Onco, Volume 2, Issue 2 (June 2022) – 5 articles

Whole genome sequencing (WGS) has helped to revolutionize biology, but the computational challenge remains for extracting valuable inferences from this information. Here, we present the cancer-associated variants from the Cancer Genome Atlas (TCGA) WGS dataset. This set of data will allow cancer researchers to further expand their analysis beyond the exomic regions of the genome to the entire genome. A total of 1342 WGS alignments available from the consortium were processed with VarScan2 and deposited to the NCI Cancer Cloud. The sample set covers 18 different cancers and reveals 157,313,519 pooled (non-unique) cancer-associated single-nucleotide variations (SNVs) across all samples. There was an average of 117,223 SNVs per sample, with a range from 1111 to 775,470 and a standard deviation of 163,273. The dataset was incorporated into BigQuery, which allows for fast access and cross-mapping, which will allow researchers to enrich their current studies with a plethora of newly available genomic data. Full article

Purpose: Glioblastoma is an aggressive and incurable brain tumor whose progression is driven in part by glioblastoma stem cells, which are also responsible for the tumor’s low therapy efficacy. The maintenance and expansion of the stem cell population is promoted by the hypoxic microenvironment, where miRNAs play fundamental roles in their survival. Methods: GBM stem-like cells were isolated from three GBM parental cell lines. The stem-like cells were then cultured under normoxic and hypoxic microenvironments followed by investigation of the in vitro “stemness” of the cells. Results: We found miR-128a-3p, 34-5p and 181a-3p to be downregulated and miR-17-5p and miR-221-3p to be upregulated in our stem-like cells compared to the GBMs. When a comparison was made between normoxia and hypoxia, a further fold downregulation was observed for miR-34-5p, miR-128a-3p and miR-181a-3p and a further upregulation was observed for miR-221-3p and 17-5p. There was an increased expression of HIF-1/2, SOX2, OCT4, VEGF, GLUT-1, BCL2 and survivin under hypoxia. Conclusion: Our data suggest that our GBMs were able to grow as stem-like cells and as spheroids. There was a differential expression of miRNAs between the stems and the GBMs and the hypoxia microenvironment influenced further dysregulation of the miRNAs and some selected genes. Full article

A complex interaction occurs between cancer cells and the extracellular matrix (ECM) in the tumour microenvironment (TME). In this study, the expressions and mutational profiles of 964 ECM-related genes and their correlations with patient overall survival (OS) in neuroblastoma, an aggressive paediatric malignancy, were investigated using cBioPortal and PCAT databases. Furthermore, extended networks comprising protein-protein, protein-long non-coding RNA (lncRNA), and protein-miRNA of 12 selected ECM-related genes were established. The higher expressions of 12 ECM-related genes, AMBN, COLQ, ELFN1, HAS3, HSPE1, LMAN1, LRP5, MUC6, RAMP2, RUVBL2, SSBP1 and UMOD in neuroblastoma patients displayed a significant correlation with patient OS, while similar associations with neuroblastoma patient risk groups, histology and MYCN amplification were obtained. Furthermore, extended gene networks formed by these 12 ECM-related genes were established using Cytoscape, STRING, MSigDB/BioGRID, GeneMANIA and Omicsnet. Finally, the implications of the 12 ECM-related genes in other cancers were revealed using GEPIA2 and the Human Pathology Atlas databases. This meta-analysis showed the significance of these 12 ECM-related genes as putative prognostic predictors in neuroblastoma and other cancers. Full article

Purpose: To study the production of angiodrastic chemokines by colonic cancer cell lines. Methods: A pro-angiogenic factor (VEGF), two angiogenic chemokines (CXCL8, CXCL6), and one angiostatic (CXCL4) chemokine were measured by ELISA in the supernatants of the colon cancer cell lines HT-29 and Caco-2. Cells were cultured for 24 h in the presence of serum from cancer patients or healthy individuals. Results were analyzed by one-way ANOVA and the General Linear Model for repeated measures. Results: Colonic epithelial cells are potent producers of angiodrastic chemokines. HT-29 and Caco-2 cells produce all four chemokines under basal conditions and 24 h after incubation with human serum. The secretion response, however, was completely different. HT-29 cells produce more CXCL8 and VEGF irrespective of culture conditions, while Caco-2 cells seem unresponsive with respect to CXCL6 and CXCL4. Moreover, HT-29 cells produce more CXCL8 and VEGF when incubated with cancer serum, contrary to Caco-2 cells which produce more CXCL4 under the same conditions. Conclusions: The two colon cancer cell lines were producers of all chemokines studied, but their responses were not uniform under similar culture conditions. CXCL8 and VEGF are differently regulated compared to CXCL4 and CXCL6 in these two cell lines Full article

Background and Objective: This paper aimed to differentiate primary cancer types from primary tumor samples on the basis of somatic point mutations (SPMs). Primary cancer site identification is necessary to perform site-specific and potentially targeted treatment. Current methods such as histopathology and lab tests cannot accurately determine cancer origin, which results in empirical patient treatment and poor survival rates. The availability of large deoxyribonucleic acid sequencing datasets has allowed scientists to examine the ability of somatic mutations to classify primary cancer sites. These datasets are highly sparse since most genes will not be mutated, have a low signal-to-noise ratio, and are often imbalanced since rare cancers have fewer samples. Methods: To overcome these limitations a sparse-input neural network (SPINN) is suggested that projects the input data in a lower-dimensional space, where the more informative genes are used for learning. To train and evaluate SPINN, an extensive dataset for SPM was collected from the cancer genome atlas containing 7624 samples spanning 32 cancer types. Different sampling strategies were performed to balance the dataset. SPINN was further validated on an independent ICGC dataset that contained 226 samples spanning four cancer types. Results and Conclusions: SPINN consistently outperformed classification algorithms such as extreme gradient boosting, deep neural networks, and support vector machines, achieving an accuracy up to 73% on independent testing data. Certain primary cancer types/subtypes (e.g., lung, brain, colon, esophagus, skin, and thyroid) were classified with an F-score > 0.80. Full article