Central dentinogenic ghost cell tumor of the maxilla: a case report with new imaging findings and review of the literature.

A dentinogenic ghost cell tumor (DGCT) is a rare benign odontogenic tumor that commonly shows characteristics of solid proliferation and has a relatively high risk of recurrence after surgical treatment. We herein report a case of a central DGCT that occurred in the maxilla and resulted in bone expansion. This study highlights new imaging findings (particularly magnetic resonance imaging) along with histopathological observations. In addition, we conducted a review of the existing literature on this rare tumor. A 37-year-old man developed swelling around the right cheek. A benign odontogenic tumor such as ameloblastoma was suspected based on the imaging examination findings (including bone expansion and the internal characteristics of the tumor) on panoramic imaging, computed tomography, and magnetic resonance imaging. The lesion was surgically excised from the right maxilla. Postoperative histopathological examination led to a definitive diagnosis of central DGCT. The tumor comprised epithelial neoplastic islands, resembling ameloblastoma, inside tight fibroconnective tissue; masses of ghost cells and formation of dentin were also observed. We had suspected that the minute high-density region around the molars on the imaging examinations represented alveolar bone change; however, it represented dentin formation. This led to difficulty diagnosing the lesion. Although DGCT may present characteristic findings on imaging examinations, its occurrence is infrequent, and in some cases, the findings may include the presence or absence of an impacted tooth without obvious calcification. The present case suggests that we should consider the possibility of an odontogenic tumor with calcification when high-density structures are observed inside the lesion.

Double-faced CX3CL1 enhances lymphangiogenesis-dependent metastasis in an aggressive subclone of oral squamous cell carcinoma.

Because cancer cells have a genetically unstable nature, they give rise to genetically different variant subclones inside a single tumor. Understanding cancer heterogeneity and subclone characteristics is crucial for developing more efficacious therapies. Oral squamous cell carcinoma (OSCC) is characterized by high heterogeneity and plasticity. On the other hand, CX3C motif ligand 1 (CX3CL1) is a double-faced chemokine with anti- and pro-tumor functions. Our study reported that CX3CL1 functioned differently in tumors with different cancer phenotypes, both in vivo and in vitro. Mouse OSCC 1 (MOC1) and MOC2 cells responded similarly to CX3CL1 in vitro. However, in vivo, CX3CL1 increased keratinization in indolent MOC1 cancer, while CX3CL1 promoted cervical lymphatic metastasis in aggressive MOC2 cancer. These outcomes were due to double-faced CX3CL1 effects on different immune microenvironments indolent and aggressive cancer created. Furthermore, we established that CX3CL1 promoted cancer metastasis via the lymphatic pathway by stimulating lymphangiogenesis and transendothelial migration of lymph-circulating tumor cells. CX3CL1 enrichment in lymphatic metastasis tissues was observed in aggressive murine and human cell lines. OSCC patient samples with CX3CL1 enrichment exhibited a strong correlation with lower overall survival rates and higher recurrence and distant metastasis rates. In conclusion, CX3CL1 is a pivotal factor that stimulates the metastasis of aggressive cancer subclones within the heterogeneous tumors to metastasize, and our study demonstrates the prognostic value of CX3CL1 enrichment in long-term monitoring in OSCC.

Effectiveness of deep learning classifiers in histopathological diagnosis of oral squamous cell carcinoma by pathologists.

The study aims to identify histological classifiers from histopathological images of oral squamous cell carcinoma using convolutional neural network (CNN) deep learning models and shows how the results can improve diagnosis. Histopathological samples of oral squamous cell carcinoma were prepared by oral pathologists. Images were divided into tiles on a virtual slide, and labels (squamous cell carcinoma, normal, and others) were applied. VGG16 and ResNet50 with the optimizers stochastic gradient descent with momentum and spectral angle mapper (SAM) were used, with and without a learning rate scheduler. The conditions for achieving good CNN performances were identified by examining performance metrics. We used ROCAUC to statistically evaluate diagnostic performance improvement of six oral pathologists using the results from the selected CNN model for assisted diagnosis. VGG16 with SAM showed the best performance, with accuracy = 0.8622 and AUC = 0.9602. The diagnostic performances of the oral pathologists statistically significantly improved when the diagnostic results of the deep learning model were used as supplementary diagnoses (p-value = 0.031). By considering the learning results of deep learning model classifiers, the diagnostic accuracy of pathologists can be improved. This study contributes to the application of highly reliable deep learning models for oral pathological diagnosis.

The Role of Hedgehog Signaling in the Melanoma Tumor Bone Microenvironment.

A crucial regulator in melanoma progression and treatment resistance is tumor microenvironments, and Hedgehog (Hh) signals activated in a tumor bone microenvironment are a potential new therapeutic target. The mechanism of bone destruction by melanomas involving Hh/Gli signaling in such a tumor microenvironment is unknown. Here, we analyzed surgically resected oral malignant melanoma specimens and observed that Sonic Hedgehog, Gli1, and Gli2 were highly expressed in tumor cells, vasculatures, and osteoclasts. We established a tumor bone destruction mouse model by inoculating B16 cells into the bone marrow space of the right tibial metaphysis of 5-week-old female C57BL mice. An intraperitoneal administration of GANT61 (40 mg/kg), a small-molecule inhibitor of Gli1 and Gli2, resulted in significant inhibition of cortical bone destruction, TRAP-positive osteoclasts within the cortical bone, and endomucin-positive tumor vessels. The gene set enrichment analysis suggested that genes involved in apoptosis, angiogenesis, and the PD-L1 expression pathway in cancer were significantly altered by the GANT61 treatment. A flow cytometry analysis revealed that PD-L1 expression was significantly decreased in cells in which late apoptosis was induced by the GANT61 treatment. These results suggest that molecular targeting of Gli1 and Gli2 may release immunosuppression of the tumor bone microenvironment through normalization of abnormal angiogenesis and bone remodeling in advanced melanoma with jaw bone invasion.

Enzyme-Cleaved Bone Marrow Transplantation Improves the Engraftment of Bone Marrow Mesenchymal Stem Cells.

Mesenchymal stem cell (MSC) therapy is a promising approach to curing bone diseases and disorders. In treating genetic bone disorders, MSC therapy is local or systemic transplantation of isolated and in vitro proliferated MSC rather than bone marrow transplantation. Recent evidence showed that bone marrow MSC engraftment to bone regeneration has been controversial in animal and human studies. Here, our modified bone marrow transplantation (BMT) method solved this problem. Like routine BMT, our modified method involves three steps: (i) isolation of bone marrow cells from the donor, (ii) whole-body lethal irradiation to the recipient, and (iii) injection of isolated bone marrow cells into irradiated recipient mice via the tail vein. The significant modification is imported at the bone marrow isolation step. While the bone marrow cells are flushed out from the bone marrow with the medium in routine BMT, we applied the enzymes' (collagenase type 4 and dispase) integrated medium to wash out the bone marrow cells. Then, cells were incubated in enzyme integrated solution at 37°C for 10 minutes. This modification designated BMT as collagenase-integrated BMT (c-BMT). Notably, successful engraftment of bone marrow MSC to the new bone formation, such as osteoblasts and chondrocytes, occurs in c-BMT mice, whereas routine BMT mice do not recruit bone marrow MSC. Indeed, flow cytometry data showed that c-BMT includes a higher proportion of LepR+, CD51+, or RUNX2+ non-hematopoietic cells than BMT. These findings suggested that c-BMT is a time-efficient and more reliable technique that ensures the disaggregation and collection of bone marrow stem cells and engraftment of bone marrow MSC to the recipient. Hence, we proposed that c-BMT might be a promising approach to curing genetic bone disorders. © 2023 The Authors. JBMR Plus published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research.

Novel Artificial Scaffold for Bone Marrow Regeneration: Honeycomb Tricalcium Phosphate.

Bone marrow is complex structure containing heterogenetic cells, making it difficult to regenerate using artificial scaffolds. In a previous study, we succeeded in developing honeycomb tricalcium phosphate (TCP), which is a cylindrical scaffold with a honeycomb arrangement of straight pores, and we demonstrated that TCP with 300 and 500 µm pore diameters (300TCP and 500TCP) induced bone marrow structure within the pores. In this study, we examined the optimal scaffold structure for bone marrow with homeostatic bone metabolism using honeycomb TCP. 300TCP and 500TCP were transplanted into rat muscle, and bone marrow formation was histologically assessed. Immunohistochemistry for CD45, CD34, Runt-related transcription factor 2 (Runx2), c-kit single staining, Runx2/N-cadherin, and c-kit/Tie-2 double staining was performed. The area of bone marrow structure, which includes CD45(+) round-shaped hematopoietic cells and CD34(+) sinusoidal vessels, was larger in 300TCP than in 500TCP. Additionally, Runx2(+) osteoblasts and c-kit(+) hematopoietic stem cells were observed on the surface of bone tissue formed within TCP. Among Runx2(+) osteoblasts, spindle-shaped N-cadherin(+) cells existed in association with c-kit(+)Tie-2(+) hematopoietic stem cells on the bone tissue formed within TCP, which formed a hematopoietic stem cell niche similar to as in vivo. Therefore, honeycomb TCP with 300 µm pore diameters may be an artificial scaffold with an optimal geometric structure as a scaffold for bone marrow formation.

Polymeric Materials, Advances and Applications in Tissue Engineering: A Review.

Tissue Engineering (TE) is an interdisciplinary field that encompasses materials science in combination with biological and engineering sciences. In recent years, an increase in the demand for therapeutic strategies for improving quality of life has necessitated innovative approaches to designing intelligent biomaterials aimed at the regeneration of tissues and organs. Polymeric porous scaffolds play a critical role in TE strategies for providing a favorable environment for tissue restoration and establishing the interaction of the biomaterial with cells and inducing substances. This article reviewed the various polymeric scaffold materials and their production techniques, as well as the basic elements and principles of TE. Several interesting strategies in eight main TE application areas of epithelial, bone, uterine, vascular, nerve, cartilaginous, cardiac, and urinary tissue were included with the aim of learning about current approaches in TE. Different polymer-based medical devices approved for use in clinical trials and a wide variety of polymeric biomaterials are currently available as commercial products. However, there still are obstacles that limit the clinical translation of TE implants for use wide in humans, and much research work is still needed in the field of regenerative medicine.

SOD3 Expression in Tumor Stroma Provides the Tumor Vessel Maturity in Oral Squamous Cell Carcinoma.

Tumor angiogenesis is one of the hallmarks of solid tumor development. The progressive tumor cells produce the angiogenic factors and promote tumor angiogenesis. However, how the tumor stromal cells influence tumor vascularization is still unclear. In the present study, we evaluated the effects of oral squamous cell carcinoma (OSCC) stromal cells on tumor vascularization. The tumor stromal cells were isolated from two OSCC patients with different subtypes: low invasive verrucous squamous carcinoma (VSCC) and highly invasive squamous cell carcinoma (SCC) and co-xenografted with the human OSCC cell line (HSC-2) on nude mice. In comparison, the CD34+ vessels in HSC-2+VSCC were larger than in HSC-2+SCC. Interestingly, the vessels in the HSC-2+VSCC expressed vascular endothelial cadherin (VE-cadherin), indicating well-formed vascularization. Our microarray data revealed that the expression of extracellular superoxide dismutase, SOD3 mRNA is higher in VSCC stromal cells than in SCC stromal cells. Moreover, we observed that SOD3 colocalized with VE-cadherin on endothelial cells of low invasive stroma xenograft. These data suggested that SOD3 expression in stromal cells may potentially regulate tumor vascularization in OSCC. Thus, our study suggests the potential interest in SOD3-related vascular integrity for a better OSCC therapeutic strategy.

Investigation of bone invasion and underlying mechanisms of oral cancer using a cell line-derived xenograft model.

The cancer stroma regulates bone invasion in oral squamous cell carcinoma (OSCC). However, data on normal stroma are limited. In the present study, the effects of gingival and periodontal ligament tissue-derived stromal cells (G-SCs and P-SCs, respectively) and human dermal fibroblasts (HDFs) on bone resorption and osteoclast activation were assessed using hematoxylin and eosin and tartrate-resistant acid phosphatase staining in a cell line-derived xenograft model. The results demonstrated that G-SCs promoted bone invasion and osteoclast activation and inhibited osteoclast proliferation following crosstalk with the human OSCC HSC-3 cell line, whereas P-SCs inhibited bone resorption and promoted osteoclast proliferation in vitro but had a minimal effect on osteoclast activation both in vitro and in vivo following crosstalk with HSC-3 cells. Furthermore, the effects of G-SCs, P-SCs and HDFs on protein expression levels of matrix metalloproteinase (MMP)-9, membrane type 1 MMP (MT1-MMP), Snail, parathyroid hormone-related peptide (PTHrP) and receptor activator of NF-κB ligand (RANKL) in HSC-3 cells in OSCC bone invasion regions were assessed using immunohistochemistry. The results demonstrated that G-SCs had a more prominent effect on the expression of MMP-9, MT1-MMP, Snail, PTHrP, and RANKL, whereas P-SCs only promoted RANKL and PTHrP expression and exerted a minimal effect on MMP-9, MT1-MMP and Snail expression. The potential genes underlying the differential effects of G-SCs and P-SCs on bone invasion in OSCC were evaluated using a microarray, which indicated that cyclin-dependent kinase 1, insulin, aurora kinase A, cyclin B1 and DNA topoisomerase II alpha underlaid these differential effects. Therefore, these results demonstrated that G-SCs promoted bone invasion in OSCC by activating osteoclasts on the bone surface, whereas P-SCs exerted an inhibitory effect. These findings could indicate a potential regulatory mechanism for bone invasion in OSCC.

Deep learning model for analyzing the relationship between mandibular third molar and inferior alveolar nerve in panoramic radiography.

In this study, the accuracy of the positional relationship of the contact between the inferior alveolar canal and mandibular third molar was evaluated using deep learning. In contact analysis, we investigated the diagnostic performance of the presence or absence of contact between the mandibular third molar and inferior alveolar canal. We also evaluated the diagnostic performance of bone continuity diagnosed based on computed tomography as a continuity analysis. A dataset of 1279 images of mandibular third molars from digital radiographs taken at the Department of Oral and Maxillofacial Surgery at a general hospital (2014-2021) was used for the validation. The deep learning models were ResNet50 and ResNet50v2, with stochastic gradient descent and sharpness-aware minimization (SAM) as optimizers. The performance metrics were accuracy, precision, recall, specificity, F1 score, and area under the receiver operating characteristic curve (AUC). The results indicated that ResNet50v2 using SAM performed excellently in the contact and continuity analyses. The accuracy and AUC were 0.860 and 0.890 for the contact analyses and 0.766 and 0.843 for the continuity analyses. In the contact analysis, SAM and the deep learning model performed effectively. However, in the continuity analysis, none of the deep learning models demonstrated significant classification performance.

Effective deep learning for oral exfoliative cytology classification.

The use of sharpness aware minimization (SAM) as an optimizer that achieves high performance for convolutional neural networks (CNNs) is attracting attention in various fields of deep learning. We used deep learning to perform classification diagnosis in oral exfoliative cytology and to analyze performance, using SAM as an optimization algorithm to improve classification accuracy. The whole image of the oral exfoliation cytology slide was cut into tiles and labeled by an oral pathologist. CNN was VGG16, and stochastic gradient descent (SGD) and SAM were used as optimizers. Each was analyzed with and without a learning rate scheduler in 300 epochs. The performance metrics used were accuracy, precision, recall, specificity, F1 score, AUC, and statistical and effect size. All optimizers performed better with the rate scheduler. In particular, the SAM effect size had high accuracy (11.2) and AUC (11.0). SAM had the best performance of all models with a learning rate scheduler. (AUC = 0.9328) SAM tended to suppress overfitting compared to SGD. In oral exfoliation cytology classification, CNNs using SAM rate scheduler showed the highest classification performance. These results suggest that SAM can play an important role in primary screening of the oral cytological diagnostic environment.

Incidence and Risk of Anti-Resorptive Agent-Related Osteonecrosis of the Jaw after Tooth Extraction: A Retrospective Study.

Bone-modifying agents (BMA) such as bisphosphonates and denosumab are frequently used for the treatment of bone metastases, osteoporosis, and multiple myeloma. BMA may lead to anti-resorptive agent-related osteonecrosis of the jaw (ARONJ). This study aimed to clarify the risk factors for and probabilities of developing ARONJ after tooth extraction in patients undergoing BMA therapy. In this study, the records of 505 target sites of 302 patients undergoing BMA who presented with mandibular fractures at the Department of Oral and Maxillofacial Surgery, Kagawa Prefectural Central Hospital, from March 2014 to January 2022, were retrospectively analyzed for the onset of ARONJ after tooth extraction. The following variables were investigated as attributes: anatomy, health status, and dental treatment. The correlation coefficient was calculated for the success or failure of endodontic surgery for each variable, the odds ratio was calculated for the upper variable, and the factors related to the onset of ARONJ were identified. The incidence rate of ARONJ was found to be 3.2%. Hypoparathyroidism was an important factor associated with ARONJ development. Thus, systemic factors are more strongly related to the onset of ARONJ after tooth extraction than local factors.

The Effectiveness of Pre-Operative Screening Tests in Determining Viral Infections in Patients Undergoing Oral and Maxillofacial Surgery.

We analyzed the rate of patients with hepatitis B virus (HBV), hepatitis C virus (HCV), or human immunodeficiency virus (HIV) infection diagnosed by pre-operative screening and estimated its cost. We retrospectively analyzed patients who underwent elective surgery at our maxillofacial surgery department between April 2014 and March 2022. We compared the number of patients with each infection identified by pre-operative screening and a pre-operative questionnaire. We also compared the prevalence of infections with varying age, sex, and oral diseases, and calculated the cost of screening per positive result. The prevalence of HBV, HCV, and HIV was 0.39% (62/15,842), 0.76% (153/15,839), and 0.07% (10/12,745), respectively. The self-reported rates were as follows: HBV, 63.4% (26/41); HCV, 50.4% (62/123); HIV, 87.5% (7/8). Differences in sex were statistically significant for all infectious diseases; age significantly affected HBV and HCV rates. There was no association between the odds ratio of oral disease and viral infections. The cost per positive result was $1873.8, $905.8, and $11,895.3 for HBV, HCV, and HIV, respectively. Although self-assessment using questionnaires is partially effective, it has inadequate screening accuracy. Formulating an auxiliary diagnosis of infectious diseases with oral diseases was challenging. The cost determined was useful for hepatitis, but not HIV.

Is attention branch network effective in classifying dental implants from panoramic radiograph images by deep learning?

Attention mechanism, which is a means of determining which part of the forced data is emphasized, has attracted attention in various fields of deep learning in recent years. The purpose of this study was to evaluate the performance of the attention branch network (ABN) for implant classification using convolutional neural networks (CNNs). The data consisted of 10191 dental implant images from 13 implant brands that cropped the site, including dental implants as pretreatment, from digital panoramic radiographs of patients who underwent surgery at Kagawa Prefectural Central Hospital between 2005 and 2021. ResNet 18, 50, and 152 were evaluated as CNN models that were compared with and without the ABN. We used accuracy, precision, recall, specificity, F1 score, and area under the receiver operating characteristics curve as performance metrics. We also performed statistical and effect size evaluations of the 30-time performance metrics of the simple CNNs and the ABN model. ResNet18 with ABN significantly improved the dental implant classification performance for all the performance metrics. Effect sizes were equivalent to "Huge" for all performance metrics. In contrast, the classification performance of ResNet50 and 152 deteriorated by adding the attention mechanism. ResNet18 showed considerably high compatibility with the ABN model in dental implant classification (AUC = 0.9993) despite the small number of parameters. The limitation of this study is that only ResNet was verified as a CNN; further studies are required for other CNN models.

Crosstalk between cancer and different cancer stroma subtypes promotes the infiltration of tumor­associated macrophages into the tumor microenvironment of oral squamous cell carcinoma.

Tumor­associated macrophages (TAMs) are linked to the progression of numerous types of cancer. However, the effects of the tumor microenvironment (TME) of oral squamous cell carcinoma (OSCC), particularly the cancer stroma on TAMs, remains to be elucidated. In the present study, the effects of verrucous SCC­associated stromal cells (VSCC­SCs), SCC­associated stromal cells (SCC­SCs) and human dermal fibroblasts (HDFs) on the differentiation, proliferation and migration of macrophages in vitro was assayed using Giemsa staining, and immunofluorescence, MTS and Transwell (migration) assays, respectively. The combined results suggested that both VSCC­SCs and SCC­SCs promoted the differentiation of macrophages into M2 type TAMs, as well as the proliferation and migration of macrophages following crosstalk with HSC­3 cells in vitro. Moreover, the SCC­SCs exerted a more prominent effect on TAMs than the VSCC­SCs. Immunohistochemical staining was used to examine the expression of CD34, CD45, CD11b and CD163 to assay the effects of VSCC­SCs, SCC­SCs and HDFs on microvessel density (MVD) and the infiltration of CD45(+) monocytes, CD11b(+) TAMs and CD163(+) M2 type macrophages. The results suggested that both VSCC­SCs and SCC­SCs promoted MVD and the infiltration of CD45(+) monocytes, CD11b(+) TAMs and CD163(+) M2 type TAMs into the TME of OSCC following crosstalk with HSC­3 cells in vivo. The SCC­SCs exerted a more prominent promoting effect than the VSCC­SCs. Finally, the potential genes underlying the differential effects of VSCC­SCs and SCC­SCs on the infiltration of TAMs were investigated using microarray analysis. The results revealed that interleukin 1ß, bone morphogenetic protein 4, interleukin 6 and C­X­C motif chemokine ligand 12 had great potential to mediate the differential effects of VSCC­SCs and SCC­SCs on TAM infiltration. On the whole, the findings presented herein, demonstrate that both VSCC­SCs and SCC­SCs promote the infiltration of TAMs into the TME of OSCC following crosstalk with HSC­3 cells; the SCC­SCs were found to exert a more prominent promoting effect. This may represent a potential regulatory mechanism for the infiltration of TAMs into the TME of OSCC.

Prognostic Factors in Endodontic Surgery Using an Endoscope: A 1 Year Retrospective Cohort Study.

This retrospective study clarified the success rate of endoscopic endodontic surgeries and identified predictors accounting for successful surgeries. In this retrospective study, 242 patients (90 males, 152 females) who underwent endoscopic endodontic surgery at a single general hospital and were diagnosed through follow-up one year later were included. Risk factors were categorized into attributes, general health, anatomy, and surgery. Then, the correlation coefficient was calculated for the success or failure of endodontic surgery for each variable, the odds ratio was calculated for the upper variable, and factors related to the surgical prognosis factor were identified. The success rate of endodontic surgery was 95.3%, showing that it was a highly predictable treatment. The top three correlation coefficients were post, age, and perilesional sclerotic signs. Among them, the presence of posts was the highest, compared with the odds ratio, which was 9.592. This retrospective study revealed the success rate and risk factors accounting for endoscopic endodontic surgeries. Among the selected clinical variables, the presence of posts was the most decisive risk factor determining the success of endodontic surgeries.

Identification of osteoporosis using ensemble deep learning model with panoramic radiographs and clinical covariates.

Osteoporosis is becoming a global health issue due to increased life expectancy. However, it is difficult to detect in its early stages owing to a lack of discernible symptoms. Hence, screening for osteoporosis with widely used dental panoramic radiographs would be very cost-effective and useful. In this study, we investigate the use of deep learning to classify osteoporosis from dental panoramic radiographs. In addition, the effect of adding clinical covariate data to the radiographic images on the identification performance was assessed. For objective labeling, a dataset containing 778 images was collected from patients who underwent both skeletal-bone-mineral density measurement and dental panoramic radiography at a single general hospital between 2014 and 2020. Osteoporosis was assessed from the dental panoramic radiographs using convolutional neural network (CNN) models, including EfficientNet-b0, -b3, and -b7 and ResNet-18, -50, and -152. An ensemble model was also constructed with clinical covariates added to each CNN. The ensemble model exhibited improved performance on all metrics for all CNNs, especially accuracy and AUC. The results show that deep learning using CNN can accurately classify osteoporosis from dental panoramic radiographs. Furthermore, it was shown that the accuracy can be improved using an ensemble model with patient covariates.

Significance of cancer stroma for bone destruction in oral squamous cell carcinoma using different cancer stroma subtypes.

Stromal cells in the tumor microenvironment (TME) can regulate the progression of numerous types of cancer; however, the bone invasion of oral squamous cell carcinoma (OSCC) has been poorly investigated. In the present study, the effect of verrucous SCC­associated stromal cells (VSCC­SCs), SCC­associated stromal cells (SCC­SCs) and human dermal fibroblasts on bone resorption and the activation of HSC­3 osteoclasts in vivo were examined by hematoxylin and eosin, AE1/3 (pan­cytokeratin) and tartrate­resistant acid phosphatase staining. In addition, the expression levels of matrix metalloproteinase (MMP)9, membrane­type 1 MMP (MT1­MMP), Snail, receptor activator of NF­κB ligand (RANKL) and parathyroid hormone­related peptide (PTHrP) in the bone invasion regions of HSC­3 cells were examined by immunohistochemistry. The results suggested that both SCC­SCs and VSCC­SCs promoted bone resorption, the activation of osteoclasts, and the expression levels of MMP9, MT1­MMP, Snail, RANKL and PTHrP. However, SCC­SCs had a more prominent effect compared with VSCC­SCs. Finally, microarray data were used to predict potential genes underlying the differential effects of VSCC­SCs and SCC­SCs on bone invasion in OSCC. The results revealed that IL1B, ICAM1, FOS, CXCL12, INS and NGF may underlie these differential effects. In conclusion, both VSCC­SCs and SCC­SCs may promote bone invasion in OSCC by enhancing the expression levels of RANKL in cancer and stromal cells mediated by PTHrP; however, SCC­SCs had a more prominent effect. These findings may represent a potential regulatory mechanism underlying the bone invasion of OSCC.

Evaluation of multi-task learning in deep learning-based positioning classification of mandibular third molars.

Pell and Gregory, and Winter's classifications are frequently implemented to classify the mandibular third molars and are crucial for safe tooth extraction. This study aimed to evaluate the classification accuracy of convolutional neural network (CNN) deep learning models using cropped panoramic radiographs based on these classifications. We compared the diagnostic accuracy of single-task and multi-task learning after labeling 1330 images of mandibular third molars from digital radiographs taken at the Department of Oral and Maxillofacial Surgery at a general hospital (2014-2021). The mandibular third molar classifications were analyzed using a VGG 16 model of a CNN. We statistically evaluated performance metrics [accuracy, precision, recall, F1 score, and area under the curve (AUC)] for each prediction. We found that single-task learning was superior to multi-task learning (all p < 0.05) for all metrics, with large effect sizes and low p-values. Recall and F1 scores for position classification showed medium effect sizes in single and multi-task learning. To our knowledge, this is the first deep learning study to examine single-task and multi-task learning for the classification of mandibular third molars. Our results demonstrated the efficacy of implementing Pell and Gregory, and Winter's classifications for specific respective tasks.

Resident stroma-secreted chemokine CCL2 governs myeloid-derived suppressor cells in the tumor microenvironment.

Accumulating evidence has shown that cancer stroma and BM-derived cells (BMDCs) in the tumor microenvironment (TME) play vital roles in tumor progression. However, the mechanism by which oral cancer stroma recruits any particular subset of BMDCs remains largely unknown. Here, we sought to identify the subset of BMDCs that is recruited by cancer stroma. We established a sequential transplantation model in BALB/c nude mice, including (a) BM transplantation of GFP-expressing cells and (b) coxenografting of patient-derived stroma (PDS; 2 cases, designated PDS1 and PDS2) with oral cancer cells (HSC-2). As controls, xenografting was performed with HSC-2 alone or in combination with normal human dermal fibroblasts (HDF). PDS1, PDS2, and HDF all promoted BMDC migration in vitro and recruitment in vivo. Multicolor immunofluorescence revealed that the PDS coxenografts recruited Arginase-1+CD11b+GR1+GFP+ cells, which are myeloid-derived suppressor cells (MDSCs), to the TME, whereas the HDF coxenograft did not. Screening using microarrays revealed that PDS1 and PDS2 expressed CCL2 mRNA (encoding C-C motif chemokine ligand 2) at higher levels than did HDF. Indeed, PDS xenografts contained significantly higher proportions of CCL2+ stromal cells and CCR2+Arginase-1+CD11b+GR1+ MDSCs (as receiver cells) than the HDF coxenograft. Consistently, a CCL2 synthesis inhibitor and a CCR2 antagonist significantly inhibited the PDS-driven migration of BM cells in vitro. Furthermore, i.p. injection of the CCR2 antagonist to the PDS xenograft models significantly reduced the CCR2+Arginase-1+CD11b+GR1+ MDSC infiltration to the TME. In conclusion, oral cancer stroma-secreted CCL2 is a key signal for recruiting CCR2+ MDSCs from BM to the TME.