The sensitivity and specificity of markers for event times

The statistical literature on assessing the accuracy of risk factors or disease markers as diagnostic tests deals almost exclusively with settings where the test, Y, is measured concurrently with disease status D. In practice, however, disease status may vary over time and there is often a time lag between when the marker is measured and the occurrence of disease. One example concerns the Framingham risk score (FR-score) as a marker for the future risk of cardiovascular events, events that occur after the score is ascertained. To evaluate such a marker, one needs to take the time lag into account since the predictive accuracy may be higher when the marker is measured closer to the time of disease occurrence. We therefore consider inference for sensitivity and specificity functions that are defined as functions of time. Semiparametric regression models are proposed. Data from a cohort study are used to estimate model parameters. One issue that arises in practice is that event times may be censored. In this research, we extend in several respects the work by Leisenring et al. (1997) that dealt only with parametric models for binary tests and uncensored data. We propose semiparametric models that accommodate continuous tests and censoring. Asymptotic distribution theory for parameter estimates is developed and procedures for making statistical inference are evaluated with simulation studies. We illustrate our methods with data from the Cardiovascular Health Study, relating the FR-score measured at enrollment to subsequent risk of cardiovascular events.     

Microarray gene expression data with linked survival phenotypes: diffuse large-B-cell lymphoma revisited

Diffuse large-B-cell lymphoma (DLBCL) is an aggressive malignancy of mature B lymphocytes and is the most common type of lymphoma in adults. While treatment advances have been substantial in what was formerly a fatal disease, less than 50% of patients achieve lasting remission. In an effort to predict treatment success and explain disease heterogeneity clinical features have been employed for prognostic purposes, but have yielded only modest predictive performance. This has spawned a series of high-profile microarray-based gene expression studies of DLBCL, in the hope that molecular-level information could be used to refine prognosis. The intent of this paper is to reevaluate these microarray-based prognostic assessments, and extend the statistical methodology that has been used in this context. Methodological challenges arise in using patients' gene expression profiles to predict survival endpoints on account of the large number of genes and their complex interdependence. We initially focus on the Lymphochip data and analysis of Rosenwald et al. (2002). After describing relationships between the analyses performed and gene harvesting (Hastie et al., 2001a), we argue for the utility of penalized approaches, in particular least angle regression-least absolute shrinkage and selection operator (Efron et al., 2004). While these techniques have been extended to the proportional hazards/partial likelihood framework, the resultant algorithms are computationally burdensome. We develop residual-based approximations that eliminate this burden yet perform similarly. Comparisons of predictive accuracy across both methods and studies are effected using time-dependent receiver operating characteristic curves. These indicate that gene expression data, in turn, only delivers modest predictions of posttherapy DLBCL survival. We conclude by outlining possibilities for further work.     

Semiparametric Models of Time-Dependent Predictive Values of Prognostic Biomarkers

Summary .  Rigorous statistical evaluation of the predictive values of novel biomarkers is critical prior to applying novel biomarkers into routine standard care. It is important to identify factors that influence the performance of a biomarker in order to determine the optimal conditions for test performance. We propose a covariate-specific time-dependent positive predictive values curve to quantify the predictive accuracy of a prognostic marker measured on a continuous scale and with censored failure time outcome. The covariate effect is accommodated with a semiparametric regression model framework. In particular, we adopt a smoothed survival time regression technique ( Dabrowska, 1997 ,  The Annals of Statistics   25, 1510–1540) to account for the situation where risk for the disease occurrence and progression is likely to change over time. In addition, we provide asymptotic distribution theory and resampling-based procedures for making statistical inference on the covariate-specific positive predictive values. We illustrate our approach with numerical studies and a dataset from a prostate cancer study.     

Identification and evaluation of a panel of serum biomarkers for predicting response to thalidomide in multiple myeloma patients

Evaluation of: Rajpal R, Dowling P, Meiller J et al. A novel panel of protein biomarkers for predicting response to thalidomide-based therapy in newly diagnosed multiple myeloma patients. Proteomics 11(8), 1391-1402 (2011). Predicting response to thalidomide-based therapy remains a challenging task faced by clinicians in the treatment of multiple myeloma. The pioneering work reported by Rajpal et al. moves one step further towards solving this challenge. They developed a proteomics-based approach that combines immunodepletion, 2D-difference gel electrophoresis analysis and mass spectrometry to search for serum proteins with expressions that show significant correlations to thalidomide treatment. This integrated approach allowed them to identify a panel of protein biomarkers. By using ELISA-based validation and strict statistical analysis, the authors have achieved an overall 84.0% predictive accuracy, with associated sensitivity and specificity values of 81.8 and 86.2%, respectively. Their methods and significant findings are reviewed within this article. This panel of biomarkers may not only guide initial therapy, but can also provide direct implications for personalized medicine in multiple myeloma patients.     

Network-based auto-probit modeling for protein function prediction

Predicting the functional roles of proteins based on various genome-wide data, such as protein-protein association networks, has become a canonical problem in computational biology. Approaching this task as a binary classification problem, we develop a network-based extension of the spatial auto-probit model. In particular, we develop a hierarchical Bayesian probit-based framework for modeling binary network-indexed processes, with a latent multivariate conditional autoregressive Gaussian process. The latter allows for the easy incorporation of protein-protein association network topologies-either binary or weighted-in modeling protein functional similarity. We use this framework to predict protein functions, for functions defined as terms in the Gene Ontology (GO) database, a popular rigorous vocabulary for biological functionality. Furthermore, we show how a natural extension of this framework can be used to model and correct for the high percentage of false negative labels in training data derived from GO, a serious shortcoming endemic to biological databases of this type. Our method performance is evaluated and compared with standard algorithms on weighted yeast protein-protein association networks, extracted from a recently developed integrative database called Search Tool for the Retrieval of INteracting Genes/proteins (STRING). Results show that our basic method is competitive with these other methods, and that the extended method-incorporating the uncertainty in negative labels among the training data-can yield nontrivial improvements in predictive accuracy.     

Evaluating prognostic accuracy of biomarkers under competing risk

To develop more targeted intervention strategies, an important research goal is to identify markers predictive of clinical events. A crucial step toward this goal is to characterize the clinical performance of a marker for predicting different types of events. In this article, we present statistical methods for evaluating the performance of a prognostic marker in predicting multiple competing events. To capture the potential time-varying predictive performance of the marker and incorporate competing risks, we define time- and cause-specific accuracy summaries by stratifying cases based on causes of failure. Such definition would allow one to evaluate the predictive accuracy of a marker for each type of event and compare its predictiveness across event types. Extending the nonparametric crude cause-specific receiver operating characteristics curve estimators by Saha and Heagerty (2010), we develop inference procedures for a range of cause-specific accuracy summaries. To estimate the accuracy measures and assess how covariates may affect the accuracy of a marker under the competing risk setting, we consider two forms of semiparametric models through the cause-specific hazard framework. These approaches enable a flexible modeling of the relationships between the marker and failure times for each cause, while efficiently accommodating additional covariates. We investigate the asymptotic property of the proposed accuracy estimators and demonstrate the finite sample performance of these estimators through simulation studies. The proposed procedures are illustrated with data from a prostate cancer prognostic study.     

Classification of multiple cancer types by multicategory support vector machines using gene expression data

MOTIVATION: High-density DNA microarray measures the activities of several thousand genes simultaneously and the gene expression profiles have been used for the cancer classification recently. This new approach promises to give better therapeutic measurements to cancer patients by diagnosing cancer types with improved accuracy. The Support Vector Machine (SVM) is one of the classification methods successfully applied to the cancer diagnosis problems. However, its optimal extension to more than two classes was not obvious, which might impose limitations in its application to multiple tumor types. We briefly introduce the Multicategory SVM, which is a recently proposed extension of the binary SVM, and apply it to multiclass cancer diagnosis problems. RESULTS: Its applicability is demonstrated on the leukemia data (Golub et al., 1999) and the small round blue cell tumors of childhood data (Khan et al., 2001). Comparable classification accuracy shown in the applications and its flexibility render the MSVM a viable alternative to other classification methods. SUPPLEMENTARY INFORMATION: http://www.stat.ohio-state.edu/~yklee/msvm.htm     

Counterpoint to “Infrared cameras overestimate skin temperature during rewarming from cold exposure”

We evaluated the paper by Maley et al. (2020) in this journal “Infrared cameras overestimate skin temperature during rewarming from cold Exposure” for the evidence provided in support of its title and conclusions. Several methodological issues were identified as well as issues with the data interpretation: 1: Only one camera was used but a conclusion was made for ‘cameras’ in general; 2: The camera accuracy (±2 °C) is too low to do a meaningful comparison without using local reference values or an on-site calibrator with higher accuracy; 3: Thermistor measurements are taken as a gold standard while issues such as temperature gradients and taping are known; 4: Both methods are not compared at the same location and a natural, physiological, gradient may be present between their locations; 5. The impact of the sensor and tape on delays in the change of the underlying tissue temperature is not considered; 6: Only a single pixel was used for the analysis, which has been shown to be problematic.We concluded that the paper by Maley et al. does not provide conclusive evidence that infrared cameras overestimate skin temperature during rewarming from cold exposure.     

Predictive accuracy and explained variation in Cox regression

We suggest a new measure of the proportion of the variation of possibly censored survival times explained by a given proportional hazards model. The proposed measure, termed V, shares several favorable properties with an earlier V1 but also improves the handling of censoring. The statistic contrasts distance measures between individual 1/0 survival processes and fitted survival curves with and without covariate information. These distance measures, Dx and D, respectively, are themselves informative as summaries of absolute rather than relative predictive accuracy. We recommend graphical comparisons of survival curves for prognostic index groups to improve the understanding of obtained values for V, Dx, and D. Their use and interpretation is exemplified for a Yorkshire lung cancer study on survival. From this and an overview for several well-known clinical data sets, we show that the likely amount of relative or absolute predictive accuracy is often low even if there are highly significant and relatively strong prognostic factors.     

Modification of the bicinchoninic acid protein assay to eliminate lipid interference in determining lipoprotein protein content.

The bicinchoninic acid (BCA) assay method for the determination of protein has been investigated for its utility in measuring the protein content of plasma lipoproteins. Although other methods, principally those based on the method of Lowry et al. (1951, J. Biol. Chem. 193, 265-275) have been extensively used for this purpose, the tolerance of the BCA method to many commonly encountered detergents and buffers offers a definite advantage over the Lowry-based methods. In this study, lipoprotein protein values obtained by the BCA method were compared to a standard modification of the Lowry et al. procedure since this assay forms the basis of much of the relevant literature. The standard BCA assay was found to overestimate the protein content of very low density lipoprotein by approximately 70% and low density lipoprotein by approximately 30%; high density lipoprotein values compared favorably. Overestimations by the BCA assay paralleled the relative phospholipid content of the lipoprotein fractions. This apparent lipid effect was eliminated by the addition of 2% sodium dodecyl sulfate to samples prior to the analysis. In the presence of this detergent, BCA assay measurements for these three lipoprotein fractions were 97, 90, and 98%, respectively, of the reference assay values.     

Breast cancer: integrating the patient with her genome

Increasingly, gene expression data are becoming the currency of the realm in assessing disease prognosis. This has been especially evident in cancer, particularly those malignancies for which tumor samples are fairly accessible and understanding prognostic factors has clear implications for treatment decisions. Recently, Pittman et al. demonstrated substantially increased accuracy of personalized disease outcome prediction in breast cancer by integrating gene-expression profile data with traditional clinical risk factors in a set of 158 breast cancer patients.     

Comparison of estimation accuracy of body density between different hydrostatics weighing methods without head submersion

This study compared the accuracy of body density (Db) estimation methods using hydrostatic weighing without complete head submersion (HW(withoutHS)) of Donnelly et al. (1988) and Donnelly and Sintek (1984) as referenced to Goldman and Buskirk's approach (1961). Donnelly et al.'s method estimates Db from a regression equation using HW(withoutHS), moreover, Donnelly and Sintek's method estimates it from HW(withoutHS) and head anthropometric variables. Fifteen Japanese males (173.8+/-4.5 cm, 63.6+/-5.4 kg, 21.2+/-2.8 years) and fifteen females (161.4+/-5.4 cm, 53.8+/-4.8 kg, 21.0+/-1.4 years) participated in this study. All the subjects were measured for head length, width and HWs under the two conditions of with and without head submersion. In order to examine the consistency of estimation values of Db, the correlation coefficients between the estimation values and the reference (Goldman and Buskirk, 1961) were calculated. The standard errors of estimation (SEE) were calculated by regression analysis using a reference value as a dependent variable and estimation values as independent variables. In addition, the systematic errors of two estimation methods were investigated by the Bland-Altman technique (Bland and Altman, 1986). In the estimation, Donnelly and Sintek's equation showed a high relationship with the reference (r=0.960, p<0.01), but had more differences from the reference compared with Donnelly et al.'s equation. Further studies are needed to develop new prediction equations for Japanese considering sex and individual differences in head anthropometry.     

New basic discoveries and frontiers in diagnosis, prognosis and prediction of response to therapy in human thyroid, urinary bladder, and prostate tumors

The fourth edition of this workshop mainly focused on three different human oncotypes, which included thyroid, urinary bladder, and prostate tumors as clinical models to gain new basic knowledge on tumor diagnosis, prognosis, and treatment. At the previous editions (Giordano et al., 2000, J Cell Physiol 183:284-287; Giordano et al., 2001, J Cell Physiol 188:274-280; Giordano et al., 2002, J Cell Physiol 191:362-365), leaders in the fields of pathology, clinical oncology, and basic research presented and discussed the most recent and prevalent findings in such neoplasms from a basic and clinical perspective. A concept that has been widely proposed is that the analysis of intrinsic biological factors displayed by primary tumors may be a valid method for diagnosing different neoplasias and for measuring both their aggressiveness and response to therapy. To date, however, no single prognostic factor, such as oncogenes, suppressor genes, or genes involved in the control of the cell cycle and/or apoptosis has yet proven to be potent enough to be used in clinical practice as a prognostic and predictive factor. The new possibility to simultaneously analyze the expression of the complete repertoire of human genes and a large number of proteins could offer a new scenario in tumor classification, allowing for the formulation of a list of genes able to define a "signature" of tumor outcome. Moreover, starting from data obtained from biomolecular tumor analyses, it has been demonstrated that with this approach, it is also possible to design future therapeutic strategies.     

Developing optimal prediction models for cancer classification using gene expression data

Microarrays can provide genome-wide expression patterns for various cancers, especially for tumor sub-types that may exhibit substantially different patient prognosis. Using such gene expression data, several approaches have been proposed to classify tumor sub-types accurately. These classification methods are not robust, and often dependent on a particular training sample for modelling, which raises issues in utilizing these methods to administer proper treatment for a future patient. We propose to construct an optimal, robust prediction model for classifying cancer sub-types using gene expression data. Our model is constructed in a step-wise fashion implementing cross-validated quadratic discriminant analysis. At each step, all identified models are validated by an independent sample of patients to develop a robust model for future data. We apply the proposed methods to two microarray data sets of cancer: the acute leukemia data by Golub et al. and the colon cancer data by Alon et al. We have found that the dimensionality of our optimal prediction models is relatively small for these cases and that our prediction models with one or two gene factors outperforms or has competing performance, especially for independent samples, to other methods based on 50 or more predictive gene factors. The methodology is implemented and developed by the procedures in R and Splus. The source code can be obtained at http://hesweb1.med.virginia.edu/bioinformatics.     

Robust generation of hepatocyte-like cells from human embryonic stem cell populations

Despite progress in modelling human drug toxicity, many compounds fail during clinical trials due to unpredicted side effects. The cost of clinical studies are substantial, therefore it is essential that more predictive toxicology screens are developed and deployed early on in drug development (Greenhough et al 2010). Human hepatocytes represent the current gold standard model for evaluating drug toxicity, but are a limited resource that exhibit variable function. Therefore, the use of immortalised cell lines and animal tissue models are routinely employed due to their abundance. While both sources are informative, they are limited by poor function, species variability and/or instability in culture (Dalgetty et al 2009). Pluripotent stem cells (PSCs) are an attractive alternative source of human hepatocyte like cells (HLCs) (Medine et al 2010). PSCs are capable of self renewal and differentiation to all somatic cell types found in the adult and thereby represent a potentially inexhaustible source of differentiated cells. We have developed a procedure that is simple, highly efficient, amenable to automation and yields functional human HLCs (Hay et al 2008 ; Fletcher et al 2008 ; Hannoun et al 2010 ; Payne et al 2011 and Hay et al 2011). We believe our technology will lead to the scalable production of HLCs for drug discovery, disease modeling, the construction of extra-corporeal devices and possibly cell based transplantation therapies.     

Survival model predictive accuracy and ROC curves

The predictive accuracy of a survival model can be summarized using extensions of the proportion of variation explained by the model, or R2, commonly used for continuous response models, or using extensions of sensitivity and specificity, which are commonly used for binary response models. In this article we propose new time-dependent accuracy summaries based on time-specific versions of sensitivity and specificity calculated over risk sets. We connect the accuracy summaries to a previously proposed global concordance measure, which is a variant of Kendall's tau. In addition, we show how standard Cox regression output can be used to obtain estimates of time-dependent sensitivity and specificity, and time-dependent receiver operating characteristic (ROC) curves. Semiparametric estimation methods appropriate for both proportional and nonproportional hazards data are introduced, evaluated in simulations, and illustrated using two familiar survival data sets.     

Importance sampling method of correction for multiple testing in affected sib-pair linkage analysis

Using the Genetic Analysis Workshop 13 simulated data set, we compared the technique of importance sampling to several other methods designed to adjust p-values for multiple testing: the Bonferroni correction, the method proposed by Feingold et al., and na?ve Monte Carlo simulation. We performed affected sib-pair linkage analysis for each of the 100 replicates for each of five binary traits and adjusted the derived p-values using each of the correction methods. The type I error rates for each correction method and the ability of each of the methods to detect loci known to influence trait values were compared. All of the methods considered were conservative with respect to type I error, especially the Bonferroni method. The ability of these methods to detect trait loci was also low. However, this may be partially due to a limitation inherent in our binary trait definitions.     

Feature ranking based on synergy networks to identify prognostic markers in DPT-1

Interaction among different risk factors plays an important role in the development and progress of complex disease, such as diabetes. However, traditional epidemiological methods often focus on analyzing individual or a few ‘essential’ risk factors, hopefully to obtain some insights into the etiology of complex disease. In this paper, we propose a systematic framework for risk factor analysis based on a synergy network, which enables better identification of potential risk factors that may serve as prognostic markers for complex disease. A spectral approximate algorithm is derived to solve this network optimization problem, which leads to a new network-based feature ranking method that improves the traditional feature ranking by taking into account the pairwise synergistic interactions among risk factors in addition to their individual predictive power. We first evaluate the performance of our method based on simulated datasets, and then, we use our method to study immunologic and metabolic indices based on the Diabetes Prevention Trial-Type 1 (DPT-1) study that may provide prognostic and diagnostic information regarding the development of type 1 diabetes. The performance comparison based on both simulated and DPT-1 datasets demonstrates that our network-based ranking method provides prognostic markers with higher predictive power than traditional analysis based on individual factors.     

Formulas predicting survival in patients with heart transplantation

Kirklin et al. (J Heart Transpl, 7 (1988) 331–336) reported survival data in 132 patients who underwent heart transplantation. Survival was evaluated by using the product-limit method of Kaplan-Meier and maximum likelihood method. In addition, the effect of pulmonary vascular resistance on survival was estimated by using multivariate analysis. A microcomputer program in BASIC for predicting the survival probability after transplantation in patients with heart transplantation is designed. The formula used in this program is derived from the survival data reported by Kirklin et al. (J Heart Transpl, 7 (1988) 331–336). A mathematical model of the ‘probacent’-probability equation and a computer program previously published by the author are employed in this study. Analysis of the computer-assisted predicted values and the data reported by Kirklin et al. (J Heart Transpl, 7 (1988) 331–336) indicates that the program is accurate and reliable with a complete agreement in expressing survival probability as a function of time after heart transplantation. The computer-assisted predictive formula can determine the relationship between the time and the survival probability and may be of value for prognostic evaluation of patients. The computer-assisted mathematical model of the ‘probacent’-probability equation may be proposed as a general approximation method to make useful predictions of probable outcomes in various biomedical phenomena.     

HLA class I nucleotide sequences, 1992

The HLA class I sequences included in this compilation are taken from publications listed in the papers: Nomenclature for factors of the HLA system, 1991 (Bodmer et al. 1992); Nomenclature for factors of the HLA system, 1990 (Bodmer et al. 1991); and Nomenclature for factors of the HLA system, 1989 (Bodmer et al. 1990). Due to the increased number of sequences we have only included sequences for exons 2, 3, and 4 in this compilation. Where discrepancies have arisen between reported sequences, the original authors have been contacted where possible, and necessary amendments to published sequences have been incorporated into this alignment. Future sequencing may identify errors in this list and we would welcome any evidence that helps to maintain the accuracy of this compilation. In the sequence alignments, identify between nucleotides is indicated by a hyphen (-). An unavailable sequence is indicated by a period (.). Gaps in the sequence are inserted to maintain the alignment between different alleles showing variation in amino acid number. *** DIRECT SUPPORT *** A4903038 00002