Google goes cancer: improving outcome prediction for cancer patients by network-based ranking of marker genes

Predicting the clinical outcome of cancer patients based on the expression of marker genes in their tumors has received increasing interest in the past decade. Accurate predictors of outcome and response to therapy could be used to personalize and thereby improve therapy. However, state of the art methods used so far often found marker genes with limited prediction accuracy, limited reproducibility, and unclear biological relevance. To address this problem, we developed a novel computational approach to identify genes prognostic for outcome that couples gene expression measurements from primary tumor samples with a network of known relationships between the genes. Our approach ranks genes according to their prognostic relevance using both expression and network information in a manner similar to Google's PageRank. We applied this method to gene expression profiles which we obtained from 30 patients with pancreatic cancer, and identified seven candidate marker genes prognostic for outcome. Compared to genes found with state of the art methods, such as Pearson correlation of gene expression with survival time, we improve the prediction accuracy by up to 7%. Accuracies were assessed using support vector machine classifiers and Monte Carlo cross-validation. We then validated the prognostic value of our seven candidate markers using immunohistochemistry on an independent set of 412 pancreatic cancer samples. Notably, signatures derived from our candidate markers were independently predictive of outcome and superior to established clinical prognostic factors such as grade, tumor size, and nodal status. As the amount of genomic data of individual tumors grows rapidly, our algorithm meets the need for powerful computational approaches that are key to exploit these data for personalized cancer therapies in clinical practice.     

Positional cloning of a novel Fanconi anemia gene, FANCD2

Fanconi anemia (FA) is a genetic disease with birth defects, bone marrow failure, and cancer susceptibility. To date, genes for five of the seven known complementation groups have been cloned. Complementation group D is heterogeneous, consisting of two distinct genes, FANCD1 and FANCD2. Here we report the positional cloning of FANCD2. The gene consists of 44 exons, encodes a novel 1451 amino acid nuclear protein, and has two protein isoforms. Similar to other FA proteins, the FANCD2 protein has no known functional domains, but unlike other known FA genes, FANCD2 is highly conserved in A. thaliana, C. elegans, and Drosophila. Retroviral transduction of the cloned FANCD2 cDNA into FA-D2 cells resulted in functional complementation of MMC sensitivity.     

Alternative intronic promoters in development and disease

Approximately 20,000 mammalian genes are estimated to encode between 250 thousand and 1 million different proteins. This enormous diversity of the mammalian proteome is caused by the ability of a single-gene locus to encode multiple protein isoforms. Protein isoforms encoded by one gene locus can be functionally distinct, and they can even have antagonistic functions. One of the mechanisms involved in creating this proteome complexity is alternative promoter usage. Alternative intronic promoters are located downstream from their canonical counterparts and drive the expression of alternative RNA isoforms that lack upstream exons. These upstream exons can encode some important functional domains, and proteins encoded by alternative mRNA isoforms can be thus functionally distinct from the full-length protein encoded by canonical mRNA isoforms. Since any misbalance of functionally distinct protein isoforms is likely to have detrimental consequences for the cell and the whole organism, their expression must be precisely regulated. Misregulation of alternative intronic promoters is frequently associated with various developmental defects and diseases including cancer, and it is becoming increasingly clear that this phenomenon deserves more attention.     

Metallothionein as a prognostic biomarker in breast cancer

Breast cancer is the most common cancer in women, with a general upward trend in incidence. Basic and clinical breast cancer research has continued at a rapid pace, in the endeavor to understand the biology of the disease so as to improve management of patients. Besides traditional pathological indicators, expression of molecular markers in breast cancer has also been comprehensively investigated. This paper will focus on the prognostic utility of metallothioneins (MTs), a family of low molecular weight metal binding proteins encoded by at least 10 functional MT genes that are associated with cell proliferation in breast cancer. Evidence that MT is a potential prognostic biomarker for breast cancer is supported by many reports in the literature. Expression of the MT protein has been detected by immunohistochemistry in a significant portion of invasive ductal breast cancers. MT expression has also been well studied in association with traditional clinico-pathological parameters of breast cancers. Generally, higher MT expression in breast cancers is predictive of worse patient outcomes. The relationship of MT isoforms to histological grade, estrogen receptor (ER) status, and prognosis will also be discussed.     

A single-nucleotide polymorphism of SMARCB1 in human breast cancers

The gene SMARCB1 has been considered a candidate for a tumor-suppressor gene. Nucleotide alterations in SMARCB1 have been reported, primarily in association with malignant rhabdoid tumor cases. We carried out a search for mutations in SMARCB1 in 60 human gastro-intestinal tract carcinoma cases, 122 breast cancer cases, and 36 human cancer cell lines. A single-nucleotide polymorphism (SNP) at codon 152 with an amino acid change (Asn to Asp) was found in 2 of 122 (1.6%) breast cancer cases, and another SNP at codon 299 without an amino acid change was found in tumor and normal tissues from 7 (5.7%) cases. Codons 152 and 299 of SMARCB1 are localized near or within the binding site for the cMYC protein. The amount of immunoprecipitated cMYC protein was reduced in two different cell lines expressing the codon 152 polymorphic SMARCB1 clone compared with those expressing wild-type SMARCB1, regardless of the identical expression of SMARCB1 protein in both cell lines. Therefore, the SNP at codon 152 is considered to be one of the coding SNPs that alters the SMARCB1-cMYC complex, which regulates various tumor-suppressor related genes against cancer. In addition, we identified three types of splicing isoforms, a 27-bp deleted gene, a 51-bp inserted gene, and a consensus gene, in both carcinoma tissues and in normal tissues; however, no clinical significance was observed for those isoforms. We found a nucleotide change at codon 152 of SMARCB1 that may alter the amount of immunoprecipitated cMYC protein, but we finally determined that SMARCB1 is highly conserved in human solid carcinomas.     

?Np73 is capable of inducing apoptosis by co-ordinately activating several BH3-only proteins

Inactivation of p53 is one of the most relevant events in human cancer, since it allows transformed cells to escape their own proliferation control and leave them irresponsive to drugs that aim to damage their DNA. When p53 falls, other members of its family may become targets to attack tumoural cells. p73 has shown capacity to mediate these attacks. However, its N-terminal truncated isoforms have been associated with oncogenesis due to their capacity to act as dominant negatives of p53 and the transactivation (TA) isoforms of p73. We previously found a relationship between the overexpression of N-terminus-truncated p73 isoform (∆Np73) and that of the proapoptotic gene Bcl-2-interacting killer (BIK). In the present report we demonstrate that ∆Np73-α has the capacity to induce apoptosis through the co-ordinated activation of a group of genes harbouring GC-rich elements in their regulatory regions. ∆Np73-α synergizes with specificity protein (Sp1) on these elements but the overall response of these genes probably depends on the additional presence of consensus p53 elements. We explore the domains of ∆Np73-α involved in this transactivation capacity and found divergences with the previously described functions for them. Moreover, we found that the transforming mutation V12 of HRas impairs this transactivation capacity of ∆Np73-α, further supporting the anti-tumoural function of this later. Our data add complexity to the action of p73 on the induction of apoptosis and tumourogenesis, opening new interpretations to the expression profile of p73 isoforms in different human neoplasias.     

Alternative promoters in expression of genetic information

Many genes are known to have several promoters. The contribution of alternative promoters to the structural and functional diversity of protein isoforms in eukaryotic cells is considered, including their role in synthesis of identical proteins from different mRNAs, generation of protein isoforms with different and even opposite functions, expression of housekeeping genes, and the variation of the recognition domains of adhesion proteins and receptors. In some cases, alternative promoters allow one gene to produce mRNAs with different open reading frames and, consequently, proteins with no amino acid sequence homology.     

The identification of six risk genes for ovarian cancer platinum response based on global network algorithm and verification analysis

Ovarian cancer is the most lethal gynaecological cancer, and resistance of platinum‐based chemotherapy is the main reason for treatment failure. The aim of the present study was to identify candidate genes involved in ovarian cancer platinum response by analysing genes from homologous recombination and Fanconi anaemia pathways. Associations between these two functional genes were explored in the study, and we performed a random walk algorithm based on reconstructed gene‐gene network, including protein‐protein interaction and co‐expression relations. Following the random walk, all genes were ranked and GSEA analysis showed that the biological functions focused primarily on autophagy, histone modification and gluconeogenesis. Based on three types of seed nodes, the top two genes were utilized as examples. We selected a total of six candidate genes (FANCA, FANCG, POLD1, KDM1A, BLM and BRCA1) for subsequent verification. The validation results of the six candidate genes have significance in three independent ovarian cancer data sets with platinum‐resistant and platinum‐sensitive information. To explore the correlation between biomarkers and clinical prognostic factors, we performed differential analysis and multivariate clinical subgroup analysis for six candidate genes at both mRNA and protein levels. And each of the six candidate genes and their neighbouring genes with a mutation rate greater than 10% were also analysed by network construction and functional enrichment analysis. In the meanwhile, the survival analysis for platinum‐treated patients was performed in the current study. Finally, the RT‐qPCR assay was used to determine the performance of candidate genes in ovarian cancer platinum response. Taken together, this research demonstrated that comprehensive bioinformatics methods could help to understand the molecular mechanism of platinum response and provide new strategies for overcoming platinum resistance in ovarian cancer treatment.     

Prognostic value of tumour microenvironment-related genes by TCGA database in rectal cancer

Rectal cancer is a common malignant tumour and the progression is highly affected by the tumour microenvironment (TME). This study intended to assess the relationship between TME and prognosis, and explore prognostic genes of rectal cancer. The gene expression profile of rectal cancer was obtained from TCGA and immune/stromal scores were calculated by Estimation of Stromal and Immune cells in Malignant Tumors using Expression data (ESTIMATE) algorithm. The correlation between immune/stromal scores and survival time as well as clinical characteristics were evaluated. Differentially expressed genes (DEGs) were identified according to the stromal/immune scores, and the functional enrichment analyses were conducted to explore functions and pathways of DEGs. The survival analyses were conducted to clarify the DEGs with prognostic value, and the protein-protein interaction (PPI) network was performed to explore the interrelation of prognostic DEGs. Finally, we validated prognostic DEGs using data from the Gene Expression Omnibus (GEO) database by PrognoScan, and we verified these genes at the protein levels using the Human Protein Atlas (HPA) databases. We downloaded gene expression profiles of 83 rectal cancer patients from The Cancer Genome Atlas (TCGA) database. The Kaplan-Meier plot demonstrated that low-immune score was associated with worse clinical outcome (P = .034), metastasis (M1 vs. M0, P = .031) and lymphatic invasion (+ vs. -, P < .001). A total of 540 genes were screened as DEGs with 539 up-regulated genes and 1 down-regulated gene. In addition, 60 DEGs were identified associated with overall survival. Functional enrichment analyses and PPI networks showed that the DEGs are mainly participated in immune process, and cytokine-cytokine receptor interaction. Finally, 19 prognostic genes were verified by GSE17536 and GSE17537 from GEO, and five genes (ADAM23, ARHGAP20, ICOS, IRF4, MMRN1) were significantly different in tumour tissues compared with normal tissues at the protein level. In summary, our study demonstrated the associations between TME and prognosis as well as clinical characteristics of rectal cancer. Moreover, we explored and verified microenvironment-related genes, which may be the potential key prognostic genes of rectal cancer. Further clinical samples and functional studies are needed to validate this finding.     

Development of an eight gene expression profile implicating human breast tumours of all grade

The goal of improving systemic treatment of breast cancers is to evolve from treating every patient with non-specific cytotoxic chemotherapy/hormonal therapy, to a more individually-tailored direct treatment. Although anatomic staging and histological grade are important prognostic factors, they often fail to predict the clinical course of this disease. This study aimed to develop a gene expression profile associated with breast cancers of differing grades. We extracted mRNA from FFPE archival breast IDC tissue samples (Grades I–III), including benign tumours. Affymetrix GeneChip^® Human Genome U133 Plus 2.0 Arrays were used to determine gene expression profiles and validated by Q-PCR. IHC was used to detect the AXIN2 protein in all tissues. From the array data, an independent group t-test revealed that 178 genes were significantly (P ≤ 0.01) differentially expressed between three grades of malignant breast tumours when compared to benign tissues. From these results, eight genes were significantly differentially expressed in more than one comparison group and are involved in processes implicated in breast cancer development and/or progression. The two most implicated candidates genes were CLD10 and ESPTI1 as their gene expression profile from the microarray analysis was replicated in Q-PCR analyses of the original tumour samples as well as in an extended population. The IHC revealed a significant association between AXIN2 protein expression and ER status. It is readily acknowledged and established that significant differences exist in gene expression between different cancer grades. Expansion of this approach may lead to an improved ability to discriminate between cancer grade and other pathological factors.     

A correlation between BCL-2 modifying factor,p53 and livin gene expressions in cancer colon patients

Accumulating evidence has revealed that livin gene and BCL-2 modifying factor (BMF) gene are closely associated with the initiation and progression of colon carcinoma by activating or suppressing multiple malignant processes. Those genes that can detect colon - cancer are a promising approach for cancer screening and diagnosis. This study aimed to evaluate correlation between livin, BMF and p53 genes expression in colon cancer tissues of patients included in the study, and their relationship with clinicopathological features and survival outcome in those patients. In this study, 50 pathologically diagnosed early cancer colon patients included and their tissue biopsy with 50 matched adjacent normal tissue, and 50 adenoma tissue specimens were analyzed for livin gene and BMF gene expressions using real time PCR. The relationship of those genes expressions with clinicopathological features, tumor markers, Time to Progression and overall survival for those patients were correlated in cancer colon group. In this study, there was a significant a reciprocal relationship between over expression of livin gene and down regulation of BMF and p53 genes in colon cancer cells. Livin mRNA was significantly higher, while BMF and p53 mRNA were significantly lower in colorectal cancer tissue compared to benign and normal colon tissue specimens (P < 0.001), however, this finding was absent between colon adenomas and normal mucosa. There was a significant association between up regulation of livin and down regulation of BMF and p53 expressions with more aggressive tumor (advanced TNM stage), rapid progression with metastasis and decreased overall survival in cancer colon patients, hence these genes can serve as significant prognostic markers of poor outcome in colon cancer patients. This work highlights the role of livin, BMF and p53 genes in colorectal tumorigenesis and the applicability of using those genes as a diagnostic and prognostic markers in patients with colon carcinoma and as a good target for cancer colon treatment in the future.     

Gene expression profiling and clinical outcome in breast cancer

Pathologic and clinical heterogeneity of breast cancer reflects the poorly documented, complex, and combinatory molecular basis of the disease and is in part responsible for therapeutic failures. The DNA microarray technique allows the analysis of RNA expression of several thousands of genes simultaneously in a sample. There are multiple potential applications of the technique in cancer research. A number of recent studies have shown the promising role of gene expression profiling in breast cancer by identifying new prognostic subclasses unidentifiable by conventional parameters and new prognostic and/or predictive gene signatures, whose predictive impact is superior to conventional histoclinical prognostic factors. In this review we describe current use of DNA microarrays in the prognosis of breast cancer. We also discuss issues that need to be addressed in the near future to allow the method to reach its full potential.     

A Meta Analysis of Pancreatic Microarray Datasets Yields New Targets as Cancer Genes and Biomarkers

The lack of specific symptoms at early tumor stages, together with a high biological aggressiveness of the tumor contribute to the high mortality rate for pancreatic cancer (PC), which has a five year survival rate of less than 5%. Improved screening for earlier diagnosis, through the detection of diagnostic and prognostic biomarkers provides the best hope of increasing the rate of curatively resectable carcinomas. Though many serum markers have been reported to be elevated in patients with PC, so far, most of these markers have not been implemented into clinical routine due to low sensitivity or specificity. In this study, we have identified genes that are significantly upregulated in PC, through a meta-analysis of large number of microarray datasets. We demonstrate that the biological functions ascribed to these genes are clearly associated with PC and metastasis, and that that these genes exhibit a strong link to pathways involved with inflammation and the immune response. This investigation has yielded new targets for cancer genes, and potential biomarkers for pancreatic cancer. The candidate list of cancer genes includes protein kinase genes, new members of gene families currently associated with PC, as well as genes not previously linked to PC. In this study, we are also able to move towards developing a signature for hypomethylated genes, which could be useful for early detection of PC. We also show that the significantly upregulated 800+ genes in our analysis can serve as an enriched pool for tissue and serum protein biomarkers in pancreatic cancer.