Consistency of convolutional neural networks in dermoscopic melanoma recognition: A prospective real-world study about the pitfalls of augmented intelligence

BACKGROUND

Deep-learning convolutional neural networks (CNNs) have outperformed even experienced dermatologists in dermoscopic melanoma detection under controlled conditions. It remains unexplored how real-world dermoscopic image transformations affect CNN robustness.

OBJECTIVES

To investigate the consistency of melanoma risk assessment by two commercially available CNNs to help formulate recommendations for current clinical use.

METHODS

A comparative cohort study was conducted from January to July 2022 at the Department of Dermatology, University Hospital Basel. Five dermoscopic images of 116 different lesions on the torso of 66 patients were captured consecutively by the same operator without deliberate rotation. Classification was performed by two CNNs (CNN-1/CNN-2). Lesions were divided into four subgroups based on their initial risk scoring and clinical dignity assessment. Reliability was assessed by variation and intraclass correlation coefficients. Excisions were performed for melanoma suspicion or two consecutively elevated CNN risk scores, and benign lesions were confirmed by expert consensus (n = 3).

RESULTS

117 repeated image series of 116 melanocytic lesions (2 melanomas, 16 dysplastic naevi, 29 naevi, 1 solar lentigo, 1 suspicious and 67 benign) were classified. CNN-1 demonstrated superior measurement repeatability for clinically benign lesions with an initial malignant risk score (mean variation coefficient (mvc): CNN-1: 49.5(±34.3)%; CNN-2: 71.4(±22.5)%; p = 0.03), while CNN-2 outperformed for clinically benign lesions with benign scoring (mvc: CNN-1: 49.7(±22.7)%; CNN-2: 23.8(±29.3)%; p = 0.002). Both systems exhibited lowest score consistency for lesions with an initial malignant risk score and benign assessment. In this context, averaging three initial risk scores achieved highest sensitivity of dignity assessment (CNN-1: 94%; CNN-2: 89%). Intraclass correlation coefficients indicated 'moderate'-to-'good' reliability for both systems (CNN-1: 0.80, 95% CI:0.71-0.87, p < 0.001; CNN-2: 0.67, 95% CI:0.55-0.77, p < 0.001).

CONCLUSIONS

Potential user-induced image changes can significantly influence CNN classification. For clinical application, we recommend using the average of three initial risk scores. Furthermore, we advocate for CNN robustness optimization by cross-validation with repeated image sets.

TRIAL REGISTRATION

ClinicalTrials.gov (NCT04605822).

Patient-derived melanoma models

Melanoma is a very aggressive, rapidly metastasizing tumor that has been studied intensively in the past regarding the underlying genetic and molecular mechanisms. More recently developed treatment modalities have improved response rates and overall survival of patients. However, the majority of patients suffer from secondary treatment resistance, which requires in depth analyses of the underlying mechanisms. Here, melanoma models based on patients-derived material may play an important role. Consequently, a plethora of different experimental techniques have been developed in the past years. Among these are 3D and 4D culture techniques, organotypic skin reconstructs, melanoma-on-chip models and patient-derived xenografts, Every technique has its own strengths but also weaknesses regarding throughput, reproducibility, and reflection of the human situation. Here, we provide a comprehensive overview of currently used techniques and discuss their use in different experimental settings.

Single-cell transcriptomics reveals prominent expression of IL-14, IL-18, and IL-32 in psoriasis

RATIONALE

Psoriasis is a chronic inflammatory skin disease involving different cytokines and chemokines.

OBJECTIVES

Here we use single-cell transcriptomic analyses to identify relevant immune cell and nonimmune cell populations for an in-depth characterization of cell types and inflammatory mediators in this disease.

METHODS

Psoriasis skin lesions of eight patients are analyzed using single-cell technology. Data are further validated by in situ hybridization (ISH) of human tissues, serum analyses of human samples and tissues of a murine model of psoriasis, and by in vitro cell culture experiments.

RESULTS

Several different immune-activated cell types with particular cytokine patterns are identified such as keratinocytes, T-helper cells, dendritic cells, macrophages, and fibroblasts. Apart from well-known factors, IL-14 (TXLNA), IL-18, and IL-32 are identified with prominent expression in individual cell types in psoriasis. The percentage of inflammatory cellular subtypes expressing IL-14, IL-18, and IL-32 was significantly higher in psoriatic skin compared with healthy control skin. These findings were confirmed by ISH of human skin samples, in a murine model of psoriasis, in human serum samples, and in in vitro experiments.

CONCLUSIONS

Taken together, we provide a differentiated view of psoriasis immune-cell phenotypes that support the role of IL-14, IL-18, and IL-32 in psoriasis pathogenesis.

MEK inhibition induces expression of differentiation marker Keratin 10 in human keratinocytes

BRAF mutant metastatic melanoma was regularly treated in the past with a BRAF inhibitor (BRAFi) alone or in combination with inhibitors of the mitogen-activated protein kinase kinase (MEKi), which is still a common treatment. This combination therapy strongly reduced the occurrence of keratoacanthomas and squamous cell carcinoma, which was frequently seen when BRAFi was used as monotherapy. Here we addressed the question whether MEK inhibition counteracts squamous cell carcinoma development in part by promoting keratinocyte differentiation. Exposure of human immortalized keratinocytes to different concentrations of MEKi revealed a significant increase in the expression of differentiation-associated keratins K10 and K1 as determined by qRT-PCR and immunofluorescence staining. Taken together, the present study suggests that in a combined treatment of melanoma with BRAFi/MEKi, MEKi reduces the incidence of squamous cell carcinomas by promoting keratinocyte differentiation under combined BRAFi/MEKi treatment in melanoma. This might open further treatment perspectives for skin cancer treatment.

Atypical fibroxanthoma and pleomorphic dermal sarcoma - gene expression analysis compared with undifferentiated cutaneous squamous cell carcinoma

BACKGROUND

The histogenetic origin of atypical fibroxanthoma (AFX) and pleomorphic dermal sarcoma (PDS) has not been definitively elucidated. In addition to a fibroblastic origin, a keratinocytic differentiation is discussed due to strong clinical, histomorphological and molecular genetic similarities with undifferentiated cutaneous squamous cell carcinoma (cSCC).

PATIENTS AND METHODS

56 cases (36 AFXs, 8 PDSs, 12 undifferentiated cSCCs) were evaluated for their clinical, histomorphological, and immunohistochemical characteristics. RNA transcriptome analysis was performed on 18 cases (6 AFXs/PDSs, 6 undifferentiated cSCCs, 6 differentiated cSCCs).

RESULTS

Clinically, the strong similarities in age, gender and tumor location were confirmed. Without further immunohistochemical staining, histomorphological differentiation between AFX/PDS and undifferentiated cSCC is often impossible. Principal component analysis of the RNA transcriptome analysis showed that AFX/PDS and differentiated cSCC each formed their own cluster, while the undifferentiated cSCCs fall in between these two groups, but without forming a cluster of their own. When examining differentially expressed genes (DEGs), the heat maps showed that there were cases within the undifferentiated cSCC that were more likely to be AFX/PDS than differentiated cSCC based on their expression profile.

CONCLUSIONS

The results provide evidence of molecular similarities between AFX/PDS and undifferentiated cSCC and suggest a common histogenetic origin.

A flexible framework for anomaly Detection via dimensionality reduction

Anomaly detection is challenging, especially for large datasets in high dimensions. Here, we explore a general anomaly detection framework based on dimensionality reduction and unsupervised clustering. DRAMA is released as a general python package that implements the general framework with a wide range of built-in options. This approach identifies the primary prototypes in the data with anomalies detected by their large distances from the prototypes, either in the latent space or in the original, high-dimensional space. DRAMA is tested on a wide variety of simulated and real datasets, in up to 3000 dimensions, and is found to be robust and highly competitive with commonly used anomaly detection algorithms, especially in high dimensions. The flexibility of the DRAMA framework allows for significant optimization once some examples of anomalies are available, making it ideal for online anomaly detection, active learning, and highly unbalanced datasets. Besides, DRAMA naturally provides clustering of outliers for subsequent analysis.

Improved diagnosis by automated macro- and micro-anatomical region mapping of skin photographs

BACKGROUND

The exact location of skin lesions is key in clinical dermatology. On one hand, it supports differential diagnosis (DD) since most skin conditions have specific predilection sites. On the other hand, location matters for dermatosurgical interventions. In practice, lesion evaluation is not well standardized and anatomical descriptions vary or lack altogether. Automated determination of anatomical location could benefit both situations.

OBJECTIVE

Establish an automated method to determine anatomical regions in clinical patient pictures and evaluate the gain in DD performance of a deep learning model (DLM) when trained with lesion locations and images.

METHODS

Retrospective study based on three datasets: macro-anatomy for the main body regions with 6000 patient pictures partially labelled by a student, micro-anatomy for the ear region with 182 pictures labelled by a student and DD with 3347 pictures of 16 diseases determined by dermatologists in clinical settings. For each dataset, a DLM was trained and evaluated on an independent test set. The primary outcome measures were the precision and sensitivity with 95% CI. For DD, we compared the performance of a DLM trained with lesion pictures only with a DLM trained with both pictures and locations.

RESULTS

The average precision and sensitivity were 85% (CI 84-86), 84% (CI 83-85) for macro-anatomy, 81% (CI 80-83), 80% (CI 77-83) for micro-anatomy and 82% (CI 78-85), 81% (CI 77-84) for DD. We observed an improvement in DD performance of 6% (McNemar test P-value 0.0009) for both average precision and sensitivity when training with both lesion pictures and locations.

CONCLUSION

Including location can be beneficial for DD DLM performance. The proposed method can generate body region maps from patient pictures and even reach surgery relevant anatomical precision, e.g. the ear region. Our method enables automated search of large clinical databases and make targeted anatomical image retrieval possible.

Melanoma 2.0. Skin cancer as a paradigm for emerging diagnostic technologies, computational modelling and artificial intelligence

We live in an unprecedented time in oncology. We have accumulated samples and cases in cohorts larger and more complex than ever before. New technologies are available for quantifying solid or liquid samples at the molecular level. At the same time, we are now equipped with the computational power necessary to handle this enormous amount of quantitative data. Computational models are widely used helping us to substantiate and interpret data. Under the label of systems and precision medicine, we are putting all these developments together to improve and personalize the therapy of cancer. In this review, we use melanoma as a paradigm to present the successful application of these technologies but also to discuss possible future developments in patient care linked to them. Melanoma is a paradigmatic case for disruptive improvements in therapies, with a considerable number of metastatic melanoma patients benefiting from novel therapies. Nevertheless, a large proportion of patients does not respond to therapy or suffers from adverse events. Melanoma is an ideal case study to deploy advanced technologies not only due to the medical need but also to some intrinsic features of melanoma as a disease and the skin as an organ. From the perspective of data acquisition, the skin is the ideal organ due to its accessibility and suitability for many kinds of advanced imaging techniques. We put special emphasis on the necessity of computational strategies to integrate multiple sources of quantitative data describing the tumour at different scales and levels.

Clinical response of CARD14-associated papulosquamous eruption to an anti-interleukin-17A antibody

Here we present another family with CARD14-associated papulosquamous eruption, which is characterized by mutations in CARD14 and skin lesions resembling psoriasis and pityriasis rubra pilaris. We show beneficial therapeutic response to anti-IL17A treatment in one patient and performed immunomonitoring of our patient, exhibiting enhanced pSTAT3 levels in T cells before treatment, which normalized after treatment. Together, our data support the pathogenic role of IL-17A in this disease, which might have consequences for future treatment decisions in this rare condition.

COVID-19 related psychological burden and potential benefits of vaccination - Data from a repeated cross-sectional survey in healthcare workers

The COVID-19 pandemic is impacting the psychological well-being, especially of health care workers, for more than two years now. Here, we followed-up on a survey we conducted at the very beginning of the pandemic, to determine potential changes in psychological strain experienced by health care workers one year later. Since our first survey in 2020, COVID-19 vaccines have been established, thus we assessed whether vaccination-status might modulate psychological burden of health care workers. We also collected data on resilience and sleep, as those might be related to successful coping. Between March and April 2021, nurses and physicians (N ​= ​286) working at the University Hospital Augsburg - with high or low exposure to COVID-19 patients - took part in an online survey. We found that fully vaccinated personnel reported lower levels of anxiety, depression, stress and exhaustion suggesting the potential positive consequences of vaccination beyond the obvious protection against a COVID-19 infection. Nurses reported more depressive symptoms, anxiety, stress and exhaustion and lower levels of job fulfilment than physicians. Individuals with high exposure to COVID-19 patients reported higher exhaustion and depersonalization. Resilience and sleep quality were significantly correlated with psychological and work-related burden, suggesting their potential role as protective resources. In general, the comparison of the present data to the survey conducted in 2020 suggests an overall increase of psychological burden in health care workers. Despite these surely alarming findings, it should be noted that being vaccinated might come along with reduced psychological strain.

Current concepts of photosensitivity in cutaneous lupus erythematosus

Cutaneous lupus erythematosus (CLE) represents a complex autoimmune disease with a broad phenotypic spectrum ranging from acute to chronic destructive cutaneous lesions. Patients with CLE exhibit high photosensitivity and ultraviolet (UV) irradiation can lead to systemic flares in systemic lupus erythematosus. However, the exact mechanisms how UV irradiation enhances cutaneous inflammation in lupus are not fully understood. Recently, new molecular mechanisms of UV-driven immune responses in CLE were identified, offering potential therapeutic approaches. Especially the induction of type I interferons, central cytokines in lupus pathogenesis which are released by various skin cells, have become the focus of current research. In this review, we describe current pathogenic concepts of photosensitivity in lupus erythematosus, including UV-driven activation of intracellular nucleic acid sensors, cellular cytokine production and immune cell activation. Furthermore, we discuss activated pathways contributing to enhanced apoptosis as well as intracellular translocation of autoantigens thereby promoting CLE upon UV light exposure.

Impact of Chronic Inflammation in Psoriasis on Bone Metabolism

Psoriasis is a chronic inflammatory disease of the skin and joints associated with several comorbidities such as arthritis, diabetes mellitus and metabolic syndrome, including obesity, hypertension and dyslipidaemia, Crohn's disease, uveitis and psychiatric and psychological diseases. Psoriasis has been described as an independent risk factor for cardiovascular diseases and thus patients with psoriasis should be monitored for the development of cardiovascular disease or metabolic syndrome. However, there is mounting evidence that psoriasis also affects the development of osteoporosis, an important metabolic disease with enormous clinical and socioeconomic impact. At present, there are still controversial opinions about the role of psoriasis in osteoporosis. A more in depth analysis of this phenomenon is of great importance for affected patients since, until now, bone metabolism is not routinely examined in psoriatic patients, which might have important long-term consequences for patients and the health system. In the present review, we summarize current knowledge on the impact of psoriatic inflammation on bone metabolism and osteoporosis.

OP0076 L-ARGININE REPROGRAMS OSTEOCLAST PURINE METABOLISM AMELIORATING BONE LOSS IN RHEUMATOID ARTHRITIS

Background

Bone erosion is a clinical feature of rheumatoid arthritis related to disease severity and poor functional prognosis. Excessive osteoclast differentiation and insufficient osteoblast function are the main reasons for the erosive process in RA. Our previous investigation indicated that L-arginine supplementation not only diminished arthritic inflammation in the serum-induced arthritis (K/BxN) model but also decreased inflammatory joints osteoclast numbers (1).

Objectives

In the present study, we aim to investigate the metabolic action of L-arginine supplementation in RA, especially on periarticular bone erosion and systemic bone loss. We plan to depict the metabolic features of TNFα induced inflammatory osteoclasts after in vitro L-arginine supplementation.

Methods

Three murine arthritis models (serum-induced arthritis (K/BxN) model, collagen-induced arthritis model, and hTNFtg mice model) were analysed in this study. L-arginine was supplemented within the drinking water after the onset of arthritis. Bone parameters for axial skeleton (spine) and peripheral skeleton (tibia) from the respective group were quantified by μCT. HE and TRAP staining were performed to address further the erosion area and osteoclast numbers in periarticular sites. In vitro osteoclast differentiation was conducted with or without L-arginine treatment, in the presence or not of TNFα activation. Seahorse and SCENITH analyses were adopted to delineate the metabolic features. JC-1 staining and transmission electron microscopy (TEM) were used to depict the mitochondria metabolism. RNA-seq and mass spectrometry (MS) were performed to investigate the underlying molecular mechanism.

Results

Inflammation was diminished in all three arthritis models after L-arginine supplementation with a significant reduction in arthritic score. Moreover, an amelioration of periarticular bone erosion, systemic bone loss, and decreased osteoclast numbers in periarticular sites were observed in arthritic mice after L-arginine treatment. L-arginine also inhibited osteoclastogenesis in vitro, particularly under TNFα activation. Seahorse and SCENITH analyses indicated TNFα promoted glycolysis while blocking mitochondria-driven oxidative phosphorylations (OXPHOS) in pre-osteoclasts. Meanwhile, JC-1 staining and TEM images also showed that TNFα decreased mitochondria membrane potential and prompted damage of mitochondria. Surprisingly, L-arginine rescued the TNFα inhibition of OXPHOS while promoting ATP production.

RNA-seq and MS data confirmed the boost of OXPHOS after L-arginine treatment under TNFα activation. To interfere with OXPHOS, L-arginine inhibited cJun thus altered arginase-1 and arginase-2 expression. Moreover, the increased ATP in L-arginine treated cells facilitated purine metabolism, especially the production of inosine and hypoxanthine, contributing to the inhibition of osteoclastogenesis. Increasing Adenosine deaminase (ADA) is essential for the production of inosine and hypoxanthine due to the decreased inhibitory regulation of the transcription factor c-Jun.

Conclusion

These data strongly demonstrated that L-arginine ameliorates bone erosion in RA through metabolic reprogramming and perturbation of purine metabolism in osteoclasts. L-arginine might therefore benefit RA therapy by reducing joint inflammation and also ameliorating bone destruction.

References

[1]Hannemann, Nicole, et al. “Transcription factor Fra-1 targets arginase-1 to enhance macrophage-mediated inflammation in arthritis.” The Journal of clinical investigation 129.7 (2019): 2669-2684.

Disclosure of Interests

Shan Cao: None declared, Rui Song: None declared, Xianyi Meng: None declared, Katerina Kachler: None declared, Maximilian Fuchs: None declared, Xinyu Meng: None declared, Yixuan Li: None declared, Verena Taudte: None declared, Meik Kunz: None declared, Ursula Schloetzer-Schrehardt: None declared, Ulrike Schleicher: None declared, Xiaoxiang Chen Speakers bureau: AbbVie, Roche and Novartis, Georg Schett Speakers bureau: AbbVie, BMS, Celgene, Janssen, Eli Lilly, Novartis, Roche and UCB, Aline Bozec: None declared.

Case Report: Diagnostic and Therapeutic Challenges in Severe Mechanobullous Epidermolysis Bullosa Acquisita

Collagen VII is the main constituent of the anchoring fibrils, important adhesive structures that attach the epidermis to the dermal extracellular matrix. Two disorders are caused by dysfunction of collagen VII, both characterized by skin and mucosa fragility, epidermolysis bullosa acquisita (EBA) and dystrophic epidermolysis bullosa (DEB). EBA and DEB share high clinical similarities with significant difference in patients’ age of onset and pathogenesis. Our patients presented with severe and recalcitrant mechanobullous EBA with characteristic DIF, IIF and ELISA diagnostics. But in both women recessive COL7A1 variants were also found, in a monoallelic state. Collagen VII from EBA keratinocytes of our cases was significantly more vulnerable to proteolytic degradation than control keratinocytes, hinting that the heterozygous pathogenic variants were sufficient to destabilize the molecule in vitro. Thus, even if the amount and functionality of mutant and normal type VII collagen polypeptides is sufficient to assure dermal-epidermal adhesion in healthy individuals, the functionally-impaired proteins are probably more prone to development of autoantibodies against them. Our work suggests that testing for COL7A1 genetic variants should be considered in patients with EBA, which either have a patient history hinting towards underlying dystrophic epidermolysis bullosa or pose therapeutic challenges.

Lupus erythematosus: correlation of clinical and histological findings and proposal for a modified disease classification

BACKGROUND

Classification of lupus erythematosus (LE) is conflicting as it is carried out from different starting points. Whereas dermatological classifications categorize LE morphologically based on specific cutaneous lesions, rheumatologic classifications are based on symptomatic aspects. Indeed, LE is a systemic autoimmune disease with variable acuity and organ involvement. All cutaneous disease patterns may occur in both limited-cutaneous and systemic LE.

PATIENTS AND METHODS

76 LE-patients with complete clinical data, clinical photographs and biopsy of cutaneous manifestations as well as paraclinical findings were retrospectively analyzed. Based on a published two-dimensional classification system that considers disease-specific skin manifestations and final disease diagnosis separately, patients' diagnoses were revised and compared with those in medical records. In addition, the extent to which patients could be clustered by diagnosis based on their LE-specific skin manifestations, corresponding histopathological changes, and paraclinical data was investigated.

RESULTS

After re-evaluation, the proportion of patients with limited-cutaneous LE decreased from 82% previously to 24%. More than two-thirds of patients indeed showed intermediate or systemic LE. Disease-specific skin manifestations, histologic characteristics and paraclinical data did not cluster with final diagnoses.

CONCLUSIONS

First, the work underlines the systemic character of the disease. Second, a two-dimensional approach can help overcome classification difficulties in LE, as skin-morphologic and symptomatic aspects can be considered separately.

Reduced Serum Levels of Bone Formation Marker P1NP in Psoriasis

Psoriasis is a chronic inflammatory disease of the skin and joints. More recent data emphasize an association with dysregulated glucose and fatty acid metabolism, obesity, elevated blood pressure and cardiac disease, summarized as metabolic syndrome. TNF-α and IL-17, central players in the pathogenesis of psoriasis, are known to impair bone formation. Therefore, the relation between psoriasis and bone metabolism parameters was investigated. Two serum markers of either bone formation-N-terminal propeptide of type I procollagen (P1NP) or bone resorption-C-terminal telopeptide of type I collagen (CTX-I)-were analyzed in a cohort of patients with psoriasis vulgaris. In patients with psoriasis, P1NP serum levels were reduced compared to gender-, age-, and body mass index-matched healthy controls. CTX-I levels were indistinguishable between patients with psoriasis and controls. Consistently, induction of psoriasis-like skin inflammation in mice decreases bone volume and activity of osteoblasts. Moreover, efficient anti-psoriatic treatment improved psoriasis severity, but did not reverse decreased P1NP level suggesting that independent of efficient skin treatment psoriasis did affect bone metabolism and might favor the development of osteoporosis. Taken together, evidence is provided that bone metabolism might be affected by psoriatic inflammation, which may have consequences for future patient counseling and disease monitoring.

Single-cell trajectories of melanoma cell resistance to targeted treatment

OBJECTIVE

Cellular heterogeneity is regarded as a major factor affecting treatment response and resistance in malignant melanoma. Recent developments in single-cell sequencing technology have provided deeper insights into these mechanisms.

METHODS

Here, we analyzed a BRAFV600E-mutant melanoma cell line by single-cell RNA-seq under various conditions: cells sensitive to BRAF inhibition with BRAF inhibitor vemurafenib and cells resistant to BRAF inhibition with vemurafenib alone or vemurafenib in combination with the MEK1/2 inhibitors cobimetinib or trametinib. Dimensionality reduction by t-distributed stochastic neighbor embedding and self-organizing maps identified distinct trajectories of resistance development clearly separating the 4 treatment conditions in cell and gene state space.

RESULTS

Trajectories associated with resistance to single-agent treatment involved cell cycle, extracellular matrix, and de-differentiation programs. In contrast, shifts detected in double-resistant cells primarily affected translation and mitogen-activated protein kinase pathway reactivation, with a small subpopulation showing markers of pluripotency. These findings were validated in pseudotime analyses and RNA velocity measurements.

CONCLUSIONS

The single-cell transcriptomic analyses reported here employed a spectrum of bioinformatics methods to identify mechanisms of melanoma resistance to single- and double-agent treatments. This study deepens our understanding of treatment-induced cellular reprogramming and plasticity in melanoma cells and identifies targets of potential relevance to the management of treatment resistance.

A Transcriptome-Wide Isoform Landscape of Melanocytic Nevi and Primary Melanomas Identifies Gene Isoforms Associated with Malignancy

Genetic splice variants have become of central interest in recent years, as they play an important role in different cancers. Little is known about splice variants in melanoma. Here, we analyzed a genome-wide transcriptomic dataset of benign melanocytic nevi and primary melanomas (n = 80) for the expression of specific splice variants. Using kallisto, a map for differentially expressed splice variants in melanoma vs. benign melanocytic nevi was generated. Among the top genes with differentially expressed splice variants were Ras-related in brain 6B (RAB6B), a member of the RAS family of GTPases, Macrophage Scavenger Receptor 1 (MSR1), Collagen Type XI Alpha 2 Chain (COLL11A2), and LY6/PLAUR Domain Containing 1 (LYPD1). The Gene Ontology terms of differentially expressed splice variants showed no enrichment for functional gene sets of melanoma vs. nevus lesions, but between type 1 (pigmentation type) and type 2 (immune response type) melanocytic lesions. A number of genes such as Checkpoint Kinase 1 (CHEK1) showed an association of mutational patterns and occurrence of splice variants in melanoma. Moreover, mutations in genes of the splicing machinery were common in both benign nevi and melanomas, suggesting a common mechanism starting early in melanoma development. Mutations in some of these genes of the splicing machinery, such as Serine and Arginine Rich Splicing Factor A3 and B3 (SF3A3, SF3B3), were significantly enriched in melanomas as compared to benign nevi. Taken together, a map of splice variants in melanoma is presented that shows a multitude of differentially expressed splice genes between benign nevi and primary melanomas. The underlying mechanisms may involve mutations in genes of the splicing machinery.

[Alopecia areata universalis under treatment with sitagliptin : Possible immunological effect of dipeptidyl peptidase-4 inhibitors?]

Ein 64-jähriger Patient entwickelte 1 Monat nach Therapieeinleitung mit Sitagliptin, einem Dipeptidylpeptidase-4(DPP‑4)-Inhibitor, und Metformin eine Alopecia universalis. Die Therapie des Diabetes wurde auf das Sitagliptin eines anderen Herstellers und Dapagliflozin umgestellt. Auf unser Anraten wurde Sitagliptin abgesetzt und eine Monotherapie mit Dapagliflozin fortgeführt. Nach 6 Wochen war eine erneute Therapie mit Sitagliptin bei unzureichend eingestelltem Diabetes notwendig. Die Alopezie persistierte. Aufgrund des immunologischen Interaktionspotenzials vermuten wir eine Assoziation zwischen DPP-4-Inhibition und der Alopezie. Der kurze therapiefreie Zeitraum scheint zu gering, um ein erneutes Haarwachstum zu beobachten. DPP‑4 kann sowohl eine Inhibition als auch Aktivierung des Immunsystems bewirken.

POS0423 NCOA3 AMPLIFIES PROFIBROTIC TRANSCRIPTIONAL PROGRAMS IN SYSTEMIC SCLEROSIS

Background:

Excessive activation of fibroblasts with a TGFβ-biased gene signature and deposition of extracellular matrix are key features of fibrotic diseases. The mechanisms underlying these transcriptional changes remain poorly understood. Deregulation, mutations and malfunctions of transcriptional co-regulators, which can interact with multiple transcription factors and enable a broad-spectrum regulation of transcriptional networks, have been implicated as driving factors in a large number of diseases and pathologies.

Objectives:

In the present study, we aimed to analyze the role of the co-regulator Nuclear Receptor Co-Activator 3 (NCOA3) in fibroblast activation and tissue fibrosis, and to evaluate a potential interaction of NCOA3 with fibrosis-relevant transcription factors.

Methods:

NCOA3 was inhibited genetically by siRNA transfection and pharmacologically by the SRC3 inhibitor-2 (SI-2). We performed bulk RNASeq of human dermal fibroblasts and in silico transcription factor binding site screening of differentially expressed genes (DEGs). The interaction of NCOA3 and TGFβ-SMAD signaling was analyzed by reporter and CoIP assays.

Results:

The expression of NCOA3 in skin biopsies of SSc patients compared to normal controls demonstrated that SSc fibroblasts express modestly, but significantly reduced levels of NCOA3, which persisted in cultured SSc fibroblasts. Stimulation of normal fibroblasts with chronically high levels of TGFβ as they also occur in fibrotic tissue remodeling strongly decreased NCOA3 expression to a similar extent as in SSc fibroblasts. Furthermore, NCOA3 expression is also deregulated in different murine models of skin fibrosis. To investigate the functional effects of decreased NCOA3 levels, we targeted the expression of NCOA3 in normal fibroblasts. SiRNA-mediated knockdown of NCOA3 ameliorated TGFβ-induced gene expression, collagen release, myofibroblast differentiation and cell proliferation. In contrast, knockdown of NCOA3 had no effects on collagen release, expression of contractile proteins or gene expression in unstimulated fibroblasts, suggesting that NCOA3 is not required for cellular homeostasis. To characterize the molecular mechanisms, we performed RNASeq upon NCOA3 knockdown. We identified 343 significant differentially expressed genes (220 downregulated and 123 upregulated with a Benjamini-Hochberg false discovery rate FDR < 0.25 and fold change > 1.5) between TGFβ-stimulated fibroblasts with and without NCOA3 knockdown (NCOA3-DEGs) including the fibrosis-relevant genes EDNRB, COL5A3, HES1, IL11 or IL33. Functional analysis of the NCOA3-DEGs showed enrichment of pathway terms such as collagen binding and extracellular matrix organization. In silico screening of the promoters of the NCOA3-DEGs for potential transcription factor binding motifs revealed binding motifs of core transcription factors of fibroblast activation and tissue fibrosis such as SMAD2/3/4, RBPJ, ZEB1, TCF4, REL, and SNAIL2 amongst the downregulated NCOA3-DEGs. Experimental validation of our biostatistical results using SMAD3 as example demonstrated a higher percentage of NCOA3-pSMAD3 double-positive fibroblasts in skin sections of SSc patients compared to healthy controls. In addition, knockdown of NCOA3 reduced TGFβ-induced SMAD-reporter activity. Furthermore, stimulation with TGFβ increased the interaction of NCOA3 with SMAD3 as analyzed by co-immunoprecipitation. Simultaneous knockdown of NCOA3 and SMAD3 showed no additional reductions compared to the single knockdowns, suggesting that NCOA3 controls SMAD3-dependent gene transcription under fibrotic conditions. Finally, inhibition of NCOA3 showed anti-fibrotic effects in different murine models of experimental skin and lung fibrosis.

Conclusion:

Our findings characterize NCOA3 as regulator of multiple pro-fibrotic transcription programs. Pharmaceutical inhibition of NCOA3 might be a strategy to interfere simultaneously with several core pro-fibrotic mediators in fibrotic diseases such as SSc.

Acknowledgements:

We thank Lena Summa, Vladyslav Fedorchenko, Wolfgang Espach and Regina Kleinlein for excellent technical assistance.

The study was funded by grants DI 1537/7-1, DI 1537/8-1, DI 1537/9-1 and -2, DI 1537/11-1, DI 1537/12-1, DI 1537/13-1, DI 1537/14-1, DI 1537/17-1, DE 2414/2-1, DE 2414/4-1, and RA 2506/3-1 of the German Research Foundation, SFB CRC1181 (project C01) and SFB TR221/ project number 324392634 (B04) of the German Research Foundation, grants J39, J40 and A64 of the IZKF in Erlangen, grant 2013.056.1 of the Wilhelm-Sander-Foundation, grants 2014_A47, 2014_A248 and 2014_A184 of the Else-Kröner-Fresenius-Foundation, grant 14-12-17-1-Bergmann of the ELAN-Foundation Erlangen, BMBF (Era-Net grant 01KT1801), MASCARA program, TP 2 and a Career Support Award of Medicine of the Ernst Jung Foundation.

Disclosure of Interests:

Clara Dees: None declared, Sebastian Poetter: None declared, Maximilian Fuchs: None declared, Christina Bergmann: None declared, Alexandru-Emil Matei: None declared, Andrea-Hermina Györfi: None declared, Alina Soare: None declared, Andreas Ramming: None declared, Paolo Ceppi: None declared, Georg Schett: None declared, Meik Kunz: None declared, Jörg H.W. Distler Consultant of: Actelion, Active Biotech, Anamar, ARXX, Bayer Pharma, Boehringer Ingelheim, Celgene, Galapagos, GSK, Inventiva, JB Therapeutics, Medac, Pfizer, RuiYi and UCB, Grant/research support from: Anamar, Active Biotech, Array Biopharma, ARXX, aTyr, BMS, Bayer Pharma, Boehringer Ingelheim, Celgene, Galapagos, GSK, Inventiva, Novartis, Sanofi-Aventis, RedX, UCB

POS0328 ENGRAILED 1 COORDINATES CYTOSKELETAL ORGANIZATION TO PROMOTE MYOFIBROBLAST DIFFERENTIATION AND FIBROTIC TISSUE REMODELING

Background:

Engrailed 1 (EN1) is a homeodomain-containing transcription factor with essential roles in embryonic development. In most cell types, the expression of EN1 is restricted to embryonic development. However, under pathological conditions, EN1 can be re-expressed to promote phenotypical adaptation. En1 is transiently expressed in the developing dermis of murine embryos in a distinct fibroblast lineage and silenced before birth (1). Former EN1-expressing cells give rise to a subpopulation of fibroblasts that has a high capacity for extracellular matrix production in adult murine skin. The role of EN1 in systemic sclerosis (SSc) was previously not explored.

Objectives:

To study the role of EN1 in the pathological activation of fibroblasts in tissue fibrosis.

Methods:

Bulk RNA-Seq and EN1 or SP1 ChIP-Seq were performed from cultured human dermal fibroblasts. The expression of EN1 was inhibited by siRNA. Cytoskeletal drugs paclitaxel, vinblastin and ROCK inhibitor (Y27632) were used to modulate the cytoskeleton in EN1 knockdown or overexpressing dermal fibroblasts. The role of EN1 in fibroblast activation was evaluated by functional experiments with EN1 knockdown or overexpression in standard 2D culture systems as well as in 3D skin equivalent models. The role of EN1 in skin fibrosis was further studied in En1fl/fl X Col6Cre mice, with fibroblast-specific knockout of En1 in three complementary mouse models: overexpression of a constitutively active TGFß-receptor I (TBRICA), bleomycin-induced skin fibrosis and TSK1 mice.

Results:

Pathologically activated dermal fibroblasts from SSc patients express higher levels of EN1 compared with age and sex matched healthy individuals in the skin and in vitro. TGFβ induces EN1 expression in fibroblasts in a SMAD3-dependent manner both in cultured fibroblasts and in murine skin. Knockdown of EN1 prevents TGFβ-induced fibroblast activation, whereas overexpression of EN1 fosters the pro-fibrotic effects of TGFβ with increased expression of αSMA, stress fibers and collagen. RNA sequencing demonstrates that EN1 induces a pro-fibrotic gene expression profile functionally related to cytoskeleton organization and ROCK activation. In silico analyses of the promoters of En1 target genes coupled with siRNA-mediated knockdown demonstrated that EN1 regulates these pro-fibrotic target genes by modulating the activity of regulatory modules that contain transcription factors of the specificity protein (SP) family. Functional experiments with selective modulators of ROCK and of microtubule polymerization confirm the coordinating role of EN1 on ROCK activity and the re-organization of cytoskeleton during myofibroblast differentiation in both conventional culture systems and 3D skin equivalents. Consistently, mice with fibroblast-specific knockout of En1 demonstrate impaired fibroblast-to-myofibroblast transition, reduced dermal thickening and impaired collagen deposition in the TBRICA, bleomycin-induced and TSK1 models.

Conclusion:

We characterize the homeodomain transcription factor EN1 as a molecular amplifier of TGFβ signaling in myofibroblast differentiation that coordinates cytoskeletal organization in a SP-dependent manner. EN1 might thus be a novel candidate for molecular targeted therapies to interfere with myofibroblast differentiation in fibrotic diseases.

References:

[1]Rinkevich Y, Walmsley GG, Hu MS, Maan ZN, Newman AM, Drukker M, et al. Skin fibrosis. Identification and isolation of a dermal lineage with intrinsic fibrogenic potential. Science. 2015;348(6232):aaa2151.

Disclosure of Interests:

Andrea-Hermina Györfi: None declared, Alexandru-Emil Matei: None declared, Maximilian Fuchs: None declared, Aleix Rius Rigau: None declared, Xuezhi Hong: None declared, ZHU Honglin: None declared, Markus Luber: None declared, Christina Bergmann: None declared, Clara Dees: None declared, Ingo Ludolph: None declared, Raymund Horch: None declared, Oliver Distler Consultant of: Actellion, AbbVie, Acceleron Pharma, Anamar, Amgen, Blade Therapeutics, CSL Behring, ChemomAb, Ergonex, Glenmark Pharma, GSK, Inventiva, Italfarmaco, iQvia, Medac, Medscape, Lilly, Sanofi, Target BioScience, UCB, Bayer, Boehringer Ingelheim, Catenion, iQone, Menarini, Mepha, Novartis, Mitsubishi, MSD, Roche, Pfizer, Georg Schett: None declared, Meik Kunz: None declared, Jörg H.W. Distler Consultant of: Actelion, Active Biotech, Anamar, ARXX, Bayer Pharma, Boehringer Ingelheim, Celgene, Galapagos, GSK, Inventiva, JB Therapeutics, Medac, Pfizer, RuiYi and UCB., Grant/research support from: Anamar, Active Biotech, Array Biopharma, aTyr, BMS, Bayer Pharma, Boehringer Ingelheim, Celgene, Galapagos, GSK, Inventiva, Novartis, Sanofi-Aventis, RedX, UCB

Melanoma Single-Cell Biology in Experimental and Clinical Settings

Cellular heterogeneity is regarded as a major factor for treatment response and resistance in a variety of malignant tumors, including malignant melanoma. More recent developments of single-cell sequencing technology provided deeper insights into this phenomenon. Single-cell data were used to identify prognostic subtypes of melanoma tumors, with a special emphasis on immune cells and fibroblasts in the tumor microenvironment. Moreover, treatment resistance to checkpoint inhibitor therapy has been shown to be associated with a set of differentially expressed immune cell signatures unraveling new targetable intracellular signaling pathways. Characterization of T cell states under checkpoint inhibitor treatment showed that exhausted CD8+ T cell types in melanoma lesions still have a high proliferative index. Other studies identified treatment resistance mechanisms to targeted treatment against the mutated BRAF serine/threonine protein kinase including repression of the melanoma differentiation gene microphthalmia-associated transcription factor (MITF) and induction of AXL receptor tyrosine kinase. Interestingly, treatment resistance mechanisms not only included selection processes of pre-existing subclones but also transition between different states of gene expression. Taken together, single-cell technology has provided deeper insights into melanoma biology and has put forward our understanding of the role of tumor heterogeneity and transcriptional plasticity, which may impact on innovative clinical trial designs and experimental approaches.

Early identification of bipolar disorder among young adults - a 22-year community birth cohort

OBJECTIVE

We set forth to build a prediction model of individuals who would develop bipolar disorder (BD) using machine learning techniques in a large birth cohort.

METHODS

A total of 3748 subjects were studied at birth, 11, 15, 18, and 22 years of age in a community birth cohort. We used the elastic net algorithm with 10-fold cross-validation to predict which individuals would develop BD at endpoint (22 years) at each follow-up visit before diagnosis (from birth up to 18 years). Afterward, we used the best model to calculate the subgroups of subjects at higher and lower risk of developing BD and analyzed the clinical differences among them.

RESULTS

A total of 107 (2.8%) individuals within the cohort presented with BD type I, 26 (0.6%) with BD type II, and 87 (2.3%) with BD not otherwise specified. Frequency of female individuals was 58.82% (n = 150) in the BD sample and 53.02% (n = 1868) among the unaffected population. The model with variables assessed at the 18-year follow-up visit achieved the best performance: AUC 0.82 (CI 0.75-0.88), balanced accuracy 0.75, sensitivity 0.72, and specificity 0.77. The most important variables to detect BD at the 18-year follow-up visit were suicide risk, generalized anxiety disorder, parental physical abuse, and financial problems. Additionally, the high-risk subgroup of BD showed a high frequency of drug use and depressive symptoms.

CONCLUSIONS

We developed a risk calculator for BD incorporating both demographic and clinical variables from a 22-year birth cohort. Our findings support previous studies in high-risk samples showing the significance of suicide risk and generalized anxiety disorder prior to the onset of BD, and highlight the role of social factors and adverse life events.

Nanoparticle-complexed antimiRs for inhibiting tumor growth and metastasis in prostate carcinoma and melanoma

Background

MiRNAs act as negative regulators of gene expression through target mRNA degradation or inhibition of its translation. In cancer, several miRNAs are upregulated and play crucial roles in tumorigenesis, making the inhibition of these oncomiRs an interesting therapeutic approach. This can be achieved by directly complementary single-stranded anti-miRNA oligonucleotides (antimiRs). A major bottleneck in antimiR therapy, however, is their efficient delivery. The nanoparticle formation with polyethylenimine (PEI) may be particularly promising, based on the PEI’s ability to electrostatically interact with oligonucleotides. This leads to their protection and supports delivery. In the present study, we explore for the first time PEI for antimiR formulation and delivery. We use the branched low molecular weight PEI F25-LMW for the complexation of different antimiRs, and analyse tumor- and metastasis-inhibitory effects of PEI/antimiR complexes in different tumor models.

Results

In prostate carcinoma, transfection of antimiRs against miR-375 and miR-141 leads to tumor cell inhibition in 2D- and 3D-models. More importantly, an in vivo tumor therapy study in prostate carcinoma xenografts reveals anti-tumor effects of the PEI/antimiR complexes. In advanced melanoma and metastasis, we identify by a microRNA screen miR-150 as a particularly relevant oncomiR candidate, and validate this result in vitro and in vivo. Again, the systemic application of PEI/antimiR complexes inhibiting this miRNA, or the previously described antimiR-638, leads to profound tumor growth inhibition. These effects are associated with the upregulation of direct miRNA target genes. In a melanoma metastasis mouse model, anti-metastatic effects of PEI/antimiR treatment are observed as well.

Conclusions

We thus describe PEI-based complexes as efficient platform for antimiR therapy, as determined in two different tumor entities using in vivo models of tumor growth or metastasis. Our study also highlights the therapeutic relevance of miR-375, miR-141, miR-150 and miR-638 as target miRNAs for antimiR-mediated inhibition.

P1077Late gadolinium enhancement with cardiac magnetic resonance imaging demonstrates left ventricular involvement is under-recognized in arrhythmogenic right ventricular cardiomyopathy

Funding Acknowledgements

The study is partially funded by Medtronic and the Minneapolis Heart Institute Foundation.

Background/Introduction: Arrhythmogenic right ventricular cardiomyopathy (ARVC) is characterized by replacement of the myocardium with fibrous and fatty tissue that may lead to an increased risk of ventricular arrhythmias and heart failure.  Although left ventricular (LV) and biventricular forms have been identified post-mortem resulting in the increased use of the term arrhythmogenic cardiomyopathy, there is only inclusion of right ventricular wall motion abnormalities in the taskforce diagnostic criteria.

Purpose

The aim of our study was to examine the utility of cardiac magnetic resonance (CMR) imaging in characterizing LV or biventricular involvement with late gadolinium enhancement (LGE) in a large cohort of patients with suspected ARVC.

Methods

Retrospective, single-institution, chart review of 76 patients diagnosed with ARVC between January 2009 and July 2019. Data collection and analysis included baseline demographics and parameters specific to diagnosis (definite, borderline, or possible) and risk stratification of ARVC based on 2019 modified taskforce criteria, as well as detailed CMR evaluation.

Results

Of the 76 patients with ARVC, 66 (87%) had at least one CMR with gadolinium administered. In that subset of patients, 27 (41%) had LGE. Of those with LGE, LV involvement was identified in 23 (85%) patients. The pattern of LGE was not localized to one myocardial region but demonstrated variable LV enhancement patterns including anterior, inferior, lateral, septal, basal, mid, apical, and from the sub-epicardium into the mid-myocardium.

Conclusions

  Left ventricular involvement reflected by LGE was identified in a high percentage of patients with suspected ARVC, and there was significant variation in the pattern of distribution in terms of region and depth of myocardial involvement.  While post-mortem examination of patients with ARVC demonstrates a high prevalence of left ventricular involvement, this study shows that CMR can consistently detect late gadolinium enhancement, and ARVC should be considered in the differential diagnosis for biventricular cardiomyopathy.  The identification of variable locations of LGE within the LV suggests there is more than one phenotype, and this imaging modality may help to clarify the implications of left ventricular involvement in disease progression.

Schmerzhafte artifizielle Ekchymosen an Hals und Gesicht

Es wird über den Fall einer 13-jährigen Patientin mit neu aufgetretenen, auf Druck schmerzhaften Hautveränderungen beidseits am Hals und im Gesicht berichtet. Auf Nachfrage traten die Effloreszenzen episodisch im Rahmen von Stresssituationen auf. Im Verlauf kam es zu einer Spontanheilung mit Narbenbildung und mehreren Rezidiven. Nach erfolgter Probebiopsie wurde ein Gardner-Diamond-Syndrom ausgeschlossen und der Verdacht auf selbstinduzierte Hautläsionen geäußert.

Influence of birth cohort on age of onset cluster analysis in bipolar I disorder

Purpose:

Two common approaches to identify subgroups of patients with bipolar disorder are clustering methodology (mixture analysis) based on the age of onset, and a birth cohort analysis. This study investigates if a birth cohort effect will influence the results of clustering on the age of onset, using a large, international database.

Methods:

The database includes 4037 patients with a diagnosis of bipolar I disorder, previously collected at 36 collection sites in 23 countries. Generalized estimating equations (GEE) were used to adjust the data for country median age, and in some models, birth cohort. Model-based clustering (mixture analysis) was then performed on the age of onset data using the residuals. Clinical variables in subgroups were compared.

Results:

There was a strong birth cohort effect. Without adjusting for the birth cohort, three subgroups were found by clustering. After adjusting for the birth cohort or when considering only those born after 1959, two subgroups were found. With results of either two or three subgroups, the youngest subgroup was more likely to have a family history of mood disorders and a first episode with depressed polarity. However, without adjusting for birth cohort (three subgroups), family history and polarity of the first episode could not be distinguished between the middle and oldest subgroups.

Conclusion:

These results using international data confirm prior findings using single country data, that there are subgroups of bipolar I disorder based on the age of onset, and that there is a birth cohort effect. Including the birth cohort adjustment altered the number and characteristics of subgroups detected when clustering by age of onset. Further investigation is needed to determine if combining both approaches will identify subgroups that are more useful for research.

On-site CAD templates reduce surgery time for complex craniostenosis repair in infants: a new method

INTRODUCTION

The surgical correction of craniostenosis in children is a time-consuming and taxing procedure. To facilitate this procedure, especially in infants with complex craniostenosis, we refined the computer-aided design and manufacturing technique (CAD/CAM) based on computed tomography (CT)-generated DICOM data. We used cutting guides and molding templates, which allowed the surgeon to reshape and fixate the supraorbital bar extracorporeally on a side table and to control the intracorporal fit without removing the template.

METHOD AND PATIENTS

To compare our traditional concept with the possibility of preoperative virtual planning (PVP) technique, the surgical treatment and courses of 16 infants with complex craniostenosis following fronto-orbital advancement (FOA) (age range 8-15 months) were analyzed in two groups (group 1: traditional, control group n = 8, group 2: CAD/CAM planned, n = 8).

RESULTS

While in both groups, the head accurately reshaped postoperatively during the follow-up; the CAD group 2 showed a significantly shorter operating time with a mean of 4 h 25 min compared with group 1 with a mean of 5 h 37 min (p = 0.038). Additionally, the CAD group 2 had a significantly lower volume of blood loss (380 ml vs. 575 ml mean, p = 0.047), lower blood transfusion volume (285 ml vs. 400 ml mean, p = 0.108), lower fresh frozen plasma (FFP) volume (140 ml vs. 275 ml mean, p = 0.019), shorter stay in the pediatric intensive care unit (PICU) (3 vs. 5 days mean (p = 0.002), and shorter total length of hospital stay (6 days vs. 8 days mean, p = 0.002).

CONCLUSION

CAD/CAM cutting guides and templates offer optimizing operative efficiency, precision, and accuracy in craniostenosis surgery in infants. As shown in this single-center observational study, the use of on-site templates significantly accelerates the reconstruction of the bandeau. The virtual 3D planning technique increases surgical precision without discernible detrimental effects.