CT-based radiomics analysis of different machine learning models for differentiating gnathic fibrous dysplasia and ossifying fibroma.

OBJECTIVE: In this study, our aim was to develop and validate the effectiveness of diverse radiomic models for distinguishing between gnathic fibrous dysplasia (FD) and ossifying fibroma (OF) before surgery. MATERIALS AND METHODS: We enrolled 220 patients with confirmed FD or OF. We extracted radiomic features from nonenhanced CT images. Following dimensionality reduction and feature selection, we constructed radiomic models using logistic regression, support vector machine, random forest, light gradient boosting machine, and eXtreme gradient boosting. We then identified the best radiomic model using receiver operating characteristic (ROC) curve analysis. After combining radiomics features with clinical features, we developed a comprehensive model. ROC curve and decision curve analysis (DCA) demonstrated the models' robustness and clinical value. RESULTS: We extracted 1834 radiomic features from CT images, reduced them to eight valuable features, and achieved high predictive efficiency, with area under curves (AUC) exceeding 0.95 for all the models. Ultimately, our combined model, which integrates radiomic and clinical data, displayed superior discriminatory ability (AUC: training cohort 0.970; test cohort 0.967). DCA highlighted its optimal clinical efficacy. CONCLUSION: Our combined model effectively differentiates between FD and OF, offering a noninvasive and efficient approach to clinical decision-making.

Clinicopathological characteristics and diagnostic accuracy of BRAF mutations in ameloblastoma: A Bayesian network analysis.

OBJECTIVE: This Bayesian network meta-analysis was performed to analyze the associations between clinicopathological characteristics and BRAF mutations in ameloblastoma (AM) patients and to evaluate the diagnostic accuracy. MATERIALS AND METHODS: Four electronic databases were searched from 2010 to 2024. The search terms used were specific to BRAF and AM. Observational studies or randomized controlled trials were considered eligible. The incidence of BRAF mutation and corresponding clinicopathological features in AM patients were subjected to Bayesian network analyses and diagnostic accuracy evaluation. RESULTS: A total of 937 AM patients from 20 studies were included. The pooled prevalence of BRAF mutations in AM patients was 72%. According to the Bayesian network analysis, BRAF mutations are more likely to occur in younger (odds ratio [OR], 2.3; credible interval [CrI]: 1.2-4.5), mandible site (OR, 3.6; 95% CrI: 2.7-5.2), and unicystic (OR, 1.6; 95% CrI: 1.1-2.4) AM patients. Similarly, higher diagnostic accuracy was found in the younger, mandible, and unicystic AM groups. CONCLUSIONS: The incidence, risk, and diagnostic accuracy of BRAF mutation in AM were greater in younger patients, those with mandible involvement, and those with unicystic AM than in patients with other clinicopathological features. In addition, there was a strong concordance in the diagnostic accuracy between molecular tests and immunohistochemical analysis.

Clinical, histopathological characteristics and malignant transformation of proliferative verrucous leukoplakia with 36 patients: a retrospective longitudinal study.

BACKGROUND: Proliferative verrucous leukoplakia (PVL), distinguished by its malignant transformation rate of 43.87% to 65.8%, stands as the oral potentially malignant disorder with the highest propensity for malignancy. PVL is marked by distinctive heterogeneity regarding the clinical or histopathological characteristics as well as prognostic factors pertinent to this condition. The purpose of this study is to compile and assess the clinicopathological features, malignant transformation, and associated risk factors in patients diagnosed with PVL. METHODS: This study is a hospital-based retrospective longitudinal study of 36 patients diagnosed with PVL from 2013 to 2023. We conducted complete clinical and histopathological evaluations of the patients. RESULTS: The cohort comprised 16 males and 20 females, yielding a male-to-female ratio of 1:1.25. The follow-up period ranged from 8 to 125 months, with an average of 47.50 months. The most common clinical type of lesion was the verrucous form (58.33%), and the gingiva was the most common site (44.44%). Each patient had between 2 to 7 lesions, averaging 3.36 per patient. During the follow-up period, twelve patients (33.3%) developed oral cancer, with an average time to malignant transformation of 35.75 months. Kaplan-Meier survival analysis indicated that patients with complaints of pain, roughness, or a rough sensation, with diabetes, and the presence of cytologic atypia histologically showed a higher risk of malignant transformation (p < 0.05). In this study, the rate of malignant transformation in the treatment group (5/23) was lower than that in the untreated group (7/13), however, no statistically significant difference (p = 0.05). CONCLUSION: The main complaints of pain, roughness, or foreign body sensation, coupled with cytologic atypia histologically are indicative of an increased risk of malignant transformation in PVL. Further research is needed to elucidate the influence of these clinicopathological parameters on the malignant progression of PVL.

Prognostic factors of palatal adenoid cystic carcinoma: A single-center analysis of 85 cases.

Objective: The purpose of this retrospective study was to describe the clinicopathological characteristics of primary adenoid cystic carcinoma (ACC) of the palate and to identify the factors affecting prognosis. Methods: The medical records of 85 patients with primary ACC of the palate treated with surgery, with or without adjuvant radiotherapy/chemotherapy, from 2009 to 2019 were reviewed. The relationship of different clinical parameters with locoregional recurrence (LR), distant metastasis (DM), and overall survival (OS) were analyzed. Results: Median follow-up time was 44.6 months. LR and DM rates were 24.7% and 25.9%, respectively, and the 5-year OS and disease-free survival (DFS) rates were 85.9% and 55.1%, respectively. Multivariate analysis showed that positive margins were independently associated with the risk of LR (p < .001). Positive margins (p = .001) and high histological grade (p = .031) were significantly associated with shorter OS. Conclusion: Positive surgical margins are a strong adverse prognostic factor affecting LR and OS in patients with ACC; apart from that, high histopathological grade is an independent predictor of poor OS. Level of Evidence: Level 3 (Prognosis - Cohort study).

FD-Net: Feature Distillation Network for Oral Squamous Cell Carcinoma Lymph Node Segmentation in Hyperspectral Imagery.

Oral squamous cell carcinoma (OSCC) has the characteristics of early regional lymph node metastasis. OSCC patients often have poor prognoses and low survival rates due to cervical lymph metastases. Therefore, it is necessary to rely on a reasonable screening method to quickly judge the cervical lymph metastastic condition of OSCC patients and develop appropriate treatment plans. In this study, the widely used pathological sections with hematoxylin-eosin (H&E) staining are taken as the target, and combined with the advantages of hyperspectral imaging technology, a novel diagnostic method for identifying OSCC lymph node metastases is proposed. The method consists of a learning stage and a decision-making stage, focusing on cancer and non-cancer nuclei, gradually completing the lesions' segmentation from coarse to fine, and achieving high accuracy. In the learning stage, the proposed feature distillation-Net (FD-Net) network is developed to segment the cancerous and non-cancerous nuclei. In the decision-making stage, the segmentation results are post-processed, and the lesions are effectively distinguished based on the prior. Experimental results demonstrate that the proposed FD-Net is very competitive in the OSCC hyperspectral medical image segmentation task. The proposed FD-Net method performs best on the seven segmentation evaluation indicators: MIoU, OA, AA, SE, CSI, GDR, and DICE. Among these seven evaluation indicators, the proposed FD-Net method is 1.75%, 1.27%, 0.35%, 1.9%, 0.88%, 4.45%, and 1.98% higher than the DeepLab V3 method, which ranks second in performance, respectively. In addition, the proposed diagnosis method of OSCC lymph node metastasis can effectively assist pathologists in disease screening and reduce the workload of pathologists.

Digital pathology-based artificial intelligence models for differential diagnosis and prognosis of sporadic odontogenic keratocysts.

Odontogenic keratocyst (OKC) is a common jaw cyst with a high recurrence rate. OKC combined with basal cell carcinoma as well as skeletal and other developmental abnormalities is thought to be associated with Gorlin syndrome. Moreover, OKC needs to be differentiated from orthokeratinized odontogenic cyst and other jaw cysts. Because of the different prognosis, differential diagnosis of several cysts can contribute to clinical management. We collected 519 cases, comprising a total of 2 157 hematoxylin and eosin-stained images, to develop digital pathology-based artificial intelligence (AI) models for the diagnosis and prognosis of OKC. The Inception_v3 neural network was utilized to train and test models developed from patch-level images. Finally, whole slide image-level AI models were developed by integrating deep learning-generated pathology features with several machine learning algorithms. The AI models showed great performance in the diagnosis (AUC = 0.935, 95% CI: 0.898-0.973) and prognosis (AUC = 0.840, 95%CI: 0.751-0.930) of OKC. The advantages of multiple slides model for integrating of histopathological information are demonstrated through a comparison with the single slide model. Furthermore, the study investigates the correlation between AI features generated by deep learning and pathological findings, highlighting the interpretative potential of AI models in the pathology. Here, we have developed the robust diagnostic and prognostic models for OKC. The AI model that is based on digital pathology shows promise potential for applications in odontogenic diseases of the jaw.

Genomic alterations in oral multiple primary cancers.

Oral squamous cell carcinoma (OSCC) is the predominant type of oral cancer, while some patients may develop oral multiple primary cancers (MPCs) with unclear etiology. This study aimed to investigate the clinicopathological characteristics and genomic alterations of oral MPCs. Clinicopathological data from patients with oral single primary carcinoma (SPC, n = 202) and oral MPCs (n = 34) were collected and compared. Copy number alteration (CNA) analysis was conducted to identify chromosomal-instability differences among oral MPCs, recurrent OSCC cases, and OSCC patients with lymph node metastasis. Whole-exome sequencing was employed to identify potential unique gene mutations in oral MPCs patients. Additionally, CNA and phylogenetic tree analyses were used to gain preliminary insights into the molecular characteristics of different primary tumors within individual patients. Our findings revealed that, in contrast to oral SPC, females predominated the oral MPCs (70.59%), while smoking and alcohol use were not frequent in MPCs. Moreover, long-term survival outcomes were poorer in oral MPCs. From a CNA perspective, no significant differences were observed between oral MPCs patients and those with recurrence and lymph node metastasis. In addition to commonly mutated genes such as CASP8, TP53 and MUC16, in oral MPCs we also detected relatively rare mutations, such as HS3ST6 and RFPL4A. Furthermore, this study also demonstrated that most MPCs patients exhibited similarities in certain genomic regions within individuals, and distinct differences of the similarity degree were observed between synchronous and metachronous oral MPCs.

Architectural and cytological features of epithelial dysplasia associated with transformation risk.

OBJECTIVES: This study explored associations between histological features of dysplasia and malignant transformation, as well as genomic copy number alterations. MATERIALS AND METHODS: Overall, 201 samples were collected from patients of oral leukoplakia. The associations of dysplastic features with malignant transformation and copy number alterations were investigated by Cox proportional hazards regression analysis and the Mann-Whitney U-test. RESULTS: Eight individual histological features, such as irregular epithelial stratification (p = 0.001), mitoses high in epithelium (p = 0.033), extension of changes along minor gland ducts (p < 0.001), etc., were associated with greater risk of malignant transformation. A model including histological features and age showed good performance for predicting malignant transformation (area under receiver operating characteristic curve: 0.806). Irregular epithelial stratification (p = 0.007), abnormal nuclear shape (p = 0.005), abnormal cell size (p = 0.004), etc. were associated with greater genomic instability. CONCLUSIONS: A Cox proportional hazards model using eight histological features and patient age reliably predicted the malignant potential of oral epithelial dysplasia. Identification of these histological features closely related to malignant transformation may aid the management of oral potentially malignant disorders and early detection of oral squamous cell carcinoma.

Neoadjuvant tislelizumab combined with chemotherapy in locally advanced oral or oropharyngeal squamous cell carcinoma: a real-world retrospective study.

Objectives: The treatment of locally advanced oral or oropharyngeal squamous cell carcinoma (LAOOPSCC) is surgery and radiotherapy or chemoradiotherapy but with unsatisfactory survival rate. Neoadjuvant programmed death-1 (PD-1) therapy are being used in several clinical trials. Therefore, in this retrospective study we aimed to determine the feasibility of neoadjuvant tislelizumab plus chemotherapy followed by surgery for LAOOPSCC. Materials and methods: The clinical data of 33 patients with LAOOPSCC who received neoadjuvant PD-1 inhibitors and chemotherapy between April 2021 and October 2022 were retrospectively analyzed. Patients with stage III-IV LAOOPSCC received tislelizumab, albumin-bound paclitaxel, and cisplatin every 3 weeks (Q3W) for two cycles, followed by surgery and adjuvant radiotherapy or concurrent chemoradiotherapy. A median follow-up period was 20 months. Results: The objective response rate (ORR) was 66.7%, with the major pathological response (MPR) rate at 54.5%, and the pathological complete response (pCR) rate was 33.3%. Sixteen patients underwent limited surgeries, and 15 patients were remitted from undergoing mandibulectomy and 9 patients were remitted from undergoing near total glossectomy or total glossectomy. A significant difference in the overall survival (OS) and disease-free survival (DFS) was observed in patients who achieved major pathological response (MPR) than who did not. The most common adverse events in neoadjuvant therapy were alopecia, decreased appetite or anorexia, leukopenia, and fatigue. Conclusion: Neoadjuvant PD-1 inhibitors combined with chemotherapy are feasible and safe, with a high pathological response and possible organ preservation in oral or oropharyngeal squamous cell carcinoma. However, further studies with a larger cohort of patients and longer follow-up period is required to strengthen our findings and evaluate the survival benefits of the treatment.

Head and neck carcinoma in children: A clinicopathological study of 42 cases.

Background/purpose: Cancer is an important part of the global burden of childhood diseases. Head and neck carcinoma in children is rare and related research is limited. This study aimed to investigate the clinicopathological features of childhood head and neck carcinoma. Materials and methods: Forty-two cases of childhood head and neck carcinoma treated in our institution were reviewed and analyzed. Results: Median age overall was 11 years. Twenty-three patients (54.8%) were male and 19 (45.2%) were female. Parotid gland location was most common (54.8%). Mucoepidermoid carcinoma and squamous cell carcinoma were the most common histological types (57.1% and 11.9%, respectively). Two patients had a history of bone marrow transplantation and two had a history of odontogenic keratocyst. The recurrence rate after treatment was 8.6%. Conclusion: Early diagnosis and treatment and close follow-up of childhood head and neck carcinoma are warranted to prevent recurrence and improve clinical outcome.

Effects of betaine supplementation on inflammatory markers: a systematic review and meta-analysis of randomised controlled trials.

Several studies have suggested that betaine is closely related to inflammatory biomarkers that contribute to the development of metabolic diseases, but the effect remains controversial. This meta-analysis aimed to assess the effects of betaine supplementation on inflammatory markers based on randomised controlled trials (RCTs). PubMed, Web of Science and ResearchGate databases were searched up to March 2023. A total of 6 RCTs with 7 intervention trials involving 277 participants were included. Betaine supplementation led to a slight reduction in levels of circulating IL-1ß of 0.65 pg/mL (95% CI, -1.23 to -0.06) with high heterogeneity (I2 = 95%). Betaine produced a small but nonsignificant reduction in levels of circulating CRP (0.33 mg/L; 95% CI, -1.79 to 1.14), IL-6 (0.47 pg/mL; 95% CI, -1.13 to 0.18) and TNF-α (0.25 pg/mL; 95% CI, -0.98 to 0.48). The present meta-analysis does not provide sufficient evidence to conclude that betaine supplementation improved the inflammation state.

Confirmation of eosinophilic sialodochitis by terminal duct biopsy.

OBJECTIVE: To analyse the histopathological features of eosinophilic sialodochitis by using terminal duct biopsy. METHODS: Sixty-five patients with suspected eosinophilic sialodochitis and four with chronic obstructive sialadenitis were prospectively enrolled. Clinical features, laboratory tests and sialograms were comparatively analysed. Terminal duct biopsy of the parotid or submandibular glands was performed concomitantly with endoscopy-assisted duct dilatation to determine the histopathological features of eosinophilic sialodochitis. RESULTS: Based on eosinophil quantification, the samples of suspected patients were scored as 'definite', 'highly suspected' and 'negative' in 26 (40%), 15 (23.1%) and 24 (36.9%) cases, respectively. Gland types and peripheral blood eosinophil counts were significantly different among these three groups. The proportions of itching glands, mucus plug exudations and elevated immunoglobulin E levels were higher in the 'definite' group than in the other two groups; however, the intergroup differences were insignificant. The primary pathological features of eosinophilic sialodochitis were abundant eosinophils and lymphocytes infiltrated around the duct, degranulation of eosinophils, extensive fibrosis and scattered mastocytes. Periductal eosinophils were not found in cases of chronic obstructive sialadenitis. CONCLUSION: Our findings suggest that terminal duct biopsy is safe and valuable for the pathological confirmation of eosinophilic sialodochitis, and can be used simultaneously with endoscopy-assisted duct dilatation.

Biomarkers of malignant transformation in oral leukoplakia: from bench to bedside. / å£è ”ç™½æ–‘ç™Œå˜ç”Ÿç‰©æ ‡å¿—ç‰©ï¼š ä»ŽåŸºç¡€ç§‘ç ”åˆ°ä¸´åºŠåº”ç”¨.

Oral leukoplakia is a common precursor lesion of oral squamous cell carcinoma, which indicates a high potential of malignancy. The malignant transformation of oral leukoplakia seriously affects patient survival and quality of life; however, it is difficult to identify oral leukoplakia patients who will develop carcinoma because no biomarker exists to predict malignant transformation for effective clinical management. As a major problem in the field of head and neck pathologies, it is imperative to identify biomarkers of malignant transformation in oral leukoplakia. In this review, we discuss the potential biomarkers of malignant transformation reported in the literature and explore the translational probabilities from bench to bedside. Although no single biomarker has yet been applied in the clinical setting, profiling for genomic instability might be a promising adjunct.

Adenoid Ameloblastoma Shares Clinicopathologic, Immunohistochemical, and Molecular Features With Dentinogenic Ghost Cell Tumor: A Comparative Analysis.

The updated classification of odontogenic tumors by the World Health Organization (WHO) has included adenoid ameloblastoma (AA) as a distinct entity. However, distinguishing between AA and dentinogenic ghost cell tumor (DGCT) can still be challenging due to their significant morphologic similarities. In this study, we aimed to compare the clinicopathologic, immunohistochemical, and molecular characteristics of AA and DGCT to aid in their differentiation and to shed light on their pathologic mechanisms. Thirteen cases of AA and 14 cases of DGCT (15 samples) were analyzed, along with 11 cases of adenomatoid odontogenic tumor (AOT) and 18 cases of conventional ameloblastoma (AM) for comparative purposes. The study found that AA and DGCT shared a similar long-term prognosis. Immunohistochemically, all cytokeratins detected, except CK8/18, were not statistically significant in differentiating AA and DGCT, while there was a statistically significant difference in the immunophenotype of CK7 and CK10/13 between AA and AM. Nuclear ß-catenin accumulation were detected in all cases of AA and DGCT, while AOTs and AMs exhibited cytoplasmic ß-catenin. Molecularly, CTNNB1 hotspot mutations were found in only 1 case of AA (1/13), but not found in the other 3 types of tumors. BRAF p.V600E mutation was positive in 2/13 (15%) AA, 1/15 (7%) DGCT, and 2/11 (18%) AOT cases. In comparison, conventional AM was positive for BRAF p.V600E mutation in 94% (17/18) of cases, while KRAS mutations were detected in 63% (7/11) of AOT cases. The study suggests that the so-called AA is a rare benign tumor that exhibits clinical, immunohistochemical, and molecular features similar to DGCTs. Based on these findings, AA should not be categorized as a standalone entity solely based on the presence of whorls/morules and cribriform/duct-like structures. Further studies are needed to investigate the pathologic mechanisms of these tumors and to identify potential therapeutic targets.

Isolation of Cells with Morphological and Spatial Information from Oral Submucous Fibrosis Samples by Laser Capture Microdissection.

Oral submucous fibrosis (OSF) is a common type of potentially malignant disorder in the oral cavity. The atrophy of epithelium and fibrosis of the lamina propria and the submucosa are often found on histopathological slides. Epithelial dysplasia, epithelial atrophy, and senescent fibroblasts have been proposed to be associated with the malignant transformation of OSF. However, because of the heterogeneity of potentially malignant oral disorders and oral squamous cell carcinoma, it is difficult to identify the specific molecular mechanisms of malignant transformation in OSF. Here, we present a method to obtain a small number of epithelial or mesenchymal cells carrying morphological data and spatial information by laser capture microdissection on formalin-fixed paraffin-embedded tissue slides. Using a microscope, we can precisely capture microscale (~500 cells) dysplastic or atrophic epithelial tissue and fibrotic subepithelial tissue. The extracted cells can be evaluated by genome or transcriptome sequencing to acquire genomic and transcriptomic data with morphological and spatial information. This approach removes the heterogeneity of bulk OSF tissue sequencing and the interference caused by cells in non-lesioned areas, allowing for precise spatial-omics analysis of OSF tissue.

PTCH/SMO gene mutations in odontogenic keratocysts and drug interventions.

BACKGROUND: Odontogenic keratocysts (OKCs) are odontogenic jaw lesions that cause destruction and dysfunction of the jawbone. OKCs can be sporadic or associated with nevoid basic cell carcinoma syndrome (NBCCS). However, the factors that initiate OKCs and the mechanism of cyst formation remain unclear. Here, we investigated the impact of PTCH1 and SMO mutations on disease progression, as well as the effects of sonic hedgehog (SHH) signaling pathway inhibitors GDC-0449 and GANT61 on OKC fibroblasts. METHODS: Eight sporadic OKC fibroblasts without gene mutations were used as the control, and six NBCCS-related fibroblasts were cultured in vitro. The effect of PTCH1 non-truncated mutation 3499G>A (p.G1167R) and SMO c.2081C>G (p.P694R) mutation on OKC fibroblast proliferation was examined by EdU assay. CCK8 and wound-healing assays detected the effects of OKC fibroblasts carrying PTCH1 c.3499G>A (p.G1167R) and SMO c.2081C>G (p.P694R) mutations on the proliferation and migration of HaCaT cells after co-culture. Quantitative real-time PCR detected the effects of GDC-0449 or GANT61 on the SHH signaling pathway in NBCCS-related OKCs with PTCH1 truncated mutations and PTCH1 c.3499G>A (p.G1167R) and/or SMO c.2081C>G (p.P694R) mutations. RESULTS: PTCH1 c.3499G>A (p.G1167R) and SMO c.2081C>G (p.P694R) promoted the proliferation of OKC fibroblasts. The proliferation and migration of HaCaT cells were affected by NBCCS-related OKC fibroblasts carrying PTCH1 c.3499G>A (p.G1167R) and SMO c.2081C>G (p.P694R) mutations. GDC-0449 significantly inhibited the SHH signaling pathway in NBCCS-related OKC fibroblasts with PTCH1 truncated mutations. An NBCCS-related OKC carrying PTCH1 c.3499G>A (p.G1167R) and SMO c.2081C>G (p.P694R) mutations were resistant to GDC-0449 but inhibited by GANT61. CONCLUSIONS: Genetic mutations in OKC fibroblasts may affect the biological behavior of epithelial and stromal cells and cause disease. GDC-0449 could be used to treat OKCs, especially NBCCS-related OKCs with PTCH1 truncated mutations. SMO c.2081C>G (p.P694R) may lead to resistance to GDC-0449; however, GANT61 may be used as an alternative inhibitor.

Concentrations and homologue patterns of SCCPs and MCCPs in the serum of the general population of adults in Hangzhou, China.

Due to their ubiquitous presence in the environment and humans, chlorinated paraffins (CPs) are a major environmental and public health concern. CPs are known to persist, bioaccumulate and potentially threaten human health, but reports on their internal exposure in the adult general population are still scarce. In this study, serum samples collected from adults living in Hangzhou, China, were quantified for SCCPs and MCCPs using GC-NCI-MS methods. A total of 150 samples were collected and subjected to analysis. ∑SCCPs were detected in 98% of the samples with a median concentration of 721 ng/g lw. MCCPs were found in all serum samples with a median concentration of 2210 ng/g lw, indicating that MCCPs were the dominant homologous group. For SCCPs and MCCPs, ∑C10 and ∑C14 were found to be the dominant carbon chain length homologues. Our results showed that age, BMI and lifestyle were not found to be significantly associated with internal exposure to CPs for the samples in this study. Based on PCA analysis, an age-specific distribution of CP homologues was observed. This suggests that internal exposure to CPs in the general population is related to exposure scenarios and history. The results of this study may contribute to a better understanding of the internal exposure to CPs in the general population and may provide a direction for the investigation of the source of exposure to CPs in the environment and daily life.

Molecular mechanisms underlying the epigallocatechin-3-gallate-mediated inhibition of oral squamous cell carcinogenesis.

OBJECTIVES: To reveal the mechanisms underlying the epigallocatechin-3-gallate (EGCG)-mediated inhibition of carcinogenesis and the related regulatory signaling pathways. DESIGN: The effect of EGCG on the proliferation of OSCC cells was examined. SuperPred, ChEMBL, Swiss TargetPrediction, DisGeNET, GeneCards, and National Center for Biotechnology Information databases were used to predict the EGCG target genes and oral leukoplakia (OL)-related, oral submucosal fibrosis (OSF)-related, and OSCC-related genes. The binding of EGCG to the target proteins was simulated using AutoDock and PyMOL. The Cancer Genome Atlas (TCGA) dataset was subjected to consensus clustering analysis to predict the downstream molecules associated with these targets, as well as their potential functions and pathways. RESULTS: EGCG significantly inhibited OSCC cell proliferation (p < 0.001). By comparing EGCG target genes with genes linked to oral potentially malignant disorder (OPMD) and OSCC, a total of eleven potential EGCG target genes were identified. Furthermore, EGCG has the capacity to bind to eleven proteins. Based on consensus clustering and enrichment analysis, it is suggested that EGCG may hinder the progression of cancer by altering the cell cycle and invasive properties in precancerous lesions of the oral cavity. Some possible strategies for modifying the cell cycle and invasive properties may include EGCG-mediated suppression of specific genes and proteins, which are associated with cancer development. CONCLUSIONS: This study investigated the molecular mechanisms and signaling pathways associated with the EGCG-induced suppression of OSCC. The identification of specific pharmacological targets of EGCG during carcinogenesis is crucial for the development of innovative combination therapies involving EGCG.

Development of a Pathomics-Based Model for the Prediction of Malignant Transformation in Oral Leukoplakia.

Accurate prognostic stratification of oral leukoplakia (OLK) with risk of malignant transformation into oral squamous cell carcinoma is crucial. We developed an objective and powerful pathomics-based model for the prediction of malignant transformation in OLK using hematoxylin and eosin (H&E)-stained images. In total, 759 H&E-stained images from multicenter cohorts were included. A training set (n = 489), validation set (n = 196), and testing set (n = 74) were used for model development. Four deep learning methods were used to train and validate the model constructed using H&E-stained images. Pathomics features generated through deep learning combined with machine learning algorithms were used to develop a pathomics-based model. Immunohistochemical staining of Ki67, p53, and PD-L1 was used to interpret the black box of the model. Pathomics-based models predicted the malignant transformation of OLK (validation set area under curve [AUC], 0.899; testing set AUC, 0.813) and significantly identified high-risk and low-risk populations. The prediction performance of malignant transformation from dysplasia grading (validation set AUC, 0.743) was lower than that of the pathomics-based model. The expressions of Ki67, p53, and PD-L1 were correlated with various pathomics features. The pathomics-based model accurately predicted the malignant transformation of OLK and may be useful for the objective and rapid assessment of the prognosis of patients with OLK.

[Application of Deep Learning in Differential Diagnosis of Ameloblastoma and Odontogenic Keratocyst Based on Panoramic Radiographs].

Objective To evaluate the accuracy of different convolutional neural networks (CNN),representative deep learning models,in the differential diagnosis of ameloblastoma and odontogenic keratocyst,and subsequently compare the diagnosis results between models and oral radiologists. Methods A total of 1000 digital panoramic radiographs were retrospectively collected from the patients with ameloblastoma (500 radiographs) or odontogenic keratocyst (500 radiographs) in the Department of Oral and Maxillofacial Radiology,Peking University School of Stomatology.Eight CNN including ResNet (18,50,101),VGG (16,19),and EfficientNet (b1,b3,b5) were selected to distinguish ameloblastoma from odontogenic keratocyst.Transfer learning was employed to train 800 panoramic radiographs in the training set through 5-fold cross validation,and 200 panoramic radiographs in the test set were used for differential diagnosis.Chi square test was performed for comparing the performance among different CNN.Furthermore,7 oral radiologists (including 2 seniors and 5 juniors) made a diagnosis on the 200 panoramic radiographs in the test set,and the diagnosis results were compared between CNN and oral radiologists. Results The eight neural network models showed the diagnostic accuracy ranging from 82.50% to 87.50%,of which EfficientNet b1 had the highest accuracy of 87.50%.There was no significant difference in the diagnostic accuracy among the CNN models (P=0.998,P=0.905).The average diagnostic accuracy of oral radiologists was (70.30±5.48)%,and there was no statistical difference in the accuracy between senior and junior oral radiologists (P=0.883).The diagnostic accuracy of CNN models was higher than that of oral radiologists (P<0.001). Conclusion Deep learning CNN can realize accurate differential diagnosis between ameloblastoma and odontogenic keratocyst with panoramic radiographs,with higher diagnostic accuracy than oral radiologists.