Predicting postoperative complications of head and neck squamous cell carcinoma in elderly patients using random forest algorithm model.

BACKGROUND: Head and Neck Squamous Cell Carcinoma (HNSCC) has a high incidence in elderly patients. The postoperative complications present great challenges within treatment and they're hard for early warning. METHODS: Data from 525 patients diagnosed with HNSCC including a training set (n = 513) and an external testing set (n = 12) in our institution between 2006 and 2011 was collected. Variables involved are general demographic characteristics, complications, disease and treatment given. Five data mining algorithms were firstly exploited to construct predictive models in the training set. Subsequently, cross-validation was used to compare the different performance of these models and the best data mining algorithm model was then selected to perform the prediction in an external testing set. RESULTS: Data from 513 patients (age > 60 y) with HNSCC in a training set was included while 44 variables were selected (P < 0.05). Five predictive models were constructed; the model with 44 variables based on the Random Forest algorithm demonstrated the best accuracy (89.084%) and the best AUC value (0.949). In an external testing set, the accuracy (83.333%) and the AUC value (0.781) were obtained by using the random forest algorithm model. CONCLUSIONS: Data mining should be a promising approach used for elderly patients with HNSCC to predict the probability of postoperative complications. Our results highlighted the potential of computational prediction of postoperative complications in elderly patients with HNSCC by using the random forest algorithm model.

Opposing functions of PLAG1 in pleomorphic adenoma: a microarray analysis of PLAG1 transgenic mice.

Over-expression of the proto-oncogene pleomorphic adenoma gene 1 (PLAG1) plays a crucial role in the formation of pleomorphic adenoma, which is the most common type of salivary gland tumor. To understand the molecular mechanisms governing PLAG1-mediated tumorigenesis, we used a microarray-based approach to identify PLAG1 target genes. We validated the expression of several genes, including Bax, Fas, p53, p21, p16, Cyclin D1, Egfr, Trail-R/DR5, c-Fos, c-myc and Igf2, by real-time RT-PCR or western blotting. Using luciferase reporter gene assays, we determined that the promoters of Bax, Fas, p53, TRAIL-R/DR5, and c-Fos were transactivated by PLAG1. PLAG1 not only activates genes that promote cell proliferation and tumor formation but also genes that inhibit these cellular processes. Therefore, we conclude that PLAG1 may play a dual role in tumor formation.

Metastatic parotid gland malignancy: A preliminary study in an eastern Chinese population.

OBJECTIVE: This study aimed to determine the incidence and clinicopathological patterns of metastatic carcinoma of the parotid gland. METHOD: Ninety patients with parotid gland metastases admitted to our hospital between January 2003 and December 2018 were included in this study. Clinical and pathological data were obtained from the medical records and follow-ups. Kaplan-Meier analysis was used to assess overall survival of patients. RESULTS: Among the 90 patients, parotid gland metastases originated from the head and neck in 86 (95.6%), from non-head and neck in 4 (4.4%), from the oral cavity in 30(33.3%), and from the eyelid in 21 (23.3%). Among the 85 cases with parotid gland lymph node metastasis, 45 (52.9%) were diagnosed with extra-lymph node metastasis. The capsule of the parotid lymph nodes was thinner than that of the cervical lymph nodes (P < 0.05). Hematogenous metastases to the parotid gland (only five cases) were rare, mainly from the non-head and neck malignancies. Patients with oral squamous cell carcinoma and meibomian adenocarcinoma with parotid metastatic disease had poorer overall survival (P < 0.05). CONCLUSION: Eastern China population analysis showed that parotid gland metastases usually arise from oral squamous cell carcinoma and eyelid, but rarely from cutaneous squamous cell carcinoma. Most cases metastasize to the parotid lymph nodes via the lymphatic system and are prone to extranodal extension with little or no facial nerve involvement. These findings have important implications for the treatment of metastatic parotid malignancies.

Lymph node involvement predicts poor prognosis in primary tongue adenoid cystic carcinoma: A preliminary study of 54 cases.

PURPOSE: Primary tongue adenoid cystic carcinoma (ACC) is extremely rare. The relationship between the patient's prognosis and the tumor's clinicopathological characteristics is uncertain. The aim of this study was to identify the prognostic factors and analyze the overall outcomes for patients with tongue ACC. PATIENTS AND METHODS: A retrospective analysis of the medical records of patients diagnosed with primary tongue ACC between 1998 and 2008 was conducted. The study comprised 54 patients, with a median age of 53.4 years. The incidence of ACC in the base of the tongue was almost 3.5 times in comparison to that in the mobile portion of the tongue. The 3-year and 5-year overall survival rates were 87.4% and 65.3%, respectively, whereas the 3-year and 5-year disease-free survival rates were 54.3% and 32.8%, respectively. Significant prognostic factors regarding overall survival in univariate analyses included tumor size and lymph node status. RESULTS: A multivariate model identified lymph node status as the only significant independent predictor of overall survival. Further analysis showed that tumor size (P = 0.034), histological grade (P = 0.021), and perineural invasion (PNI, P = 0.023) were three important factors affecting lymph node metastasis. Patients with lymph node metastasis have a higher tendency of developing distant metastasis (P < 0.01). CONCLUSIONS: Lymph node status was a significant determinant of prognosis in primary tongue ACC and could be used for a rational design of treatment strategies in future.

In Situ Release of VEGF Enhances Osteogenesis in 3D Porous Scaffolds Engineered with Osterix-Modified Adipose-Derived Stem Cells.

Adipose-derived stem cells (ADSCs) can differentiate into various cell types and thus have great potential for regenerative medicine. Herein, rat ADSCs were isolated; transduced with lentiviruses expressing Osterix (Osx), a transcriptional factor essential for osteogenesis. Osx overexpression upregulated key osteogenesis-related genes, such as special AT-rich binding protein 2, alkaline phosphatase, osteocalcin, and osteopontin, at both mRNA and protein levels. In addition, mineral nodule formation and alkaline phosphatase activity were enhanced in Osx-overexpressing ADSCs. The expression of dickkopf-related protein 1, a potent Wnt signaling pathway inhibitor, was also increased, whereas that of ß-catenin, an intracellular signal transducer in the Wnt pathway, was decreased. ß-catenin expression was partially recovered by treatment with lithium chloride, a canonical Wnt pathway activator. The Osx-expressing ADSCs were then combined with 3D gelatin-coated porous poly(ɛ-caprolactone) scaffolds with a unique release prolife of entrapped recombinant human vascular endothelial growth factor (VEGF). The controlled release of VEGF promoted osteogenic differentiation capacity in vitro. When the scaffold-ADSC complexes were transplanted into rat calvarial critical-sized defects, more bone formed on the gelatin/VEGF-coated scaffolds than on other scaffold types. Taken together, the results indicate that, Osx-overexpression promotes ADSCs' osteogenesis both in vitro and in vivo, which could be enhanced by release of VEGF.