Assessing the clinical utility of genomic expression data across human cancers

Integrating Somatic Mutations for Breast Cancer Survival Prediction Using Machine Learning Methods

Breast cancer is the most common malignancy in women, and because it has a high mortality rate, it is urgent to develop computational methods to increase the accuracy of breast cancer survival predictive models. Although multi-omics data such as gene expression have been extensively used in recent studies, the accurate prognosis of breast cancer remains a challenge. Somatic mutations are another important and promising data source for studying cancer development, and its effect on the prognosis of breast cancer remains to be further explored. Meanwhile, these omics datasets are high-dimensional and redundant. Therefore, we adopted multiple kernel learning (MKL) to efficiently integrate somatic mutation to currently molecular data including gene expression, copy number variation (CNV), methylation, and protein expression data for the prediction of breast cancer survival. Before integration, the maximum relevance minimum redundancy (mRMR) feature selection method was utilized to select features that present high relevance to survival and low redundancy among themselves for each type of data. The experimental results demonstrated that the proposed method achieved the most optimal performance and there was a remarkable improvement in the prediction performance when somatic mutations were included, indicating that somatic mutations are critical for improving breast cancer survival predictions. Moreover, mRMR was superior to other feature selection methods used in previous studies. Furthermore, MKL outperformed the other traditional classifiers in multi-omics data integration. Our analysis indicated that through employing promising omics data such as somatic mutations and harnessing the power of proper feature selection methods and effective integration frameworks, the breast cancer survival predictive accuracy can be further increased, thereby providing a more optimal clinical diagnosis and more effective treatment for breast cancer patients.

Potential relationship between the selenoproteome and cancer

The role of the selenoproteome, which is the collection of all proteins containing selenium in an organism, in cancer development, growth and progression requires further investigation, due to the importance of selenium in both cancer and immune system function. Data about the selenoproteome, including its differential expression, single nucleotide variations, copy number variations, methylation, pathways and overall survival (OS) in five leading types of cancer are available from the GSCALite website. Subsequent to the analysis of these datasets, it was revealed that there was increased expression of GPX3 in stomach adenocarcinoma and lung squamous cell carcinoma, SELENOV in oesophageal carcinoma, GPX8 and GPX4 in colon adenocarcinoma, TXNRD1 and SEPHS1 in hepatocellular carcinoma and GPX8 in lung adenocarcinoma were associated with poor survival. Decreased gene expression of SELENOP was indicated in liver hepatocellular carcinoma and GPX3, and SELENOW, SELENOK, SELENBP1 and SECISBP2 in lung adenocarcinoma were associated with a poor prognosis. OS data suggested that hypermethylation of GPX4 in colon adenocarcinoma, GPX8 in lung squamous cell carcinoma, GPX1 in stomach adenocarcinoma and GPX3 in lung adenocarcinoma was associated with low survival, as is hypomethylation of GPX5 in lung adenocarcinoma. The selenoproteome is heterogeneous, especially in its effect on the OS of patients with cancer. The present study demonstrated that the roles of GPX4 in colon adenocarcinoma, SCLY and SELENOV in oesophageal carcinoma, SEPHS1 in liver hepatocellular carcinoma, SELENOK in lung cancer, as well as SELENOM and SELENOW in stomach adenocarcinoma requires further research. The present study may lead to the identification of novel biomarkers or potential therapeutic targets for use in the treatment of cancers, such as colon adenocarcinoma, oesophageal carcinoma, liver hepatocellular carcinoma, lung cancer and stomach adenocarcinoma.

Stratification of Breast Cancer by Integrating Gene Expression Data and Clinical Variables

Breast cancer is a heterogeneous disease. Although gene expression profiling has led to the definition of several subtypes of breast cancer, the precise discovery of the subtypes remains a challenge. Clinical data is another promising source. In this study, clinical variables are utilized and integrated to gene expressions for the stratification of breast cancer. We adopt two phases: gene selection and clustering, where the integration is in the gene selection phase; only genes whose expressions are most relevant to each clinical variable and least redundant among themselves are selected for further clustering. In practice, we simply utilize maximum relevance minimum redundancy (mRMR) for gene selection and k-means for clustering. We compare the results of our method with those of two commonly used only expression-based breast cancer stratification methods: prediction analysis of microarray 50 (PAM50) and highest variability (HV). The result is that our method outperforms them in identifying subtypes significantly associated with five-year survival and recurrence time. Specifically, our method identified recurrence-associated breast cancer subtypes that were not identified by PAM50 and HV. Additionally, our analysis discovered three survival-associated luminal-A subgroups and two survival-associated luminal-B subgroups. The study indicates that screening clinically relevant gene expressions yields improved breast cancer stratification.

Comprehensive analysis of the tumor immune micro-environment in non-small cell lung cancer for efficacy of checkpoint inhibitor

Characterizing the molecular immune subtype and micro-environment of lung cancer is necessary to understand immunogenic interactions between infiltrating immune and stromal cells, and how tumor cells overcome immune checkpoint blockades. This study seeks to identify computational methodologies for subtyping gene expression-based tumor-immune micro-environment interactions, which differentiate non-small cell lung cancer (NSCLC) into immune-defective and immune-competent subtypes. Here, 101 lung squamous cell carcinomas (LUSCs) and 87 lung adenocarcinomas (LUADs) tumor samples have been analyzed. Several micro-environmental factors differentially induce LUAD or LUSC immune subtypes, as well as immune checkpoint expression. In particular, tumor-associated macrophages (TAMs) are key immune cells play a vital role in inflammation and cancer micro-environments of LUSCs; whereas, regulatory B cells are immunosuppressive and tumorigenic in LUADs. Additionally, cytolytic activity upon CD8⁺ T cell activation is decreased by the abundance of B cells and macrophages in immune-competent subtypes. Therefore, identifying immune subtypes in lung cancer and their impact on tumor micro-environment will lead to clinical tools for assessing LUADs and LUSCs in patients, as well as maximize the efficacy of immune checkpoint inhibitors.

Developing a genetic signature to predict drug response in ovarian cancer

There is a lack of personalized treatment options for women with recurrent platinum-resistant ovarian cancer. Outside of bevacizumab and a group of poly ADP-ribose polymerase inhibitors, few options are available to women that relapse. We propose that efficacious drug combinations can be determined via molecular characterization of ovarian tumors along with pre-established pharmacogenomic profiles of repurposed compounds. To that end, we selectively performed multiple two-drug combination treatments in ovarian cancer cell lines that included reactive oxygen species inducers and HSP90 inhibitors. This allowed us to select cell lines that exhibit disparate phenotypes of proliferative inhibition to a specific drug combination of auranofin and AUY922. We profiled altered mechanistic responses from these agents in both reactive oxygen species and HSP90 pathways, as well as investigated PRKCI and lncRNA expression in ovarian cancer cell line models. Generation of dual multi-gene panels implicated in resistance or sensitivity to this drug combination was produced using RNA sequencing data and the validity of the resistant signature was examined using high-density RT-qPCR. Finally, data mining for the prevalence of these signatures in a large-scale clinical study alluded to the prevalence of resistant genes in ovarian tumor biology. Our results demonstrate that high-throughput viability screens paired with reliable in silico data can promote the discovery of effective, personalized therapeutic options for a currently untreatable disease.

The correlation analysis of miRNAs and target genes in metastasis of cervical squamous cell carcinoma

Aim: This study was intended to identify the metastasis-related miRNAs and target genes in cervical squamous cell carcinoma. Materials & methods: The mRNA and miRNA next-generation sequencing data were downloaded. Differential expression analysis was carried out, followed by target gene prediction of differentially expressed miRNAs. The biological function of differentially expressed genes was performed. Validation was carried out by survival analysis and qRT-PCR. Results: N4BP3 were associated with the survival time of patients. Hsa-mir-451 and hsa-mir-486 were related to tumor differentiation stage. Validated expression of hsa-mir-24–2, hsa-mir-582, NOTCH1, PIP4K2B, DIP2B and IGFBP5 was consistent with the bioinformatics analysis. Conclusion: Alterations of miRNAs and target genes may be useful in understanding the metastasis mechanisms of cervical squamous cell carcinoma.