Ranking prognosis markers in cancer genomic studies

Plasma ALU-247, ALU-115, and cfDNA Integrity as Diagnostic and Prognostic Biomarkers for Breast Cancer

Diagnosis of breast cancer (BC) by using sensitive and specific biomarkers is necessary. Cell-free DNA is a candidate biomarker in various cancers. Contrasting, shorted uniformed DNA released from apoptotic non-diseased cells, DNA released from malignant cells varies in size. DNA integrity is a ratio between 247 and 115 bp. So, this study was designed to investigate the role of plasma ALU-247, ALU-115, and DNA integrity as possible diagnostic and prognostic markers in BC patients as compared to plasma CA15.3. The concentrations of selected parameters were determined for 40 patients with BC (2 stage I, 31 stage II, 2 stage III, and 5 stage IV) and 10 healthy volunteers by quantitative real-time PCR and ELISA. The sensitivities of ALU-247, ALU-115, and cfDI as biomarkers for BC were evaluated and compared with CA15.3. Also, disease-free survival and overall survival were estimated. For all parameters, the concentrations in patients were significantly higher than in the control group; association with tumor stage and high sensitivities was observed. The studied parameters failed to predict survival or relapse in BC patients before surgery. Plasma ALU-247, ALU-115, and DNA integrity may prove to have clinical utility in BC diagnosis. Elevated preoperative CA15.3 was shown to be directly related to tumor burden, which may improve its diagnostic capability. Those selected parameters could be effectively used together with plasma CA15.3 for BC screening at early stage. Furthermore, both ALU-247 and ALU-115 seem to be preoperative prognostic markers for BC.

Joint Covariate Detection on Expression Profiles for Selecting Prognostic miRNAs in Glioblastoma

An important application of expression profiles is to stratify patients into high-risk and low-risk groups using limited but key covariates associated with survival outcomes. Prior to that, variables considered to be associated with survival outcomes are selected. A combination of single variables, each of which is significantly related to survival outcomes, is always regarded to be candidates for posterior patient stratification. Instead of individually significant variables, a combination that contains not only significant but also insignificant variables is supposed to be concentrated on. By means of bottom-up enumeration on each pair of variables, we propose a joint covariate detection strategy to select candidates that not only correspond to close association with survival outcomes but also help to make a clear stratification of patients. Experimental results on a publicly available dataset of glioblastoma multiforme indicate that the selected pair composed of an individually significant and an insignificant miRNA keeps a better performance than the combination of significant single variables. The selected miRNA pair is ultimately regarded to be associated with the prognosis of glioblastoma multiforme by further pathway analysis.

Boosting the concordance index for survival data--a unified framework to derive and evaluate biomarker combinations

The development of molecular signatures for the prediction of time-to-event outcomes is a methodologically challenging task in bioinformatics and biostatistics. Although there are numerous approaches for the derivation of marker combinations and their evaluation, the underlying methodology often suffers from the problem that different optimization criteria are mixed during the feature selection, estimation and evaluation steps. This might result in marker combinations that are suboptimal regarding the evaluation criterion of interest. To address this issue, we propose a unified framework to derive and evaluate biomarker combinations. Our approach is based on the concordance index for time-to-event data, which is a non-parametric measure to quantify the discriminatory power of a prediction rule. Specifically, we propose a gradient boosting algorithm that results in linear biomarker combinations that are optimal with respect to a smoothed version of the concordance index. We investigate the performance of our algorithm in a large-scale simulation study and in two molecular data sets for the prediction of survival in breast cancer patients. Our numerical results show that the new approach is not only methodologically sound but can also lead to a higher discriminatory power than traditional approaches for the derivation of gene signatures.

On the validity of time-dependent AUC estimators

Recent developments in molecular biology have led to the massive discovery of new marker candidates for the prediction of patient survival. To evaluate the predictive value of these markers, statistical tools for measuring the performance of survival models are needed. We consider estimators of discrimination measures, which are a popular approach to evaluate survival predictions in biomarker studies. Estimators of discrimination measures are usually based on regularity assumptions such as the proportional hazards assumption. Based on two sets of molecular data and a simulation study, we show that violations of the regularity assumptions may lead to over-optimistic estimates of prediction accuracy and may therefore result in biased conclusions regarding the clinical utility of new biomarkers. In particular, we demonstrate that biased medical decision making is possible even if statistical checks indicate that all regularity assumptions are satisfied.

Multicategory reclassification statistics for assessing improvements in diagnostic accuracy

In this paper, we extend the definitions of the net reclassification improvement (NRI) and the integrated discrimination improvement (IDI) in the context of multicategory classification. Both measures were proposed in Pencina and others (2008. Evaluating the added predictive ability of a new marker: from area under the receiver operating characteristic (ROC) curve to reclassification and beyond. Statistics in Medicine 27, 157-172) as numeric characterizations of accuracy improvement for binary diagnostic tests and were shown to have certain advantage over analyses based on ROC curves or other regression approaches. Estimation and inference procedures for the multiclass NRI and IDI are provided in this paper along with necessary asymptotic distributional results. Simulations are conducted to study the finite-sample properties of the proposed estimators. Two medical examples are considered to illustrate our methodology.

Adjusting confounders in ranking biomarkers: a model-based ROC approach

High-throughput studies have been extensively conducted in the research of complex human diseases. As a representative example, consider gene-expression studies where thousands of genes are profiled at the same time. An important objective of such studies is to rank the diagnostic accuracy of biomarkers (e.g. gene expressions) for predicting outcome variables while properly adjusting for confounding effects from low-dimensional clinical risk factors and environmental exposures. Existing approaches are often fully based on parametric or semi-parametric models and target evaluating estimation significance as opposed to diagnostic accuracy. Receiver operating characteristic (ROC) approaches can be employed to tackle this problem. However, existing ROC ranking methods focus on biomarkers only and ignore effects of confounders. In this article, we propose a model-based approach which ranks the diagnostic accuracy of biomarkers using ROC measures with a proper adjustment of confounding effects. To this end, three different methods for constructing the underlying regression models are investigated. Simulation study shows that the proposed methods can accurately identify biomarkers with additional diagnostic power beyond confounders. Analysis of two cancer gene-expression studies demonstrates that adjusting for confounders can lead to substantially different rankings of genes.

Risk Factors of Non-Hodgkin Lymphoma

INTRODUCTION: Despite decades of intensive research, Non-Hodgkin Lymphoma (NHL) remains poorly understood and is largely incurable. NHL is a heterogeneous group of malignancies with multiple subtypes, each of which has distinct morphologic, immunophenotypic, and clinical features. Identifying the risk factors for NHL may improve our understanding of the underlying biological mechanisms and have an impact on clinical practice. AREAS COVERED: This article provides a review of several aspects of NHL, including epidemiology and subtype classification, clinical, environmental, genetic, and genomic risk factors identified for etiology and prognosis, and available statistical and bioinformatics tools for identification of genetic and genomic risk factors from the analysis of high-throughput studies. EXPERT OPINION: Multiple clinical and environmental risk factors have been identified. However, they have failed to provide practically effective prediction. Genetic and genomic risk factors identified from high-throughput studies have suffered a lack of reproducibility. The identification of genetic/genomic risk factors demands innovative statistical and bioinformatics tools. Although multiple analysis methods have been developed, there is still room for improvement. There is a critical need for well-designed, prospective, large-scale pangenomic studies.