Dataset Information

Design of a multi-signature ensemble classifier predicting neuroblastoma patients' outcome.

ABSTRACT: BACKGROUND: Neuroblastoma is the most common pediatric solid tumor of the sympathetic nervous system. Development of improved predictive tools for patients stratification is a crucial requirement for neuroblastoma therapy. Several studies utilized gene expression-based signatures to stratify neuroblastoma patients and demonstrated a clear advantage of adding genomic analysis to risk assessment. There is little overlapping among signatures and merging their prognostic potential would be advantageous. Here, we describe a new strategy to merge published neuroblastoma related gene signatures into a single, highly accurate, Multi-Signature Ensemble (MuSE)-classifier of neuroblastoma (NB) patients outcome. METHODS: Gene expression profiles of 182 neuroblastoma tumors, subdivided into three independent datasets, were used in the various phases of development and validation of neuroblastoma NB-MuSE-classifier. Thirty three signatures were evaluated for patients' outcome prediction using 22 classification algorithms each and generating 726 classifiers and prediction results. The best-performing algorithm for each signature was selected, validated on an independent dataset and the 20 signatures performing with an accuracy > = 80% were retained. RESULTS: We combined the 20 predictions associated to the corresponding signatures through the selection of the best performing algorithm into a single outcome predictor. The best performance was obtained by the Decision Table algorithm that produced the NB-MuSE-classifier characterized by an external validation accuracy of 94%. Kaplan-Meier curves and log-rank test demonstrated that patients with good and poor outcome prediction by the NB-MuSE-classifier have a significantly different survival (p < 0.0001). Survival curves constructed on subgroups of patients divided on the bases of known prognostic marker suggested an excellent stratification of localized and stage 4s tumors but more data are needed to prove this point. CONCLUSIONS: The NB-MuSE-classifier is based on an ensemble approach that merges twenty heterogeneous, neuroblastoma-related gene signatures to blend their discriminating power, rather than numeric values, into a single, highly accurate patients' outcome predictor. The novelty of our approach derives from the way to integrate the gene expression signatures, by optimally associating them with a single paradigm ultimately integrated into a single classifier. This model can be exported to other types of cancer and to diseases for which dedicated databases exist.

SUBMITTER: Cornero A

PROVIDER: S-EPMC3314564 | biostudies-literature | 2012

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Design of a multi-signature ensemble classifier predicting neuroblastoma patients' outcome.

Cornero Andrea A Acquaviva Massimo M Fardin Paolo P Versteeg Rogier R Schramm Alexander A Eva Alessandra A Bosco Maria Carla MC Blengio Fabiola F Barzaghi Sara S Varesio Luigi L

BMC bioinformatics 20120328

<h4>Background</h4>Neuroblastoma is the most common pediatric solid tumor of the sympathetic nervous system. Development of improved predictive tools for patients stratification is a crucial requirement for neuroblastoma therapy. Several studies utilized gene expression-based signatures to stratify neuroblastoma patients and demonstrated a clear advantage of adding genomic analysis to risk assessment. There is little overlapping among signatures and merging their prognostic potential would be ad ...[more]

PMID: 22536959

Similar Datasets

Project description:BackgroundNeuroblastoma is the most common extracranial solid tumor of childhood, arising from the sympathetic nervous system. High-risk neuroblastoma (HRNB) remains a major therapeutic challenge with low survival rates despite the intensification of therapy. This study aimed to develop a malignant-cell marker gene signature (MMGS) that might serve as a prognostic indicator in HRNB patients.MethodsMulti-omics datasets, including mRNA expression (single-cell and bulk), DNA methylation, and clinical information of HRNB patients, were used to identify prognostic malignant cell marker genes. MMGS was established by univariate Cox analysis, LASSO, and stepwise multivariable Cox regression analysis. Kaplan-Meier (KM) curve and time-dependent receiver operating characteristic curve (tROC) were used to evaluate the prognostic value and performance of MMGS, respectively. MMGS further verified its reliability and accuracy in the independent validation set. Finally, the characteristics of functional enrichment, tumor immune features, and inflammatory activity between different MMGS risk groups were also investigated.ResultsWe constructed a prognostic model consisting of six malignant cell maker genes (MAPT, C1QTNF4, MEG3, NPW, RAMP1, and CDT1), which stratified patients into ultra-high-risk (UHR) and common-high-risk (CHR) group. Patients in the UHR group had significantly worse overall survival (OS) than those in the CHR group. MMGS was verified as an independent predictor for the OS of HRNB patients. The area under the curve (AUC) values of MMGS at 1-, 3-, and 5-year were 0.78, 0.693, and 0.618, respectively. Notably, functional enrichment, tumor immune features, and inflammatory activity analyses preliminarily indicated that the poor prognosis in the UHR group might result from the dysregulation of the metabolic process and immunosuppressive microenvironment.ConclusionThis study established a novel six-malignant cell maker gene prognostic model that can be used to predict the prognosis of HRNB patients, which may provide new insight for the treatment and personalized monitoring of HRNB patients.

Project description:Immunotherapy for neuroblastoma remains unsatisfactory due to heterogeneity and weak immunogenicity. Exploring powerful signatures for the evaluation of immunotherapy outcomes remain the primary purpose. We constructed a ferroptosis-related gene (FRG) signature by least absolute shrinkage and selection operator and Cox regression, identified 10 independent prognostic FRGs in a training cohort (GSE62564), and then verified them in an external validation cohort (TCGA). Associated with clinical factors, the signature accurately predicts overall survival of 3, 5, and 10 years. An independent prognostic nomogram, which included FRG risk, age, stage of the International Neuroblastoma Staging System, and an MYCN status, was constructed. The area under the curves showed satisfactory prognostic predicting performance. Through bulk RNA-seq and proteomics data, we revealed the relationship between hub genes and the key onco-promoter MYCN gene and then validated the results in MYCN-amplified and MYCN-non-amplified cell lines with qRT-PCR. The FRG signature significantly divided patients into high- and low-risk groups, and the differentially expressed genes between the two groups were enriched in immune actions, autophagy, and carcinogenesis behaviors. The low-risk group embodied higher positive immune component infiltration and a higher expression of immune checkpoints with a more favorable immune cytolytic activity (CYT). We verified the predictive power of this signature with data from melanoma patients undergoing immunotherapy, and the predictive power was satisfactory. Gene mutations were closely related to the signature and prognosis. AURKA and PRKAA2 were revealed to be nodal hub FRGs in the signature, and both were shown to have significantly different expressions between the INSS stage IV and other stages after immunohistochemical validation. With single-cell RNA-seq analysis, we found that genes related to T cells were enriched in TNFA signaling and interferon-γ hallmark. In conclusion, we constructed a ferroptosis-related gene signature that can predict the outcomes and work in evaluating the effects of immunotherapy.

Dataset Information

Design of a multi-signature ensemble classifier predicting neuroblastoma patients' outcome.

Publications

Design of a multi-signature ensemble classifier predicting neuroblastoma patients' outcome.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets