Dataset Information

Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context.

ABSTRACT: BACKGROUND: Different microarray studies have compiled gene lists for predicting outcomes of a range of treatments and diseases. These have produced gene lists that have little overlap, indicating that the results from any one study are unstable. It has been suggested that the underlying pathways are essentially identical, and that the expression of gene sets, rather than that of individual genes, may be more informative with respect to prognosis and understanding of the underlying biological process. RESULTS: We sought to examine the stability of prognostic signatures based on gene sets rather than individual genes. We classified breast cancer cases from five microarray studies according to the risk of metastasis, using features derived from predefined gene sets. The expression levels of genes in the sets are aggregated, using what we call a set statistic. The resulting prognostic gene sets were as predictive as the lists of individual genes, but displayed more consistent rankings via bootstrap replications within datasets, produced more stable classifiers across different datasets, and are potentially more interpretable in the biological context since they examine gene expression in the context of their neighbouring genes in the pathway. In addition, we performed this analysis in each breast cancer molecular subtype, based on ER/HER2 status. The prognostic gene sets found in each subtype were consistent with the biology based on previous analysis of individual genes. CONCLUSIONS: To date, most analyses of gene expression data have focused at the level of the individual genes. We show that a complementary approach of examining the data using predefined gene sets can reduce the noise and could provide increased insight into the underlying biological pathways.

SUBMITTER: Abraham G

PROVIDER: S-EPMC2895626 | biostudies-literature | 2010

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context.

Abraham Gad G Kowalczyk Adam A Loi Sherene S Haviv Izhak I Zobel Justin J

BMC bioinformatics 20100525

<h4>Background</h4>Different microarray studies have compiled gene lists for predicting outcomes of a range of treatments and diseases. These have produced gene lists that have little overlap, indicating that the results from any one study are unstable. It has been suggested that the underlying pathways are essentially identical, and that the expression of gene sets, rather than that of individual genes, may be more informative with respect to prognosis and understanding of the underlying biolog ...[more]

PMID: 20500821

Similar Datasets

Project description:BackgroundIn the ongoing battle against breast cancer, a leading cause of cancer-related mortality among women globally, the urgent need for innovative prognostic markers and therapeutic targets is undeniable. This study pioneers an advanced methodology by integrating machine learning techniques to unveil a vascular mimicry signature, offering predictive insights into breast cancer outcomes. Vascular mimicry refers to the phenomenon where cancer cells mimic blood vessel formation absent of endothelial cells, a trait associated with heightened tumor aggression and diminished response to conventional treatments.MethodsThe study's comprehensive analysis spanned data from over 6,000 breast cancer patients across 12 distinct datasets, incorporating both proprietary clinical data and single-cell data from 7 patients, accounting for a total of 43,095 cells. By employing an integrative strategy that utilized 10 machine learning algorithms across 108 unique combinations, the research scrutinized 100 existing breast cancer signatures. Empirical validation was sought through immunohistochemistry assays, alongside explorations into potential immunotherapeutic and chemotherapeutic avenues.ResultsThe investigation successfully identified six genes related to vascular mimicry from multi-center cohorts, laying the groundwork for a novel predictive model. This model outstripped the prognostic accuracy of traditional clinical and molecular indicators in forecasting recurrence and mortality risks. High-risk individuals identified by our model faced worse outcomes. Further validation through IHC assays in 30 patients underscored the model's extensive applicability. Notably, the model unveiled varying therapeutic responses; low-risk patients might achieve greater benefits from immunotherapy, whereas high-risk patients demonstrated a particular sensitivity to certain chemotherapies, such as ispinesib.ConclusionsThis model marks a significant step forward in the precise evaluation of breast cancer prognosis and therapeutic responses across different patient groups. It heralds the possibility of refining patient outcomes through tailored treatment strategies, accentuating the potential of machine learning in revolutionizing cancer prognosis and management.

Project description:The extensive and intricate relationships between circadian rhythm and cancer have been reported in numerous studies. However, in breast cancer (BC), the potential role of circadian clock-related genes (CCRGs) in prognosis prediction has not been fully clarified. The transcriptome data and clinical information were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus databases. A CCRGs-based risk signature was established by differential expression analysis, univariate, Lasso and multivariate Cox regression analyses. we conducted a gene set enrichment analysis (GSEA) between groups. A nomogram integrating independent clinical factors and risk score was generated and evaluated by calibration curves and decision curve analysis (DCA). Differentially expression analysis revealed 80 differentially expressed CCRGs, and 27 of them were significantly associated with the overall survival (OS) of BC. BC can be classified into 4 molecular subtypes with significant differences in prognosis based on the 27 CCRGs. Three prognostic CCRGs, including desmocollin 1 (DSC1), LEF1, and protocadherin 9 (PCDH9), were identified to be independent risk factors of BC prognosis and were used to construct a risk score model. BC patients were divided into high- and low-risk groups, and there were significant differences in prognosis between the 2 groups both in the training and validation cohorts. It was found that patients in different groups of race, status, or T stage had significant levels of risk score. Furthermore, patients of different risk levels exhibit varying degrees of sensitivity to vinorelbine, lapatinib, metformin, and vinblastine. GSEA showed that in the high-risk group, immune response-related activities were dramatically repressed whereas cilium-related processes were significantly stimulated. Cox regression analysis demonstrated that age, N stage, radiotherapy and the risk score were independent prognostic risk factors of BC, and a nomogram was established based on these variables. The nomogram exerted a favorable concordance index (0.798) as well as calibration performance, which strongly supports the clinical application of the nomogram. Our study indicated the disruption of the expression of CCRGs in BC and built a favorable prognostic risk model based on 3 independent prognostic CCRGs. These genes may be applied as candidate molecular targets for the diagnosis and therapy of BC.

Project description:As the malignancy with the highest global incidence, breast cancer represents a significant threat to women's health. Recent advances have shed light on the importance of mitochondrial function in cancer, particularly in metabolic reprogramming within tumors. Recognizing this, we developed a novel risk signature based on mitochondrial-related genes to improve prognosis prediction and risk stratification in breast cancer patients. In this study, transcriptome data and clinical features of breast cancer samples were extracted from two sources: the TCGA, serving as the training set, and the METABRIC, used as the independent validation set. We developed the signature using LASSO-Cox regression and assessed its prognostic efficacy via ROC curves. Furthermore, the signature was integrated with clinical features to create a Nomogram model, whose accuracy was validated through clinical calibration curves and decision curve analysis. To further elucidate prognostic variations between high and low-risk groups, we conducted functional enrichment and immune infiltration analyses. Additionally, the study encompassed a comparison of mutation landscapes and drug sensitivity, providing a comprehensive understanding of the differing characteristics in these groups. Conclusively, we established a risk signature comprising 8 mitochondrial-related genes-ACSL1, ALDH2, MTHFD2, MRPL13, TP53AIP1, SLC1A1, ME3, and BCL2A1. This signature was identified as an independent risk predictor for breast cancer patient survival, exhibiting a significant high hazard ratio (HR = 3.028, 95%CI 2.038-4.499, P < 0.001). Patients in the low-risk group showed a more favorable prognosis, with enhanced immune infiltration, distinct mutation landscapes, and greater sensitivity to anti-tumor drugs. In contrast, the high-risk group exhibited an adverse trend in these aspects. This risk signature represents a novel and effective prognostic indicator, suggesting valuable insights for patient stratification in breast cancer.

Project description:BackgroundOvarian cancer (OC) is the most fatal gynaecological malignancy and has a poor prognosis. Glycosylation, the biosynthetic process that depends on specific glycosyltransferases (GTs), has recently attracted increasing importance due to the vital role it plays in cancer. In this study, we aimed to determine whether OC patients could be stratified by glycosyltransferase gene profiles to better predict the prognosis and efficiency of immune checkpoint blockade therapies (ICBs).MethodsWe retrieved transcriptome data across 420 OC and 88 normal tissue samples using The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) databases, respectively. An external validation dataset containing 185 OC samples was downloaded from the Gene Expression Omnibus (GEO) database. Knockdown and pathway prediction of B4GALT5 were conducted to investigate the function and mechanism of B4GALT5 in OC proliferation, migration and invasion.ResultsA total of 50 differentially expressed GT genes were identified between OC and normal ovarian tissues. Two clusters were stratified by operating consensus clustering, but no significant prognostic value was observed. By applying the least absolute shrinkage and selection operator (LASSO) Cox regression method, a 6-gene signature was built that classified OC patients in the TCGA cohort into a low- or high-risk group. Patients with high scores had a worse prognosis than those with low scores. This risk signature was further validated in an external GEO dataset. Furthermore, the risk score was an independent risk predictor, and a nomogram was created to improve the accuracy of prognostic classification. Notably, the low-risk OC patients exhibited a higher degree of antitumor immune cell infiltration and a superior response to ICBs. B4GALT5, one of six hub genes, was identified as a regulator of proliferation, migration and invasion in OC.ConclusionTaken together, we established a reliable GT-gene-based signature to predict prognosis, immune status and identify OC patients who would benefit from ICBs. GT genes might be a promising biomarker for OC progression and a potential therapeutic target for OC.

Dataset Information

Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context.

Publications

Prediction of breast cancer prognosis using gene set statistics provides signature stability and biological context.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets