Predicting cervical lymph node metastasis in OSCC based on computed tomography imaging genomics.

BACKGROUND: To investigate the correlation between computed tomography (CT) radiomic characteristics and key genes for cervical lymph node metastasis (LNM) in oral squamous cell carcinoma (OSCC). METHODS: The region of interest was annotated at the edge of the primary tumor on enhanced CT images from 140 patients with OSCC and obtained radiomic features. Ribonucleic acid (RNA) sequencing was performed on pathological sections from 20 patients. the DESeq software package was used to compare differential gene expression between groups. Weighted gene co-expression network analysis was used to construct co-expressed gene modules, and the KEGG and GO databases were used for pathway enrichment analysis of key gene modules. Finally, Pearson correlation coefficients were calculated between key genes of enriched pathways and radiomic features. RESULTS: Four hundred and eighty radiomic features were extracted from enhanced CT images of 140 patients; seven of these correlated significantly with cervical LNM in OSCC (p < 0.01). A total of 3527 differentially expressed RNAs were screened from RNA sequencing data of 20 cases. original_glrlm_RunVariance showed significant positive correlation with most long noncoding RNAs. CONCLUSIONS: OSCC cervical LNM is related to the salivary hair bump signaling pathway and biological process. Original_glrlm_RunVariance correlated with LNM and most differentially expressed long noncoding RNAs.

Prediction of 5-year overall survival of tongue cancer based machine learning.

OBJECTIVE: We aimed to develop a 5-year overall survival prediction model for patients with oral tongue squamous cell carcinoma based on machine learning methods. SUBJECTS AND METHODS: The data were obtained from electronic medical records of 224 OTSCC patients at the PLA General Hospital. A five-year overall survival prediction model was constructed using logistic regression, Support Vector Machines, Decision Tree, Random Forest, Extreme Gradient Boosting, and Light Gradient Boosting Machine. Model performance was evaluated according to the area under the curve (AUC) of the receiver operating characteristic curve. The output of the optimal model was explained using the Python package (SHapley Additive exPlanations, SHAP). RESULTS: After passing through the grid search and secondary modeling, the Light Gradient Boosting Machine was the best prediction model (AUC = 0.860). As explained by SHapley Additive exPlanations, N-stage, age, systemic inflammation response index, positive lymph nodes, plasma fibrinogen, lymphocyte-to-monocyte ratio, neutrophil percentage, and T-stage could perform a 5-year overall survival prediction for OTSCC. The 5-year survival rate was 42%. CONCLUSION: The Light Gradient Boosting Machine prediction model predicted 5-year overall survival in OTSCC patients, and this predictive tool has potential prognostic implications for patients with OTSCC.

Predictive value of systemic inflammatory response-related indices for survival in tongue cancer.

OBJECTIVES: We assessed the potential prognostic significance of the preoperative systemic inflammation index, platelet-to-lymphocyte ratio, and neutrophil-to-lymphocyte ratio in patients who underwent surgical resection. SUBJECTS AND METHODS: This retrospective study included 224 patients with clinicopathologically confirmed squamous carcinoma of the tongue who underwent surgery between August 2009 and December 2017. The optimal cut-off values for the indices were determined by receiver operating characteristic curves. Correlations between the indices and clinicopathological variables were determined by Pearson chi-square or Fisher exact tests. The Kaplan-Meier test was used to compare overall survival between groups (high and low values); the log-rank or Breslow test was used to assess differences in survival. Univariate and multivariate Cox regression models were used to analyze predictive values of the indices as independent indicators of overall survival. Bilateral p values of <0.05 were considered statistically significant. RESULTS: Significant association was found between the indices and sex, tissue grade, tumor location, and lymph nodes metastases (p < 0.05). On Kaplan-Meier analysis, patients with lower values of the indices had longer overall survival (p < 0.05). Univariate and multivariate Cox regression models identified age, lymph node metastases, and neutrophil-to-lymphocyte ratio as independent predictors of overall survival. CONCLUSION: The studied indices have potential prognostic significance in patients with squamous tongue cancer.

Neoadjuvant immunoradiotherapy in patients with locally advanced oral cavity squamous cell carcinoma: a retrospective study.

BACKGROUND: Although anti-programmed death receptor-1 (PD-1) agents have been evaluated in the neoadjuvant setting for the treatment of locally advanced head and neck cancer, including oral cavity squamous cell carcinoma (OCSCC), the overall response rate is modest. The aim of the present study was to evaluate the efficacy and safety of neoadjuvant nivolumab in combination with stereotactic body radiotherapy (SBRT) for the treatment of locally advanced OCSCC. METHODS: OCSCC patients who underwent surgical resection within 6 months of treatment with nivolumab plus SBRT from December 2018 to February 2021 were analyzed retrospectively. RESULTS: All 30 eligible patients enrolled in this study well tolerated the neoadjuvant treatment with no serious adverse events (AEs). Of them, 27 patients (90.0%) achieved R0 resection, and 5 patients (16.7%) experienced procedure-associated complications. The complete response (CR), partial response (PR) and stable disease (SD) were 10.0%, 46.7% and 43.3% respectively. The major pathological response (MPR), complete pathological response (pCR) and clinical to pathological downstaging rate were 60.0%, 33.3% and 83.3% respectively. During the median follow-up period of 13.5 months, 26 patients (86.7%) who underwent surgical resection remained alive. The disease-free survival (DFS) and overall survival (OS) at 24 months were 70.4% and 76.4% respectively. CONCLUSIONS: Neoadjuvant nivolumab plus SBRT is safe and efficacious, and could be used as a potential neoadjuvant option for the treatment of patients with locally advanced OCSCC.

Clinical Response to Neoadjuvant Immunotherapy Combined with Targeted Therapy and Chemotherapy in Oral Squamous Cell Carcinoma: Experience in Three Patients.

Immune checkpoint inhibitors (ICIs) pembrolizumab and nivolumab have been approved for the treatment of head and neck squamous cell carcinoma (HNSCC) and used in neoadjuvant immunotherapy in clinical trials. However, combination of ICIs with targeted therapy and chemotherapy was rarely used in pre-surgical HNSCC patients. Herein, we encountered three cases of patients with oral squamous cell carcinoma (OSCC) who all had good responses to neoadjuvant immunotherapy (anti-PD-1 inhibitors) combined with nimotuzumab (anti-EGFR monoclonal antibody) plus paclitaxel. Both Case 1 and Case 2 underwent the same neoadjuvant therapeutic combination (nivolumab, nimotuzumab and paclitaxel) and exhibited a marked tumor shrinkage even complete disappearance by radiological evaluation. Moreover, pathological response was observed in post-surgical tissues of Case 1. Additionally, Case 3 with tongue squamous cell carcinoma also had satisfactory tumor regression (complete healing of his tongue ulcer upon treatment) after receiving similar neoadjuvant therapy with sintilimab (another PD-1 inhibitor), nimotuzumab and paclitaxel. We characterized their potential causes behind favorable treatment outcomes. While there were differences in driver mutations and tumor mutation burden (TMB) identified in pre-treatment tumor tissues among the three patients, numerous CD68+ (macrophages) infiltrates were common for all the cases. Of note, the majority (>80%) of the total macrophages were molecularly defined as PD-L1-positive macrophages. Given the high expression of PD-L1 in macrophages is associated with better immunotherapy outcomes, we propose that the high proportion of CD68+PD-L1+ cells in total macrophages alone could serve as a promising biomarker for neoadjuvant immunotherapy in combination with other therapies in HNSCC.

A deep learning model using convolutional neural networks for caries detection and recognition with endoscopes.

Background: Caries are common, especially in economically undeveloped countries with limited access to medical resources. Sometimes patient cannot even realize that they have oral problems until they feel obvious pain. Deep convolutional neural networks (CNNs) have been widely adopted for medical image analysis and management and have yielded some progress in stomatology while the endoscopes are cheap and easily used in daily life for families or other non-medical situations. Therefore, we created a deep learning model to detect and recognize caries using endoscopic images. Methods: We used 194 images of non-caries and 1,059 images of permanent molar and premolar caries to build a classification and a segmentation model in patients of endoscope images from the Department of Stomatology of People's Liberation Army General Hospital (PLAGH). A classification model combined with an end-to-end semantic segmentation model, DeepLabv3+ was used for segmenting the caries, then we evaluated with a 5-fold cross-validation protocol whereby each fold was used once. Results: In the classification model, the mean area under the curve (AUC) [90% confidence interval (CI)] was 0.9897 (0.9821-0.9956) (P<0.01) In the segmentation model, the mean accuracy was 0.9843 (0.9820-0.9871), the recall was 0.6996 (0.6810-0.7194), the specificity was 0.9943 (0.9937-0.9954), the Dice coefficient was 0.7099 (0.6948-0.7343), and the intersection over union (IoU) was 0.5779 (0.5646-0.6006). Conclusions: We used a deep learning model to monitor caries and encourage their early diagnosis and treatment.

[Expression and clinical significance of BTG-1 in 78 patients with oral squamous cell carcinoma].

PURPOSE: To investigate the expression and clinical significance of B cell translocation gene 1 (BTG1) in oral squamous cell carcinoma(OSCC). METHODS: Immunohistochemical staining was used to detect the expression of BTG1 protein in 78 cases of OSCC tissues, 78 adjacent tissues, 20 normal oral mucosa tissues, and 80 cervical lymph nodes. Western blot and real-time quantitative PCR were used to detect BTG1 protein and mRNA expression levels in 78 OSCC tissues and adjacent tissues. Kaplan-Meier survival analysis, Log rank test and Cox regression analysis were performed with SPSS 21.0 software package. RESULTS: The expression level of BTG1 in OSCC and cervical positive lymph nodes was significantly lower than that in normal tissues and negative lymph nodes adjacent to the cancer, and the expression of BTG1 in poorly differentiated OSCC was significantly lower than that in highly differentiated OSCC (P<0.05); Survival analysis showed the progression-free survival (PFS) and overall survival time(OS) of BTG1 low-expression group were significantly lower than those of high-expression group (P<0.05). Cox regression analysis showed that the degree of tumor differentiation, cervical lymph node metastasis, and BTG1 expression all affected patients' prognosis. CONCLUSIONS: BTG1 is lowerly expressed in OSCC, with expression related to TNM stage and differentiation of OSCC but no relation with gender, age, and tumor location, including that BTG1 may be involved in the occurrence, development and prognosis of OSCC.