifying hypothesis for the functional integration of melanoma cells into the neural tissue of the brain. Additionally, we describe how neural crest-based regulatory pathways are reactivated in the adult brain and how tumor–host coherence is developed as a function of the shared thermodynamic and informational constraints placed on both tumor and host. We also describe how our proposed conceptual model allows for the understanding of therapeutic interventions as selective disruptions of the neural, metabolic, and immunological couplings that support metastatic adaptation. Full article

Oncolytic viruses represent an emerging class of therapeutic agents that have the potential to transform the care of patients with melanoma. In this narrative review, we describe the evolution of oncolytic virus approaches. We begin by describing early investigations using wild type viruses and then the development of sophisticated Herpes simplex virus 1 (HSV-1) variant constructs such as talimogene laherparepvec (T-VEC) and vusolimogene oderparepvec (Replimune-1, RP1), which incorporate deletions of viral genes and expression of human or synthetic transgenes to promote tumor selectivity, dendritic cell recruitment, antigen presentation, and stimulation of systemic anti-tumor immune responses. We review the status of clinical trials of oncolytic viruses in melanoma, highlight regulatory challenges, and describe important concepts and key remaining questions within the field. While T-VEC remains the only Food and Drug Administration (FDA)-approved oncolytic virus for melanoma treatment, ongoing research focusing on next-generation viral constructs and combination strategies aims to further improve clinical outcomes and expand the applicability of oncolytic virus therapy in melanoma. Full article

Melanoma shows heterogeneity across body sites like skin, acral skin, and the uvea, driven by molecular characteristics and genetic variations. However, comparative studies exploring the heterogeneity of melanoma across different anatomical sites remain limited, hindering a comprehensive understanding of its underlying biology. We proposed a research framework through bioinformatics to analyze the tumor ecosystems of cutaneous, acral, and uveal melanoma, from molecular characteristics to genetic variations at single-cell resolution. We found that oxidative phosphorylation (OXPHOS) is a critical driver of tumor cell evolution, with abnormal ribosomal gene and tumor suppressor expression observed in uveal melanoma (UM). Additionally, we screened for potential drug targets and drugs against tumor cells. In the immune microenvironment, acral melanoma (AM) and UM exhibit stronger immunosuppressive characteristics compared to cutaneous melanoma (CM). OXPHOS contributes to T cell cytotoxicity dysregulation in CM and AM, while interferon-γ is crucial in UM. Tumor cells may also induce T cell dysfunction through biological signals such as MIF-CD74 and HLA-E-NKG2A. This study offers valuable insights into melanoma heterogeneity, providing a comprehensive research framework for understanding the distinct molecular and immune characteristics of CM, AM, and UM, and potentially guiding the development of therapeutic strategies tailored to each melanoma subtype. Full article

Melanoma represents a worldwide public health problem due to its high incidence and mortality rates. Despite the advances in melanoma therapy, not all patients respond to single or combined therapy because of primary or acquired resistance to the anti-tumor agents. Recently, positive results have been reported since the specific monoclonal antibodies, such as Ipilimumab (Ipi) and Nivolumab (Niv), were included in therapeutic protocols as immune checkpoint inhibitors. The evolution of neoplastic diseases and the therapeutic approaches in cancer involve several biological processes, including apoptosis, DNA progression through cell cycle phases, the release of pro-inflammatory cytokines, and changes in the expression of melanoma genes. Therefore, the potential modulation of these processes and associated molecules, due to single or combined treatments with oncolytic drugs like Carboplatin and Paclitaxel, checkpoint inhibitors such as Ipi and Niv, or natural bioactive compounds like Resveratrol or Quercetin, could represent a great benefit in melanoma treatment, contributing to the decrease or even reversal of the drug resistance in melanoma cells. Full article

Accurate skin lesion classification is crucial for the early detection of malignant lesions, including melanoma, as well as improved patient outcomes. While convolutional neural networks (CNNs) excel at capturing local morphological features, they struggle with global context modeling essential for comprehensive lesion assessment. Vision transformers address this limitation but suffer from quadratic computational complexity O(n²), hindering deployment in resource-constrained clinical environments. We propose DermaMamba, a novel dual-branch fusion architecture that integrates CNN-based local feature extraction with Vision Mamba (VMamba) for efficient global context modeling with linear complexity O(n). Our approach introduces a state space fusion mechanism with adaptive weighting that dynamically balances local and global features based on lesion characteristics. We incorporate medical domain knowledge through multi-directional scanning strategies and ABCDE (Asymmetry, Border irregularity, Color variation, Diameter, Evolution) rule feature integration. Extensive experiments on the ISIC dataset show that DermaMamba achieves 92.1% accuracy, 91.7% precision, 91.3% recall, and 91.5% mac-F1 score, which outperforms the best baseline by 2.0% accuracy with 2.3× inference speedup and 40% memory reduction. The improvements are statistically significant based on a significance test (p < 0.001, Cohen’s d > 0.8), with greater than 79% confidence also preserved on challenging boundary cases. These results establish DermaMamba as an effective solution bridging diagnostic accuracy and computational efficiency for clinical deployment. Full article

Background and Objectives: Lymph node management in cutaneous melanoma has undergone a paradigm shift, transitioning from routine complete lymph node dissection (CLND) to a more selective, individualized approach. This narrative review explores the historical evolution, current evidence and clinical guidelines surrounding lymphadenectomy for a patient with Stage III of melanoma. Materials and Methods: A comprehensive literature search was conducted across PubMed, Scopus and Web of Science, focusing on randomized controlled trials, meta-analyses and updated international guidelines published in the past 15 years. Results: Traditional surgical approaches favored radical lymphadenectomy for regional disease control. However, pivotal trials such as the Multicenter Selective Lymphadenectomy Trial II (MSLT-II) and German Dermatologic Cooperative Oncology Group Selective Lymphadenectomy Trial (DeCOG-SLT) demonstrated no survival advantage from immediate CLND following a positive sentinel lymph node biopsy (SLNB), underscoring increased surgical morbidity. Consequently, guidelines from Associazione Italiana di Oncologia Medica (AIOM), the European Society for Medical Oncology (ESMO), and the National Comprehensive Cancer Network (NCCN) now endorse SLNB as the standard for nodal staging, reserving CLND for select high-risk cases. Conclusions: The role of lymphadenectomy in melanoma is increasingly becoming selective, shaped by tumor burden, nodal involvement and response to systemic therapy. SLNB remains central to staging and treatment planning, while CLND is no longer routine. Continued clinical trials and integration with immunotherapy will further refine surgical strategies in melanoma care. Full article

The accurate segmentation of pigmented skin lesions is a critical prerequisite for reliable melanoma detection, yet approximately 30% of lesions exhibit fuzzy or poorly defined borders. This ambiguity makes the definition of a single contour unreliable and limits the effectiveness of computer-assisted diagnosis (CAD) systems. While clinical assessment based on the ABCDE criteria (asymmetry, border, color, diameter, and evolution), dermoscopic imaging, and scoring systems remains the standard, these methods are inherently subjective and vary with clinician experience. We address this challenge by reframing segmentation into three distinct regions: background, border, and lesion core. These regions are delineated using superpixels generated via the Simple Linear Iterative Clustering (SLIC) algorithm, which provides meaningful structural units for analysis. Our contributions are fourfold: (1) redefining lesion borders as regions, rather than sharp lines; (2) generating superpixel-level embeddings with a transformer-based autoencoder; (3) incorporating these embeddings as features for superpixel classification; and (4) integrating neighborhood information to construct enhanced feature vectors. Unlike pixel-level algorithms that often overlook boundary context, our pipeline fuses global class information with local spatial relationships, significantly improving precision and recall in challenging border regions. An evaluation on the HAM10000 melanoma dataset demonstrates that our superpixel–RAG–transformer (region adjacency graph) pipeline achieves exceptional performance (100% F1 score, accuracy, and precision) in classifying background, border, and lesion core superpixels. By transforming raw dermoscopic images into region-based structured representations, the proposed method generates more informative inputs for downstream deep learning models. This strategy not only advances melanoma analysis but also provides a generalizable framework for other medical image segmentation and classification tasks. Full article