Automatized self-supervised learning for skin lesion screening.

Melanoma, the deadliest form of skin cancer, has seen a steady increase in incidence rates worldwide, posing a significant challenge to dermatologists. Early detection is crucial for improving patient survival rates. However, performing total body screening (TBS), i.e., identifying suspicious lesions or ugly ducklings (UDs) by visual inspection, can be challenging and often requires sound expertise in pigmented lesions. To assist users of varying expertise levels, an artificial intelligence (AI) decision support tool was developed. Our solution identifies and characterizes UDs from real-world wide-field patient images. It employs a state-of-the-art object detection algorithm to locate and isolate all skin lesions present in a patient's total body images. These lesions are then sorted based on their level of suspiciousness using a self-supervised AI approach, tailored to the specific context of the patient under examination. A clinical validation study was conducted to evaluate the tool's performance. The results demonstrated an average sensitivity of 95% for the top-10 AI-identified UDs on skin lesions selected by the majority of experts in pigmented skin lesions. The study also found that the tool increased dermatologists' confidence when formulating a diagnosis, and the average majority agreement with the top-10 AI-identified UDs reached 100% when assisted by our tool. With the development of this AI-based decision support tool, we aim to address the shortage of specialists, enable faster consultation times for patients, and demonstrate the impact and usability of AI-assisted screening. Future developments will include expanding the dataset to include histologically confirmed melanoma and validating the tool for additional body regions.

Clinical, Morphological and Molecular Features of Spitz tumors.

Spitz tumors represent a heterogeneous group of challenging melanocytic neoplasms, displaying a range of biological behaviors, spanning from benign lesions, Spitz nevi (SN) to Spitz melanomas (SM), with intermediate lesions in between known as atypical Spitz tumors (AST). They are histologically characterized by large epithelioid and/or spindled melanocytes arranged in fascicles or nests, often associated with characteristic epidermal hyperplasia and fibrovascular stromal changes. In the last decade, the detection of mutually exclusive structural rearrangements involving receptor tyrosine kinases ROS1, ALK, NTRK1, NTRK2, NTRK3, RET, MET, serine threonine kinases BRAF and MAP3K8, or HRAS mutation, led to a clinical, morphological and molecular based classification of Spitz tumors. The recognition of some reproducible histological features can help dermatopathologist in assessing these lesions and can provide clues to predict the underlying molecular driver. In this review, we will focus on clinical and morphological findings in molecular Spitz tumor subgroups.

Histopathology of skin melanocytic lesions.

Melanocytic lesions are instable tumors, the genome of which and its changes determinate their morphology and biological properties. Intermediate lesions share histomorphological features of both, nevi and melanoma. Melanocytomas represent a group of them separated on the basis of recent molecular-biological studies. The article summarizes benign, intermediate, malignant and combined melanocytic skin lesions and offers practical recommendations for diagnosis.

Multiple Primary Melanoma Associated with CDKN2A Mutation-Case Report and Review of the Literature.

The CDKN2A gene remains understudied in melanoma compared to BRAF alterations. Inactivation of this tumor suppressor gene through homozygous deletions in the 9p21 chromosomal region leads to cellular proliferation and disrupts pro-apoptotic pathways. Genetic changes in CDKN2A are linked to multiple primary melanomas (MPM), with patients diagnosed with melanoma facing an elevated risk of developing additional primaries. We present the rare case of a 72-year-old Caucasian woman with nine metastasizing melanomas across diverse anatomical sites, posing a diagnostic challenge. Initial diagnosis in 2022 revealed ulcerated superficial spreading melanomas, progressing to intradermal and papillary dermal populations with neurotropism and angiotropism by early 2023. Lymph node metastases were identified, classifying the condition as pT3b N3b. Subsequent assessments in April 2023 revealed clinically suspicious melanocytic lesions diagnosed as intradermal and traumatized junctional nevi. In late 2023, cutaneous pigmented lesions and subcutaneous metastases were confirmed as nodular nevoid low-CSD multiple melanomas. Fluorescence in situ hybridization testing revealed homozygous CDKN2A deletion, necessitating close multidisciplinary collaboration for an optimized care plan for effective monitoring and intervention in this intricate clinical scenario. In summary, this case report highlights the diagnostic challenges of MPM in a single patient. Stressing the importance of immuno-histochemistry and CDKN2A genetic testing, our findings underscore the crucial role of these tools in accurately distinguishing malignant melanocytic proliferations from nevi and characterizing MPM cases.

Influence of Anatomic Locations of T1b Melanomas and Patient Age on Outcomes.

BACKGROUND: There are no guidelines on when to more strongly recommend sentinel lymph node biopsy (SLNB) for T1b melanomas. OBJECTIVE: To examine whether anatomic locations of T1b melanomas and patient age influence metastases. METHODS: We conducted a retrospective study using data from two hospitals in Los Angeles County from January 2010 through January 2020. RESULTS: Out of 620 patients with primary melanomas, 566 melanomas were staged based on the American Joint Committee on Cancer 8th edition melanoma staging. Forty-one were T1b, of which 13 were located on the face/ear/scalp and 28 were located elsewhere. T1b melanomas located on the face/ear/scalp had an increased risk of lymph node or distant metastasis compared with other anatomic sites (31% vs 3.6%, P=0.028). For all melanomas, the risk of lymph node or distant metastasis decreased with age of 64 years or greater (P<0.001 and P=0.034). For T1b melanomas, the risk of distant metastasis increased with increasing age (P=0.047). LIMITATIONS: Data were from a single county.  Conclusion: T1b melanomas of the face/ear/scalp demonstrated a higher risk of lymph node or distant metastasis and may help guide the recommendation of SLNB, imaging, and surveillance. Younger patients may be more strongly considered for SLNB and older patients with T1b melanomas may warrant imaging.  J Drugs Dermatol. 2024;23(5):306-310. doi:10.36849/JDD.7667.

[Circumscribed choroidal hemangioma and non-pigmented choroidal melanoma: clinical, instrumental and molecular genetic features]. / Otgranichennaya gemangioma i bespigmentnaya melanoma khorioidei: kliniko-instrumental'nye i molekulyarno-geneticheskie osobennosti.

Circumscribed choroidal hemangioma (CCH) and early non-pigmented choroidal melanoma (CM) have similar clinical, ultrasound and morphometric features, which in some cases makes their differential diagnosis difficult. There are few studies in the literature devoted to a comparative analysis of the molecular genetic features of CCH and non-pigmented CM, and the results of those studies are contradictory. PURPOSE: This study attempts to develop a method of non-invasive molecular genetic differential diagnostics of CCH and non-pigmented CM. MATERIAL AND METHODS: Based on the results of clinical and instrumental examination methods, 60 patients (60 eyes) with CCH (n=30) and non-pigmented CM (n=30) were included in this prospective study. The control group consisted of 30 individuals without intraocular tumors. Mutations in the GNAQ/GNA11 genes were determined by real-time PCR using the analysis of genomic circulating tumor DNA isolated from peripheral blood plasma. The average follow-up period was 12.1±1.8 months. RESULTS: The study revealed a significant association of mutations in exons 4 and 5 of the GNAQ/GNA11 genes with the presence of non-pigmented CM (27/30; 90%). These mutations were not detected in the group of patients with CCH. Mutations in exons 4 and 5 of the GNAQ/GNA11 genes were also not detected in the control group of healthy individuals. CONCLUSION: This study proposes a method of non-invasive and low-cost differential diagnostics based on molecular genetic analysis and detection of mutations in exons 4 and 5 of the GNAQ and GNA11 genes, which are specific for CM (90%).

BRAF-Mutated Melanoma Journey in Latin America: Expert Recommendations From Diagnosis to Treatment.

OBJECTIVES: • Gather a panel of Latin American experts in testing and treating BRAF-melanoma. • Describe the current landscape of BRAF-mutated melanoma in Latin America. • Outline the current gaps in testing and recommend improvements for testing and treating BRAF-mutated melanoma in the region. INTRODUCTION: Melanoma prevalence in Latin America is lower than in high- and middle-income countries. However, recent data indicate that the region's incidence and mortality are rising, with more stage IV patients being diagnosed. According to international clinical practice guidelines, conducting BRAF-mutation testing in patients with stage III or stage IV melanoma and high-risk resected disease is imperative. Still, BRAF-mutation testing and targeted therapies are inconsistently available in the region. METHODS: Americas Health Foundation convened a meeting of Latin American experts on BRAF-mutated melanoma to develop guidelines and recommendations for diagnosis through treatment. RESULTS AND CONCLUSIONS: Some recommendations for improving diagnostics through improving access and reducing the cost of BRAF-mutation testing, enhancing efficiency in pathology laboratories, and creating country-specific local guidelines. The panel also gave treatment recommendations for neo-adjuvant therapy, adjuvant therapy, and therapy for patients with metastatic disease in Latin America.

Early diagnosis of melanoma: a randomized trial assessing the impact of the transmission of photographs taken with a smartphone from the general practitioner to the dermatologist on the time to dermatological consultation.

BACKGROUND: Difficulty obtaining a dermatological consultation is an obstacle to the early diagnosis of melanoma. On the one hand, patients survival depends on the lesion thickness at the time of diagnosis. On the other hand, dermatologists treat many patients with benign lesions. Optimizing patient care pathways is a major concern. The aim of the present study was to assess whether the e-mail transmission of photographs of suspected melanoma lesions between general practitioners (GPs) and dermatologists reduces the time to dermatological consultation for patients whose suspicious skin lesions ultimately require resection. METHODS: We conducted a cluster-randomized controlled study in primary care involving 51 French GPs between April 2017 and August 2019. A total of 250 patients referred to a dermatologist for a suspected melanoma lesion were included GPs were randomized to either the smartphone arm or the usual care arm. In the smartphone arm, the GPs referred patients to the dermatologist by sending 2 photographs of the suspicious lesion using their smartphone. The dermatologist then had to set up an appointment at an appropriate time. In the usual care arm, GPs referred patients to a dermatologist according to their usual practice. The primary outcome was the time to dermatological consultation for patients whose lesion ultimately required resection. RESULTS: 57 GPs volunteered were randomized (27 to the smartphone arm, and 30 to the usual care arm). A total of 125 patients were included in each arm (mean age: 49.8 years; 53% women) and followed 8 months. Twenty-three dermatologists participated in the study. The time to dermatological consultation for patients whose suspicious skin lesion required resection was 56.5 days in the smartphone arm and 63.7 days in the usual care arm (mean adjusted time reduction: -18.5 days, 95% CI [-74.1;23.5], p = .53). CONCLUSIONS: The e-mail transmission of photographs from GPs to dermatologists did not improve the dermatological management of patients whose suspicious skin lesions ultimately required resection. Further research is needed to validate quality criteria that might be useful for tele-expertise in dermatology. TRIAL REGISTRATION: Registered on ClinicalTrials.gov under reference number NCT03137511 (May 2, 2017).

Prognostic biomarkers in melanoma: a 2023 update from clinical trials in different therapeutic scenarios.

INTRODUCTION: Over the past decade, significant advancements in the field of melanoma have included the introduction of a new staging system and the development of immunotherapy and targeted therapies, leading to changes in substage classification and impacting patient prognosis. Despite these strides, early detection remains paramount. The quest for dependable prognostic biomarkers is ongoing, given melanoma's unpredictable nature, especially in identifying patients at risk of relapse. Reliable biomarkers are critical for informed treatment decisions. AREAS COVERED: This review offers a comprehensive review of prognostic biomarkers in the context of clinical trials for immunotherapy and targeted therapy. It explores different clinical scenarios, including adjuvant, metastatic, and neo-adjuvant settings. Key findings suggest that tumor mutational burden, PD-L1 expression, IFN-γ signature, and immune-related factors are promising biomarkers associated with improved treatment responses. EXPERT OPINION: Identifying practical prognostic factors for melanoma therapy is challenging due to the tumor's heterogeneity. Promising biomarkers include tumor mutational burden (TMB), circulating tumor DNA, and those characterizing the tumor microenvironment, especially the immune component. Future research should prioritize large-scale, prospective studies to validate and standardize these biomarkers, emphasizing clinical relevance and real-world applicability. Easily accessible biomarkers have the potential to enhance the precision and effectiveness of melanoma management.

[Predicting the effectiveness of organ-preserving treatment of choroidal melanoma using optical coherence tomography data]. / Prognozirovanie effektivnosti organosokhrannogo lecheniya melanomy khorioidei po dannym opticheskoi kogerentnoi tomografii.

Optical coherence tomography (OCT) is currently widely used for the diagnosis of choroidal melanoma (CM), but the problem of predicting the outcomes of planned CM treatment remains unsolved. PURPOSE: This study was conducted to identify OCT signs that adversely affect the outcome of organ-preserving CM treatment. MATERIAL AND METHODS: OCT scan images of 30 patients who underwent organ-preserving treatment and were under observation were selected for this study. Brachytherapy (BT) as monotherapy was performed in 27 patients (in 2 cases - twice, and in 1 case - three times), in one patient - in combination with the previous transpupillary thermotherapy (TTT). Multiple TTT (4 sessions within 4 months) as monotherapy were performed in 2 patients. In 9 cases, a single organ-preserving treatment (BT - 6 patients, TTT - 3 patients) was ineffective. In these cases, the effectiveness of the first stage of organ-preserving treatment was taken into account. RESULTS: Seven signs of an unfavorable prognosis of the performed treatment were identified by analyzis of tomograms and statistical processing of the obtained data. These signs include: the presence of intraretinal edema, detachment of the neuroepithelium (NED) over the tumor, including with a break in the photoreceptors, accumulation of transudate over the tumor, the presence of large cysts, intraretinal cavities and NED near the tumor (secondary retinal detachment). A combination of three or more signs were observed in all cases of inefficiency of the first stage of treatment. Most often, intraretinal edema and NED over the tumor were combined with the accumulation of subretinal transudate and NED near the tumor. The presence of 6 or all 7 signs took place in cases of a negative therapeutic effect after local destruction. CONCLUSION: When planning organ-preserving CM treatment, in addition to biometric parameters, it is necessary to pay special attention to the identification of such morphological signs as NED over and near the tumor, accumulation of transudate under the NED, the presence of intraretinal edema, large intraretinal cysts and cavities.

Malignant melanoma in a 12-year-old boy 17 months after completing hepatoblastoma treatment.

BACKGROUND: Melanoma is rare as a secondary malignant neoplasm among childhood cancer survivors. CASE: We report a case of a 12-year-old boy who developed malignant melanoma with systemic metastases 17 months after completing treatment for hepatoblastoma. The diagnosis was made unexpectedly based on a bone marrow examination. The patient did not respond to immune checkpoint inhibitor therapy and died 6 weeks after being diagnosed with melanoma. Whole-exome sequencing to examine 103 genes associated with cancer predisposition did not identify any germ-line variants. CONCLUSION: This case study provides a unique example of melanoma in a childhood cancer survivor following hepatoblastoma treatment but does not identify any candidate variant to link hepatoblastoma and melanoma.

Automated Prediction of Malignant Melanoma using Two-Stage Convolutional Neural Network.

PURPOSE: A skin lesion refers to an area of the skin that exhibits anomalous growth or distinctive visual characteristics compared to the surrounding skin. Benign skin lesions are noncancerous and generally pose no threat. These irregular skin growths can vary in appearance. On the other hand, malignant skin lesions correspond to skin cancer, which happens to be the most prevalent form of cancer in the United States. Skin cancer involves the unusual proliferation of skin cells anywhere on the body. The conventional method for detecting skin cancer is relatively more painful. METHODS: This work involves the automated prediction of skin cancer and its types using two stage Convolutional Neural Network (CNN). The first stage of CNN extracts low level features and second stage extracts high level features. Feature selection is done using these two CNN and ABCD (Asymmetry, Border irregularity, Colour variation, and Diameter) technique. The features extracted from the two CNNs are fused with ABCD features and fed into classifiers for the final prediction. The classifiers employed in this work include ensemble learning methods such as gradient boosting and XG boost, as well as machine learning classifiers like decision trees and logistic regression. This methodology is evaluated using the International Skin Imaging Collaboration (ISIC) 2018 and 2019 dataset. RESULTS: As a result, the first stage CNN which is used for creation of new dataset achieved an accuracy of 97.92%. Second stage CNN which is used for feature selection achieved an accuracy of 98.86%. Classification results are obtained for both with and without fusion of features. CONCLUSION: Therefore, two stage prediction model achieved better results with feature fusion.

Detecting normal and cancer skin cells via glycosylation and adhesion signatures: A path to enhanced microfluidic phenotyping.

Recruiting circulating cells based on interactions between surface receptors and corresponding ligands holds promise for capturing cells with specific adhesive properties. Our study investigates the adhesion of skin cells to specific lectins, particularly focusing on advancements in lectin-based biosensors with diagnostic potential. We explore whether we can successfully capture normal skin (melanocytes and keratinocytes) and melanoma (WM35, WM115, WM266-4) cells in a low-shear flow environment by coating surfaces with lectins. Specifically, we coated surfaces with Dolichos biflorus (DBA) and Maackia Amurensis (MAL) lectins, which were used to detect and capture specific skin cells from the flow of cell mixture. Alterations in glycan expression (confirmed by fluorescent microscopy) demonstrated that DBA binds predominantly to normal skin cells, while MAL interacts strongly with melanoma cells. Assessing adhesion under static and dynamic low-shear stress conditions (up to 30 mPa) underscores the reliability of DBA and MAL as markers for discriminating specific cell type. Melanocytes and keratinocytes adhere to DBA-coated surfaces, while melanoma cells prefer MAL-coated surfaces. A comprehensive analysis encompassing cell shape, cytoskeleton, and focal adhesions shows the independence of our approach from the inherent characteristics of cells, thus demonstrating its robustness. Our results carry practical implications for lectin-biosensor designs, emphasizing the significance of glycan-based discrimination of pathologically altered cells. Combined with microfluidics, it demonstrates the value of cell adhesion as a discriminant of cancer-related changes, with potential applications spanning diagnostics, therapeutic interventions, and advanced biomedical technologies.

Malignant melanoma arising in a burn scar.

Secondary malignancies are rare but devastating complications of longstanding burn scars. Squamous cell carcinoma is the most common, followed by basal cell carcinoma and melanomas. There are fewer than 50 total reported cases of malignant melanomas arising in burn scars. We report a case of malignant melanoma arising within a longstanding burn scar confirmed by histology, FISH, and PRAME staining to further characterize melanomas arising in burn scars and to illustrate the diagnostic challenges they present.

Metastasis diagnosis using attenuated total reflection-Fourier transform infra-red (ATR-FTIR) spectroscopy.

The suitability of Fourier transform infrared spectroscopy as a metastasis prognostic tool has not been reported for some cancer types. Our main aim was to show spectroscopic differences between live un-preprocessed cancer cells of different metastatic levels. Spectra of four cancer cell pairs, including colon cancer (SW480, SW620); human melanoma (WM115, WM266.4); murine melanoma (B16F01, B16F10); and breast cancer (MCF7, MDA-MB-231); each pair having the same genetic background, but different metastatic level were analyzed in the regions 1400-1700 cm-1 and 3100-3500 cm-1 using Principal Component Analysis, curve fitting, multifractal dimension and receiver operating characteristic (ROC) curves. The results show spectral markers I1540/I1473, I1652/I1473, [Formula: see text], and multifractal dimension of the spectral images are significantly different for the cells based on their metastatic levels. ROC curve analysis showed good diagnostic performance of the spectral markers in separating cells based on metastatic degree, with areas under the ROC curves having 95% confidence interval lower limits greater than 0.5 for most instances. These spectral features can be important in predicting the probability of metastasis in primary tumors, providing useful guidance for treatment planning. Our markers are effective in differentiating metastatic levels without sample fixation or drying and therefore could be compactible for future use in in-vivo procedures involving spectroscopic cancer diagnosis.

Multimodal MALDI imaging mass spectrometry for improved diagnosis of melanoma.

Imaging mass spectrometry (IMS) provides promising avenues to augment histopathological investigation with rich spatio-molecular information. We have previously developed a classification model to differentiate melanoma from nevi lesions based on IMS protein data, a task that is challenging solely by histopathologic evaluation. Most IMS-focused studies collect microscopy in tandem with IMS data, but this microscopy data is generally omitted in downstream data analysis. Microscopy, nevertheless, forms the basis for traditional histopathology and thus contains invaluable morphological information. In this work, we developed a multimodal classification pipeline that uses deep learning, in the form of a pre-trained artificial neural network, to extract the meaningful morphological features from histopathological images, and combine it with the IMS data. To test whether this deep learning-based classification strategy can improve on our previous results in classification of melanocytic neoplasia, we utilized MALDI IMS data with collected serial H&E stained sections for 331 patients, and compared this multimodal classification pipeline to classifiers using either exclusively microscopy or IMS data. The multimodal pipeline achieved the best performance, with ROC-AUCs of 0.968 vs. 0.938 vs. 0.931 for the multimodal, unimodal microscopy and unimodal IMS pipelines respectively. Due to the use of a pre-trained network to perform the morphological feature extraction, this pipeline does not require any training on large amounts of microscopy data. As such, this framework can be readily applied to improve classification performance in other experimental settings where microscopy data is acquired in tandem with IMS experiments.