A comparative study of survival models for breast cancer prognostication based on microarray data: does a single gene beat them all?
Ontology highlight
ABSTRACT: Motivation
Survival prediction of breast cancer (BC) patients independently of treatment, also known as prognostication, is a complex task since clinically similar breast tumors, in addition to be molecularly heterogeneous, may exhibit different clinical outcomes. In recent years, the analysis of gene expression profiles by means of sophisticated data mining tools emerged as a promising technology to bring additional insights into BC biology and to improve the quality of prognostication. The aim of this work is to assess quantitatively the accuracy of prediction obtained with state-of-the-art data analysis techniques for BC microarray data through an independent and thorough framework.Results
Due to the large number of variables, the reduced amount of samples and the high degree of noise, complex prediction methods are highly exposed to performance degradation despite the use of cross-validation techniques. Our analysis shows that the most complex methods are not significantly better than the simplest one, a univariate model relying on a single proliferation gene. This result suggests that proliferation might be the most relevant biological process for BC prognostication and that the loss of interpretability deriving from the use of overcomplex methods may be not sufficiently counterbalanced by an improvement of the quality of prediction.Availability
The comparison study is implemented in an R package called survcomp and is available from http://www.ulb.ac.be/di/map/bhaibeka/software/survcomp/.
SUBMITTER: Haibe-Kains B
PROVIDER: S-EPMC2553442 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundThe development of clinical -omic biomarkers for predicting patient prognosis has mostly focused on multi-gene models. However, several studies have described significant weaknesses of multi-gene biomarkers. Indeed, some high-profile reports have even indicated that multi-gene biomarkers fail to consistently outperform simple single-gene ones. Given the continual improvements in -omics technologies and the availability of larger, better-powered datasets, we revisited this "single-gene hypothesis" using new techniques and datasets.ResultsBy deeply sampling the population of available gene sets, we compare the intrinsic properties of single-gene biomarkers to multi-gene biomarkers in twelve different partitions of a large breast cancer meta-dataset. We show that simple multi-gene models consistently outperformed single-gene biomarkers in all twelve partitions. We found 270 multi-gene biomarkers (one per ~11,111 sampled) that always made better predictions than the best single-gene model.ConclusionsThe single-gene hypothesis for breast cancer does not appear to retain its validity in the face of improved statistical models, lower-noise genomic technology and better-powered patient cohorts. These results highlight that it is critical to revisit older hypotheses in the light of newer techniques and datasets.
| S-EPMC6215649 | biostudies-literature
Project description:Microarray gene expression data sets are jointly analyzed to increase statistical power. They could either be merged together or analyzed by meta-analysis. For a given ensemble of data sets, it cannot be foreseen which of these paradigms, merging or meta-analysis, works better. In this article, three joint analysis methods, Z-score normalization, ComBat and the inverse normal method (meta-analysis) were selected for survival prognosis and risk assessment of breast cancer patients. The methods were applied to eight microarray gene expression data sets, totaling 1324 patients with two clinical endpoints, overall survival and relapse-free survival. The performance derived from the joint analysis methods was evaluated using Cox regression for survival analysis and independent validation used as bias estimation. Overall, Z-score normalization had a better performance than ComBat and meta-analysis. Higher Area Under the Receiver Operating Characteristic curve and hazard ratio were also obtained when independent validation was used as bias estimation. With a lower time and memory complexity, Z-score normalization is a simple method for joint analysis of microarray gene expression data sets. The derived findings suggest further assessment of this method in future survival prediction and cancer classification applications.
| S-EPMC5863785 | biostudies-literature
Project description:BackgroundWith the growing abundance of microarray data, statistical methods are increasingly needed to integrate results across studies. Two common approaches for meta-analysis of microarrays include either combining gene expression measures across studies or combining summaries such as p-values, probabilities or ranks. Here, we compare two Bayesian meta-analysis models that are analogous to these methods.ResultsTwo Bayesian meta-analysis models for microarray data have recently been introduced. The first model combines standardized gene expression measures across studies into an overall mean, accounting for inter-study variability, while the second combines probabilities of differential expression without combining expression values. Both models produce the gene-specific posterior probability of differential expression, which is the basis for inference. Since the standardized expression integration model includes inter-study variability, it may improve accuracy of results versus the probability integration model. However, due to the small number of studies typical in microarray meta-analyses, the variability between studies is challenging to estimate. The probability integration model eliminates the need to model variability between studies, and thus its implementation is more straightforward. We found in simulations of two and five studies that combining probabilities outperformed combining standardized gene expression measures for three comparison values: the percent of true discovered genes in meta-analysis versus individual studies; the percent of true genes omitted in meta-analysis versus separate studies, and the number of true discovered genes for fixed levels of Bayesian false discovery. We identified similar results when pooling two independent studies of Bacillus subtilis. We assumed that each study was produced from the same microarray platform with only two conditions: a treatment and control, and that the data sets were pre-scaled.ConclusionThe Bayesian meta-analysis model that combines probabilities across studies does not aggregate gene expression measures, thus an inter-study variability parameter is not included in the model. This results in a simpler modeling approach than aggregating expression measures, which accounts for variability across studies. The probability integration model identified more true discovered genes and fewer true omitted genes than combining expression measures, for our data sets.
| S-EPMC1851021 | biostudies-literature
Project description:BACKGROUND: An estimated 12% of females in the United States will develop breast cancer in their lifetime. Although, there are advances in treatment options including surgery and chemotherapy, breast cancer is still the second most lethal cancer in women. Thus, there is a clear need for better methods to predict prognosis for each breast cancer patient. With the advent of large genetic databases and the reduction in cost for the experiments, researchers are faced with choosing from a large pool of potential prognostic markers from numerous breast cancer gene expression profile studies. METHODS: Five microarray datasets related to breast cancer were examined using gene set analysis and the cancers were categorized into different subtypes using a scoring system based on genetic pathway activity. RESULTS: We have observed that significant genes in the individual studies show little reproducibility across the datasets. From our comparative analysis, using gene pathways with clinical variables is more reliable across studies and shows promise in assessing a patient's prognosis. CONCLUSIONS: This study concludes that, in light of clinical variables, there are significant gene pathways in common across the datasets. Specifically, several pathways can further significantly stratify patients for survival. These candidate pathways should help to develop a panel of significant biomarkers for the prognosis of breast cancer patients in a clinical setting.
| S-EPMC2972286 | biostudies-literature
Project description:BackgroundGlioblastoma (GBM) remains associated with a dismal prognoses despite standard therapies. While population-level survival statistics are established, generating individualized prognosis remains challenging. We aim to develop machine learning (ML) models that generate personalized survival predictions for GBM patients to enhance prognostication.MethodsAdult patients with histologically confirmed IDH-wildtype GBM from the National Cancer Database (NCDB) were analyzed. ML models were developed with TabPFN, TabNet, XGBoost, LightGBM, and Random Forest algorithms to predict mortality at 6, 12, 18, and 24 months postdiagnosis. SHapley Additive exPlanations (SHAP) were employed to enhance the interpretability of the models. Models were primarily evaluated using the area under the receiver operating characteristic (AUROC) values, and the top-performing models indicated by the highest AUROCs for each outcome were deployed in a web application that was created for individualized predictions.ResultsA total of 7537 patients were retrieved from the NCDB. Performance evaluation revealed the top-performing models for each outcome were built using the TabPFN algorithm. The TabPFN models yielded mean AUROCs of 0.836, 0.78, 0.732, and 0.724 in predicting 6, 12, 18, and 24 month mortality, respectively.ConclusionsThis study establishes ML models tailored to individual patients to enhance GBM prognostication. Future work should focus on external validation and dynamic updating as new data emerge.
| S-EPMC11232516 | biostudies-literature
Project description:BackgroundMicroarray technologies are widely used to quantify the abundance of transcripts corresponding to thousands of genes. To maximise the robustness of transcriptome results, we have tested the performance and reproducibility of rat and mouse gene expression data obtained with Affymetrix, Illumina and Operon platforms.ResultsWe present a thorough analysis of the degree of reproducibility provided by analysing the transcriptomic profile of the same animals of several experimental groups under different popular microarray technologies in different tissues. Concordant results from inter- and intra-platform comparisons were maximised by testing many popular computational methods for generating fold changes and significances and by only considering oligonucleotides giving high expression levels. The choice of Affymetrix signal extraction technique was shown to have the greatest effect on the concordance across platforms. In both species, when choosing optimal methods, the agreement between data generated on the Affymetrix and Illumina was excellent; this was verified using qRT-PCR on a selection of genes present on all platforms.ConclusionThis study provides an extensive assessment of analytical methods best suited for processing data from different microarray technologies and can assist integration of technologically different gene expression datasets in biological systems.
| S-EPMC2652496 | biostudies-literature
Project description:BackgroundRecent technological advances in high-throughput data collection allow for experimental study of increasingly complex systems on the scale of the whole cellular genome and proteome. Gene network models are needed to interpret the resulting large and complex data sets. Rationally designed perturbations (e.g., gene knock-outs) can be used to iteratively refine hypothetical models, suggesting an approach for high-throughput biological system analysis. We introduce an approach to gene network modeling based on a scalable linear variant of fuzzy logic: a framework with greater resolution than Boolean logic models, but which, while still semi-quantitative, does not require the precise parameter measurement needed for chemical kinetics-based modeling.ResultsWe demonstrated our approach with exhaustive search for fuzzy gene interaction models that best fit transcription measurements by microarray of twelve selected genes regulating the yeast cell cycle. Applying an efficient, universally applicable data normalization and fuzzification scheme, the search converged to a small number of models that individually predict experimental data within an error tolerance. Because only gene transcription levels are used to develop the models, they include both direct and indirect regulation of genes.ConclusionBiological relationships in the best-fitting fuzzy gene network models successfully recover direct and indirect interactions predicted from previous knowledge to result in transcriptional correlation. Fuzzy models fit on one yeast cell cycle data set robustly predict another experimental data set for the same system. Linear fuzzy gene networks and exhaustive rule search are the first steps towards a framework for an integrated modeling and experiment approach to high-throughput "reverse engineering" of complex biological systems.
| S-EPMC514698 | biostudies-literature
Project description:DNA microarray studies identified gene expression signatures predictive of metastatic relapse in early breast cancer. Standard feature selection procedures applied to reduce the set of predictive genes did not take into account the correlation between genes. In this paper, we studied the performances of three high-dimensional regression methods - CoxBoost, LASSO (Least Absolute Shrinkage and Selection Operator), and Elastic net - to identify prognostic signatures in patients with early breast cancer.We analyzed three public retrospective datasets, including a total of 384 patients with axillary lymph node-negative breast cancer. The Amsterdam van't Veer's training set of 78 patients was used to determine the optimal gene sets and classifiers using sensitivity thresholds resulting in mis-classification of no more than 10% of the poor-prognosis group. To ensure the comparability between different methods, an automatic selection procedure was used to determine the number of genes included in each model. The van de Vijver's and Desmedt's datasets were used as validation sets to evaluate separately the prognostic performances of our classifiers. The results were compared to the original Amsterdam 70-gene classifier.The automatic selection procedure reduced the number of predictive genes up to a minimum of six genes. In the two validation sets, the three models (Elastic net, LASSO, and CoxBoost) led to the definition of genomic classifiers predicting the 5-year metastatic status with similar performances, with respective 59, 56, and 54% accuracy, 83, 75, and 83% sensitivity, and 53, 52, and 48% specificity in the Desmedt's dataset. In comparison, the Amsterdam 70-gene signature showed 45% accuracy, 97% sensitivity, and 34% specificity. The gene overlap and the classification concordance between the three classifiers were high. All the classifiers added significant prognostic information to that provided by the traditional prognostic factors and showed a very high overlap with respect to gene ontologies (GOs) associated with genes overexpressed in the predicted poor-prognosis vs. good-prognosis classes and centred on cell proliferation. Interestingly, all classifiers reported high sensitivity to predict the 4-year status of metastatic disease.High-dimensional regression methods are attractive in prognostic studies because finding a small subset of genes may facilitate the transfer to the clinic, and also because they strengthen the robustness of the model by limiting the selection of false-positive predictive genes. With only six genes, the CoxBoost classifier predicted the 4-year status of metastatic disease with 93% sensitivity. Selecting a few genes related to ontologies other than cell proliferation might further improve the overall sensitivity performance.
| S-EPMC4426954 | biostudies-literature
Project description:BackgroundGenome-wide expression signatures are emerging as potential marker for overall survival and disease recurrence risk as evidenced by recent commercialization of gene expression based biomarkers in breast cancer. Similar predictions have recently been carried out using genome-wide copy number alterations and microRNAs. Existing software packages for microarray data analysis provide functions to define expression-based survival gene signatures. However, there is no software that can perform survival analysis using SNP array data or draw survival curves interactively for expression-based sample clusters.ResultsWe have developed the survival analysis module in the dChip software that performs survival analysis across the genome for gene expression and copy number microarray data. Built on the current dChip software's microarray analysis functions such as chromosome display and clustering, the new survival functions include interactive exploring of Kaplan-Meier (K-M) plots using expression or copy number data, computing survival p-values from the log-rank test and Cox models, and using permutation to identify significant chromosome regions associated with survival.ConclusionsThe dChip survival module provides user-friendly way to perform survival analysis and visualize the results in the context of genes and cytobands. It requires no coding expertise and only minimal learning curve for thousands of existing dChip users. The implementation in Visual C++ also enables fast computation. The software and demonstration data are freely available at http://dchip-surv.chenglilab.org.
| S-EPMC3068974 | biostudies-literature
Project description:BackgroundIn this study, we developed integrative, personalized prognostic models for breast cancer recurrence and overall survival (OS) that consider receptor subtypes, epidemiological data, quality of life (QoL), and treatment.MethodsA total of 15 314 women with stage I to III invasive primary breast cancer treated at The University of Texas MD Anderson Cancer Center between 1997 and 2012 were used to generate prognostic models by Cox regression analysis in a two-stage study. Model performance was assessed by calculating the area under the curve (AUC) and calibration analysis and compared with Nottingham Prognostic Index (NPI) and PREDICT.ResultsHost characteristics were assessed for 10 809 women as the discovery population (median follow-up?=?6.09 years, 1144 recurrence and 1627 deaths) and 4505 women as the validation population (median follow-up?=?7.95 years, 684 recurrence and 1095 deaths). In addition to the known clinical/pathological variables, the model for recurrence included alcohol consumption while the model for OS included smoking status and physical component summary score. The AUCs for recurrence and OS were 0.813 and 0.810 in the discovery and 0.807 and 0.803 in the validation, respectively, compared with AUCs of 0.761 and 0.753 in discovery and 0.777 and 0.751 in validation for NPI. Our model further showed better calibration compared with PREDICT. We also developed race-specific and receptor subtype-specific models with comparable AUCs. Racial disparity was evident in the distributions of many risk factors and clinical presentation of the disease.ConclusionsOur integrative prognostic models for breast cancer exhibit high discriminatory accuracy and excellent calibration and are the first to incorporate receptor subtype and epidemiological and QoL data.
| S-EPMC6279311 | biostudies-literature