Prospective multicenter study using artificial intelligence to improve dermoscopic melanoma diagnosis in patient care.

BACKGROUND: Early detection of melanoma, a potentially lethal type of skin cancer with high prevalence worldwide, improves patient prognosis. In retrospective studies, artificial intelligence (AI) has proven to be helpful for enhancing melanoma detection. However, there are few prospective studies confirming these promising results. Existing studies are limited by low sample sizes, too homogenous datasets, or lack of inclusion of rare melanoma subtypes, preventing a fair and thorough evaluation of AI and its generalizability, a crucial aspect for its application in the clinical setting. METHODS: Therefore, we assessed "All Data are Ext" (ADAE), an established open-source ensemble algorithm for detecting melanomas, by comparing its diagnostic accuracy to that of dermatologists on a prospectively collected, external, heterogeneous test set comprising eight distinct hospitals, four different camera setups, rare melanoma subtypes, and special anatomical sites. We advanced the algorithm with real test-time augmentation (R-TTA, i.e., providing real photographs of lesions taken from multiple angles and averaging the predictions), and evaluated its generalization capabilities. RESULTS: Overall, the AI shows higher balanced accuracy than dermatologists (0.798, 95% confidence interval (CI) 0.779-0.814 vs. 0.781, 95% CI 0.760-0.802; p = 4.0e-145), obtaining a higher sensitivity (0.921, 95% CI 0.900-0.942 vs. 0.734, 95% CI 0.701-0.770; p = 3.3e-165) at the cost of a lower specificity (0.673, 95% CI 0.641-0.702 vs. 0.828, 95% CI 0.804-0.852; p = 3.3e-165). CONCLUSION: As the algorithm exhibits a significant performance advantage on our heterogeneous dataset exclusively comprising melanoma-suspicious lesions, AI may offer the potential to support dermatologists, particularly in diagnosing challenging cases.

Melanoma is a type of skin cancer that can spread to other parts of the body, often resulting in death. Early detection improves survival rates. Computational tools that use artificial intelligence (AI) can be used to detect melanoma. However, few studies have checked how well the AI works on real-world data obtained from patients. We tested a previously developed AI tool on data obtained from eight different hospitals that used different types of cameras, which also included images taken of rare melanoma types and from a range of different parts of the body. The AI tool was more likely to correctly identify melanoma than dermatologists. This AI tool could be used to help dermatologists diagnose melanoma, particularly those that are difficult for dermatologists to diagnose.

Expression of the plasma membrane citrate carrier (pmCiC) in human cancerous tissues-correlation with tumour aggressiveness.

We have recently shown that cancer cells of various origins take up extracellular citrate through the plasma membrane citrate carrier (pmCiC), a specific plasma membrane citrate transporter. Extracellular citrate is required to support cancer cell metabolism, in particular fatty acid synthesis, mitochondrial activity, protein synthesis and histone acetylation. In addition, cancer cells tend to acquire a metastatic phenotype in the presence of extracellular citrate. Our recent study also showed that cancer-associated stromal cells synthesise and release citrate and that this process is controlled by cancer cells. In the present study, we evaluated the expression of pmCiC, fibroblast activation protein-α (FAP) and the angiogenesis marker cluster of differentiation 31 (CD31) in human cancer tissues of different origins. In the cohort studied, we found no correlation between disease stage and the expression of FAP or CD31. However, we have identified a clear correlation between pmCiC expression in cancer cells and cancer-associated stroma with tumour stage. It can be concluded that pmCiC is increased in cancer cells and in cancer-supporting cells in the tumour microenvironment at the later stages of cancer development, particularly at the metastatic sites. Therefore, pmCiC expression has the potential to serve as a prognostic marker, although further studies are needed.

Retrospective Single-Center Case Study of Clinical Variables and the Degree of Actinic Elastosis Associated with Rare Skin Cancers.

(1) Background: Rare skin cancers include epithelial, neuroendocrine, and hematopoietic neoplasias as well as cutaneous sarcomas. Ultraviolet (UV) radiation and sunburns are important drivers for the incidence of certain cutaneous sarcomas; however, the pathogenetic role of UV light is less clear in rare skin cancers compared to keratinocyte cancer and melanoma. In this study, we compared the degree of actinic elastosis (AE) as a surrogate for lifetime UV exposure among selected rare skin cancers (atypical fibroxanthoma [AFX], pleomorphic dermal sarcoma [PDS], dermatofibrosarcoma protuberans [DFSP], Kaposi sarcoma [KS], Merkel cell carcinoma [MCC], and leiomyosarcoma [LMS]) while taking into account relevant clinical variables (age, sex, and body site). (2) Methods: We newly established a semi-quantitative score for the degree of AE ranging from 0 = none to 3 = total loss of elastic fibers (basophilic degeneration) and multiplied it by the perilesional vertical extent (depth), measured histometrically (tumor-associated elastosis grade (TEG)). We matched the TEG of n = 210 rare skin cancers from 210 patients with their clinical variables. (3) Results: TEG values were correlated with age and whether tumors arose on UV-exposed body sites. TEG values were significantly higher in AFX and PDS cases compared to all other analyzed rare skin cancer types. As expected, TEG values were low in DFSP and KS, while MCC cases exhibited intermediate TEG values. (4) Conclusions: High cumulative UV exposure is more strongly associated with AFX/PDS and MCC than with other rare skin cancers. These important results expand the available data associated with rare skin cancers while also offering insight into the value of differentiating among these tumor types based on their relationship with sun exposure, potentially informing preventative, diagnostic and/or therapeutic approaches.

Fistulising skin metastases in Crohn's disease: a case report and review of the literature.

BACKGROUND: Metastatic Crohn's disease is a rare disorder characterized by various granulomatous skin lesions that occur independently of gastrointestinal tract involvement. However, currently there is no standardized care or specific treatment. Therapeutic approaches include immunosuppressive agents, such as corticosteroids, azathioprine, and monoclonal antibodies targeting inflammatory cytokines like tumor necrosis factor (TNF). CASE PRESENTATION: We present a case of a 29-year-old western European woman with significant blind ending abdominal subcutaneous fistulas and abscesses, who sought evaluation in the dermatology department. Histological examination revealed multiple epithelioid cell granulomas. There was no evidence of infectious or rheumatologic diseases such as sarcoidosis. The tentative diagnosis was metastatic Crohn's disease, which was not related to an intestinal manifestation of the disease. The patient responded to infliximab but had to discontinue it due to an allergic reaction. Subsequent adalimumab treatment failed to induce clinical remission; thus, therapy was switched to ustekinumab, resulting in a positive response. Written informed consent for publication of their clinical details and clinical images was obtained from the patient. For our study more than 1600 publications were screened for cases of metastatic Crohn's disease on PubMed database. 59 case reports with 171 patients were included in the analysis and evaluated for localization, diagnostic and therapeutic approaches, and complications and were summarized in this review. CONCLUSION: The successful ustekinumab treatment of a patient with metastatic Crohn's disease underscores the potential of this minimally investigated therapeutic option, highlighting the need for future treatment guidelines given the increasing prevalence of such cases.

Complete lymph node dissection in cutaneous melanoma patients with positive sentinel lymph node: Outcome and predictors in a retrospective cohort study over 16 years.

BACKGROUND: In melanoma treatment, complete lymph node dissection (CLND) has been considered the therapeutic gold standard in patients with positive sentinel lymph node biopsy (SLNB). This long-held approach was revised in 2017, with recent evidence questioning the therapeutic benefit of CLND in malignant melanoma (MM) therapy. In this study, we aimed to fill this knowledge gap by retrospectively analyzing the impact of CLND on MM patients' survival. METHODS: We retrospectively analyzed the multi-center population-based Clinical Cancer Registry at the Tumor Center Regensburg (TUDOK) database (2004-2020) to identify patients who had been diagnosed with SLN-positive MM and underwent (non)invasive management thereof. Patient cohorts were subdivided according to the treatment received (CLND and waiving CLND). Primary outcomes included overall survival (OS), recurrence-free survival (RFS), and cumulative recurrence rate. RESULTS: We identified 1143 MM patients, of whom 126 (11.0%) had positive SLN status. CLND was waived in the majority of SLN-positive MM cases (n = 71; 56.3%), with 55 (43.7%) patients undergoing CLND. Univariable and multivariable Cox regression revealed no significant advantage for CLND patients compared to non-CLND patients in OS (HR=0.970, p = 0.915 and HR=1.295, p = 0.479, respectively), RFS (HR=1.050, p = 0.849 and HR=1.220, p = 0.544, respectively), and cumulative recurrence rate (HR=1.234, p = 0.441 and HR=1.220, p = 0.544), respectively). CONCLUSION: We found that CLND had no significant impact on patient survival and MM recurrence rate, thus corroborating the validity of current clinical guidelines.

Federated Learning for Decentralized Artificial Intelligence in Melanoma Diagnostics.

Importance: The development of artificial intelligence (AI)-based melanoma classifiers typically calls for large, centralized datasets, requiring hospitals to give away their patient data, which raises serious privacy concerns. To address this concern, decentralized federated learning has been proposed, where classifier development is distributed across hospitals. Objective: To investigate whether a more privacy-preserving federated learning approach can achieve comparable diagnostic performance to a classical centralized (ie, single-model) and ensemble learning approach for AI-based melanoma diagnostics. Design, Setting, and Participants: This multicentric, single-arm diagnostic study developed a federated model for melanoma-nevus classification using histopathological whole-slide images prospectively acquired at 6 German university hospitals between April 2021 and February 2023 and benchmarked it using both a holdout and an external test dataset. Data analysis was performed from February to April 2023. Exposures: All whole-slide images were retrospectively analyzed by an AI-based classifier without influencing routine clinical care. Main Outcomes and Measures: The area under the receiver operating characteristic curve (AUROC) served as the primary end point for evaluating the diagnostic performance. Secondary end points included balanced accuracy, sensitivity, and specificity. Results: The study included 1025 whole-slide images of clinically melanoma-suspicious skin lesions from 923 patients, consisting of 388 histopathologically confirmed invasive melanomas and 637 nevi. The median (range) age at diagnosis was 58 (18-95) years for the training set, 57 (18-93) years for the holdout test dataset, and 61 (18-95) years for the external test dataset; the median (range) Breslow thickness was 0.70 (0.10-34.00) mm, 0.70 (0.20-14.40) mm, and 0.80 (0.30-20.00) mm, respectively. The federated approach (0.8579; 95% CI, 0.7693-0.9299) performed significantly worse than the classical centralized approach (0.9024; 95% CI, 0.8379-0.9565) in terms of AUROC on a holdout test dataset (pairwise Wilcoxon signed-rank, P < .001) but performed significantly better (0.9126; 95% CI, 0.8810-0.9412) than the classical centralized approach (0.9045; 95% CI, 0.8701-0.9331) on an external test dataset (pairwise Wilcoxon signed-rank, P < .001). Notably, the federated approach performed significantly worse than the ensemble approach on both the holdout (0.8867; 95% CI, 0.8103-0.9481) and external test dataset (0.9227; 95% CI, 0.8941-0.9479). Conclusions and Relevance: The findings of this diagnostic study suggest that federated learning is a viable approach for the binary classification of invasive melanomas and nevi on a clinically representative distributed dataset. Federated learning can improve privacy protection in AI-based melanoma diagnostics while simultaneously promoting collaboration across institutions and countries. Moreover, it may have the potential to be extended to other image classification tasks in digital cancer histopathology and beyond.

Dermatologist-like explainable AI enhances trust and confidence in diagnosing melanoma.

Artificial intelligence (AI) systems have been shown to help dermatologists diagnose melanoma more accurately, however they lack transparency, hindering user acceptance. Explainable AI (XAI) methods can help to increase transparency, yet often lack precise, domain-specific explanations. Moreover, the impact of XAI methods on dermatologists' decisions has not yet been evaluated. Building upon previous research, we introduce an XAI system that provides precise and domain-specific explanations alongside its differential diagnoses of melanomas and nevi. Through a three-phase study, we assess its impact on dermatologists' diagnostic accuracy, diagnostic confidence, and trust in the XAI-support. Our results show strong alignment between XAI and dermatologist explanations. We also show that dermatologists' confidence in their diagnoses, and their trust in the support system significantly increase with XAI compared to conventional AI. This study highlights dermatologists' willingness to adopt such XAI systems, promoting future use in the clinic.

Adjuvant immunotherapy with nivolumab versus observation in completely resected Merkel cell carcinoma (ADMEC-O): disease-free survival results from a randomised, open-label, phase 2 trial.

BACKGROUND: Merkel cell carcinoma (MCC) is an immunogenic but aggressive skin cancer. Even after complete resection and radiation, relapse rates are high. PD-1 and PD-L1 checkpoint inhibitors showed clinical benefit in advanced MCC. We aimed to assess efficacy and safety of adjuvant immune checkpoint inhibition in completely resected MCC (ie, a setting without an established systemic standard-of-care treatment). METHODS: In this multicentre phase 2 trial, patients (any stage, Eastern Cooperative Oncology Group performance status 0-1) at 20 academic medical centres in Germany and the Netherlands with completely resected MCC lesions were randomly assigned 2:1 to receive nivolumab 480 mg every 4 weeks for 1 year, or observation, stratified by stage (American Joint Committee on Cancer stages 1-2 vs stages 3-4), age (<65 vs ≥65 years), and sex. Landmark disease-free survival (DFS) at 12 and 24 months was the primary endpoint, assessed in the intention-to-treat populations. Overall survival and safety were secondary endpoints. This planned interim analysis was triggered when the last-patient-in was followed up for more than 1 year. This study is registered with ClinicalTrials.gov (NCT02196961) and with the EU Clinical Trials Register (2013-000043-78). FINDINGS: Between Oct 1, 2014, and Aug 31, 2020, 179 patients were enrolled (116 [65%] stage 3-4, 122 [68%] ≥65 years, 111 [62%] male). Stratification factors (stage, age, sex) were balanced across the nivolumab (n=118) and internal control group (observation, n=61); adjuvant radiotherapy was more common in the control group. At a median follow-up of 24·3 months (IQR 19·2-33·4), median DFS was not reached (between-groups hazard ratio 0·58, 95% CI 0·30-1·12); DFS rates in the nivolumab group were 85% at 12 months and 84% at 24 months, and in the observation group were 77% at 12 months and 73% at 24 months. Overall survival results were not yet mature. Grade 3-4 adverse events occurred in 48 [42%] of 115 patients who received at least one dose of nivolumab and seven [11%] of 61 patients in the observation group. No treatment-related deaths were reported. INTERPRETATION: Adjuvant therapy with nivolumab resulted in an absolute risk reduction of 9% (1-year DFS) and 10% (2-year DFS). The present interim analysis of ADMEC-O might suggest clinical use of nivolumab in this area of unmet medical need. However, overall survival events rates, with ten events in the active treatment group and six events in the half-the-size observation group, are not mature enough to draw conclusions. The explorative data of our trial support the continuation of ongoing, randomised trials in this area. ADMEC-O suggests that adjuvant immunotherapy is clinically feasible in this area of unmet medical need. FUNDING: Bristol Myers Squibb.

[Multilocular pyoderma gangrenosum : Association with primary manifestation of primary biliary cholangitis]. / Multilokuläres Pyoderma gangraenosum : Assoziation mit der Erstmanifestation einer primär biliären Cholangitis.

A 16-year-old female patient with previously diagnosed acne vulgaris was transferred to our clinic in reduced general condition with rapidly progressive and extremely painful ulcerations. In the laboratory exam, inflammatory parameters were highly elevated, but she was normothermic. Based on the findings, we diagnosed multilocular pyoderma gangrenosum. Further investigations established the diagnosis of primary biliary cholangitis as the underlying condition. Treatment with systemic corticosteroids was initiated and we started therapy with ursodeoxycholic acid. This led to improvement within a few days. PAPA-syndrome (pyogenic arthritis, pyoderma gangrenosum and acne vulgaris) could be ruled out by genetic analysis.

Indurated erythema of abdominal skin: An unusual presentation of metastatic endometrial carcinoma-Case report with literature review.

Carcinoma erysipelatoides (CE) is a rare clinical manifestation of cutaneous metastasis, which mimics inflammatory conditions such as erysipelas. Depending on the site of the originating tumour, unusual manifestations involving different sites of the body may occur. We herein report a case of a 60-year-old female patient with metastatic endometrial carcinoma presenting as CE of the abdominal skin and the inguinal folds. Even though the diagnosis of advanced malignancy had been established before and she was currently receiving chemotherapy (carboplatin and paclitaxel), the clinical appearance closely resembled fungal (candidal intertrigo) and consecutively bacterial (erysipelas) infection, which resulted in treatment with antimycotics and antibiotics at first. Dermatohistopathological examination of skin biopsies revealed a diffuse and nodular infiltrate of pleomorphic atypical tumour cells with strong expression of cytokeratin 7 and PAX8, also detectable within lymphatic vessels. Therapy comprised antiseptic ointments to prevent superinfection, palliative electron beam radiation and supportive care. Since there were no targetable KRAS-, NRAS- and BRAF-gene mutations, systemic therapy was switched to checkpoint inhibition (pembrolizumab) in combination with lenvatinib. The overall prognosis of cutaneous metastasis of endometrial carcinoma is dismal with most patients succumbing to disease within few months. Similarly, our patient died after 3 months due to sepsis in the course of malignant pleural effusion. We aim to highlight the possibility of unusual sites of CE and the risk of respective clinical misdiagnoses.

Degree of Actinic Elastosis Is a Surrogate of Exposure to Chronic Ultraviolet Radiation and Correlates More Strongly with Cutaneous Squamous Cell Carcinoma than Basal Cell Carcinoma.

(1) Background: Keratinocyte cancer (KC) is associated with exposure to ultraviolet (UV) radiation. However, data are controversial as to whether chronic UV exposure or high intermittent UV exposure are key drivers of carcinogenesis in cutaneous squamous cell carcinoma (cSCC) and basal cell carcinoma (BCC). Prolonged sun exposure of the skin causes photo-aging, which is associated with actinic elastosis, a condition characterized by the degeneration of elastin in the upper dermis, which is assessable via conventional histology. In this study, we aimed to compare the degree of actinic elastosis in different types of KC with regard to various patient characteristics. (2) Methods: We defined a semiquantitative score for the degree of actinic elastosis ranging from 0 = none to 3 = total loss of elastic fibers (basophilic degeneration). The extent was measured histometrically by two independent dermatohistopathologists in the immediate vicinity of 353 KC. The scores were merged and matched with tumor types (cSCC and BCC with subtypes), and clinical variables such as body site, sex and age. (3) Results: As expected, the degree of actinic elastosis correlated with age. However, it was significantly higher in cSCC compared to BCC irrespective of age, sex, body site and tumor subtypes. (4): Conclusions: Lifetime sun exposure may be estimated via routine histology using this scoring technique for actinic elastosis as a surrogate marker. cSCCs are more strongly associated with chronic UV exposure than BCCs, even in sun-exposed localizations such as the face.

Critical evaluation of molecular tumour board outcomes following 2 years of clinical practice in a Comprehensive Cancer Centre.

BACKGROUND: Recently, molecular tumour boards (MTBs) have been integrated into the clinical routine. Since their benefit remains debated, we assessed MTB outcomes in the Comprehensive Cancer Center Ostbayern (CCCO) from 2019 to 2021. METHODS AND RESULTS: In total, 251 patients were included. Targeted sequencing was performed with PCR MSI-evaluation and immunohistochemistry for PD-L1, Her2, and mismatch repair enzymes. 125 treatment recommendations were given (49.8%). High-recommendation rates were achieved for intrahepatic cholangiocarcinoma (20/30, 66.7%) and gastric adenocarcinoma (10/16, 62.5%) as opposed to colorectal cancer (9/36, 25.0%) and pancreatic cancer (3/18, 16.7%). MTB therapies were administered in 47 (18.7%) patients, while 53 (21.1%) received alternative treatment regimens. Thus 37.6% of recommended MTB therapies were implemented (47/125 recommendations). The clinical benefit rate (complete + partial + mixed response + stable disease) was 50.0% for MTB and 63.8% for alternative treatments. PFS2/1 ratios were 34.6% and 16.1%, respectively. Significantly improved PFS could be achieved for m1A-tier-evidence-based MTB therapies (median 6.30 months) compared to alternative treatments (median 2.83 months; P = 0.0278). CONCLUSION: The CCCO MTB yielded a considerable recommendation rate, particularly in cholangiocarcinoma patients. The discrepancy between the low-recommendation rates in colorectal and pancreatic cancer suggests the necessity of a weighted prioritisation of entities. High-tier recommendations should be implemented predominantly.

Subtypes of Melanomas Associated with Different Degrees of Actinic Elastosis in Conventional Histology, Irrespective of Age and Body Site, Suggesting Chronic Ultraviolet Light Exposure as Driver for Lentigo Maligna Melanoma and Nodular Melanoma.

(1) Background: Ultraviolet (UV) radiation and sunburns are associated with an increased incidence of acquired nevi and melanomas. However, the data are controversial as to whether chronic UV exposure or high intermittent UV exposure is the major carcinogenic factor in melanocytic tumors. In this study, we compared the degree of actinic elastosis (AE) as a surrogate for lifetime UV exposure in nevi and different clinical melanoma subtypes (i.e., superficial spreading melanoma (SSM), nodular malignant melanoma (NMM), acral lentiginous melanoma (ALM), and lentigo maligna melanoma (LMM)) with respect to clinical variables (age, sex, and body site). (2) Methods: We defined a semi-quantitative score for the degree of AE ranging from 0 = none to 3 = total loss of elastic fibers (basophilic degeneration) and multiplied it by the perilesional vertical extent (depth), measured histometrically (tumor-associated elastosis grade (TEG)). We matched the TEG of n = 595 melanocytic lesions from 559 patients with their clinical variables. (3) Results: The TEG was correlated with age and UV-exposed body sites. Furthermore, the TEG was significantly higher in LMM than in all other types of melanomas and the TEG in NMM was higher than in SSM, irrespective of patient age and tumor site. (4) Conclusions: High cumulative UV exposure is more strongly associated with LMM and NMM than with other melanoma subtypes.

The Role of Citrate Homeostasis in Merkel Cell Carcinoma Pathogenesis.

Merkel cell carcinoma (MCC) is a rare but highly aggressive tumor of the skin with a poor prognosis. The factors driving this cancer must be better understood in order to discover novel targets for more effective therapies. In the search for targets, we followed our interest in citrate as a central and critical metabolite linked to fatty acid synthesis in cancer development. A key to citrate uptake in cancer cells is the high expression of the plasma membrane citrate transporter (pmCiC), which is upregulated in the different adenocarcinoma types tested so far. In this study, we show that the pmCiC is also highly expressed in Merkel cell carcinoma cell lines by western blot and human tissues by immunohistochemistry staining. In the presence of extracellular citrate, MCC cells show an increased proliferation rate in vitro; a specific pmCiC inhibitor (Na+-gluconate) blocks this citrate-induced proliferation. Furthermore, the 3D in vivo Chick Chorioallantoic Membrane (CAM) model showed that the application of Na+-gluconate also decreases Merkel cell carcinoma growth. Based on our results, we conclude that pmCiC and extracellular citrate uptake should be considered further as a potential novel target for the treatment of Merkel cell carcinoma.