Spatial Immunophenotyping from Whole-Slide Multiplexed Tissue Imaging Using Convolutional Neural Networks.

The multiplexed immunofluorescence (mIF) platform enables biomarker discovery through the simultaneous detection of multiple markers on a single tissue slide, offering detailed insights into intratumor heterogeneity and the tumor-immune microenvironment at spatially resolved single cell resolution. However, current mIF image analyses are labor-intensive, requiring specialized pathology expertise which limits their scalability and clinical application. To address this challenge, we developed CellGate, a deep-learning (DL) computational pipeline that provides streamlined, end-to-end whole-slide mIF image analysis including nuclei detection, cell segmentation, cell classification, and combined immuno-phenotyping across stacked images. The model was trained on over 750,000 single cell images from 34 melanomas in a retrospective cohort of patients using whole tissue sections stained for CD3, CD8, CD68, CK-SOX10, PD-1, PD-L1, and FOXP3 with manual gating and extensive pathology review. When tested on new whole mIF slides, the model demonstrated high precision-recall AUC. Further validation on whole-slide mIF images of 9 primary melanomas from an independent cohort confirmed that CellGate can reproduce expert pathology analysis with high accuracy. We show that spatial immuno-phenotyping results using CellGate provide deep insights into the immune cell topography and differences in T cell functional states and interactions with tumor cells in patients with distinct histopathology and clinical characteristics. This pipeline offers a fully automated and parallelizable computing process with substantially improved consistency for cell type classification across images, potentially enabling high throughput whole-slide mIF tissue image analysis for large-scale clinical and research applications.

Region of Interest Detection in Melanocytic Skin Tumor Whole Slide Images-Nevus and Melanoma.

Automated region of interest detection in histopathological image analysis is a challenging and important topic with tremendous potential impact on clinical practice. The deep learning methods used in computational pathology may help us to reduce costs and increase the speed and accuracy of cancer diagnosis. We started with the UNC Melanocytic Tumor Dataset cohort which contains 160 hematoxylin and eosin whole slide images of primary melanoma (86) and nevi (74). We randomly assigned 80% (134) as a training set and built an in-house deep learning method to allow for classification, at the slide level, of nevi and melanoma. The proposed method performed well on the other 20% (26) test dataset; the accuracy of the slide classification task was 92.3% and our model also performed well in terms of predicting the region of interest annotated by the pathologists, showing excellent performance of our model on melanocytic skin tumors. Even though we tested the experiments on a skin tumor dataset, our work could also be extended to other medical image detection problems to benefit the clinical evaluation and diagnosis of different tumors.

Spontaneous multifocal pyogenic granulomas.

Cutaneous pyogenic granulomas (PGs) are common, benign vascular tumors of uncertain pathogenesis; however, a growing body of literature suggests that the formation of PGs may be secondary to genetic alterations in both the Ras/Raf/MAPK and PI3K/Akt/mTOR pathways. We present three cases of spontaneous multifocal PGs that first presented in infancy, were not associated with other vascular anomalies or discernable etiology, harbored somatic genetic variants in the Ras/Raf/MAPK pathway (NRAS n = 2, FGFR1 n = 1), were refractory to treatment with beta-blockers and mTOR inhibitors, and responded best to pulsed dye laser. We propose the term "spontaneous multifocal PGs" to describe this entity.

Standardized Protocols for Clinical and Histopathological Characterization of Hidradenitis Suppurativa Tissue Specimens.

Methods for describing and reporting the clinical and histologic characteristics of cutaneous tissue samples from patients with hidradenitis suppurativa (HS) are not currently standardized, limiting clinicians' and scientists' ability to uniformly record, report, and communicate about the characteristics of tissue used in translational experiments. A recently published consensus statement outlined morphological definitions of typical HS lesions, but no consensus has been reached regarding clinical characterization and examination of HS tissue samples. In this study, we aimed to establish a protocol for reporting histopathologic and clinical characteristics of HS tissue specimens. This study was conducted from May 2023 to August 2023. Experts in clinical care, dermatopathology, and translational research were recruited, and a modified Delphi technique was used to develop a protocol for histologic reporting and clinical characterization of submitted tissue specimens from patients with HS. A total of 27 experts participated (14 dermatologists, 3 fellowship-trained dermatopathologists, 3 plastic surgeons, 3 general surgeons, and 4 research scientists) in creating and reviewing protocols for the clinical and histopathological examination of HS tissue specimens. The protocols were formatted as a synoptic report and will help to consistently classify specimens in biobanks on the basis of histologic features and more accurately report and select samples used in translational research projects.

Two Cases of Cutaneous Sarcomatoid Squamous Cell Carcinoma Resembling Cutaneous Giant Cell Tumor of Soft Tissue.

ABSTRACT: Cutaneous sarcomatoid squamous cell carcinoma is well-described with histology resembling pleomorphic undifferentiated sarcoma featuring collagenous or myxoid stroma with or without elements of keratinizing squamous carcinoma. This report presents 2 cases of dedifferentiated squamous cell carcinoma (SCC) composed of sheets of malignant mononuclear cells with malignant osteoclast-like multinucleated giant cells, extravasated blood, and hemosiderin resembling cutaneous giant cell tumor (cGCT). In the first case, an exophytic facial mass of a 96-year-old woman removed by shave showing extensive cGCT-like tumor but with microscopic elements of SCC in situ and positivity for cytokeratin 5/6 in the malignant spindle cells and SCC. The second case involved a 32-year-old man with a pedunculated penile mass removed by shave biopsy, displaying malignant cytology resembling cGCT, focal staining for cytokeratin AE1/AE3 and p63, and CD68 highlighting the osteoclast-like giant cells. Molecular analysis revealed CDKN2A, TP53, and TERT. Upon reexcision, case 2 showed focally invasive keratinizing SCC associated with differentiated penile intraepithelial neoplasia and lichen sclerosus. Skin specimens with an exophytic mass histologically resembling cGCT but with malignant cytology should be meticulously evaluated for elements of SCC. Molecular analysis, detecting mutations like H3F3 or HMGA2-NCOR2 fusion, can aid in distinguishing cutaneous sarcomatoid squamous cell carcinoma from GCT bone or GCT soft tissue.

Region of Interest Detection in Melanocytic Skin Tumor Whole Slide Images - Nevus & Melanoma.

Automated region of interest detection in histopathological image analysis is a challenging and important topic with tremendous potential impact on clinical practice. The deep-learning methods used in computational pathology may help us to reduce costs and increase the speed and accuracy of cancer diagnosis. We started with the UNC Melanocytic Tumor Dataset cohort that contains 160 hematoxylin and eosin whole-slide images of primary melanomas (86) and nevi (74). We randomly assigned 80% (134) as a training set and built an in-house deep-learning method to allow for classification, at the slide level, of nevi and melanomas. The proposed method performed well on the other 20% (26) test dataset; the accuracy of the slide classification task was 92.3% and our model also performed well in terms of predicting the region of interest annotated by the pathologists, showing excellent performance of our model on melanocytic skin tumors. Even though we tested the experiments on the skin tumor dataset, our work could also be extended to other medical image detection problems to benefit the clinical evaluation and diagnosis of different tumors.

Lymphangitic Melanomatosis: Case Report of Intralymphatic Spread of Melanoma in a 66-year-old Man.

ABSTRACT: Melanoma with lymphatic invasion has been associated with increased risk of metastasis, but the mechanisms and clinical implications are poorly understood. Although current reports have documented angiotropic spread of melanoma and suggest lymphatic spread of melanoma to increase the likelihood of metastasis, to our knowledge, lymphangitic metastatic melanoma resembling cutaneous carcinomatosis or presenting with facial hyperpigmentation has not been described. In this case report, we describe extensive cutaneous intralymphatic spread of melanoma, or lymphangitic melanomatosis, producing macular skin pigmentation in a 66-year-old man.

Unusual clinical presentations of cutaneous squamous cell involving distal extremities in four patients: the importance of tissue sampling.

INTRODUCTION: cSCC is the second most common cutaneous malignancy worldwide behind basal cell carcinoma. Typically, SCC is diagnosed early before it infiltrates local subcutaneous tissue or metastasizes. However, unusual presentations are possible and can lead to delayed treatment and possibly worse outcomes. MATERIALS AND METHODS: All patients were White of non-Hispanic or Latino decent. Three-quarters of the cases were male, and a quarter female. The age range was 45 to 78 years. The documented sizes of lesions ranged from 6 cm to 10 cm in diameter. Three of the cases were initially diagnosed as nonhealing wounds, and one was diagnosed as cellulitis. RESULTS: The authors observed that SCC can present unusually by mimicking nonhealing infected ulcers or skin infections such as cellulitis. Over 18 months, the authors' practice recorded 4 cases of cSCC that were initially treated as persistent infections, which potentially lead to worse outcomes. CONCLUSION: These cases provide patterns and clues to potentially expedite the diagnosis and treatment of cSCC. Any skin lesion thought to be infectious but not responding to treatment should undergo tissue sampling.

Photodistributed Toxic Epidermal Necrolysis: Case Report and Review of Current Literature.

Importance: Cases of photodistributed Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) have been infrequently reported since the first documented case in 1989. This emerging clinical entity and its underlying mechanism have yet to be fully characterized. Objective: To report a case of photodistributed SJS/TEN and highlight similarities to other cases reported in the literature. Design, Setting, and Participants: Case report and literature review of published cases of photodistributed SJS/TEN. The case report describes a 29-year-old woman with recent lamotrigine and trimethoprim-sulfamethoxazole exposure who developed TEN in a photodistributed pattern 1 day after prolonged sun exposure. A search of PubMed using the keywords toxic epidermal necrolysis, Stevens-Johnson syndrome, photo-distributed, photo-induced, and sun-exposed was performed to identify other cases reported in the literature. Results: Literature review revealed 8 previously reported cases of healthy individuals with known drug and UV radiation (UVR) exposures who subsequently developed SJS or TEN with photodistribution. Cases reviewed were skewed demographically to young women aged 19 to 48 years (8 of 9 patients) with all cases reporting UVR exposure 24 to 72 hours prior to the onset of symptoms. Conclusions and Relevance: Photodistributed TEN has been increasingly described in the literature and may represent a distinct variant of SJS/TEN. While the pathogenesis remains unclear, the role of UVR as a "second hit" is suggested by the data presented in the cases documented thus far.