TYK2 inhibition with deucravacitinib ameliorates erosive oral lichen planus.

Erosive oral lichen planus (OLP) is a challenging disease. This T cell driven disorder frequently shows a treatment unresponsive course and strongly limits patients' quality of life. The disease lacks FDA or EMA approved drugs for its treatment and the efficacy of the commonly administered treatments (i.e. topical and systemic steroids, steroid sparing agents) is often only partial. Although the etiopathogenesis of the disease still needs to be fully elucidated, recent advances helped to identify interferon-É£ (IFN-É£) as a pivotal cytokine in OLP pathogenesis, thus making the interference with its signalling a therapeutic target. Janus kinase (JAK) inhibitors therefore gained relevance for their inhibitory effect on IFN-É£ signalling. While some drugs such as abrocitinib, upadacitinib, tofacitinib directly interfere with IFN-É£ signalling through blockade of JAK1 and/or JAK2, deucravacitinib, a selective TYK-2 inhibitor indirectly interferes on IFN-É£ activation through interference with interleukin (IL)-12, a potent promotor for Th1/IFN-É£ responses. This mechanism of action makes deucravacitinib a candidate drug for the treatment of OLP. Here we provide initial evidence that deucravacitinib 6 mg daily has a beneficial effect in three patients with oral OLP.

Non-biologic immunosuppressive drugs for inflammatory and autoimmune skin diseases.

Non-biologic immunosuppressive drugs, such as azathioprine, dapsone or methotrexate are fundamental treatment options for a wide range of autoimmune and chronic inflammatory skin diseases. Some of these drugs were initially used for malignancies (e.g., azathioprine or methotrexate) or infectious diseases (e.g., hydroxychloroquine or dapsone) but are nowadays mostly used for their immunosuppressive/immunomodulating action. Although dermatologists have years of clinical experience with these drugs, some of the mechanisms of action are not fully understood and are the subject of research. Although these drugs are commonly used, lack of experience or knowledge regarding their safety profiles and management leads to skepticism among physicians. Here, we summarize the mechanism of action and detailed management of adverse effects of the most commonly used immunosuppressive drugs for skin diseases. Furthermore, we discuss the management of these drugs during pregnancy and breastfeeding, as well as their interaction and handling during vaccination.

Pathogenic relevance of antibodies against desmoglein 3 in patients with oral lichen planus.

BACKGROUND AND OBJECTIVES: Oral lichen planus (OLP) is a T cell driven disorder that significantly impairs patients' quality of life. Previous reports suggest that both cellular and humoral activities against desmoglein (dsg) 1 and 3 may be involved in OLP pathogenesis. Here, we aim to analyze the frequency of occurrence and pathological significance of anti-dsg antibodies in a large cohort of OLP patients. MATERIALS AND METHODS: OLP patients were screened for anti-dsg antibodies by enzyme-linked immunosorbent assay in three tertiary referral centers. OLP sera with anti-dsg antibodies were further analyzed by Western blot and dispase-based keratinocyte dissociation assay (DDA) to identify the targeted dsg ectodomains and to assess their pathogenicity. RESULTS: Of 151-screened individuals with OLP, only four patients (2.6%) with erosive OLP showed serum IgG against dsg1/3. Western blot analysis with recombinant dsg3 ectodomains revealed preferential recognition of the extracellular domain 5. By DDA with spontaneously immortalized human keratinocytes, none of the sera from these four patients induced acantholysis. CONCLUSIONS: Activation of humoral immunity occurs prevalently in patients with erosive OLP, probably due to epitope spreading. OLP serum antibodies are unable to induce loss of intercellular adhesion in vitro, strongly suggesting that they are not disease causing but rather an epiphenomenon.

Acidosis-mediated increase in IFN-γ-induced PD-L1 expression on cancer cells as an immune escape mechanism in solid tumors.

Immune checkpoint inhibitors have revolutionized cancer therapy, yet the efficacy of these treatments is often limited by the heterogeneous and hypoxic tumor microenvironment (TME) of solid tumors. In the TME, programmed death-ligand 1 (PD-L1) expression on cancer cells is mainly regulated by Interferon-gamma (IFN-γ), which induces T cell exhaustion and enables tumor immune evasion. In this study, we demonstrate that acidosis, a common characteristic of solid tumors, significantly increases IFN-γ-induced PD-L1 expression on aggressive cancer cells, thus promoting immune escape. Using preclinical models, we found that acidosis enhances the genomic expression and phosphorylation of signal transducer and activator of transcription 1 (STAT1), and the translation of STAT1 mRNA by eukaryotic initiation factor 4F (elF4F), resulting in an increased PD-L1 expression. We observed this effect in murine and human anti-PD-L1-responsive tumor cell lines, but not in anti-PD-L1-nonresponsive tumor cell lines. In vivo studies fully validated our in vitro findings and revealed that neutralizing the acidic extracellular tumor pH by sodium bicarbonate treatment suppresses IFN-γ-induced PD-L1 expression and promotes immune cell infiltration in responsive tumors and thus reduces tumor growth. However, this effect was not observed in anti-PD-L1-nonresponsive tumors. In vivo experiments in tumor-bearing IFN-γ-/- mice validated the dependency on immune cell-derived IFN-γ for acidosis-mediated cancer cell PD-L1 induction and tumor immune escape. Thus, acidosis and IFN-γ-induced elevation of PD-L1 expression on cancer cells represent a previously unknown immune escape mechanism that may serve as a novel biomarker for anti-PD-L1/PD-1 treatment response. These findings have important implications for the development of new strategies to enhance the efficacy of immunotherapy in cancer patients.

Federated Learning for Decentralized Artificial Intelligence in Melanoma Diagnostics.

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

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

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