TY - JOUR A1 - Maron, Roman C. A1 - Haggenmüller, Sarah A1 - von Kalle, Christof A1 - Utikal, Jochen S. A1 - Meier, Friedegund A1 - Gellrich, Frank F. A1 - Hauschild, Axel A1 - French, Lars E. A1 - Schlaak, Max A1 - Ghoreschi, Kamran A1 - Kutzner, Heinz A1 - Heppt, Markus V. A1 - Haferkamp, Sebastian A1 - Sondermann, Wiebke A1 - Schadendorf, Dirk A1 - Schilling, Bastian A1 - Hekler, Achim A1 - Krieghoff-Henning, Eva A1 - Kather, Jakob N. A1 - Fröhling, Stefan A1 - Lipka, Daniel B. A1 - Brinker, Titus J. T1 - Robustness of convolutional neural networks in recognition of pigmented skin lesions JF - European Journal of Cancer N2 - Background A basic requirement for artificial intelligence (AI)–based image analysis systems, which are to be integrated into clinical practice, is a high robustness. Minor changes in how those images are acquired, for example, during routine skin cancer screening, should not change the diagnosis of such assistance systems. Objective To quantify to what extent minor image perturbations affect the convolutional neural network (CNN)–mediated skin lesion classification and to evaluate three possible solutions for this problem (additional data augmentation, test-time augmentation, anti-aliasing). Methods We trained three commonly used CNN architectures to differentiate between dermoscopic melanoma and nevus images. Subsequently, their performance and susceptibility to minor changes (‘brittleness’) was tested on two distinct test sets with multiple images per lesion. For the first set, image changes, such as rotations or zooms, were generated artificially. The second set contained natural changes that stemmed from multiple photographs taken of the same lesions. Results All architectures exhibited brittleness on the artificial and natural test set. The three reviewed methods were able to decrease brittleness to varying degrees while still maintaining performance. The observed improvement was greater for the artificial than for the natural test set, where enhancements were minor. Conclusions Minor image changes, relatively inconspicuous for humans, can have an effect on the robustness of CNNs differentiating skin lesions. By the methods tested here, this effect can be reduced, but not fully eliminated. Thus, further research to sustain the performance of AI classifiers is needed to facilitate the translation of such systems into the clinic. KW - artificial intelligence KW - machine learning KW - deep learning KW - neural networks KW - dermatology KW - skin neoplasms KW - melanoma KW - nevus Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-370245 VL - 145 ER - TY - JOUR A1 - Tappenbeck, Nils A1 - Schröder, Hannes M. A1 - Niebergall-Roth, Elke A1 - Hassinger, Fathema A1 - Dehio, Ulf A1 - Dieter, Kathrin A1 - Kraft, Korinna A1 - Kerstan, Andreas A1 - Esterlechner, Jasmina A1 - Frank, Natasha Y. A1 - Scharffetter-Kochanek, Karin A1 - Murphy, George F. A1 - Orgill, Dennis P. A1 - Beck, Joachim A1 - Frank, Markus H. A1 - Ganss, Christoph A1 - Kluth, Mark A. T1 - In vivo safety profile and biodistribution of GMP-manufactured human skin-derived ABCB5-positive mesenchymal stromal cells for use in clinical trials JF - Cytotherapy N2 - Background aims Human dermal ABCB5-expressing mesenchymal stromal cells (ABCB5+ MSCs) represent a promising candidate for stem cell–based therapy of various currently uncurable diseases in several fields of regenerative medicine. We have developed and validated a method to isolate, from human skin samples, and expand ABCB5+ MSCs that meet the guideline criteria of the International Society for Cellular Therapy. We are able to process these cells into a Good Manufacturing Practice–conforming, MSC-based advanced-therapy medicinal product. Methods To support the development of ABCB5+ MSCs for potential therapeutic topical, intramuscular and intravenous administration, we have tested our product in a series of Good Laboratory Practice–compliant nonclinical in-vivo studies addressing all relevant aspects of biosafety, including potential long-term persistence and proliferation, distribution to nontarget tissues, differentiation into undesired cell types, ectopic tissue formation, tumor formation and local tissue reaction. Results (i) Subcutaneous application of 1 × 107 ABCB5+ MSCs/animal and intravenous application of 2 × 106 ABCB5+ MSCs/animal, respectively, to immunocompromised mice did not result in safety-relevant biodistribution, persistence or proliferation of the cells; (ii) three monthly subcutaneous injections of ABCB5+ MSCs at doses ranging from 1 × 105 to 1 × 107 cells/animal and three biweekly intravenous injections of 2 × 106 ABCB5+ MSCs/animal, respectively, to immunocompromised mice were nontoxic and revealed no tumorigenic potential; and (iii) intramuscular injection of 5 × 106 ABCB5+ MSCs/animal to immunocompromised mice was locally well tolerated. Discussion The present preclinical in vivo data demonstrate the local and systemic safety and tolerability of a novel advanced-therapy medicinal product based on human skin-derived ABCB5+ MSCs. KW - stromal cells KW - stem cells KW - MSC KW - biodistribution KW - safety KW - ABCB5 KW - GMP KW - tumorigenicity KW - toxicity KW - persistence Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-240456 VL - 21 ER - TY - JOUR A1 - Sondermann, Wiebke A1 - Utikal, Jochen Sven A1 - Enk, Alexander H. A1 - Schadendorf, Dirk A1 - Klode, Joachim A1 - Hauschild, Axel A1 - Weichenthal, Michael A1 - French, Lars E. A1 - Berking, Carola A1 - Schilling, Bastian A1 - Haferkamp, Sebastian A1 - Fröhling, Stefan A1 - von Kalle, Christof A1 - Brinker, Titus J. T1 - Prediction of melanoma evolution in melanocytic nevi via artificial intelligence: A call for prospective data JF - European Journal of Cancer N2 - Recent research revealed the superiority of artificial intelligence over dermatologists to diagnose melanoma from images. However, 30–50% of all melanomas and more than half of those in young patients evolve from initially benign lesions. Despite its high relevance for melanoma screening, neither clinicians nor computers are yet able to reliably predict a nevus’ oncologic transformation. The cause of this lies in the static nature of lesion presentation in the current standard of care, both for clinicians and algorithms. The status quo makes it difficult to train algorithms (and clinicians) to precisely assess the likelihood of a benign skin lesion to transform into melanoma. In addition, it inhibits the precision of current algorithms since ‘evolution’ image features may not be part of their decision. The current literature reveals certain types of melanocytic nevi (i.e. ‘spitzoid’ or ‘dysplastic’ nevi) and criteria (i.e. visible vasculature) that, in general, appear to have a higher chance to transform into melanoma. However, owing to the cumulative nature of oncogenic mutations in melanoma, a more fine-grained early morphologic footprint is likely to be detectable by an algorithm. In this perspective article, the concept of melanoma prediction is further explored by the discussion of the evolution of melanoma, the concept for training of such a nevi classifier and the implications of early melanoma prediction for clinical practice. In conclusion, the authors believe that artificial intelligence trained on prospective image data could be transformative for skin cancer diagnostics by (a) predicting melanoma before it occurs (i.e. pre-in situ) and (b) further enhancing the accuracy of current melanoma classifiers. Necessary prospective images for this research are obtained via free mole-monitoring mobile apps. KW - melanoma KW - skin cancer KW - artificial Intelligence KW - deep learning KW - prediction Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-239263 VL - 119 ER - TY - JOUR A1 - Brinker, Titus J. A1 - Hekler, Achim A1 - Enk, Alexander H. A1 - Berking, Carola A1 - Haferkamp, Sebastian A1 - Hauschild, Axel A1 - Weichenthal, Michael A1 - Klode, Joachim A1 - Schadendorf, Dirk A1 - Holland-Letz, Tim A1 - von Kalle, Christof A1 - Fröhling, Stefan A1 - Schilling, Bastian A1 - Utikal, Jochen S. T1 - Deep neural networks are superior to dermatologists in melanoma image classification JF - European Journal of Cancer N2 - Background Melanoma is the most dangerous type of skin cancer but is curable if detected early. Recent publications demonstrated that artificial intelligence is capable in classifying images of benign nevi and melanoma with dermatologist-level precision. However, a statistically significant improvement compared with dermatologist classification has not been reported to date. Methods For this comparative study, 4204 biopsy-proven images of melanoma and nevi (1:1) were used for the training of a convolutional neural network (CNN). New techniques of deep learning were integrated. For the experiment, an additional 804 biopsy-proven dermoscopic images of melanoma and nevi (1:1) were randomly presented to dermatologists of nine German university hospitals, who evaluated the quality of each image and stated their recommended treatment (19,296 recommendations in total). Three McNemar's tests comparing the results of the CNN's test runs in terms of sensitivity, specificity and overall correctness were predefined as the main outcomes. Findings The respective sensitivity and specificity of lesion classification by the dermatologists were 67.2% (95% confidence interval [CI]: 62.6%–71.7%) and 62.2% (95% CI: 57.6%–66.9%). In comparison, the trained CNN achieved a higher sensitivity of 82.3% (95% CI: 78.3%–85.7%) and a higher specificity of 77.9% (95% CI: 73.8%–81.8%). The three McNemar's tests in 2 × 2 tables all reached a significance level of p < 0.001. This significance level was sustained for both subgroups. Interpretation For the first time, automated dermoscopic melanoma image classification was shown to be significantly superior to both junior and board-certified dermatologists (p < 0.001). KW - deep learning KW - melanoma KW - skin cancer KW - artificial intelligence Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-220539 VL - 119 ER - TY - JOUR A1 - Brinker, Titus J. A1 - Hekler, Achim A1 - Hauschild, Axel A1 - Berking, Carola A1 - Schilling, Bastian A1 - Enk, Alexander H. A1 - Haferkamp, Sebastian A1 - Karoglan, Ante A1 - von Kalle, Christof A1 - Weichenthal, Michael A1 - Sattler, Elke A1 - Schadendorf, Dirk A1 - Gaiser, Maria R. A1 - Klode, Joachim A1 - Utikal, Jochen S. T1 - Comparing artificial intelligence algorithms to 157 German dermatologists: the melanoma classification benchmark JF - European Journal of Cancer N2 - Background Several recent publications have demonstrated the use of convolutional neural networks to classify images of melanoma at par with board-certified dermatologists. However, the non-availability of a public human benchmark restricts the comparability of the performance of these algorithms and thereby the technical progress in this field. Methods An electronic questionnaire was sent to dermatologists at 12 German university hospitals. Each questionnaire comprised 100 dermoscopic and 100 clinical images (80 nevi images and 20 biopsy-verified melanoma images, each), all open-source. The questionnaire recorded factors such as the years of experience in dermatology, performed skin checks, age, sex and the rank within the university hospital or the status as resident physician. For each image, the dermatologists were asked to provide a management decision (treat/biopsy lesion or reassure the patient). Main outcome measures were sensitivity, specificity and the receiver operating characteristics (ROC). Results Total 157 dermatologists assessed all 100 dermoscopic images with an overall sensitivity of 74.1%, specificity of 60.0% and an ROC of 0.67 (range = 0.538–0.769); 145 dermatologists assessed all 100 clinical images with an overall sensitivity of 89.4%, specificity of 64.4% and an ROC of 0.769 (range = 0.613–0.9). Results between test-sets were significantly different (P < 0.05) confirming the need for a standardised benchmark. Conclusions We present the first public melanoma classification benchmark for both non-dermoscopic and dermoscopic images for comparing artificial intelligence algorithms with diagnostic performance of 145 or 157 dermatologists. Melanoma Classification Benchmark should be considered as a reference standard for white-skinned Western populations in the field of binary algorithmic melanoma classification. KW - benchmark KW - artificial intelligence KW - deep learning KW - melanoma Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-220569 VL - 111 ER - TY - JOUR A1 - Kervarrec, Thibault A1 - Samimi, Mahtab A1 - Guyétant, Serge A1 - Sarma, Bhavishya A1 - Chéret, Jérémy A1 - Blanchard, Emmanuelle A1 - Berthon, Patricia A1 - Schrama, David A1 - Houben, Roland A1 - Touzé, Antoine T1 - Histogenesis of Merkel Cell Carcinoma: A Comprehensive Review JF - Frontiers in Oncology N2 - Merkel cell carcinoma (MCC) is a primary neuroendocrine carcinoma of the skin. This neoplasia features aggressive behavior, resulting in a 5-year overall survival rate of 40%. In 2008, Feng et al. identified Merkel cell polyomavirus (MCPyV) integration into the host genome as the main event leading to MCC oncogenesis. However, despite identification of this crucial viral oncogenic trigger, the nature of the cell in which MCC oncogenesis occurs is actually unknown. In fact, several hypotheses have been proposed. Despite the large similarity in phenotype features between MCC tumor cells and physiological Merkel cells (MCs), a specialized subpopulation of the epidermis acting as mechanoreceptor of the skin, several points argue against the hypothesis that MCC derives directly from MCs. Alternatively, MCPyV integration could occur in another cell type and induce acquisition of an MC-like phenotype. Accordingly, an epithelial as well as a fibroblastic or B-cell origin of MCC has been proposed mainly based on phenotype similarities shared by MCC and these potential ancestries. The aim of this present review is to provide a comprehensive review of the current knowledge of the histogenesis of MCC. KW - merkel cell polyomavirus (MCPyV) KW - epithelial KW - fibroblast KW - B cell KW - Merkel cell carcinoma (MCC) KW - histogenesis KW - origin Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-325733 VL - 9 ER - TY - JOUR A1 - Scholz, S. L. A1 - Cosgarea, I. A1 - Süßkind, D. A1 - Murali, R. A1 - Möller, I. A1 - Reis, H. A1 - Leonardelli, S. A1 - Schilling, B. A1 - Schimming, T. A1 - Hadaschik, E. A1 - Franklin, C. A1 - Paschen, A. A1 - Sucker, A. A1 - Steuhl, K. P. A1 - Schadendorf, D. A1 - Westekemper, H. A1 - Griewank, K. G. T1 - NF1 mutations in conjunctival melanoma JF - British Journal of Cancer N2 - Background Conjunctival melanoma is a potentially deadly eye tumour. Despite effective local therapies, tumour recurrence and metastasis remain frequent. The genetics of conjunctival melanomas remain incompletely understood. Methods A large cohort of 63 conjunctival melanomas was screened for gene mutations known to be important in other melanoma subtypes by targeted next-generation sequencing. Mutation status was correlated with patient prognosis. Results Frequent mutations in genes activating the MAP kinase pathway were identified. NF1 mutations were most frequent (n = 21, 33%). Recurrent activating mutations were also identified in BRAF (n = 16, 25%) and RAS genes (n = 12, 19%; 11 NRAS and 1 KRAS). Conclusions Similar to cutaneous melanomas, conjunctival melanomas can be grouped genetically into four groups: BRAF-mutated, RAS-mutated, NF1-mutated and triple wild-type melanomas. This genetic classification may be useful for assessment of therapeutic options for patients with metastatic conjunctival melanoma KW - cancer genetics KW - eye cancer Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-233329 VL - 118 ER - TY - JOUR A1 - Huttelmaier, Johanna A1 - Benoit, Sandra A1 - Goebeler, Matthias T1 - Comorbidity in bullous pemphigoid: up-date and clinical implications JF - Frontiers in Immunology N2 - Bullous pemphigoid is the most common autoimmune blistering disease in industrialized countries and particularly affects the elderly. In this patient population, comorbid diseases are frequent and may complicate management and treatment of bullous pemphigoid. A better understanding why distinct diseases are more frequent in bullous pemphigoid patients may lead to new pathophysiological insights and - as a consequence - result in better patient care. The association of bullous pemphigoid with neurological and psychiatric diseases is well known and confirmed by several case-control studies. Association with further diseases such as malignancy and metabolic diseases are still discussed controversially. In recent years new relationships between bullous pemphigoid and autoimmune as well as inflammatory skin diseases have been reported. This review provides a systematic overview on studies addressing comorbidity in bullous pemphigoid patients. Increasing the awareness of both, common and rare comorbid diseases, may enable clinicians to optimize patient support and individualized treatment of bullous pemphigoid. KW - bullous pemphigoid KW - autoimmune skin blistering disease KW - comorbidity KW - neurologic disease KW - metabolic disease KW - malignancy KW - inflammatory disease Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-321671 VL - 14 ER - TY - JOUR A1 - Houben, Roland T1 - Reduced frequency of migraine attacks following coronavirus disease 2019: a case report JF - Journal of Medical Case Reports N2 - Background Severe acute respiratory syndrome coronavirus 2 is a virus affecting different organs and causing a wide variety and severity of symptoms. Headache as well as loss of smell and taste are the most frequently reported neurological manifestations of coronavirus disease 2019 induced by severe acute respiratory syndrome coronavirus 2. Here we report on a patient with chronic migraine and medication overuse headache, who experienced remarkable mitigation of migraine following coronavirus disease 2019. Case presentation For many years prior to the severe acute respiratory syndrome coronavirus 2 infection, a 57-year-old Caucasian male suffered from very frequent migraine attacks and for control of headaches he had been taking triptans almost daily. In the 16-month period before the outbreak of coronavirus disease 2019, triptan was taken 98% of the days with only a 21-day prednisolone-supported triptan holiday, which, however, had no longer-lasting consequences on migraine frequency. Upon severe acute respiratory syndrome coronavirus 2 infection, the patient developed only mild symptoms including fever, fatigue, and headache. Directly following recovery from coronavirus disease 2019, the patient surprisingly experienced a period with largely reduced frequency and severity of migraine attacks. Indeed, during 80 days following coronavirus disease 2019, migraine as well as triptan usage were restricted to only 25% of the days, no longer fulfilling criteria of a chronic migraine and medication overuse headache. Conclusion Severe acute respiratory syndrome coronavirus 2 infection might be capable of triggering mitigation of migraine. KW - migraine KW - triptan KW - severe acute respiratory syndrome coronavirus 2 KW - coronavirus disease 2019 KW - case report Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-357327 VL - 17 ER - TY - JOUR A1 - Haake, Markus A1 - Haack, Beatrice A1 - Schäfer, Tina A1 - Harter, Patrick N. A1 - Mattavelli, Greta A1 - Eiring, Patrick A1 - Vashist, Neha A1 - Wedekink, Florian A1 - Genssler, Sabrina A1 - Fischer, Birgitt A1 - Dahlhoff, Julia A1 - Mokhtari, Fatemeh A1 - Kuzkina, Anastasia A1 - Welters, Marij J. P. A1 - Benz, Tamara M. A1 - Sorger, Lena A1 - Thiemann, Vincent A1 - Almanzar, Giovanni A1 - Selle, Martina A1 - Thein, Klara A1 - Späth, Jacob A1 - Gonzalez, Maria Cecilia A1 - Reitinger, Carmen A1 - Ipsen-Escobedo, Andrea A1 - Wistuba-Hamprecht, Kilian A1 - Eichler, Kristin A1 - Filipski, Katharina A1 - Zeiner, Pia S. A1 - Beschorner, Rudi A1 - Goedemans, Renske A1 - Gogolla, Falk Hagen A1 - Hackl, Hubert A1 - Rooswinkel, Rogier W. A1 - Thiem, Alexander A1 - Romer Roche, Paula A1 - Joshi, Hemant A1 - Pühringer, Dirk A1 - Wöckel, Achim A1 - Diessner, Joachim E. A1 - Rüdiger, Manfred A1 - Leo, Eugen A1 - Cheng, Phil F. A1 - Levesque, Mitchell P. A1 - Goebeler, Matthias A1 - Sauer, Markus A1 - Nimmerjahn, Falk A1 - Schuberth-Wagner, Christine A1 - Felten, Stefanie von A1 - Mittelbronn, Michel A1 - Mehling, Matthias A1 - Beilhack, Andreas A1 - van der Burg, Sjoerd H. A1 - Riedel, Angela A1 - Weide, Benjamin A1 - Dummer, Reinhard A1 - Wischhusen, Jörg T1 - Tumor-derived GDF-15 blocks LFA-1 dependent T cell recruitment and suppresses responses to anti-PD-1 treatment JF - Nature Communications N2 - Immune checkpoint blockade therapy is beneficial and even curative for some cancer patients. However, the majority don’t respond to immune therapy. Across different tumor types, pre-existing T cell infiltrates predict response to checkpoint-based immunotherapy. Based on in vitro pharmacological studies, mouse models and analyses of human melanoma patients, we show that the cytokine GDF-15 impairs LFA-1/β2-integrin-mediated adhesion of T cells to activated endothelial cells, which is a pre-requisite of T cell extravasation. In melanoma patients, GDF-15 serum levels strongly correlate with failure of PD-1-based immune checkpoint blockade therapy. Neutralization of GDF-15 improves both T cell trafficking and therapy efficiency in murine tumor models. Thus GDF-15, beside its known role in cancer-related anorexia and cachexia, emerges as a regulator of T cell extravasation into the tumor microenvironment, which provides an even stronger rationale for therapeutic anti-GDF-15 antibody development. KW - cancer microenvironment KW - immunotherapy KW - T cells KW - tumour immunology Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-357333 VL - 14 ER -