Estimating covariate‐adjusted measures of diagnostic accuracy based on pooled biomarker assessments

There is a need for epidemiological and medical researchers to identify new biomarkers (biological markers) that are useful in determining exposure levels and/or for the purposes of disease detection. Often this process is stunted by high testing costs associated with evaluating new biomarkers. Traditionally, biomarker assessments are individually tested within a target population. Pooling has been proposed to help alleviate the testing costs, where pools are formed by combining several individual specimens. Methods for using pooled biomarker assessments to estimate discriminatory ability have been developed. However, all these procedures have failed to acknowledge confounding factors. In this paper, we propose a regression methodology based on pooled biomarker measurements that allow the assessment of the discriminatory ability of a biomarker of interest. In particular, we develop covariate‐adjusted estimators of the receiver‐operating characteristic curve, the area under the curve, and Youden's index. We establish the asymptotic properties of these estimators and develop inferential techniques that allow one to assess whether a biomarker is a good discriminator between cases and controls, while controlling for confounders. The finite sample performance of the proposed methodology is illustrated through simulation. We apply our methods to analyze myocardial infarction (MI) data, with the goal of determining whether the pro‐inflammatory cytokine interleukin‐6 is a good predictor of MI after controlling for the subjects' cholesterol levels.     

Capturing the population structure of microparasites: using ITS‐sequence data and a pooled DNA approach

The internal transcribed spacer (ITS) region of nuclear ribosomal DNA is a common marker not only for the molecular identification of different taxa and strains, but also for the analysis of population structure of wild microparasite communities. Importantly, the multicopy nature of this region allows the amplification of low‐quantity samples of the target DNA, a common problem in studies on unicellular, unculturable microparasites. We analysed ITS sequences from the protozoan parasite Caullerya mesnili (class Ichthyosporea) infecting waterflea (Daphnia) hosts, across several host population samples. We showed that analysing representative ITS‐types [as identified by statistical parsimony network analysis (SPN)] is a suitable method to address relevant polymorphism. The spatial patterns were consistent regardless of whether parasite DNA was extracted from individual hosts or pooled host samples. Remarkably, the efficiency in detecting different sequence types was even higher after sample pooling. As shown by simulations, an easily manageable number of sequences from pooled DNA samples are sufficient to resolve the spatial population structure in this system. In summary, the ITS region analysed from pooled DNA samples can provide valuable insights into the spatial and temporal dynamics of microparasites. Moreover, the application of SPN analysis is a good alternative to the well‐established neighbour‐joining method (NJ) for the identification of representative ITS‐types. SPN can even outperform NJ by joining most of the singleton sequences to representative sequence clusters.     

Association of T174M polymorphism of angiotensinogen gene with essential hypertension: A meta-analysis

The association between T174M polymorphism of angiotensinogen gene and essential hypertension risk remains controversial. We herein performed a meta-analysis to achieve a reliable estimation of their relationship. All the studies published up to May 2013 on the association between T174M polymorphism and essential hypertension risk were identified by searching the electronic repositories PubMed, MEDLINE and EMBASE, Springer, Elsevier Science Direct, Cochrane Library and Google Scholar. Data were extracted and pooled odds ratios (ORs) with 95% confidence intervals (95% CIs) were calculated. Ultimately, nine eligible studies, including 2188 essential hypertension cases and 2459 controls, were enrolled in this meta-analysis. No significant associations were found under the overall ORs for M-allele comparison (M vs. T, pooled OR 0.92, 95% CI 0.62–1.37), MM vs. TT (pooled OR 0.86, 95% CI 0.29–2.51), TM vs. TT n (pooled OR 0.91, 95% CI 0.63–1.32), recessive model (MM vs. TT+TM, pooled OR 0.89, 95% CI 0.35–2.30), dominant model (MM+TM vs. TT, pooled OR 0.91, 95% CI 0.60–1.38) between T174M polymorphism and risk for essential hypertension. This meta-analysis suggested that the T174M polymorphism of the angiotensinogen gene might not be associated with the susceptibility of essential hypertension in Asian or European populations.     

From heterogeneous healthcare data to disease-specific biomarker networks: A hierarchical Bayesian network approach

In this work, we introduce an entirely data-driven and automated approach to reveal disease-associated biomarker and risk factor networks from heterogeneous and high-dimensional healthcare data. Our workflow is based on Bayesian networks, which are a popular tool for analyzing the interplay of biomarkers. Usually, data require extensive manual preprocessing and dimension reduction to allow for effective learning of Bayesian networks. For heterogeneous data, this preprocessing is hard to automatize and typically requires domain-specific prior knowledge. We here combine Bayesian network learning with hierarchical variable clustering in order to detect groups of similar features and learn interactions between them entirely automated. We present an optimization algorithm for the adaptive refinement of such group Bayesian networks to account for a specific target variable, like a disease. The combination of Bayesian networks, clustering, and refinement yields low-dimensional but disease-specific interaction networks. These networks provide easily interpretable, yet accurate models of biomarker interdependencies. We test our method extensively on simulated data, as well as on data from the Study of Health in Pomerania (SHIP-TREND), and demonstrate its effectiveness using non-alcoholic fatty liver disease and hypertension as examples. We show that the group network models outperform available biomarker scores, while at the same time, they provide an easily interpretable interaction network.     

A flexible approach to measurement error correction in case-control studies

SUMMARY: We investigate the use of prospective likelihood methods to analyze retrospective case-control data where some of the covariates are measured with error. We show that prospective methods can be applied and the case-control sampling scheme can be ignored if one adequately models the distribution of the error-prone covariates in the case-control sampling scheme. Indeed, subject to this, the prospective likelihood methods result in consistent estimates and information standard errors are asymptotically correct. However, the distribution of such covariates is not the same in the population and under case-control sampling, dictating the need to model the distribution flexibly. In this article, we illustrate the general principle by modeling the distribution of the continuous error-prone covariates using the skewnormal distribution. The performance of the method is evaluated through simulation studies, which show satisfactory results in terms of bias and coverage. Finally, the method is applied to the analysis of two data sets which refer, respectively, to a cholesterol study and a study on breast cancer.     

Use of summary measures to adjust for informative missingness in repeated measures data with random effects

We discuss how to apply the conditional informative missing model of Wu and Bailey (1989, Biometrics 45, 939-955) to the setting where the probability of missing a visit depends on the random effects of the primary response in a time-dependent fashion. This includes the case where the probability of missing a visit depends on the true value of the primary response. Summary measures for missingness that are weighted sums of the indicators of missed visits are derived for these situations. These summary measures are then incorporated as covariates in a random effects model for the primary response. This approach is illustrated by analyzing data collected from a trial of heroin addicts where missed visits are informative about drug test results. Simulations of realistic experiments indicate that these time-dependent summary measures also work well under a variety of informative censoring models. These summary measures can achieve large reductions in estimation bias and mean squared errors relative to those obtained by using other summary measures.     

Cohort case-control design and analysis for clustered failure-time data

Cohort case-control design is an efficient and economical design to study risk factors for disease incidence or mortality in a large cohort. In the last few decades, a variety of cohort case-control designs have been developed and theoretically justified. These designs have been exclusively applied to the analysis of univariate failure-time data. In this work, a cohort case-control design adapted to multivariate failure-time data is developed. A risk set sampling method is proposed to sample controls from nonfailures in a large cohort for each case matched by failure time. This method leads to a pseudolikelihood approach for the estimation of regression parameters in the marginal proportional hazards model (Cox, 1972, Journal of the Royal Statistical Society, Series B 34, 187-220), where the correlation structure between individuals within a cluster is left unspecified. The performance of the proposed estimator is demonstrated by simulation studies. A bootstrap method is proposed for inferential purposes. This methodology is illustrated by a data example from a child vitamin A supplementation trial in Nepal (Nepal Nutrition Intervention Project-Sarlahi, or NNIPS).     

A pooling approach to detect signatures of selective sweeps in genome scans using microsatellites

We have evaluated a pooling approach that can reduce the number of polymerase chain reactions in a screen for selective sweeps by more than an order of magnitude. We show that the complex peak pattern that results from pooling of all samples from a given population is a faithful reflection of the composite pattern of the individual alleles, although with an under‐representation of the larger alleles. Candidate loci for selective sweeps can be identified by visual inspection of the pool patterns. We have also implemented a software tool, which can find suitable microsatellite loci in the vicinity of annotated genes.     

A generalized calibrated Bayesian hierarchical modeling approach to basket trials with multiple endpoints

A basket trial simultaneously evaluates a treatment in multiple cancer subtypes, offering an effective way to accelerate drug development in multiple indications. Many basket trials are designed and monitored based on a single efficacy endpoint, primarily the tumor response. For molecular targeted or immunotherapy agents, however, a single efficacy endpoint cannot adequately characterize the treatment effect. It is increasingly important to use more complex endpoints to comprehensively assess the risk–benefit profile of such targeted therapies. We extend the calibrated Bayesian hierarchical modeling approach to monitor phase II basket trials with multiple endpoints. We propose two generalizations, one based on the latent variable approach and the other based on the multinomial–normal hierarchical model, to accommodate different types of endpoints and dependence assumptions regarding information sharing. We introduce shrinkage parameters as functions of statistics measuring homogeneity among subgroups and propose a general calibration approach to determine the functional forms. Theoretical properties of the generalized hierarchical models are investigated. Simulation studies demonstrate that the monitoring procedure based on the generalized approach yields desirable operating characteristics.     

Amyotrophic lateral sclerosis and occupational exposure to electromagnetic fields

In an hypothesis-generating case-control study of amyotrophic lateral sclerosis, lifetime occupational histories were obtained. The patients (n = 28) were clinic based. The occupational exposure of interest in this report is electromagnetic fields (EMFs). This is the first and so far the only exposure analyzed in this study. Occupational exposure up to 2 years prior to estimated disease symptom onset was used for construction of exposure indices for cases. Controls (n = 32) were blood and nonblood relatives of cases. Occupational exposure for controls was through the same age as exposure for the corresponding cases. Twenty (71%) cases and 28 (88%) controls had at least 20 years of work experience covering the exposure period. The occupational history and task data were used to classify blindly each occupation for each subject as having high, medium/high, medium, medium/low, or low EMF exposure, based primarily on data from an earlier and unrelated study designed to obtain occupational EMF exposure information on workers in “electrical” and “nonelectrical” jobs. By using the length of time each subject spent in each occupation through the exposure period, two indices of exposure were constructed: total occupational exposure (E₁) and average occupational exposure (E₂). For cases and controls with at least 20 years of work experience, the odds ratio (OR) for exposure at the 75th percentile of the E₁ case exposure data relative to minimum exposure was 7.5 (P < 0.02; 95% Cl, 1.4–38.1) and the corresponding OR for E₂ was 5.5 (P < 0.02; 95% CI, 1.3–22.5). For all cases and controls, the ORs were 2.5 (P < 0.1; 95% CI, 0.9–8.1) for E₁ and 2.3 (P = 0.12; 95% CI, 0.8–6.6) for E₂. This study should be considered an hypothesis-generating study. Larger studies, using incident cases and improved exposure assessment, should be undertaken. Bioelectromagnetics 18:28–35, 1997. © 1997 Wiley-Liss, Inc.     

A coalescence approach to gene conversion

In this paper we develop a coalescent model with intralocus gene conversion. Such models are of increasing importance in the analysis of intralocus variability and linkage disequilibrium. We derive the distribution of the waiting time until a gene conversion event occurs in a sample in terms of the distribution of the length of the transferred segment, zeta. We do not assume any specific form of the distribution of zeta. Further, given that a gene conversion event occurs we find the distribution of (sigma, tau), the end points of the transferred segment and derive results on correlations between local trees in positions chi(1) and chi(2). Among other results we show that the correlation between the branch lengths of two local trees in the coalescent with gene conversion (and no recombination) decreases toward a nonzero constant when the distance between chi(1) and chi(2) increases. Finally, we show that a model including both recombination and gene conversion might account for the lack of intralocus associations found in, e.g., Drosophila melanogaster.     

Monitoring futility and efficacy in phase II trials with Bayesian posterior distributions—A calibration approach

A multistage single arm phase II trial with binary endpoint is considered. Bayesian posterior probabilities are used to monitor futility in interim analyses and efficacy in the final analysis. For a beta‐binomial model, decision rules based on Bayesian posterior probabilities are converted to “traditional” decision rules in terms of number of responders among patients observed so far. Analytical derivations are given for the probability of stopping for futility and for the probability to declare efficacy. A workflow is presented on how to select the parameters specifying the Bayesian design, and the operating characteristics of the design are investigated. It is outlined how the presented approach can be transferred to statistical models other than the beta‐binomial model.     

A shrinkage approach for estimating a treatment effect using intermediate biomarker data in clinical trials

In clinical trials, a biomarker (S ) that is measured after randomization and is strongly associated with the true endpoint (T) can often provide information about T and hence the effect of a treatment (Z ) on T. A useful biomarker can be measured earlier than T and cost less than T. In this article, we consider the use of S as an auxiliary variable and examine the information recovery from using S for estimating the treatment effect on T, when S is completely observed and T is partially observed. In an ideal but often unrealistic setting, when S satisfies Prentice's definition for perfect surrogacy, there is the potential for substantial gain in precision by using data from S to estimate the treatment effect on T. When S is not close to a perfect surrogate, it can provide substantial information only under particular circumstances. We propose to use a targeted shrinkage regression approach that data-adaptively takes advantage of the potential efficiency gain yet avoids the need to make a strong surrogacy assumption. Simulations show that this approach strikes a balance between bias and efficiency gain. Compared with competing methods, it has better mean squared error properties and can achieve substantial efficiency gain, particularly in a common practical setting when S captures much but not all of the treatment effect and the sample size is relatively small. We apply the proposed method to a glaucoma data example.     

Predicting time to peak occurrence of insect life-stages usingregression models calibrated from stage-frequency data and ancillary stage-mortality data

1 Integrated Pest Management programmes often require predictions of peak occurrence of particular insect life‐stages to optimize the timing of population monitoring and control operations. 2 Given a known or estimated starting time for a synchronously developing pest population, predictive models estimated from stage‐frequency data alone can only predict the times of peak occurrence assuming a constant mortality rate across stages. 3 Here, continuation ratio regression models of relative stage frequencies estimated from stage‐frequency data are combined with a stage‐specific model of mortality estimated from ancillary mortality data to allow prediction of time of peak occurrence. 4 To calculate time of peak occurrence new mathematical derivations are given for continuation ratio models. 5 The models are used to predict the time of peak occurrence in degree‐day units for each of the first to third larval instars of the Tasmanian Eucalyptus leaf beetle Chrysophtharta bimaculata (Olivier) (Coleoptera: Chrysomelidae), a serious defoliator of Eucalyptus regnans and E. nitens plantations in Tasmania.