Obesity-induced dysregulation of skin-resident PPARγ+ Treg cells promotes IL-17A-mediated psoriatic inflammation.

Obesity is a major risk factor for psoriasis, but how obesity disrupts the regulatory mechanisms that keep skin inflammation in check is unclear. Here, we found that skin was enriched with a unique population of CD4+Foxp3+ regulatory T (Treg) cells expressing the nuclear receptor peroxisome proliferation-activated receptor gamma (PPARγ). PPARγ drove a distinctive transcriptional program and functional suppression of IL-17A+ γδ T cell-mediated psoriatic inflammation. Diet-induced obesity, however, resulted in a reduction of PPARγ+ skin Treg cells and a corresponding loss of control over IL-17A+ γδ T cell-mediated inflammation. Mechanistically, PPARγ+ skin Treg cells preferentially took up elevated levels of long-chain free fatty acids in obese mice, which led to cellular lipotoxicity, oxidative stress, and mitochondrial dysfunction. Harnessing the anti-inflammatory properties of these PPARγ+ skin Treg cells could have therapeutic potential for obesity-associated inflammatory skin diseases.

Sustained release hydrogel for durable locoregional chemoimmunotherapy for BRAF-mutated melanoma.

The wide prevalence of BRAF mutations in diagnosed melanomas drove the clinical advancement of BRAF inhibitors in combination with immune checkpoint blockade for treatment of advanced disease. However, deficits in therapeutic potencies and safety profiles motivate the development of more effective strategies that improve the combination therapy's therapeutic index. Herein, we demonstrate the benefits of a locoregional chemoimmunotherapy delivery system, a novel thermosensitive hydrogel comprised of gelatin and Pluronic® F127 components already widely used in humans in both commercial and clinical products, for the co-delivery of a small molecule BRAF inhibitor with immune checkpoint blockade antibody for the treatment of BRAF-mutated melanoma. In vivo evaluation of administration route and immune checkpoint target effects revealed intratumoral administration of antagonistic programmed cell death protein 1 antibody (aPD-1) lead to potent antitumor therapy in combination with BRAF inhibitor vemurafenib. The thermosensitive F127-g-Gelatin hydrogel that was evaluated in multiple murine models of BRAF-mutated melanoma that facilitated prolonged local drug release within the tumor (>1 week) substantially improved local immunomodulation, tumor control, rates of tumor response, and animal survival. Thermosensitive F127-g-Gelatin hydrogels thus improve upon the clinical benefits of vemurafenib and aPD-1 in a locoregional chemoimmunotherapy approach for the treatment of BRAF-mutated melanoma.

Explaining the black-box smoothly-A counterfactual approach.

We propose a BlackBox Counterfactual Explainer, designed to explain image classification models for medical applications. Classical approaches (e.g., , saliency maps) that assess feature importance do not explain how imaging features in important anatomical regions are relevant to the classification decision. Such reasoning is crucial for transparent decision-making in healthcare applications. Our framework explains the decision for a target class by gradually exaggerating the semantic effect of the class in a query image. We adopted a Generative Adversarial Network (GAN) to generate a progressive set of perturbations to a query image, such that the classification decision changes from its original class to its negation. Our proposed loss function preserves essential details (e.g., support devices) in the generated images. We used counterfactual explanations from our framework to audit a classifier trained on a chest X-ray dataset with multiple labels. Clinical evaluation of model explanations is a challenging task. We proposed clinically-relevant quantitative metrics such as cardiothoracic ratio and the score of a healthy costophrenic recess to evaluate our explanations. We used these metrics to quantify the counterfactual changes between the populations with negative and positive decisions for a diagnosis by the given classifier. We conducted a human-grounded experiment with diagnostic radiology residents to compare different styles of explanations (no explanation, saliency map, cycleGAN explanation, and our counterfactual explanation) by evaluating different aspects of explanations: (1) understandability, (2) classifier's decision justification, (3) visual quality, (d) identity preservation, and (5) overall helpfulness of an explanation to the users. Our results show that our counterfactual explanation was the only explanation method that significantly improved the users' understanding of the classifier's decision compared to the no-explanation baseline. Our metrics established a benchmark for evaluating model explanation methods in medical images. Our explanations revealed that the classifier relied on clinically relevant radiographic features for its diagnostic decisions, thus making its decision-making process more transparent to the end-user.

Automated detection of premalignant oral lesions on whole slide images using convolutional neural networks.

INTRODUCTION: Oral epithelial dysplasia (OED) is a precursor lesion to oral squamous cell carcinoma, a disease with a reported overall survival rate of 56 percent across all stages. Accurate detection of OED is critical as progression to oral cancer can be impeded with complete excision of premalignant lesions. However, previous research has demonstrated that the task of grading of OED, even when performed by highly trained experts, is subject to high rates of reader variability and misdiagnosis. Thus, our study aims to develop a convolutional neural network (CNN) model that can identify regions suspicious for OED whole-slide pathology images. METHODS: During model development, we optimized key training hyperparameters including loss function on 112 pathologist annotated cases between the training and validation sets. Then, we compared OED segmentation and classification metrics between two well-established CNN architectures for medical imaging, DeepLabv3+ and UNet++. To further assess generalizability, we assessed case-level performance of a held-out test set of 44 whole-slide images. RESULTS: DeepLabv3+ outperformed UNet++ in overall accuracy, precision, and segmentation metrics in a 4-fold cross validation study. When applied to the held-out test set, our best performing DeepLabv3+ model achieved an overall accuracy and F1-Score of 93.3 percent and 90.9 percent, respectively. CONCLUSION: The present study trained and implemented a CNN-based deep learning model for identification and segmentation of oral epithelial dysplasia (OED) with reasonable success. Computer assisted detection was shown to be feasible in detecting premalignant/precancerous oral lesions, laying groundwork for eventual clinical implementation.

Cutaneous findings in myotonic dystrophy.

Myotonic dystrophy types 1 and 2 are a group of complex genetic disorders resulting from the expansion of (CTG)n nucleotide repeats in the DMPK gene. In addition to the hallmark manifestations of myotonia and skeletal muscle atrophy, myotonic dystrophy also affects a myriad of other organs including the heart, lungs, as well as the skin. The most common cutaneous manifestations of myotonic dystrophy are early male frontal alopecia and adult-onset pilomatricomas. Myotonic dystrophy also increases the risk of developing malignant skin diseases such as basal cell carcinoma and melanoma. To aid in the diagnosis and treatment of myotonic dystrophy related skin conditions, it is important for the dermatologist to become cognizant of the common and rare cutaneous manifestations of this genetic disorder. We performed a PubMed search using the key terms "myotonic dystrophy" AND "cutaneous" OR "skin" OR "dermatologic" AND "manifestation" OR "finding." The resulting publications were manually reviewed for additional relevant publications, and subsequent additional searches were performed as needed, especially regarding the molecular mechanisms of pathogenesis. In this review, we aim to provide an overview of myotonic dystrophy types 1 and 2 and summarize their cutaneous manifestations as well as potential mechanisms of pathogenesis.

Deep Learning Prediction of Voxel-Level Liver Stiffness in Patients with Nonalcoholic Fatty Liver Disease.

PURPOSE: To reconstruct virtual MR elastography (MRE) images based on traditional MRI inputs with a machine learning algorithm. MATERIALS AND METHODS: In this single-institution, retrospective study, 149 patients (mean age, 58 years ± 12 [standard deviation]; 71 men) with nonalcoholic fatty liver disease who underwent MRI and MRE between January 2016 and January 2019 were evaluated. Nine conventional MRI sequences and clinical data were used to train a convolutional neural network to reconstruct MRE images at the per-voxel level. The architecture was further modified to accept multichannel three-dimensional inputs and to allow inclusion of clinical and demographic information. Liver stiffness and fibrosis category (F0 [no fibrosis] to F4 [significant fibrosis]) of reconstructed images were assessed by using voxel- and patient-level agreement by correlation, sensitivity, and specificity calculations; in addition, classification by receiver operator characteristic analyses was performed, and Dice score was used to evaluate hepatic stiffness locality. RESULTS: The model for predicting liver stiffness incorporated four image sequences (precontrast T1-weighted liver acquisition with volume acquisition [LAVA] water and LAVA fat, 120-second-delay T1-weighted LAVA water, and single-shot fast spin-echo T2 weighted) and clinical data. The model had a patient-level and voxel-level correlation of 0.50 ± 0.05 and 0.34 ± 0.03, respectively. By using a stiffness threshold of 3.54 kPa to make a binary classification into no fibrosis or mild fibrosis (F0-F1) versus clinically significant fibrosis (F2-F4), the model had sensitivity of 80% ± 4, specificity of 75% ± 5, accuracy of 78% ± 3, area under the receiver operating characteristic curve of 84 ± 0.04, and a Dice score of 0.74. CONCLUSION: The generation of virtual elastography images is feasible by using conventional MRI and clinical data with a machine learning algorithm.Keywords: MR Imaging, Abdomen/GI, Liver, Cirrhosis, Computer Applications/Virtual Imaging, Experimental Investigations, Feature Detection, Classification, Reconstruction Algorithms, Supervised Learning, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2021.

Critical role of interferons in gastrointestinal injury repair.

The etiology of ulcerative colitis is poorly understood and is likely to involve perturbation of the complex interactions between the mucosal immune system and the commensal bacteria of the gut, with cytokines acting as important cross-regulators. Here we use IFN receptor-deficient mice in a dextran sulfate sodium (DSS) model of acute intestinal injury to study the contributions of type I and III interferons (IFN) to the initiation, progression and resolution of acute colitis. We find that mice lacking both types of IFN receptors exhibit enhanced barrier destruction, extensive loss of goblet cells and diminished proliferation of epithelial cells in the colon following DSS-induced damage. Impaired mucosal healing in double IFN receptor-deficient mice is driven by decreased amphiregulin expression, which IFN signaling can up-regulate in either the epithelial or hematopoietic compartment. Together, these data underscore the pleiotropic functions of IFNs and demonstrate that these critical antiviral cytokines also support epithelial regeneration following acute colonic injury.

Cannabinoids in dermatologic surgery.

Though known as a medicinal herb for centuries, the recent legalization of cannabinoids across many states has ushered in a new era where cannabinoids have become a popular treatment option among clinicians and patients alike. Cannabinoids have demonstrated efficacy in wound healing, reducing inflammation, ameliorating pain, and have shown potential as an antitumor agent. As a result, cannabinoids have been rapidly woven into the fabric of modern medicine. However, the utility of cannabinoids in dermatologic surgery has not been explored to date. In this article, we review the current literature to discuss the potential impact of cannabinoid use in dermatologic surgery.

Interactive Classification of Whole-Slide Imaging Data for Cancer Researchers.

Whole-slide histology images contain information that is valuable for clinical and basic science investigations of cancer but extracting quantitative measurements from these images is challenging for researchers who are not image analysis specialists. In this article, we describe HistomicsML2, a software tool for learn-by-example training of machine learning classifiers for histologic patterns in whole-slide images. This tool improves training efficiency and classifier performance by guiding users to the most informative training examples for labeling and can be used to develop classifiers for prospective application or as a rapid annotation tool that is adaptable to different cancer types. HistomicsML2 runs as a containerized server application that provides web-based user interfaces for classifier training, validation, exporting inference results, and collaborative review, and that can be deployed on GPU servers or cloud platforms. We demonstrate the utility of this tool by using it to classify tumor-infiltrating lymphocytes in breast carcinoma and cutaneous melanoma. SIGNIFICANCE: An interactive machine learning tool for analyzing digital pathology images enables cancer researchers to apply this tool to measure histologic patterns for clinical and basic science studies.

Incorporating External Information in Tissue Subtyping: A Topic Modeling Approach.

Probabilistic topic models, have been widely deployed for various applications such as learning disease or tissue subtypes. Yet, learning the parameters of such models is usually an ill-posed problem and may result in losing valuable information about disease severity. A common approach is to add a discriminative loss term to the generative model's loss in order to learn a representation that is also predictive of disease severity. However, finding a balance between these two losses is not straightforward. We propose an alternative way in this paper. We develop a framework which allows for incorporating external covariates into the generative model's approximate posterior. These covariates can have more discriminative power for disease severity compared to the representation that we extract from the posterior distribution. For instance, they can be features extracted from a neural network which predicts disease severity from CT images. Effectively, we enforce the generative model's approximate posterior to reside in the subspace of these discriminative covariates. We illustrate our method's application on a large-scale lung CT study of Chronic Obstructive Pulmonary Disease (COPD), a highly heterogeneous disease. We aim at identifying tissue subtypes by using a variant of topic model as a generative model. We quantitatively evaluate the predictive performance of the inferred subtypes and demonstrate that our method outperforms or performs on par with some reasonable baselines. We also show that some of the discovered subtypes are correlated with genetic measurements, suggesting that the identified subtypes may characterize the disease's underlying etiology.

Blockade of immune checkpoints in lymph nodes through locoregional delivery augments cancer immunotherapy.

Systemic administration of immune checkpoint blockade (ICB) monoclonal antibodies (mAbs) can unleash antitumor functions of T cells but is associated with variable response rates and off-target toxicities. We hypothesized that antitumor efficacy of ICB is limited by the minimal accumulation of mAb within tissues where antitumor immunity is elicited and regulated, which include the tumor microenvironment (TME) and secondary lymphoid tissues. In contrast to systemic administration, intratumoral and intradermal routes of administration resulted in higher mAb accumulation within both the TME and its draining lymph nodes (LNs) or LNs alone, respectively. The use of either locoregional administration route resulted in pronounced T cell responses from the ICB therapy, which developed in the secondary lymphoid tissues and TME of treated mice. Targeted delivery of mAb to tumor-draining lymph nodes (TdLNs) alone was associated with enhanced antitumor immunity and improved therapeutic effects compared to conventional systemic ICB therapy, and these effects were sustained at reduced mAb doses and comparable to those achieved by intratumoral administration. These data suggest that locoregional routes of administration of ICB mAb can augment ICB therapy by improving immunomodulation within TdLNs.

Propranolol exhibits activity against hemangiomas independent of beta blockade.

Propranolol is a widely used beta blocker that consists of a racemic mixture of R and S stereoisomers. Only the S stereoisomer has significant activity against the beta-adrenergic receptor. A fortuitous clinical observation was made in an infant who received propranolol for cardiac disease, and regression of a hemangioma of infancy was noted. This has led to the widespread use of propranolol for the treatment of large and life-threatening hemangiomas of infancy. Infants receiving propranolol require monitoring to ensure that they do not suffer from side effects related to beta blockade. The exact mechanism of activity of propranolol in hemangioma of infancy is unknown. In this study, we treated hemangioma stem cells with both beta blockade active S- and inactive R-propranolol and looked for genes that were coordinately regulated by this treatment. Among the genes commonly downregulated, Angiopoietin-like 4 (ANGPTL4) was among the most regulated. We confirmed that propranolol isomers downregulated ANGPTL4 in endothelial cells, with greater downregulation of ANGPTL4 using the beta blockade inactive R-propranolol. ANGPTL4 is present in human hemangiomas of infancy. Finally, R-propranolol inhibited the growth of bEnd.3 hemangioma cells in vivo. The implication of this is that hemangioma growth can be blocked without the side effects of beta blockade. Given that humans have been exposed to racemic propranolol for decades and thus to R-propranolol, clinical development of R-propranolol for hemangiomas of infancy and other angiogenic diseases is warranted.

Phosphoinositide 3-Kinase Signaling Can Modulate MHC Class I and II Expression.

Molecular events activating the PI3K pathway are frequently detected in human tumors and the activation of PI3K signaling alters numerous cellular processes including tumor cell proliferation, survival, and motility. More recent studies have highlighted the impact of PI3K signaling on the cellular response to interferons and other immunologic processes relevant to antitumor immunity. Given the ability of IFNγ to regulate antigen processing and presentation and the pivotal role of MHC class I (MHCI) and II (MHCII) expression in T-cell-mediated antitumor immunity, we sought to determine the impact of PI3K signaling on MHCI and MHCII induction by IFNγ. We found that the induction of cell surface MHCI and MHCII molecules by IFNγ is enhanced by the clinical grade PI3K inhibitors dactolisib and pictilisib. We also found that PI3K inhibition increases STAT1 protein levels following IFNγ treatment and increases accessibility at genomic STAT1-binding motifs. Conversely, we found that pharmacologic activation of PI3K signaling can repress the induction of MHCI and MHCII molecules by IFNγ, and likewise, the loss of PTEN attenuates the induction of MHCI, MHCII, and STAT1 by IFNγ. Consistent with these in vitro studies, we found that within human head and neck squamous cell carcinomas, intratumoral regions with high phospho-AKT IHC staining had reduced MHCI IHC staining. IMPLICATIONS: Collectively, these findings demonstrate that MHC expression can be modulated by PI3K signaling and suggest that activation of PI3K signaling may promote immune escape via effects on antigen presentation.

Palladium based nanoparticles for the treatment of advanced melanoma.

IGF1R and CD44 are overexpressed in most advanced melanomas so we designed chemotherapeutic nanoparticles to target those receptors. Tris(dibenzylideneacetone)dipalladium (Tris DBA-Pd) is a novel inhibitor of N-myristoyltransferase 1 (NMT-1) and has proven in vivo activity against melanoma. However, poor solubility impairs its effectiveness. To improve its therapeutic efficacy and overcome drug resistance in advanced melanomas, we synthesized Tris DBA-Pd hyaluronic acid nanoparticles (Tris DBA-Pd HANP) and evaluated them against in vivo xenografts of LM36R, an aggressive BRAF mutant human melanoma resistant to BRAF inhibitors. We treated xenografted mice in four arms: empty HANPs, free Tris DBA-Pd, Tris DBA-Pd HANPs, and Tris DBA-Pd HANPs with IGF1R antibody. The Tris DBA-Pd HANP group was the most responsive to treatment and showed the greatest depletion of CD44-positive cells on IHC. Surprisingly, the HANP containing IGF1R antibody was less effective than particles without antibody, possibly due to steric hindrance of IGF1R and CD44 binding. Tris DBA-Pd nanoparticles are an effective therapy for CD44-positive tumors like melanoma, and further development of these nanoparticles should be pursued.

Reversible secretome and signaling defects in diabetic mesenchymal stem cells from peripheral arterial disease patients.

OBJECTIVE: Regenerative medicine seeks to stall or to reverse the pathologic consequences of chronic diseases. Many people with diabetes have peripheral arterial disease (PAD), which increases their already high risk of major amputation. Cellular therapies are a promising regenerative medicine approach to PAD that can be used to focally inject regenerative cells to endangered tissue beds. Mesenchymal stem cells (MSCs) are known to promote tissue regeneration through stromal support and paracrine stimulation of new blood vessels (angiogenesis). Whereas little is known about human diabetic MSCs (dMSCs), particularly those from patients with PAD, dMSCs have a limited expansion capacity but can be improved with human platelet lysate (PL) supplementation. PL is rich in many growth factors, including epidermal growth factor (EGF), which is known to be important to cell proliferation and survival signaling pathways. We hypothesize that dMSCs have a reversible defect in EGF receptor pathways. The objective of this work was to test this hypothesis using dMSCs from PAD patients. METHODS: The secretome expression of EGF and prominent angiogens was characterized from bone marrow (BM)-derived and adipose tissue-derived (ATD) dMSCs from five patients (six limbs) undergoing major amputation. Western blot was used to characterize the AKT and extracellular signal-regulated protein kinases 1 and 2 expression in dMSCs under standard culture (5% fetal bovine serum plus fibroblast growth factor 2 [FGF2]), 5% human PL, or 5% fetal bovine serum plus EGF. Healthy donor MSCs were control cells. The angiogenic activity of BM- and ATD-dMSCs was tested on human umbilical vein endothelial cells (ECs). Paired t-test, analysis of variance, and Kruskal-Wallis tests were used as appropriate. RESULTS: Both BM- and ATD-dMSCs had typical MSC surface marker expression and similar expansion profiles, and they did not express EGF in their secretome. PL supplementation of dMSCs improved AKT signaling, but they were resistant to FGF2 activation of extracellular signal-regulated protein kinases 1 and 2. EGF supplementation led to similar AKT expression as with PL, but PL had greater phosphorylation of AKT at 30 and 60 minutes. The conditioned media from both BM- and ATD-dMSCs had robust levels of prominent angiogens (vascular endothelial growth factor, monocyte chemoattractant protein 1, hepatocyte growth factor), which stimulated EC proliferation and migration, and the co-culture of dMSCs with ECs led to significantly longer EC sprouts in three-dimensional gel than EC-alone pellets. CONCLUSIONS: PL and EGF supplementation improves AKT expression in dMSCs over that of FGF2, but PL improved pAKT over that of EGF. Thus, PL supplementation strategies may improve AKT signaling, which could be important to MSC survival in cellular therapies. Furthermore, BM- and ATD-dMSCs have similar secretomes and robust in vitro angiogenic activity, which supports pursuing dMSCs from both reservoirs in regenerative medicine strategies.

Targeted Therapies: Immunologic Effects and Potential Applications Outside of Cancer.

Two pharmacologic approaches that are currently at the forefront of treating advanced cancer are those that center on disrupting critical growth/survival signaling pathways within tumor cells (commonly referred to as "targeted therapies") and those that center on enhancing the capacity of a patient's immune system to mount an antitumor response (immunotherapy). Maximizing responses to both of these approaches requires an understanding of the oncogenic events present in a given patient's tumor and the nature of the tumor-immune microenvironment. Although these 2 modalities were developed and initially used independently, combination regimens are now being tested in clinical trials, underscoring the need to understand how targeted therapies influence immunologic events. Translational studies and preclinical models have demonstrated that targeted therapies can influence immune cell trafficking, the production of and response to chemokines and cytokines, antigen presentation, and other processes relevant to antitumor immunity and immune homeostasis. Moreover, because these and other effects of targeted therapies occur in nonmalignant cells, targeted therapies are being evaluated for use in applications outside of oncology.

Skin cancer screening: recommendations for data-driven screening guidelines and a review of the US Preventive Services Task Force controversy.

Melanoma is usually apparent on the skin and readily detected by trained medical providers using a routine total body skin examination, yet this malignancy is responsible for the majority of skin cancer-related deaths. Currently, there is no national consensus on skin cancer screening in the USA, but dermatologists and primary care providers are routinely confronted with making the decision about when to recommend total body skin examinations and at what interval. The objectives of this paper are: to propose rational, risk-based, data-driven guidelines commensurate with the US Preventive Services Task Force screening guidelines for other disorders; to compare our proposed guidelines to recommendations made by other national and international organizations; and to review the US Preventive Services Task Force's 2016 Draft Recommendation Statement on skin cancer screening.

HMG-CoA synthase 1 is a synthetic lethal partner of BRAFV600E in human cancers.

Contributions of metabolic changes to cancer development and maintenance have received increasing attention in recent years. Although many human cancers share similar metabolic alterations, it remains unclear whether oncogene-specific metabolic alterations are required for tumor development. Using an RNAi-based screen targeting the majority of the known metabolic proteins, we recently found that oncogenic BRAFV600E up-regulates HMG-CoA lyase (HMGCL), which converts HMG-CoA to acetyl-CoA and a ketone body, acetoacetate, that selectively enhances BRAFV600E-dependent MEK1 activation in human cancer. Here, we identified HMG-CoA synthase 1 (HMGCS1), the upstream ketogenic enzyme of HMGCL, as an additional "synthetic lethal" partner of BRAFV600E Although HMGCS1 expression did not correlate with BRAFV600E mutation in human melanoma cells, HMGCS1 was selectively important for proliferation of BRAFV600E-positive melanoma and colon cancer cells but not control cells harboring active N/KRAS mutants, and stable knockdown of HMGCS1 only attenuated colony formation and tumor growth potential of BRAFV600E melanoma cells. Moreover, cytosolic HMGCS1 that co-localized with HMGCL and BRAFV600E was more important than the mitochondrial HMGCS2 isoform in BRAFV600E-expressing cancer cells in terms of acetoacetate production. Interestingly, HMGCL knockdown did not affect HMGCS1 expression levels, whereas HMGCS1 knockdown caused a compensating increase in HMGCL protein level because of attenuated protein degradation. However, this increase did not reverse the reduced ketogenesis in HMGCS1 knockdown cells. Mechanistically, HMGCS1 inhibition decreased intracellular acetoacetate levels, leading to reduced BRAFV600E-MEK1 binding and consequent MEK1 activation. We conclude that the ketogenic HMGCS1-HMGCL-acetoacetate axis may represent a promising therapeutic target for managing BRAFV600E-positive human cancers.