Gene Expression Signatures Identify Novel Therapeutics for Metastatic Pancreatic Neuroendocrine Tumors.

PURPOSE: Pancreatic neuroendocrine tumors (pNETs) are uncommon malignancies noted for their propensity to metastasize and comparatively favorable prognosis. Although both the treatment options and clinical outcomes have improved in the past decades, most patients will die of metastatic disease. New systemic therapies are needed. EXPERIMENTAL DESIGN: Tissues were obtained from 43 patients with well-differentiated pNETs undergoing surgery. Gene expression was compared between primary tumors versus liver and lymph node metastases using RNA-Seq. Genes that were selectively elevated at only one metastatic site were filtered out to reduce tissue-specific effects. Ingenuity pathway analysis (IPA) and the Connectivity Map (CMap) identified drugs likely to antagonize metastasis-specific targets. The biological activity of top identified agents was tested in vitro using two pNET cell lines (BON-1 and QGP-1). RESULTS: A total of 902 genes were differentially expressed in pNET metastases compared with primary tumors, 626 of which remained in the common metastatic profile after filtering. Analysis with IPA and CMap revealed altered activity of factors involved in survival and proliferation, and identified drugs targeting those pathways, including inhibitors of mTOR, PI3K, MEK, TOP2A, protein kinase C, NF-kB, cyclin-dependent kinase, and histone deacetylase. Inhibitors of MEK and TOP2A were consistently the most active compounds. CONCLUSIONS: We employed a complementary bioinformatics approach to identify novel therapeutics for pNETs by analyzing gene expression in metastatic tumors. The potential utility of these drugs was confirmed by in vitro cytotoxicity assays, suggesting drugs targeting MEK and TOP2A may be highly efficacious against metastatic pNETs. This is a promising strategy for discovering more effective treatments for patients with pNETs.

Machine learning with the TCGA-HNSC dataset: improving usability by addressing inconsistency, sparsity, and high-dimensionality.

BACKGROUND: In the era of precision oncology and publicly available datasets, the amount of information available for each patient case has dramatically increased. From clinical variables and PET-CT radiomics measures to DNA-variant and RNA expression profiles, such a wide variety of data presents a multitude of challenges. Large clinical datasets are subject to sparsely and/or inconsistently populated fields. Corresponding sequencing profiles can suffer from the problem of high-dimensionality, where making useful inferences can be difficult without correspondingly large numbers of instances. In this paper we report a novel deployment of machine learning techniques to handle data sparsity and high dimensionality, while evaluating potential biomarkers in the form of unsupervised transformations of RNA data. We apply preprocessing, MICE imputation, and sparse principal component analysis (SPCA) to improve the usability of more than 500 patient cases from the TCGA-HNSC dataset for enhancing future oncological decision support for Head and Neck Squamous Cell Carcinoma (HNSCC). RESULTS: Imputation was shown to improve prognostic ability of sparse clinical treatment variables. SPCA transformation of RNA expression variables reduced runtime for RNA-based models, though changes to classifier performance were not significant. Gene ontology enrichment analysis of gene sets associated with individual sparse principal components (SPCs) are also reported, showing that both high- and low-importance SPCs were associated with cell death pathways, though the high-importance gene sets were found to be associated with a wider variety of cancer-related biological processes. CONCLUSIONS: MICE imputation allowed us to impute missing values for clinically informative features, improving their overall importance for predicting two-year recurrence-free survival by incorporating variance from other clinical variables. Dimensionality reduction of RNA expression profiles via SPCA reduced both computation cost and model training/evaluation time without affecting classifier performance, allowing researchers to obtain experimental results much more quickly. SPCA simultaneously provided a convenient avenue for consideration of biological context via gene ontology enrichment analysis.

RABL6A inhibits tumor-suppressive PP2A/AKT signaling to drive pancreatic neuroendocrine tumor growth.

Hyperactivated AKT/mTOR signaling is a hallmark of pancreatic neuroendocrine tumors (PNETs). Drugs targeting this pathway are used clinically, but tumor resistance invariably develops. A better understanding of factors regulating AKT/mTOR signaling and PNET pathogenesis is needed to improve current therapies. We discovered that RABL6A, a new oncogenic driver of PNET proliferation, is required for AKT activity. Silencing RABL6A caused PNET cell-cycle arrest that coincided with selective loss of AKT-S473 (not T308) phosphorylation and AKT/mTOR inactivation. Restoration of AKT phosphorylation rescued the G1 phase block triggered by RABL6A silencing. Mechanistically, loss of AKT-S473 phosphorylation in RABL6A-depleted cells was the result of increased protein phosphatase 2A (PP2A) activity. Inhibition of PP2A restored phosphorylation of AKT-S473 in RABL6A-depleted cells, whereas PP2A reactivation using a specific small-molecule activator of PP2A (SMAP) abolished that phosphorylation. Moreover, SMAP treatment effectively killed PNET cells in a RABL6A-dependent manner and suppressed PNET growth in vivo. The present work identifies RABL6A as a new inhibitor of the PP2A tumor suppressor and an essential activator of AKT in PNET cells. Our findings offer what we believe is a novel strategy of PP2A reactivation for treatment of PNETs as well as other human cancers driven by RABL6A overexpression and PP2A inactivation.

Introduction of ultrasound into gross anatomy curriculum: perceptions of medical students.

BACKGROUND: The exposure to ultrasound technology during medical school education is highly variable across institutions. OBJECTIVES: The objectives of this study were to assess medical students' perceptions of ultrasound use to teach Gross Anatomy along with traditional teaching methods, and determine their ability to identify sonographic anatomy after focused didactic sessions. METHODS: Prospective observational study. Phase I of the study included three focused ultrasound didactic sessions integrated into Gross Anatomy curriculum. During Phase II, first-year medical students completed a questionnaire. RESULTS: One hundred nine subjects participated in this study; 96% (95% confidence interval [CI] 92-99%) agreed that ultrasound-based teaching increased students' knowledge of anatomy acquired through traditional teaching methods. Ninety-two percent (95% CI 87-97%) indicated that ultrasound-based teaching increases confidence to perform invasive procedures in the future. Ninety-one percent (95% CI 85-96%) believed that it is feasible to integrate ultrasound into the current Anatomy curriculum. Ninety-eight percent (95% CI 95-100%) of medical students accurately identified vascular structures on ultrasound images of normal anatomy of the neck. On a scale of 1 to 10, the average confidence level reported in interpreting the images was 7.4 (95% CI 7.1-7.7). Overall, 94% (95% CI 91-99%) accurately answered questions about ultrasound fundamentals and sonographic anatomy. CONCLUSIONS: The majority of medical students believed that it is feasible and beneficial to use ultrasound in conjunction with traditional teaching methods to teach Gross Anatomy. Medical students were very accurate in identifying sonographic vascular anatomy of the neck after brief didactic sessions.

Identification and initial characterization of 6,000 expressed sequenced tags (ESTs) from rat normal-growing cartilage and swarm rat chondrosarcoma cDNA libraries.

Chondrosarcoma is the second most common type of skeletal malignancy with a survival rate at five years for histological grade III of only 29 percent. The development of a reliable chondrosarcoma animal model could enable the study of tumor growth and progression, the effect of the host on tumor behavior, and the effectiveness of various therapeutic modalities. The Swarm rat chondrosarcoma is a tumor tissue line that has been maintained through the years by serial subcutaneous injections, and the histochemical characteristics of the tumor have remained essentially similar in all transplants over the years. This study was designed to initiate the characterization of the Swarm rat chondrosarcoma model by gene expression profiling as compared to normal-growing rat cartilage. Analysis of the gene expression from both libraries revealed a complex pattern of gene expression, including many genes not yet reported to be expressed by chondrocytes. It suggests that the biochemical characterization of growing cartilage and chondrosarcoma reported to date has only begun to describe the complexity of these tissues.