Supervised training for oral potentially malignant disorder (OPMD) image segmentation requires expensive annotated data, particularly scarce in remote regions with few medical specialists. This study aims to propose a self-guided pre-training method to automatically extract features from unlabeled images, enhancing the performance of OPMD lesion segmentation. This study utilized 3,417 OPMD photographs from ZJUSS as the internal dataset and two independent external datasets from WCHS and CS-SJTU for validation. The internal labeled dataset was evaluated using five independent patient-level train/validation/test splits to prevent patient-level data leakage. We proposed a multiscale separable attention masked image model, MS-SAMIM, and a self-guided mask generation method. To enhance robustness, we incorporated a teacher–student consistency constraint and a contrastive learning constraint. Our model achieved 84.35

Oral ulcers, particularly in diabetic patients, pose significant clinical challenges due to impaired healing conditions and high susceptibility to infection. Considering the highly dynamic, moist, and complex multi-bacterial environment of the oral cavity, there is an urgent need to develop a bioactive formulation with wet-adhesion and antibacterial properties to manage diabetic oral ulcers. Here, we developed a novel multifunctional hydrogel (polyvinyl alcohol-borate-cytidine-Ag+, PBCA) based on the i-motif structure of C-Ag+-C coordination, boronate ester bonds formed between PVA and cytidine, and glucocorticoid loading. The hydrogel exhibited excellent physicochemical properties, with an adhesion force of up to 46.6 ± 2.1 kPa. It can adhere to the oral mucosa of rats for over one hour and features continuous silver ion release and structural stability. In vitro studies confirmed its broad-spectrum antibacterial efficacy against both bacterial and fungal oral pathogens, alongside outstanding cytocompatibility with human oral keratinocytes and mouse fibroblasts. In vivo experiments using an oral ulcer model in diabetic rats demonstrated that PBCA hydrogel significantly accelerated mucosal healing, enhanced collagen deposition, promoted angiogenesis, and reduced local inflammation without inducing systemic organ toxicity or skin and oral mucosa irritation. These findings suggest that PBCA hydrogel provides a promising material-based strategy for the treatment of complex chronic oral ulcers under impaired healing conditions.

Background Mucosal-dominant pemphigus vulgaris (MD-PV) and low-risk mucous membrane pemphigoid (LR-MMP) are autoimmune bullous diseases primarily affecting the oral mucosa, yet their oral microbiome profiles remain inadequately characterized.Methods Using 16S rRNA sequencing of saliva from 21 MD-PV patients, 26 LR-MMP patients, and 14 healthy controls (HC), we analyzed microbial diversity, differential taxa (LEfSe), and functional potential (PICRUSt2/BugBase).Results Both patient groups showed significant microbial restructuring without major richness changes, featuring increased Firmicutes and decreased Proteobacteria. Disease-specific signatures included Flavobacteriia enrichment in MD-PV and Coriobacteriia/Actinobacteria in LR-MMP. Shared metabolic alterations involved "Biosynthesis of amino acids," "Phosphotransferase system," and "Ribosome" pathways, while distinct activations included "all-trans-farnesol biosynthesis" in MD-PV and "peptidoglycan biosynthesis" in LR-MMP. Phenotype prediction revealed increased Gram-positive bacteria and reduced pathogenic and stress-tolerant taxa. Microbial dysbiosis scores positively correlated with clinical disease severity.Conclusion Our study identifies distinct oral microbial dysbiosis patterns in MD-PV and LR-MMP, with conserved functional shifts and disease-specific metabolic adaptations. The microbiota-severity correlation highlights its potential role in disease mechanisms, offering new insights for therapeutic exploration.Trial Registration Chinese Clinical Trial Registry (ChiCTR): ChiCTR2500105460

OBJECTIVE:To evaluate the quality and readability of large language models (LLMs) when responding to Frequently Asked Questions (FAQs) about oral lichen planus (OLP). METHODS:We evaluated the responses of three LLMs (ChatGPT-4o, Gemini 2.0 Flash Experimental, and Copilot) to 13 patient-centered FAQs about OLP. Questions were identified using query tools, and answers were assessed by 14 oral medicine experts using the Quality Assessment of Medical Artificial Intelligence (QAMAI) tool. Readability was analyzed with the Flesch Reading Ease (FRE) and Flesch-Kincaid Grade Level (FKG) tools. RESULTS:All LLMs provided generally accurate and relevant responses, with median QAMAI scores indicating "good" to "very good" quality. ChatGPT achieved slightly higher completeness, particularly for questions on OLP definition and treatment. The reference provision was inconsistent across all chatbots. Readability analysis revealed that most responses required college-level literacy, with ChatGPT producing the most complex texts, Gemini occasionally achieving more accessible outputs, and Copilot situated in an intermediate position. CONCLUSIONS:LLMs may have potential as adjunctive tools for patient education in OLP, although they remain limited by incomplete information, inconsistent references, and suboptimal readability. Future research should incorporate longitudinal LLMs evaluations and training to develop models delivering accurate, accessible information, tailored to users' literacy levels.

Rauch–Steindl syndrome is a rare autosomal dominant genetic disorder caused by pathogenic variants in the NSD2 gene, characterized by growth retardation, intellectual disability, and characteristic facial features. Although systemic manifestations have been increasingly recognized, the orofacial phenotype of this syndrome remains insufficiently defined. This report aims to provide a detailed description of the orofacial findings in a patient with Rauch–Steindl syndrome. Here, we report a 17-year-old male with growth retardation, mild intellectual disability, and characteristic facial features. Oral examination revealed a white nodular lesion at the midline of the dorsal tongue, multiple permanent tooth agenesis, and abnormal tooth morphology. Whole-exome sequencing combined with copy number variation analysis identified an approximately 10.13 kb deletion in the 4p16.3 region, encompassing exons 12–14 of the NSD2 gene. According to the American College of Medical Genetics and Genomics guidelines, this deletion was classified as pathogenic, thereby supporting the diagnosis of Rauch–Steindl syndrome. Based on the clinical and mycological findings, the lingual lesion was diagnosed as nodular median rhomboid glossitis with chronic Candida infection. The lesion partially regressed after topical antifungal therapy but persisted at the one-year follow-up. This case expands the phenotypic spectrum of Rauch–Steindl syndrome by delineating its orofacial manifestations, particularly dental agenesis and abnormal tooth development. The occurrence of nodular median rhomboid glossitis may represent a secondary consequence of altered immune homeostasis in a patient with an NSD2 variant, rather than a direct phenotypic manifestation of Rauch–Steindl syndrome. Our findings further support the multisystem involvement associated with NSD2 haploinsufficiency and highlight the importance of comprehensive dental evaluation in patients with this syndrome. Further studies with larger cohorts are warranted to systematically characterize the oral phenotype associated with Rauch–Steindl syndrome.

Abstract Oral leukoplakia (OLK) is a common potentially malignant oral disorder with high risk of malignant transformation. While photodynamic therapy (PDT) offers a minimally invasive treatment for OLK, some patients show resistance to PDT and the mechanisms remain unclear. This study aims to identify key regulatory pathways driving PDT resistance in OLK. Single-cell RNA sequencing of OLK samples (three PDT-sensitive, three PDT-resistant) revealed significant NRF2 upregulation in resistant tissues. Validation across two independent cohorts (n = 117) confirmed that p-NRF2 levels were significantly elevated in PDT-resistant cases, exhibiting strong predictive power for treatment response (AUC > 0.8). Mechanistically, NRF2 promotes CTNNB1 transcription, activates WNT signaling, modulates reactive oxygen species responses, and regulates keratinization, collectively contributing to PDT resistance. In a 4NQO-induced OLK mouse model, NRF2 inhibition combined with PDT effectively reversed OLK lesions and restored mucosal histology. These findings establish p-NRF2 as a valuable biomarker for guiding PDT regimens in OLK patients, reveal NRF2’s role in mediating PDT resistance via the WNT signaling pathway, and highlight NRF2 inhibition as a promising strategy to enhance PDT efficacy.

Chronic inflammation of the oral mucosa could affect daily living and even threaten systemic health. Unlike periodontitis, oral lichen planus, a common oral chronic inflammatory disease, has diverse clinical manifestations and can progress to malignancy. Hence, this study aimed to investigate the mechanism of oral chronic inflammation using single-cell RNA sequencing (scRNA-seq), spatial transcriptome, a large clinical follow-up cohort with bulk RNA sequencing, cytological experiments, and multiplex immunohistochemistry. We found that epithelial pyroptosis-induced triggering receptor expressed on myeloid cell-1 (TREM1)+ macrophages activated pathogenic T helper cell 17 via interleukin-1β, to spur the inflammatory development of oral mucosal epithelium. Besides, we established a spatiotemporal interactional online database, Oral-Gut Axis Mucosal Immune Atlas (ORGUAMIA), to uncover the extensive pro-inflammatory role of epithelial pyroptosis-induced TREM1+ macrophages in chronic digestive tract disorders. In summary, this study highlights the role of epithelial pyroptosis-induced TREM1+ macrophages accelerating mucosal epithelial inflammation and offers ORGUAMIA as a tool for researchers using scRNA-seq and spatial transcriptome without technological barriers.

BACKGROUND:Dyskeratosis congenita (DC) is a rare and fatal disease, presenting with a classic triad of skin pigmentation, nail dystrophy and oral leukoplakia. However, diagnosing DC is challenging based solely on the protean manifestations and multisystemic involvement. Therefore, it is urgent to identify an early feature facilitating initial suspicion of DC. RESULTS:In this study, we enrolled a cohort of six male children diagnosed with DC, all of whom exhibited erosions or ulcers on the tongue, while five of them did not display the complete classic triad. Strikingly, oral erosions or ulcers have never been included in any existing clinical diagnostic criteria for DC. Through a retrospective analysis, we further demonstrated that extensive and persistent tongue ulceration emerges as an early and practicable clinical marker, provoking suspicion of DC even in the absence of the classic triad. CONCLUSIONS:Our findings challenge prevailing diagnostic criteria and advocates for an expanded consideration of tongue ulceration as a primary and indicative manifestation of DC, thereby affording a strategic advantage for early detection and intervention of this lethal disease.

This study aimed to evaluate the efficacy of a Wnt pathway activator, lithium chloride (LiCl), and LiCl-stimulated human primary periodontal ligament cells (hPDLCs) cell sheets integrated with medical-grade melt-electrowritten polycaprolactone (mPCL) scaffolds for in vivo periodontal regeneration. This study followed a three-stage experimental design. The first stage consisted of local administration of various concentrations of LiCl (10, 100, and 1000 mM) in rat periodontal defects for 3 weeks to evaluate its dosage effects on in vivo periodontal regeneration. Next, the impact of LiCl on cementogenic and osteogenic differentiation of human primary periodontal ligament cells (hPDLCs) was examined in vitro. Finally, LiCl-pretreated hPDLCs cell sheets, combined with mPCL scaffolds, were transplanted into rat periodontal defects for 4 weeks. Micro-CT, histology, and immunohistochemistry analyses were conducted to assess in vivo periodontal tissue formation. Local administration of LiCl (10 mM and 100 mM) promoted the regeneration of alveolar bone, cementum, and obliquely oriented PDL Sharpey fibres in the rat defect model. In vitro, 10 mM LiCl enhanced cementogenic and osteogenic differentiation in hPDLCs. Moreover, when implanting a 10 mM LiCl-treated hPDLCs cell sheet combined with a PCL scaffold into experimentally induced rat periodontal defects, significant regeneration occurred, including the formation of alveolar bone and cementum, along with Sharpey’s fibres inserted into the new cementum and alveolar bone. This study demonstrates the potential of combining LiCl-stimulated hPDLCs cell sheets with mPCL scaffolds to enhance periodontal regeneration, highlighting the significance of LiCl as a promising therapeutic agent for periodontal regenerative treatment and tissue engineering strategies.

Significance Photodynamic therapy (PDT) is effective for treating oral leukoplakia (OLK), but hyperkeratosis of the lesion often reduces its efficacy and the recurrence rate is high. This study compared the difference of short-term efficacy and recurrence risk between laser-assisted PDT and conventional PDT for treating OLK. Approach Logistic regression analysis was used for comparison of short-term efficacy between conventional and laser-assisted PDT. Cox regression was used for comparison of recurrence risks between the two groups. Results There was no statistically significant difference in short-term efficacy between conventional and laser-assisted PDT (P>0.05), but laser-assisted PDT could achieve greater area reduction per treatment (t = 2.125, P = 0.033) and reduced recurrence risk (HR = 0.301, 95% CI: 0.160-0.866). Conclusions Laser-assisted PDT was more efficient in treating OLK than conventional PDT and could reduce the risk of OLK recurrence after treatment.

Significance Oral squamous cell carcinoma (OSCC) often develops from oral potentially malignant disorder (OPMD). Even after complete resection of the tumor and surrounding OPMD lesion, recurrence is common. In this study, we reported the short-term and long-term efficacy of photodynamic therapy (PDT) for OPMD developed after surgical treatment of OSCC. Approach Clinicopathological features and follow-up data of the patients were collected and analyzed. Results 15 OPMD cases developed after surgical treatment of OSCC and treated with PDT were collected. 4 had complete response after PDT and 8 had partial response. In an average follow-up time of 20.2 months, 8 cases develop OSCC again. Conclusions PDT is safe and effective for OPMDs after surgical treatment of OSCC, the minimal invasiveness of PDT makes it preferable in preserving the structure and function of oral cavity, however, frequent monitoring is still needed due to the high risk of malignant transformation of these lesions.

OBJECTIVES:Introduce a new intrapapillary capillary loop (IPCL) classification using narrow-band imaging (NBI) to diagnose early malignant transformation in oral potentially malignant disorders (OPMDs). METHODS:In the exploratory phase, NBI images from 241 patients with OPMDs and oral squamous cell carcinoma (OSCC) were assessed for three basic IPCL features-dilation, unevenness, and tortuosity-to identify patterns associated with OPMDs malignant transformation. Chi-square analysis and logistic regression differentiated these patterns between low-risk (no/mild dysplasia) and high-risk (moderate/severe dysplasia, invasive carcinoma) lesions, establishing a new classification system. In the validation phase, the system was applied to 90 patients with histopathological results confirming diagnostic accuracy. RESULTS:Four distinct IPCL types were identified: normal, dilated, irregular, and chaotic, with increasing odds ratios for high-risk lesions. The ROC curve had an AUC of 0.913, with irregular and chaotic types providing the best diagnostic performance, achieving sensitivities and specificities of 84.2% and 88.4%, respectively. In the validation phase, the new classification achieved an overall diagnostic accuracy of 84.4% and substantial inter-observer consistency (κ = 0.694). CONCLUSION:The new IPCL classification effectively identifies high-risk oral lesions, enhancing the diagnostic utility of NBI for early detection of OPMDs malignant transformations.

Background The grading of oral epithelial dysplasia is often time-consuming for oral pathologists and the results are poorly reproducible between observers. In this study, we aimed to establish an objective, accurate and useful detection and grading system for oral epithelial dysplasia in the whole-slides of oral leukoplakia.Methods Four convolutional neural networks were compared using the image patches from 56 whole-slide of oral leukoplakia labeled by pathologists as the gold standard. Sequentially, feature detection models were trained, validated and tested with 1,000 image patches using the optimal network. Lastly, a comprehensive system named E-MOD-plus was established by combining feature detection models and a multiclass logistic model.Results EfficientNet-B0 was selected as the optimal network to build feature detection models. In the internal dataset of whole-slide images, the prediction accuracy of E-MOD-plus was 81.3% (95% confidence interval: 71.4-90.5%) and the area under the receiver operating characteristic curve was 0.793 (95% confidence interval: 0.650 to 0.925); in the external dataset of 229 tissue microarray images, the prediction accuracy was 86.5% (95% confidence interval: 82.4-90.0%) and the area under the receiver operating characteristic curve was 0.669 (95% confidence interval: 0.496 to 0.843).Conclusions E-MOD-plus was objective and accurate in the detection of pathological features as well as the grading of oral epithelial dysplasia, and had potential to assist pathologists in clinical practice.

BACKGROUND:Serum response factor (SRF) and myocardial-associated transcription factor-A (MRTF-A) had different regulatory effects on the tumorigenesis and development in different cancers. However, the role of MRTF-A/SRF in oral squamous cell carcinoma (OSCC) remains to be determined. METHODS:CCK-8 assay, cell scratch experiment, and transwell invasion assay were conducted to investigate the effects of MRTF-A/SRF on biological behavior of OSCC cells. The expression pattern and prognostic value of MRTF-A/SRF in OSCC were analyzed based on cBioPortal website and TCGA database. Protein-protein interaction network was visualized to identify protein functions. Go and KEGG pathway analyses were performed to investigate related pathways. The effect of MRTF-A/SRF on epithelial-mesenchymal transformation (EMT) of OSCC cells was explored by western blot assay. RESULTS:Overexpression of MRTF-A/SRF inhibited the proliferation, migration, and invasion of OSCC cells in vitro. High expression of SRF was related to better prognosis of OSCC patients on hard palate, alveolar ridge, and oral tongue. Besides, overexpression of MRTF-A/SRF inhibited the EMT of OSCC cells. CONCLUSION:SRF was closely related to the prognosis of OSCC. High expression of SRF and its co-activator MRTF-A inhibited proliferation, migration, and invasion of OSCC cells in vitro, possibly via EMT suppression.

BACKGROUND:The treatment of oral leukoplakia (OLK) with aminolaevulinic acid photodynamic therapy (ALA-PDT) is widespread. Nonetheless, there is variation in efficacy. Therefore, this study constructed a model for predicting the short-term efficacy and recurrence of OLK after ALA-PDT. METHODS:The short-term efficacy and recurrence of ALA-PDT were calculated by statistical analysis, and the relevant influencing factors were analyzed by Logistic regression and COX regression model. Finally, prediction models for total response (TR) rate, complete response (CR) rate and recurrence in OLK patients after ALA-PDT treatment were established. Features from pathology sections were extracted using deep learning autoencoder and combined with clinical variables to improve prediction performance of the model. RESULTS:The logistic regression analysis showed that the non-homogeneous (OR: 4.911, P: 0.023) OLK and lesions with moderate to severe epithelial dysplasia (OR: 4.288, P: 0.042) had better short-term efficacy. The area under receiver operating characteristic curve (AUC) of CR, TR and recurrence predict models after the ALA-PDT treatment of OLK patients is 0.872, 0.718, and 0.564, respectively. Feature extraction revealed an association between inflammatory cell infiltration in the lamina propria and recurrence after PDT. Combining clinical variables and deep learning improved the performance of recurrence model by more than 30 %. CONCLUSIONS:ALA-PDT has excellent short-term efficacy in the management of OLK but the recurrence rate was high. Prediction model based on clinicopathological characteristics has excellent predictive effect for short-term efficacy but limited effect for recurrence. The use of deep learning and pathology images greatly improves predictive value of the models.

OBJECTIVE:To explore the clinical epidemiological characteristics of oral lichen planus (OLP) and risk factors for erosive/ulcerative OLP. MATERIALS AND METHODS:Patients diagnosed with OLP from 11 different hospitals were included in the study. Descriptive statistical methods were used to explore the clinical epidemiological characteristics and logistic regression, sensitivity analysis, and subgroup analysis were utilized to explore the risk factors for erosive/ulcerative OLP. RESULTS:The average age of patients was 49.2 ± 13.3 years, and 61.4% of the patients were women. The ratios of patients with reticular, hyperemic/erythematous, and erosive/ulcerative lesions were 47.9%, 27.8%, and 24.2%, respectively. Analysis of risk factors for erosive/ulcerative OLP identified the following variables: age, course of disease of 12 months or more, II°-III° dental calculus, hypertension, diabetes, and heart disease, as well as regions of habitation. Subgroup analysis showed significant differences in risk factors for erosive/ulcerative OLP in patients with and without risk behaviors. CONCLUSION:The clinical epidemiological characteristics of patients with OLP in the Chinese population in this study are basically consistent with existing reports in developed countries. And we identified clinical characteristics associated with erosive/ulcerative OLP through clinical epidemiological analysis.

OBJECTIVE:Proliferative verrucous leukoplakia (PVL) is characterized by a spectrum of clinicopathological features and a high risk of malignant transformation. In this study, we aimed to delineate the dynamic changes in molecular signature during PVL progression and identify the potential cell subtypes that play a key role in the premalignant evolution of PVL.METHODS:We performed single-cell RNA sequencing on three biopsy samples from a large PVL lesion. These samples exhibited a histopathological continuum of PVL progression.RESULTS:By analyzing the transcriptome profiles of 27,611 cells from these samples, we identified ten major cell lineages and revealed that cellular remodeling occurred during the progression of PVL lesions, including epithelial, stromal, and immune cells. Epithelial cells are shifted to tumorigenic states and secretory patterns at the premalignant stage. Immune cells showed growing immunosuppressive phenotypes during PVL progression. Remarkably, two novel cell subtypes INSR+ endothelial cells and ASPN+ fibroblasts, were discovered and may play vital roles in microenvironment remodeling, such as angiogenesis and stromal fibrosis, which are closely involved in malignant transformation.CONCLUSION:Our work is the first to depict the cellular landscape of PVL and speculate that disease progression may be driven by functional remodeling of multiple cell subtypes.