Expression of the Embryonic Cancer Stem Cells' Biomarkers SOX2 and OCT3/4 in Oral Leukoplakias and Squamous Cell Carcinomas: A Preliminary Study.

INTRODUCTION: Cancer stem cells (CSCs) are incriminated for initiating the process of carcinogenesis either de novo or through the transformation of oral potentially malignant disorders (OPMDs) to oral squamous cell carcinoma (OSCC). The aim of this study was to detect the expression of embryonic-type CSC markers OCT3/4 and SOX2 in OSCCs and oral leukoplakias (OLs), the most common of OPMDs. MATERIALS AND METHODS: The study type is experimental, and the study design is characterized as semiquantitative research, which belongs to the branch of experimental research. The experiment was conducted in the Department of Oral Medicine/Pathology, School of Dentistry, Aristotle University of Thessaloniki, Greece. This study focuses on the semiquantitative immunohistochemical (IHC) pattern of expression of CSCs protein-biomarkers SOX2 and OCT3/4, in paraffin embedded samples of 21 OSCCs of different grades of differentiation and 30 cases of OLs with different grades of dysplasia, compared to five cases of normal oral mucosa in both terms of cells' stain positivity and intensity. Statistical analysis was performed through SPSS 2017 Pearson Chi-square and the significance level was set at 0.05 (p=0.05). The expression of the respective genes of SOX2 and OCT3/4 was studied through quantitative polymerase chain reaction (qPCR), in paraffin-embedded samples of 12 cases of OLs with mild/non dysplasia and 19 cases moderately/poorly differentiated OSCCs(n=19) and five normal mucosa using the Independent Paired T-test. RESULTS: The genes SOX2 and Oct3/4 were expressed in all examined cases although no statistically significant correlations among normal, OL and OSCC, were established. A nuclear/membrane staining of OCT3/4 was noticed only in three out of 21 OSCCs but in none of OLs or normal cases (without statistical significance). A characteristic nuclear staining of SOX2 was noticed in the majority of the samples, mostly in the basal and parabasal layers of the epithelium. SOX2 was significantly detected in the OSCCs group (strong positivity in 17/21) than in the OL group (30 cases, mostly mildly stained) (p-value=0.007), and the normal oral epithelium (mild stained, p=0.065). Furthermore, SOX2 was overexpressed in well differentiated OSCCs group (5/OSCCs, strongly stained) rather than in mildly dysplastic and non-dysplastic OLs samples (14/OLs, mildly stained) (p-value =0.035). CONCLUSION: The characteristic expression of SOX2 but not of OCT3/4 in OLs' and OSCCs' lesions suggests the presence of neoplastic cells with certain CSC characteristics whose implication in the early stages of oral tumorigenesis could be further evaluated. The clinical use of SOX2, as prognostic factor, requires further experimental evaluation in larger number of samples.

A novel lightweight deep convolutional neural network for early detection of oral cancer.

OBJECTIVES: To develop a lightweight deep convolutional neural network (CNN) for binary classification of oral lesions into benign and malignant or potentially malignant using standard real-time clinical images. METHODS: A small deep CNN, that uses a pretrained EfficientNet-B0 as a lightweight transfer learning model, was proposed. A data set of 716 clinical images was used to train and test the proposed model. Accuracy, specificity, sensitivity, receiver operating characteristics (ROC) and area under curve (AUC) were used to evaluate performance. Bootstrapping with 120 repetitions was used to calculate arithmetic means and 95% confidence intervals (CIs). RESULTS: The proposed CNN model achieved an accuracy of 85.0% (95% CI: 81.0%-90.0%), a specificity of 84.5% (95% CI: 78.9%-91.5%), a sensitivity of 86.7% (95% CI: 80.4%-93.3%) and an AUC of 0.928 (95% CI: 0.88-0.96). CONCLUSIONS: Deep CNNs can be an effective method to build low-budget embedded vision devices with limited computation power and memory capacity for diagnosis of oral cancer. Artificial intelligence (AI) can improve the quality and reach of oral cancer screening and early detection.

Sore or Swollen Lips Part 3: Diagnosis and Treatment.

This series of three papers reviews the causes, diagnosis and differential diagnosis, and outlines the management of sore and/or swollen lips. Clinical relevance: Sore and/or swollen lips are not uncommon, often have a local cause but may reflect a systemic disease. The previous 2 papers in the series discussed their causes. This paper reviews their diagnosis and treatment.

Sore or Swollen Lips Part 1- Causes and Diagnosis.

This series of three papers reviews the causes, diagnosis and differential diagnosis, and outlines the management of sore and/or swollen lips. Clinical relevance: Sore and/or swollen lips are not uncommon, often have a local cause but may reflect a systemic disease. This first part of a series of three papers reviews their causes and diagnosis alphabetically, for ease of reference.

Sore or Swollen Lips Part 2: Systemic Causes.

This series of three papers reviews the causes, diagnosis and differential diagnosis, and outlines the management of sore and/or swollen lips. Clinical relevance: Sore and/or swollen lips are not uncommon, often have a local cause, but may reflect a systemic disease. The previous article reviewed important causes, from actinic to contact cheilitis, while this paper starts with drug-induced cheilitis and completes that alphabetical list.