Gene expression profiling for prognosis using Cox regression

Orchestrating information across tissues via a novel multitask GAT framework to improve quantitative gene regulation relation modeling for survival analysis

Survival analysis is critical to cancer prognosis estimation. High-throughput technologies facilitate the increase in the dimension of genic features, but the number of clinical samples in cohorts is relatively small due to various reasons, including difficulties in participant recruitment and high data-generation costs. Transcriptome is one of the most abundantly available OMIC (referring to the high-throughput data, including genomic, transcriptomic, proteomic and epigenomic) data types. This study introduced a multitask graph attention network (GAT) framework DQSurv for the survival analysis task. We first used a large dataset of healthy tissue samples to pretrain the GAT-based HealthModel for the quantitative measurement of the gene regulatory relations. The multitask survival analysis framework DQSurv used the idea of transfer learning to initiate the GAT model with the pretrained HealthModel and further fine-tuned this model using two tasks i.e. the main task of survival analysis and the auxiliary task of gene expression prediction. This refined GAT was denoted as DiseaseModel. We fused the original transcriptomic features with the difference vector between the latent features encoded by the HealthModel and DiseaseModel for the final task of survival analysis. The proposed DQSurv model stably outperformed the existing models for the survival analysis of 10 benchmark cancer types and an independent dataset. The ablation study also supported the necessity of the main modules. We released the codes and the pretrained HealthModel to facilitate the feature encodings and survival analysis of transcriptome-based future studies, especially on small datasets. The model and the code are available at http://www.healthinformaticslab.org/supp/.

Sparse modelling of cancer patients' survival based on genomic copy number alterations

Copy number alterations (CNA) are structural variation in the genome, in which some regions exhibit more or less than the normal two chromosomal copies. This genomic CNA profile provides critical information in tumour progression and is therefore informative for patients' survival. It is currently a statistical challenge to model patients' survival using their genomic CNA profiles while at the same time identify regions in the genome that are associated with patients' survival. Some methods have been proposed, including Cox proportional hazard (PH) model with ridge, lasso, or elastic net penalties. However, these methods do not take the general dependencies between genomic regions into account and produce results that are difficult to interpret. In this paper, we extend the elastic net penalty by introducing additional penalty that takes into account general dependencies between genomic regions. This new model produces smooth parameter estimates while simultaneously performs variable selection via sparse solution. The results indicate that the proposed method shows a better prediction performance than other models in our simulation study, while enabling us to investigate regions in the genome that are associated with the patients' survival with sensible interpretation. We illustrate the method using a real dataset from a lung cancer cohort and simulated data.

Comparison of pathway and gene-level models for cancer prognosis prediction

BACKGROUND

Cancer prognosis prediction is valuable for patients and clinicians because it allows them to appropriately manage care. A promising direction for improving the performance and interpretation of expression-based predictive models involves the aggregation of gene-level data into biological pathways. While many studies have used pathway-level predictors for cancer survival analysis, a comprehensive comparison of pathway-level and gene-level prognostic models has not been performed. To address this gap, we characterized the performance of penalized Cox proportional hazard models built using either pathway- or gene-level predictors for the cancers profiled in The Cancer Genome Atlas (TCGA) and pathways from the Molecular Signatures Database (MSigDB).

RESULTS

When analyzing TCGA data, we found that pathway-level models are more parsimonious, more robust, more computationally efficient and easier to interpret than gene-level models with similar predictive performance. For example, both pathway-level and gene-level models have an average Cox concordance index of ~ 0.85 for the TCGA glioma cohort, however, the gene-level model has twice as many predictors on average, the predictor composition is less stable across cross-validation folds and estimation takes 40 times as long as compared to the pathway-level model. When the complex correlation structure of the data is broken by permutation, the pathway-level model has greater predictive performance while still retaining superior interpretative power, robustness, parsimony and computational efficiency relative to the gene-level models. For example, the average concordance index of the pathway-level model increases to 0.88 while the gene-level model falls to 0.56 for the TCGA glioma cohort using survival times simulated from uncorrelated gene expression data.

CONCLUSION

The results of this study show that when the correlations among gene expression values are low, pathway-level analyses can yield better predictive performance, greater interpretative power, more robust models and less computational cost relative to a gene-level model. When correlations among genes are high, a pathway-level analysis provides equivalent predictive power compared to a gene-level analysis while retaining the advantages of interpretability, robustness and computational efficiency.

Sparse partial least-squares regression for high-throughput survival data analysis

The partial least-square (PLS) method has been adapted to the Cox's proportional hazards model for analyzing high-dimensional survival data. But because the latent components constructed in PLS employ all predictors regardless of their relevance, it is often difficult to interpret the results. In this paper, we propose a new formulation of sparse PLS (SPLS) procedure for survival data to allow simultaneous sparse variable selection and dimension reduction. We develop a computing algorithm for SPLS by modifying an iteratively reweighted PLS algorithm and illustrate the method with the Swedish and the Netherlands Cancer Institute breast cancer datasets. Through the numerical studies, we find that our SPLS method generally performs better than the standard PLS and sparse Cox regression methods in variable selection and prediction.

Network-based survival analysis reveals subnetwork signatures for predicting outcomes of ovarian cancer treatment

Cox regression is commonly used to predict the outcome by the time to an event of interest and in addition, identify relevant features for survival analysis in cancer genomics. Due to the high-dimensionality of high-throughput genomic data, existing Cox models trained on any particular dataset usually generalize poorly to other independent datasets. In this paper, we propose a network-based Cox regression model called Net-Cox and applied Net-Cox for a large-scale survival analysis across multiple ovarian cancer datasets. Net-Cox integrates gene network information into the Cox's proportional hazard model to explore the co-expression or functional relation among high-dimensional gene expression features in the gene network. Net-Cox was applied to analyze three independent gene expression datasets including the TCGA ovarian cancer dataset and two other public ovarian cancer datasets. Net-Cox with the network information from gene co-expression or functional relations identified highly consistent signature genes across the three datasets, and because of the better generalization across the datasets, Net-Cox also consistently improved the accuracy of survival prediction over the Cox models regularized by or . This study focused on analyzing the death and recurrence outcomes in the treatment of ovarian carcinoma to identify signature genes that can more reliably predict the events. The signature genes comprise dense protein-protein interaction subnetworks, enriched by extracellular matrix receptors and modulators or by nuclear signaling components downstream of extracellular signal-regulated kinases. In the laboratory validation of the signature genes, a tumor array experiment by protein staining on an independent patient cohort from Mayo Clinic showed that the protein expression of the signature gene FBN1 is a biomarker significantly associated with the early recurrence after 12 months of the treatment in the ovarian cancer patients who are initially sensitive to chemotherapy. Net-Cox toolbox is available at http://compbio.cs.umn.edu/Net-Cox/.

Network-based computational models are attracting increasing attention in studying cancer genomics because molecular networks provide valuable information on the functional organizations of molecules in cells. Survival analysis mostly with the Cox proportional hazard model is widely used to predict or correlate gene expressions with time to an event of interest (outcome) in cancer genomics. Surprisingly, network-based survival analysis has not received enough attention. In this paper, we studied resistance to chemotherapy in ovarian cancer with a network-based Cox model, called Net-Cox. The experiments confirm that networks representing gene co-expression or functional relations can be used to improve the accuracy and the robustness of survival prediction of outcome in ovarian cancer treatment. The study also revealed subnetwork signatures that are enriched by extracellular matrix receptors and modulators and the downstream nuclear signaling components of extracellular signal-regulators, respectively. In particular, FBN1, which was detected as a signature gene of high confidence by Net-Cox with network information, was validated as a biomarker for predicting early recurrence in platinum-sensitive ovarian cancer patients in laboratory.

High-dimensional Cox models: the choice of penalty as part of the model building process

The Cox proportional hazards regression model is the most popular approach to model covariate information for survival times. In this context, the development of high-dimensional models where the number of covariates is much larger than the number of observations (p>>n) is an ongoing challenge. A practicable approach is to use ridge penalized Cox regression in such situations. Beside focussing on finding the best prediction rule, one is often interested in determining a subset of covariates that are the most important ones for prognosis. This could be a gene set in the biostatistical analysis of microarray data. Covariate selection can then, for example, be done by L(1)-penalized Cox regression using the lasso (Tibshirani (1997). Statistics in Medicine 16, 385-395). Several approaches beyond the lasso, that incorporate covariate selection, have been developed in recent years. This includes modifications of the lasso as well as nonconvex variants such as smoothly clipped absolute deviation (SCAD) (Fan and Li (2001). Journal of the American Statistical Association 96, 1348-1360; Fan and Li (2002). The Annals of Statistics 30, 74-99). The purpose of this article is to implement them practically into the model building process when analyzing high-dimensional data with the Cox proportional hazards model. To evaluate penalized regression models beyond the lasso, we included SCAD variants and the adaptive lasso (Zou (2006). Journal of the American Statistical Association 101, 1418-1429). We compare them with "standard" applications such as ridge regression, the lasso, and the elastic net. Predictive accuracy, features of variable selection, and estimation bias will be studied to assess the practical use of these methods. We observed that the performance of SCAD and adaptive lasso is highly dependent on nontrivial preselection procedures. A practical solution to this problem does not yet exist. Since there is high risk of missing relevant covariates when using SCAD or adaptive lasso applied after an inappropriate initial selection step, we recommend to stay with lasso or the elastic net in actual data applications. But with respect to the promising results for truly sparse models, we see some advantage of SCAD and adaptive lasso, if better preselection procedures would be available. This requires further methodological research.

Gene expression profiling for survival prediction in pediatric rhabdomyosarcomas: a report from the children's oncology group

PURPOSE

We investigated whether tumors from diagnostic biopsies of primary rhabdomyosarcoma (RMS) contain relevant prognostic information in the form of gene expression signatures that can be used to model and predict outcome of patients.

PATIENTS AND METHODS

A 22,000-probe set microarray was used to evaluate 120 RMS specimens and correlate gene expression patterns to survival. Multivariate gene expression models or metagenes were developed using cross-validated Cox regression proportional hazards modeling and were evaluated using Kaplan-Meier analysis.

RESULTS

A 34-metagene, based on expression patterns of 34 genes, was highly predictive of outcome. It was not highly correlated with individual clinical risk factors such as patient age, stage, tumor size, or histology. However, it was correlated with a risk classification used by the Children's Oncology Group and the biologic subsets of alveolar histology tumors.

CONCLUSION

These data support further evaluation of RMS metagenes to discriminate patients with good prognosis from those with poor prognosis, with the potential to direct risk-adapted therapy.

Pathway analysis using random forests with bivariate node-split for survival outcomes

MOTIVATION

There is great interest in pathway-based methods for genomics data analysis in the research community. Although machine learning methods, such as random forests, have been developed to correlate survival outcomes with a set of genes, no study has assessed the abilities of these methods in incorporating pathway information for analyzing microarray data. In general, genes that are identified without incorporating biological knowledge are more difficult to interpret. Correlating pathway-based gene expression with survival outcomes may lead to biologically more meaningful prognosis biomarkers. Thus, a comprehensive study on how these methods perform in a pathway-based setting is warranted.

RESULTS

In this article, we describe a pathway-based method using random forests to correlate gene expression data with survival outcomes and introduce a novel bivariate node-splitting random survival forests. The proposed method allows researchers to identify important pathways for predicting patient prognosis and time to disease progression, and discover important genes within those pathways. We compared different implementations of random forests with different split criteria and found that bivariate node-splitting random survival forests with log-rank test is among the best. We also performed simulation studies that showed random forests outperforms several other machine learning algorithms and has comparable results with a newly developed component-wise Cox boosting model. Thus, pathway-based survival analysis using machine learning tools represents a promising approach in dissecting pathways and for generating new biological hypothesis from microarray studies.

AVAILABILITY

R package Pwayrfsurvival is available from URL: http://www.duke.edu/~hp44/pwayrfsurvival.htm.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

The additive hazards model with high-dimensional regressors

This paper considers estimation and prediction in the Aalen additive hazards model in the case where the covariate vector is high-dimensional such as gene expression measurements. Some form of dimension reduction of the covariate space is needed to obtain useful statistical analyses. We study the partial least squares regression method. It turns out that it is naturally adapted to this setting via the so-called Krylov sequence. The resulting PLS estimator is shown to be consistent provided that the number of terms included is taken to be equal to the number of relevant components in the regression model. A standard PLS algorithm can also be constructed, but it turns out that the resulting predictor can only be related to the original covariates via time-dependent coefficients. The methods are applied to a breast cancer data set with gene expression recordings and to the well known primary biliary cirrhosis clinical data.

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

Background

Censored data are increasingly common in many microarray studies that attempt to relate gene expression to patient survival. Several new methods have been proposed in the last two years. Most of these methods, however, are not available to biomedical researchers, leading to many re-implementations from scratch of ad-hoc, and suboptimal, approaches with survival data.

Results

We have developed SignS (Signatures for Survival data), an open-source, freely-available, web-based tool and R package for gene selection, building molecular signatures, and prediction with survival data. SignS implements four methods which, according to existing reviews, perform well and, by being of a very different nature, offer complementary approaches. We use parallel computing via MPI, leading to large decreases in user waiting time. Cross-validation is used to asses predictive performance and stability of solutions, the latter an issue of increasing concern given that there are often several solutions with similar predictive performance. Biological interpretation of results is enhanced because genes and signatures in models can be sent to other freely-available on-line tools for examination of PubMed references, GO terms, and KEGG and Reactome pathways of selected genes.

Conclusion

SignS is the first web-based tool for survival analysis of expression data, and one of the very few with biomedical researchers as target users. SignS is also one of the few bioinformatics web-based applications to extensively use parallelization, including fault tolerance and crash recovery. Because of its combination of methods implemented, usage of parallel computing, code availability, and links to additional data bases, SignS is a unique tool, and will be of immediate relevance to biomedical researchers, biostatisticians and bioinformaticians.

Cross-validated Cox regression on microarray gene expression data

This paper describes how penalized Cox regression, in combination with cross-validated partial likelihood can be employed to obtain reliable survival prediction models for high dimensional microarray data. The suggested procedure is demonstrated on a breast cancer survival data set consisting of 295 tumours as collected in the National Cancer Institute in Amsterdam and previously reported in more general papers. The main aim of this paper it to show how generally accepted biostatistical procedures can be employed to analyse high-dimensional data.

Predicting survival outcomes using subsets of significant genes in prognostic marker studies with microarrays

Background

Genetic markers hold great promise for refining our ability to establish precise prognostic prediction for diseases. The development of comprehensive gene expression microarray technology has allowed the selection of relevant marker genes from a large pool of candidate genes in early-phased, developmental prognostic marker studies. The primary analytical task in such studies is to select a small fraction of relevant genes, typically from a list of significant genes, for further investigation in subsequent studies.

Results

We develop a methodology for predicting survival outcomes using subsets of significant genes in prognostic marker studies with microarrays. Key components in this methodology include building prediction models, assessing predictive performance of prediction models, and assessing significance of prediction results. As particular specifications, we assume Cox proportional hazard models with a compound covariate. For assessing predictive accuracy, we propose to use the cross-validated log partial likelihood. To assess significance of prediction results, we apply permutation procedures in cross-validated prediction. As an additional key component peculiar to prognostic prediction, we also consider incorporation of standard prognostic factors. The methodology is evaluated using both simulated and real data.

Conclusion

The developed methodology for prognostic prediction using a subset of significant genes can provide new insights based on predictive capability, possibly incorporating standard prognostic factors, in selecting a fraction of relevant genes for subsequent studies.

Benefit assessment of therapeutic products: the Centers for Education and Research on Therapeutics

The ability to manage risk depends critically on an understanding of the degree to which a known risk is balanced by the probability of a clinical benefit. Despite the massive emphasis on risk and risk management in the past few years and the long-term focus on defining benefit in the regulatory system, considerable uncertainty remains about the methods of defining benefit and how to operationalize this knowledge. In this 'think tank,' part of a larger series on risk management, issues were divided into those that can be identified before a study is initiated, those that commonly arise after a study is completed, biomarkers and surrogates, use of benefit findings in defining quality and performance indicators, implementation of findings into health systems and formularies, and methods of comparative trials. Key categories for the establishment of a research agenda to fill in gaps in our understanding of assessing benefit were developed by the group.

Testing association of a pathway with survival using gene expression data

MOTIVATION

A recent surge of interest in survival as the primary clinical endpoint of microarray studies has called for an extension of the Global Test methodology to survival.

RESULTS

We present a score test for association of the expression profile of one or more groups of genes with a (possibly censored) survival time. Groups of genes may be pathways, areas of the genome, clusters from a cluster analysis or all genes on a chip. The test allows one to test hypotheses about the influence of these groups of genes on survival directly, without the intermediary of single gene testing. The test is based on the Cox proportional hazards model and is calculated using martingale residuals. It is possible to adjust the test for the presence of covariates. We also present a diagnostic graph to assist in the interpretation of the test result, visualizing the influence of genes. The test is applied to a tumor dataset, revealing pathways from the gene ontology database that are associated with survival of patients.

AVAILABILITY

The Global Test for survival has been incorporated into the R-package globaltest (version 3.0), available at http://www.bioconductor.org