SignS: a parallelized, open-source, freely available, web-based tool for gene selection and molecular signatures for survival and censored data

TGF-β-induced IGFBP-3 is a key paracrine factor from activated pericytes that promotes colorectal cancer cell migration and invasion

The crosstalk between cancer cells and the tumor microenvironment has been implicated in cancer progression and metastasis. Fibroblasts and immune cells are widely known to be attracted to and modified by cancer cells. However, the role of pericytes in the tumor microenvironment beyond endothelium stabilization is poorly understood. Here, we report that pericytes promoted colorectal cancer (CRC) cell proliferation, migration, invasion, stemness, and chemoresistance in vitro, as well as tumor growth in a xenograft CRC model. We demonstrate that coculture with human CRC cells induced broad transcriptomic changes in pericytes, mostly associated with TGF-β receptor activation. The prognostic value of a TGF-β response signature in pericytes was analyzed in CRC patient data sets. This signature was found to be a good predictor of CRC relapse. Moreover, in response to stimulation by CRC cells, pericytes expressed high levels of TGF-β1, initiating an autocrine activation loop. Investigation of secreted mediators and underlying molecular mechanisms revealed that IGFBP-3 is a key paracrine factor from activated pericytes affecting CRC cell migration and invasion. In summary, we demonstrate that the interplay between pericytes and CRC cells triggers a vicious cycle that stimulates pericyte cytokine secretion, in turn increasing CRC cell tumorigenic properties. Overall, we provide another example of how cancer cells can manipulate the tumor microenvironment.

Vitamin D receptor expression and associated gene signature in tumour stromal fibroblasts predict clinical outcome in colorectal cancer

OBJECTIVE

Colorectal cancer (CRC) is a major health concern. Vitamin D deficiency is associated with high CRC incidence and mortality, suggesting a protective effect of vitamin D against this disease. Given the strong influence of tumour stroma on cancer progression, we investigated the potential effects of the active vitamin D metabolite 1α,25-dihydroxyvitamin D₃ (1,25(OH)₂D₃) on CRC stroma.

DESIGN

Expression of vitamin D receptor (VDR) and two 1,25(OH)₂D₃ target genes was analysed in 658 patients with CRC with prolonged clinical follow-up. 1,25(OH)₂D₃ effects on primary cultures of patient-derived colon normal fibroblasts (NFs) and cancer-associated fibroblasts (CAFs) were studied using collagen gel contraction and migration assays and global gene expression analyses. Publicly available data sets (n=877) were used to correlate the 1,25(OH)₂D₃-associated gene signature in CAFs with CRC outcome.

RESULTS

High VDR expression in tumour stromal fibroblasts was associated with better overall survival (OS) and progression-free survival in CRC, independently of its expression in carcinoma cells. 1,25(OH)₂D₃ inhibited the protumoural activation of NFs and CAFs and imposed in CAFs a 1,25(OH)₂D₃-associated gene signature that correlated with longer OS and disease-free survival in CRC. Furthermore, expression of two genes from the signature, CD82 and S100A4, correlated with stromal VDR expression and clinical outcome in our cohort of patients with CRC.

CONCLUSIONS

1,25(OH)₂D₃ has protective effects against CRC through the regulation of stromal fibroblasts. Accordingly, expression of VDR and 1,25(OH)₂D₃-associated gene signature in stromal fibroblasts predicts a favourable clinical outcome in CRC. Therefore, treatment of patients with CRC with VDR agonists could be explored even in the absence of VDR expression in carcinoma cells.

Integrative genome-wide gene expression profiling of clear cell renal cell carcinoma in Czech Republic and in the United States

Gene expression microarray and next generation sequencing efforts on conventional, clear cell renal cell carcinoma (ccRCC) have been mostly performed in North American and Western European populations, while the highest incidence rates are found in Central/Eastern Europe. We conducted whole-genome expression profiling on 101 pairs of ccRCC tumours and adjacent non-tumour renal tissue from Czech patients recruited within the "K2 Study", using the Illumina HumanHT-12 v4 Expression BeadChips to explore the molecular variations underlying the biological and clinical heterogeneity of this cancer. Differential expression analysis identified 1650 significant probes (fold change ≥2 and false discovery rate <0.05) mapping to 630 up- and 720 down-regulated unique genes. We performed similar statistical analysis on the RNA sequencing data of 65 ccRCC cases from the Cancer Genome Atlas (TCGA) project and identified 60% (402) of the downregulated and 74% (469) of the upregulated genes found in the K2 series. The biological characterization of the significantly deregulated genes demonstrated involvement of downregulated genes in metabolic and catabolic processes, excretion, oxidation reduction, ion transport and response to chemical stimulus, while simultaneously upregulated genes were associated with immune and inflammatory responses, response to hypoxia, stress, wounding, vasculature development and cell activation. Furthermore, genome-wide DNA methylation analysis of 317 TCGA ccRCC/adjacent non-tumour renal tissue pairs indicated that deregulation of approximately 7% of genes could be explained by epigenetic changes. Finally, survival analysis conducted on 89 K2 and 464 TCGA cases identified 8 genes associated with differential prognostic outcomes. In conclusion, a large proportion of ccRCC molecular characteristics were common to the two populations and several may have clinical implications when validated further through large clinical cohorts.

miRNA expression in diffuse large B-cell lymphoma treated with chemoimmunotherapy

Diffuse large B-cell lymphoma (DLBCL) prognostication requires additional biologic markers. miRNAs may constitute markers for cancer diagnosis, outcome, or therapy response. In the present study, we analyzed the miRNA expression profile in a retrospective multicenter series of 258 DLBCL patients uniformly treated with chemoimmunotherapy. Findings were correlated with overall survival (OS) and progression-free survival (PFS). miRNA and gene-expression profiles were studied using microarrays in an initial set of 36 cases. A selection of miRNAs associated with either DLBCL molecular subtypes (GCB/ABC) or clinical outcome were studied by multiplex RT-PCR in a test group of 240 cases with available formalin-fixed, paraffin-embedded (FFPE) diagnostic samples. The samples were divided into a training set (123 patients) and used to derive miRNA-based and combined (with IPI score) Cox regression models in an independent validation series (117 patients). Our model based on miRNA expression predicts OS and PFS and improves upon the predictions based on clinical variables. Combined models with IPI score identified a high-risk group of patients with a 2-year OS and a PFS probability of < 50%. In summary, a precise miRNA signature is associated with poor clinical outcome in chemoimmunotherapy-treated DLBCL patients. This information improves upon IPI-based predictions and identifies a subgroup of candidate patients for alternative therapeutic regimens.

DNA methylation profiles and their relationship with cytogenetic status in adult acute myeloid leukemia

BACKGROUND

Aberrant promoter DNA methylation has been shown to play a role in acute myeloid leukemia (AML) pathophysiology. However, further studies to discuss the prognostic value and the relationship of the epigenetic signatures with defined genomic rearrangements in acute myeloid leukemia are required.

METHODOLOGY/PRINCIPAL FINDINGS

We carried out high-throughput methylation profiling on 116 de novo AML cases and we validated the significant biomarkers in an independent cohort of 244 AML cases. Methylation signatures were associated with the presence of a specific cytogenetic status. In normal karyotype cases, aberrant methylation of the promoter of DBC1 was validated as a predictor of the disease-free and overall survival. Furthermore, DBC1 expression was significantly silenced in the aberrantly methylated samples. Patients with chromosome rearrangements showed distinct methylation signatures. To establish the role of fusion proteins in the epigenetic profiles, 20 additional samples of human hematopoietic stem/progenitor cells (HSPC) transduced with common fusion genes were studied and compared with patient samples carrying the same rearrangements. The presence of MLL rearrangements in HSPC induced the methylation profile observed in the MLL-positive primary samples. In contrast, fusion genes such as AML1/ETO or CBFB/MYH11 failed to reproduce the epigenetic signature observed in the patients.

CONCLUSIONS/SIGNIFICANCE

Our study provides a comprehensive epigenetic profiling of AML, identifies new clinical markers for cases with a normal karyotype, and reveals relevant biological information related to the role of fusion proteins on the methylation signature.

Mantle cell lymphoma: transcriptional regulation by microRNAs

Mantle cell lymphoma (MCL) pathogenesis is still partially unexplained. We investigate the importance of microRNA (miRNA) expression as an additional feature that influences MCL pathway deregulation and may be useful for predicting patient outcome. Twenty-three MCL samples, eight cell lines and appropriate controls were screened for their miRNAs and gene expression profiles and DNA copy-number changes. MCL patients exhibit a characteristic signature that includes 117 miRNA (false discovery rate <0.05). Combined analysis of miRNAs and the gene expression profile, paired with bioinformatics target prediction (miRBase and TargetScan), revealed a series of genes and pathways potentially targeted by a small number of miRNAs, including essential pathways for lymphoma survival such as CD40, mitogen-activated protein kinase and NF-kappaB. Functional validation in MCL cell lines demonstrated NF-kappaB subunit nuclear translocation to be regulated by the expression of miR-26a. The expression of 12 selected miRNAs was studied by quantitative PCR in an additional series of 54 MCL cases. Univariate analysis identified a single miRNA, miR-20b, whose lack of expression distinguished cases with a survival probability of 56% at 60 months. In summary, using a novel bioinformatics approach, this study identified miRNA changes that contribute to MCL pathogenesis and markers of potential utility in MCL diagnosis and clinical prognostication.

CMA: a comprehensive Bioconductor package for supervised classification with high dimensional data

BACKGROUND

For the last eight years, microarray-based classification has been a major topic in statistics, bioinformatics and biomedicine research. Traditional methods often yield unsatisfactory results or may even be inapplicable in the so-called "p >> n" setting where the number of predictors p by far exceeds the number of observations n, hence the term "ill-posed-problem". Careful model selection and evaluation satisfying accepted good-practice standards is a very complex task for statisticians without experience in this area or for scientists with limited statistical background. The multiplicity of available methods for class prediction based on high-dimensional data is an additional practical challenge for inexperienced researchers.

RESULTS

In this article, we introduce a new Bioconductor package called CMA (standing for "Classification for MicroArrays") for automatically performing variable selection, parameter tuning, classifier construction, and unbiased evaluation of the constructed classifiers using a large number of usual methods. Without much time and effort, users are provided with an overview of the unbiased accuracy of most top-performing classifiers. Furthermore, the standardized evaluation framework underlying CMA can also be beneficial in statistical research for comparison purposes, for instance if a new classifier has to be compared to existing approaches.

CONCLUSION

CMA is a user-friendly comprehensive package for classifier construction and evaluation implementing most usual approaches. It is freely available from the Bioconductor website at (http://bioconductor.org/packages/2.3/bioc/html/CMA.html).

PaLS: filtering common literature, biological terms and pathway information

Many biological experiments and their subsequent analysis yield lists of genes or proteins that can potentially be important to the prognosis or diagnosis of certain diseases (e.g. cancer). Nowadays, information about the function of those genes or proteins may be already gathered in some databases, but it is essential to understand if some of the members of those lists have a function in common or if they belong to the same metabolic pathway. To help researchers filter those genes or proteins that have such information in common, we have developed PaLS (pathway and literature strainer, http://pals.bioinfo.cnio.es). PaLS takes a list or a set of lists of gene or protein identifiers and shows which ones share certain descriptors. Four publicly available databases have been used for this purpose: PubMed, which links genes with those articles that make reference to them; Gene Ontology, an annotated ontology of terms related to the cellular component, biological process or molecular function where those genes or proteins are involved; KEGG pathways and Reactome pathways. Those descriptors among these four sources of information that are shared by more members of the list (or lists) are highlighted by PaLS.