Journal of The European Academy of Dermatology and Venereology

Background Current therapeutic guidelines for immune‐mediated inflammatory diseases (IMIDs) contraindicate the administration of tumour necrosis factor inhibitors (TNFis) in advanced heart failure (HF) and highlight the potential risk for new‐onset HF. The current evidence has low certainty, and the results of studies in the past two decades have even challenged the recommendations regarding the impact of TNFis on the risk of HF. Objectives The objective was to systematically synthesize data on the risk of HF in TNFi‐treated groups compared to non‐treated controls in IMIDs. Methods A systematic search was conducted in August 2023. Randomized controlled trials (RCTs) and non‐randomized observational studies of IMID patients comparing groups receiving TNFis to non‐TNFi‐exposed controls were included. The outcome was the incidence of worsening, de novo, and composite HF. Random‐effects meta‐analysis was conducted using risk ratios (RR) with 95% confidence intervals (CIs) to pool data. Results The systematic search identified 49 studies, with 45 included in the quantitative analysis. For the worsening of HF, the pooled results of non‐randomized studies showed no statistically significant risk‐increasing effect of TNFis (RR = 1.18, 95% CI: 0.69–2.00). Analyses from both RCTs and non‐randomized data indicated no increased risk of de novo HF in the TNFi‐group compared to controls (RR = 0.87, 95% CI: 0.60–1.25 and RR = 0.86, 95% CI: 0.64–1.14, respectively). Similarly, no increased risk was found for composite (worsening and de novo) HF in the TNFi‐treated group versus controls, pooling non‐randomized data. Conclusions Our findings indicate that TNFi‐treated IMID patients do not have an increased risk for developing de novo HF, and no statistically significant risk enhancement could be observed in the risk of worsening HF with TNFis. Updating IMID guidelines should be considered regarding TNFis' non‐risk increasing effect on de novo HF, whereas further validating data on the risk of worsening HF in TNFi‐treated IMID patients is needed.

Background Line‐field confocal optical coherence tomography (LC‐OCT) provides in vivo images with cellular resolution for melanocytic lesions. To date, no descriptions have associated cellular discohesion forming clefts (‘clefting’) with malignancy in melanocytic lesions using in vivo imaging techniques such as reflectance confocal microscopy (RCM) or LC‐OCT. Objectives This study aimed to describe the feature of ‘clefting’ in melanocytic lesions and its correlation with histopathology and RCM. Methods Consecutive benign, atypical and malignant melanocytic lesions were scanned with LC‐OCT at the Hospital Clinic of Barcelona, Hôpital Erasme and University Hospital of Saint‐Etienne. Images of body and facial lesions were reviewed according to study criteria. Only lesions with a univocal or unambiguous diagnosis (clinically or histologically) were selected. Results Clefting was found in 36 of the 200 analysed lesions. It was present in 28 of the 70 melanomas (40%) and in eight of the 130 benign lesions (6%). The odds ratio for malignancy associated with the presence of clefting in a melanocytic lesion was 10.03 (CI 4.07–27.5; p < 0.001). The magnitude of clefting also showed significant differences: marked clefting was present in 17 of the 70 melanomas (24%) and in only two benign lesions (2%), with an OR of 20.22 (CI 4.55–186.2; p < 0.001). No differences were observed regarding the presence of this feature in different locations. It was strongly associated with other malignant features, such as epidermal consumption, nests in the epidermis and disruption of the dermo‐epidermal junction. Conclusions This study highlights that clefting observed in melanocytic lesions is not an artefact created by sample preparation, but a specific feature related to malignancy. Moreover, marked clefting is apparently associated with deeper melanoma invasion.

Background Vitiligo (non‐segmental) is a chronic autoimmune skin disease leading to white patches. Although vitiligo has been delineated previously, the endophenotypes of the disease are still to be defined. Objective The aim of this study was to investigate, without prior hypotheses, whether different clinical phenotypes of vitiligo could be identified and differ based on their demographic and clinical characteristics. Methods This retrospective, non‐interventional, single‐centre study included patients with vitiligo. Demographic and clinical characteristics of vitiligo were recorded. Hierarchical cluster analysis was then performed. Results A total of 399 patients were included. The cluster analysis classified the patients into five clusters as follows: (1) ‘highly active patients’ with elevated frequency of disease activity signs, receiving combinatorial topical and systemic therapy; (2) mild vitiligo patients with younger patients, low surface area involved and absence of signs of disease activity; (3) extensive vitiligo patients, with older patients and the largest surface area involved; (4) and (5) mild or moderate to severe disease, with localization of the lesions on site of chronic skin friction (type 2A koebner phenomenon). Conclusion Patients with vitiligo can be classified into five phenotypes, with two defining the highly active and extensive patients.

Background One of the most promising and rapidly advancing research areas in recent years is using dermoscopic images for automatic diagnosis with artificial intelligence and machine learning methods. Objective This study aimed to synthesize the existing studies for the clinical use of applications made with artificial intelligence methods and to summarize the predictive performance of deep learning and hybrid models‐based algorithms in all these studies with a large‐scale meta‐analysis. Method The literature review was conducted between January 2006 and May 2024, and meta‐analysis data were created by scanning the Web of Science (WOS), Scopus and MEDLINE databases. This study followed the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis (PRISMA) checklist. Results A total of 2722 articles were evaluated. Data from 78 diagnostic tests from 39 primary studies meeting the inclusion and exclusion criteria were assessed. The pooled SROC overall model AUC was 0.96 [95% CI: 0.94–0.98], sensitivity was 0.89 [95% CI: 0.85–0.91] and specificity was 0.92 [95% CI: 0.90–0.94]. In the subgroup analyses, the pooled AUC was 0.98 [95% CI: 0.96–0.99] for HYBRID models. Conclusion Recent studies have suggested that artificial intelligence algorithms and machine learning methods should be used extensively in medicine to assist physicians, especially in diagnosing melanoma. The ability of HYBRID model algorithms to predict diseases is promising. In particular, the performance of HYBRID models was found to be high. This information can assist clinicians in interpreting the most appropriate algorithms for diagnosing melanoma.

Background Although chronic spontaneous urticaria (CSU) is a common form of chronic urticaria (CU), its impact on overall disease burden and treatment patterns in real‐world settings requires further understanding. Objectives To assess prevalence, treatment patterns, disease profile and burden in patients with CSU in five European countries (EU5: France, Germany, Italy, Spain and the United Kingdom [UK]). Methods Data were from the 2020 EU5 National Health and Wellness Survey (NHWS). Age‐ and sex‐adjusted prevalence of diagnosed CSU was estimated for the overall EU5 and by country. Sociodemographic, health characteristics, treatment history and dermatology‐specific quality of life were described for the CSU cohort. Patient‐reported outcomes, including the Short Form 12‐item Survey (SF‐12v2) mental (MCS), physical component (PCS) summary scores; SF‐6D; EQ‐5D; General Anxiety Disorder‐7 (GAD‐7); Patient Health Questionnaire‐9 (PHQ‐9); Work Productivity and Activity Impairment (WPAI) and healthcare resource use (HRU), were summarized for the CSU cohort, overall and by country and descriptively compared to the full NHWS sample. Results Of 62,319 respondents, 794 patients were diagnosed with CU. Among these, 519 had CSU. Weighted prevalence of CSU in EU5 was 0.92%. Roughly half of the CSU cohort was treated with prescription and/or over‐the‐counter medication. Among the CSU cohort, 36.7% of patients experienced a very large to extremely large disease impact (Dermatology Life Quality Index [DLQI] > 10); most had poorly controlled disease (71.6%). Mean MCS, PCS, SF‐6D and EQ‐5D scores were worse for CSU versus the full NHWS sample. A higher proportion of the CSU cohort, versus the full NHWS sample, had GAD‐7 and PHQ‐9 scores of ≥5. Mild–severe anxiety, depression and WPAI scores were highest in the UK. HRU was higher in the CSU cohort than in the full NHWS sample. Conclusion CSU negatively impacts patients' lives. Yet, many patients remain inadequately treated and uncontrolled. These results highlight a need for comprehensive care approaches to improve patient outcomes.

Background Acral melanoma (AM) shows different genomic profiles based on BRAF status, and the data on the association with BRAF and immune checkpoint molecules are lacking. Objectives This study aimed to identify the significance of BRAF mutation in AMs, exploring expression patterns of immune checkpoint molecules and transcriptome profiles related to tumour immunity according to BRAF status. Methods Immunohistochemical (IHC) staining of BRAF, Programmed death‐1 (PD‐1), Lymphocyte‐activation gene‐3 (LAG‐3) and T‐Cell Immunoglobulin and mucin domain‐3 (TIM‐3) was performed on AM tissues. Through spatial transcriptome analysis, the correlation between BRAF expression and immune checkpoint molecule expression was examined. Analysis on differentially expressed genes along with pathway analysis and immune cell deconvolution was performed comparing BRAF‐high and BRAF‐low groups at the mRNA level. Results In IHC and spatial transcriptome analysis, BRAF positivity was associated with high expression of PD‐1, LAG‐3 and TIM‐3. Among a total of 144 patients, positive IHC results for BRAF (p < 0.01), PD‐1 (p < 0.01), LAG‐3 (p < 0.01) and TIM‐3 (p < 0.01) were significantly associated with histopathologic traits including pathological subtypes, cytomorphology, pagetoid spread and nest formation. In spatial transcriptome analysis, the expression level of LAG‐3 showed a significant association with the expression level of BRAF (p < 0.01). Pathways related to anti‐tumour immunity were significantly downregulated in the BRAF‐high group. In immune cell deconvolution, endothelial cells (p = 0.001), mast cells (p = 0.014) and neutrophils (p = 0.004) were significantly higher in the BRAF‐high group. Conclusions BRAF mutation in AM is associated with increased expression of immune checkpoint molecules, supporting the use of immunotherapy for BRAF‐mutant AM in clinical practice. Different tumour microenvironments regarding tumour immunity in BRAF‐mutant AM may explain the poor prognosis.

Background Interleukin (IL)‐23p19 and IL‐17 inhibitors have demonstrated high efficacy for psoriasis in randomized controlled trials, though real‐world data, particularly for risankizumab (IL‐23p19 inhibitor) and brodalumab (IL‐17 receptor (IL‐17R) inhibitor), is limited. Objectives To assess drug survival of IL‐23p19 and IL‐17 inhibitors compared to other biologics for psoriasis. Methods We conducted a cohort study using data from the British Association of Dermatologists Biologics and Immunomodulators Register (BADBIR) from November 2007 to June 2023. Multivariable flexible parametric models assessed drug survival, with discontinuation due to ineffectiveness and adverse effects reported separately. The primary outcome measure was the absolute difference in restricted mean survival time at 2 years, referred to as adjusted survival time, between all comparators. Results Among 19,034 treatment courses (median follow‐up: 2.3 years), treatments included adalimumab (tumour necrosis factor‐alpha (TNF‐a) inhibitor, n = 6,815), ustekinumab (IL‐12/23p40 inhibitor, n = 5,639), secukinumab (IL‐17A inhibitor, n = 3,051), ixekizumab (IL‐17A inhibitor, n = 1,072), brodalumab (n = 367), guselkumab (IL‐23p19 inhibitor, n = 1,258) and risankizumab (n = 832). Guselkumab and risankizumab had the highest adjusted survival times (years [interquartile ranges]) for effectiveness (1.93 [1.91–1.95] and 1.93 [1.90–1.96], respectively). Risankizumab had the highest survival for safety (1.94 [1.92–1.96]) followed by guselkumab (1.92 [1.90–1.94]) and ustekinumab (1.92 [1.91–1.93]). Brodalumab showed lower adjusted survival time for effectiveness (1.75 [1.69–1.81]) than most biologics except secukinumab and adalimumab; and similar survival for safety (1.85 [1.81–1.90]) compared to IL‐17A inhibitors and adalimumab. In patients with psoriatic arthritis, ustekinumab showed reduced drug survival. Prior biologic exposure was associated with a dose–response reduction in survival which was significantly larger for IL‐17 inhibitors. Conclusions Guselkumab and risankizumab have the most favourable drug survival for effectiveness, with comparable safety to ustekinumab, and more favourable than other BADBIR biologics. Longer drug survival may reduce treatment burden by minimizing treatment switches, clinic visits and disease flares, supporting IL‐23p19 inhibitors as a practical long‐term option for psoriasis.

Background Early‐stage mycosis fungoides (MF) is often an indolent disease with a favourable prognosis, though 25% progress to advanced stages. Patients with ≥10% body surface area (BSA) involvement have a worse prognosis than those with <10%, but the impact of other BSA levels remains unclear. Objectives This study aimed to determine whether additional BSA cut‐off points higher than 10% provide additional prognostic information. Methods This study included 401 patients with early‐stage MF of the Leiden University Medical Centre and a validation cohort of 602 patients from the PROCLIPI database. Different percentages of lesional BSA were analysed both in the total groups and in subgroups of patients with only patch‐stage disease or with patch–plaques‐stage disease, and correlated with survival. Results Both cohorts showed a progressive and gradual decline in overall survival (OS) with each 10% increase in BSA, reaching a 5‐year OS below 50% (LUMC) or 70% (PROCLIPI) for BSA ≥40%. In both cohorts, patients with a BSA of 10%–39% had a significantly worse OS than those with a BSA of less than 10%, but a significantly better OS than those with ≥40%. These differences in OS were only found in patients with patches and plaques and not in patients with only patches. Conclusions An additional cut‐off point of 40% involved BSA has prognostic significance in patients with early patch/plaque‐stage MF and may be included in future updates of the clinical staging system. In these patients, systemic therapies should be considered.

Background Chronic cutaneous graft‐versus‐host disease (ccGVHD) is a debilitating complication of allogeneic haematopoietic stem cell transplantation, manifesting as either sclerotic or lichenoid eruptions (lGVHD). Although frequent, the pathogenesis of lGVHD remains mainly unknown and represents a therapeutic challenge. Objectives This study aims to decipher the immunological mechanisms underlying lichenoid GVHD and to evaluate the effect of blocking the Type I interferons (IFN‐I) pathway in a mice model of ccGVHD on local inflammation. Methods First, we performed single‐cell gene expression analysis (scRNA‐Seq) of human lGVHD skin samples (n = 3) compared with healthy control (n = 4) and analysed the distribution and inflammatory signatures of immune cells. Results were confronted with bulk‐RNA‐Seq data of an independent cohort of lGVHD (n = 8) previously published by our group to compare the predicted immune population within lGVHD skin through deconvolution analyses. The IFN‐I score was assessed through quantitative PCR on lGVHD lesions (n = 13) compared with HC (n = 10). Secondly, we analysed publicly available skin microarray data from a mouse model of ccGVHD, in which allografted mice were treated with an anti‐IFNAR1 antibody targeting the Type I interferon receptor. Results ScRNA‐Seq analysis of human immune cells revealed a strong Type I and II interferons signature in lGVHD skin, along with an increased expression of Th2/Th17 and activation markers in T cells and macrophages, respectively. Consistent with our human data, murine skin GVHD samples were characterized by interferon and allograft inflammatory responses, which were greatly prevented by early infusions of anti‐IFNAR1 antibody after allograft. Bioinformatic analyses highlighted interferons, along with Th2 and Th17 markers, as putative key regulators in mice skin inflammation that were also diminished or abrogated consequently to early anti‐IFNAR1 antibody infusions. Conclusion Together, our data suggests that interferon pathways, and in particular Type I via IFNAR1, may be promising therapeutic targets for the treatment of lichenoid chronic GVHD.