Where Artificial Intelligence Can Take Us in the Management and Understanding of Cancerization Fields

Squamous cell carcinoma and its precursor lesion actinic keratosis are often found together in areas of skin chronically exposed to sun, otherwise called cancerisation fields. The clinical assessment of cancerisation fields and the correct diagnosis of lesions within these fields is usually challenging for dermatologists. The recent adoption of skin cancer diagnostic imaging techniques, particularly LC-OCT, helps clinicians in guiding treatment decisions of cancerization fields in a non-invasive way. The combination of artificial intelligence and non-invasive skin imaging opens up many possibilities as AI can perform tasks impossible for humans in a reasonable amount of time. In this text we review past examples of the application of AI to dermatological images for actinic keratosis/squamous cell carcinoma diagnosis, and we discuss about the prospects of the application of AI for the characterization and management of cancerization fields.

Line-field confocal optical coherence tomography in melanocytic and non-melanocytic skin tumors

INTRODUCTION

Line-field confocal optical coherence tomography (LC-OCT) is a recently introduced, non-invasive skin imaging technique combining the technical advantages of reflectance confocal microscopy and conventional OCT in terms of isotropic resolution and in-tissue penetration. Several studies have been published so far about the use of LC-OCT in melanocytic and non-melanocytic skin tumors. The aim of this review was to summarize the currently available data on the use of LC-OCT for benign and malignant melanocytic and non-melanocytic skin tumors.

EVIDENCE ACQUISITION

We searched scientific databases for any literature published up to 30^th April 2023 and concerning the use of LC-OCT for melanocytic and non-melanocytic skin tumors. Identified papers were evaluated, and relevant information was extracted.

EVIDENCE SYNTHESIS

A total of 29 studies were found including original articles, short reports, and letters to the Editor: 6 applied to melanocytic skin tumors, 22 to non-melanocytic skin tumors and 1 to both. The use of LC-OCT increased the diagnostic accuracy for melanocytic and non-melanocytic skin lesions. The highest diagnostic performance was found for basal cell carcinoma (BCC), but significant improvements in the diagnostic accuracy were also detected for the differentiation of actinic keratosis (AK) from squamous cell carcinoma (SCC) and of melanoma from nevi. The LC-OCT features of other skin tumors were also described and successfully correlated with histopathology.

CONCLUSIONS

LC-OCT proved to increase the diagnostic accuracy for melanocytic and non-melanocytic skin lesions, thanks to the combination of high resolution/penetration, 3D reconstructions, and integrated dermoscopy. Although BCC seems the most suitable tumors for LC-OCT examination, the device is extremely performant for the differentiation of AK from SCC and the discrimination of melanoma from nevi as well. Additional studies on diagnostic performance and new investigations about the presurgical assessment of tumor margins with LC-OCT and its association with human and artificial intelligence algorithms are in progress.

Chemerin plasma levels are increased in COVID-19 patients and are an independent risk factor of mortality

Background

Chemerin is an extracellular protein with chemotactic activities and its expression is increased in various diseases such as metabolic syndrome and inflammatory conditions. Its role in lung pathology has not yet been extensively studied but both known pro- and anti-inflammatory properties have been observed. The aim of our study was to evaluate the involvement of the chemerin/ChemR23 system in the physiopathology of COVID-19 with a particular focus on its prognostic value.

Methods

Blood samples from confirmed COVID-19 patients were collected at day 1, 5 and 14 from admission to Erasme Hospital (Brussels – Belgium). Chemerin concentrations and inflammatory biomarkers were analyzed in the plasma. Blood cells subtypes and their expression of ChemR23 were determined by flow cytometry. The expression of chemerin and ChemR23 was evaluated on lung tissue from autopsied COVID-19 patients by immunohistochemistry (IHC).

Results

21 healthy controls (HC) and 88 COVID-19 patients, including 40 in intensive care unit (ICU) were included. Plasma chemerin concentration were significantly higher in ICU patients than in HC at all time-points analyzed (p<0.0001). Moreover, they were higher in deceased patients compared to survivors (p<0.05). Logistic univariate regression and multivariate analysis demonstrated that chemerin level at day 14 of admission was an independent risk factor for death. Accordingly, chemerin levels correlated with inflammatory biomarkers such as C-reactive protein and tumor necrosis factor α. Finally, IHC analysis revealed a strong expression of ChemR23 on smooth muscle cells and chemerin on myofibroblasts in advanced acute respiratory distress syndrome (ARDS).

Discussion

Increased plasma chemerin levels are a marker of severity and may predict death of COVID-19 patients. However, multicentric studies are needed, before chemerin can be considered as a biomarker of severity and death used in daily clinical practice. Further studies are also necessary to identify the precise mechanisms of the chemerin/ChemR23 system in ARDS secondary to viral pneumonia.

Distinct Expression Patterns of Interleukin-22 Receptor 1 on Blood Hematopoietic Cells in SARS-CoV-2 Infection

The new pandemic virus SARS-CoV-2 is characterized by uncontrolled hyper-inflammation in severe cases. As the IL-22/IL-22R1 axis was reported to be involved in inflammation during viral infections, we characterized the expression of IL-22 receptor1, IL-22 and IL-22 binding protein in COVID-19 patients. Blood samples were collected from 19 non-severe and 14 severe patients on the day they presented (D0), at D14, and six months later, and from 6 non-infected controls. The IL-22R1 expression was characterized by flow cytometry. Results were related to HLA-DR expression of myeloid cells, to plasma concentrations of different cytokines and chemokines and NK cells and T lymphocytes functions characterized by their IFN-γ, IL-22, IL-17A, granzyme B and perforin content. The numbers of IL-22R1⁺ classical, intermediate, and non-classical monocytes and the proportions of IL-22R1⁺ plasmacytoid DC (pDC), myeloid DC1 and DC2 (mDC1, mDC2) were higher in patients than controls at D0. The proportions of IL-22R1⁺ classical and intermediate monocytes, and pDC and mDC2 remained high for six months. High proportions of IL-22R1⁺ non-classical monocytes and mDC2 displayed HLA-DR^high expression and were thus activated. Multivariate analysis for all IL-22R1⁺ myeloid cells discriminated the severity of the disease (AUC=0.9023). However, correlation analysis between IL-22R1⁺ cell subsets and plasma chemokine concentrations suggested pro-inflammatory effects of some subsets and protective effects of others. The numbers of IL-22R1⁺ classical monocytes and pDC were positively correlated with pro-inflammatory chemokines MCP-1 and IP-10 in severe infections, whereas IL-22R1⁺ intermediate monocytes were negatively correlated with IL-6, IFN-α and CRP in non-severe infections. Moreover, in the absence of in vitro stimulation, NK and CD4⁺ T cells produced IFN-γ and IL-22, and CD4⁺ and CD8⁺ T cells produced IL-17A. CD4⁺ T lymphocytes also expressed IL-22R1, the density of its expression defining two different functional subsets. In conclusion, we provide the first evidence that SARS-CoV-2 infection is characterized by an abnormal expression of IL22R1 on blood myeloid cells and CD4⁺ T lymphocytes. Our results suggest that the involvement of the IL-22R1/IL-22 axis could be protective at the beginning of SARS-CoV-2 infection but could shift to a detrimental response over time.

Functional reprogramming of monocytes in acute and convalescent severe COVID-19 patients

Severe COVID-19 disease is associated with dysregulation of the myeloid compartment during acute infection. Survivors frequently experience long-lasting sequelae, but little is known about the eventual persistence of this immune alteration. Herein, we evaluated TLR-induced cytokine responses in a cohort of mild to critical patients during acute or convalescent phases (n = 97). In the acute phase, we observed impaired cytokine production by monocytes in the patients with the most severe COVID-19. This capacity was globally restored in convalescent patients. However, we observed increased responsiveness to TLR1/2 ligation in patients who recovered from severe disease, indicating that these cells display distinct functional properties at the different stages of the disease. In patients with acute severe COVID-19, we identified a specific transcriptomic and epigenomic state in monocytes that can account for their functional refractoriness. The molecular profile of monocytes from recovering patients was distinct and characterized by increased chromatin accessibility at activating protein 1 (AP1) and MAF loci. These results demonstrate that severe COVID-19 infection has a profound impact on the differentiation status and function of circulating monocytes, during both the acute and the convalescent phases, in a completely distinct manner. This could have important implications for our understanding of short- and long-term COVID-19–related morbidity.

SP-D and CC-16 Pneumoproteins' Kinetics and Their Predictive Role During SARS-CoV-2 Infection

Background

Surfactant protein D (SP-D) and pulmonary club cell protein 16 (CC-16) are called “pneumoproteins” and are involved in host defense against oxidative stress, inflammation, and viral outbreak. This study aimed to determine the predictive value of these pneumoproteins on the incidence of acute respiratory distress syndrome (ARDS) or death in patients with coronavirus disease-2019 (COVID-19).

Methods

This retrospective study included 87 patients admitted to an emergency department. Blood samples were collected on three time points (days 1, 5, and 14 from hospital admission). SP-D and CC-16 serum levels were determined, and univariate and multivariate analyses considering confounding variables (age, body mass index, tobacco use, dyspnea, hypertension, diabetes mellitus, neutrophil-to-lymphocyte ratio) were performed.

Results

Based on the multivariate analysis, SP-D level on D1 was positively and slightly correlated with subsequent development of ARDS, independent of body mass index, dyspnea, and diabetes mellitus. CC-16 level on D1 was modestly and positively correlated with fatal outcome. A rise in SP-D between D1 and D5 and D1 and D14 had a strong negative association with incidence of ARDS. These associations were independent of tobacco use and neutrophil-to-lymphocyte ratio.

Conclusions

Overall, our data reveal that increase in SP-D levels is a good prognostic factor for patients with COVID-19, and that initial CC-16 levels correlated with slightly higher risk of death. SP-D and CC-16 may prove useful to predict outcomes in patients with COVID-19.

Line-field confocal optical coherence tomography as a tool for three-dimensional in vivo quantification of healthy epidermis: A pilot study

Epidermal three-dimensional (3D) topography/quantification has not been completely characterized yet. The recently developed line-field confocal optical coherence tomography (LC-OCT) provides real-time, high-resolution, in-vivo 3D imaging of the skin. This pilot study aimed at quantifying epidermal metrics (epidermal thicknesses, dermal-epidermal junction [DEJ] undulation and keratinocyte number/shape/size) using 3D LC-OCT. For each study participant (8 female, skin-type-II, younger/older volunteers), seven body sites were imaged with LC-OCT. Epidermal metrics were calculated by segmentations and measurements assisted by artificial intelligence (AI) when appropriate. Thicknesses of epidermis/SC, DEJ undulation and keratinocyte nuclei volume varied across body sites. Evidence of keratinocyte maturation was observed in vivo: keratinocyte nuclei being small/spherical near the DEJ and flatter/elliptical near the skin surface. Skin microanatomy can be quantified by combining LC-OCT and AI. This technology could be highly relevant to understand aging processes and conditions linked to epidermal disorders. Future clinical/research applications are to be expected in this scenario.