Search

Gliomas are characterized by a malignant phenotype with proliferation, cell cycle arrest and invasion. To explore the biological consequences of epigenetically regulated miRNAs, we performed a microarray-based screening (whose expression was affected by 5-AZA treatment) followed by bisulfite sequencing validation. We found that miR-134 as an epigenetically regulated suppressor gene with prognostic value in gliomas. MicroRNA-134 was downregulated in high-grade gliomas, especially in GBM samples. Functional studies in vitro and in vivo in mouse models showed that overexpression of miR-134 was sufficient to reduce cell cycle arrest, cell proliferation and invasion. Target analysis and functional assays correlated the malignant phenotype with miR-134 target gene KRAS, an established upstream regulator of ERK and AKT pathways. Overall, our results highlighted a role for miR-134 in explaining the malignant phenotype of gliomas and suggested its relevance as a target to develop for early diagnostics and therapy. [ABSTRACT FROM AUTHOR]

Measuring the phylogenetic information content of data has a long history in systematics. Here we explore a Bayesian approach to information content estimation. The entropy of the posterior distribution compared with the entropy of the prior distribution provides a natural way to measure information content. If the data have no information relevant to ranking tree topologies beyond the information supplied by the prior, the posterior and prior will be identical. Information in data discourages consideration of some hypotheses allowed by the prior, resulting in a posterior distribution that is more concentrated (has lower entropy) than the prior. We focus on measuring information about tree topology using marginal posterior distributions of tree topologies. We show that both the accuracy and the computational efficiency of topological information content estimation improve with use of the conditional clade distribution, which also allows topological information content to be partitioned by clade. We explore two important applications of our method: providing a compelling definition of saturation and detecting conflict among data partitions that can negatively affect analyses of concatenated data. [ABSTRACT FROM AUTHOR]

We investigate diagnostic measures for assessing the influence of observations and model misspecification on the Cox regression model when there are missing covariate data. Our diagnostics include case-deletion measures, conditional martingale residuals, and score residuals. The Q-distance is introduced to examine the effects of deleting individual observations on the estimates of finite- and infinite-dimensional parameters. Conditional martingale residuals are used to construct goodness-of-fit statistics for testing misspecification of the model assumptions. A resampling method is developed to approximate the p-values of the goodness-of-fit statistics. We conduct simulation studies to evaluate our methods, and analyse a real dataset to illustrate their use. [ABSTRACT FROM AUTHOR]

Expressed sequence tag (EST) sequencing is a one-pass sequencing reading of cloned cDNAs derived from a certain tissue. The frequency of unique tags among different unbiased cDNA libraries is used to infer the relative expression level of each tag. In this article, we propose a hierarchical multinomial model with a nonlinear Dirichlet prior for the EST data with multiple libraries and multiple types of tissues. A novel hierarchical prior is developed and the properties of the proposed prior are examined. An efficient Markov chain Monte Carlo algorithm is developed for carrying out the posterior computation. We also propose a new selection criterion for detecting which genes are differentially expressed between two tissue types. Our new method with the new gene selection criterion is demonstrated via several simulations to have low false negative and false positive rates. A real EST data set is used to motivate and illustrate the proposed method. [ABSTRACT FROM AUTHOR]

The marginal likelihood is commonly used for comparing different evolutionary models in Bayesian phylogenetics and is the central quantity used in computing Bayes Factors for comparing model fit. A popular method for estimating marginal likelihoods, the harmonic mean (HM) method, can be easily computed from the output of a Markov chain Monte Carlo analysis but often greatly overestimates the marginal likelihood. The thermodynamic integration (TI) method is much more accurate than the HM method but requires more computation. In this paper, we introduce a new method, steppingstone sampling (SS), which uses importance sampling to estimate each ratio in a series (the “stepping stones”) bridging the posterior and prior distributions. We compare the performance of the SS approach to the TI and HM methods in simulation and using real data. We conclude that the greatly increased accuracy of the SS and TI methods argues for their use instead of the HM method, despite the extra computation needed. [ABSTRACT FROM PUBLISHER]

Background We investigated whether androgen deprivation therapy (ADT) use is associated with an increased risk of death from cardiovascular causes in patients treated for localized prostate cancer. Methods From the Cancer of the Prostate Strategic Urologic Research Endeavor database, data on 3262 patients treated with radical prostatectomy and 1630 patients treated with external beam radiation therapy, brachytherapy, or cryotherapy for localized prostate cancer were included in this analysis. Competing risks regression analyses were performed to assess whether use of ADT was associated with a shorter time to death from cardiovascular causes after controlling for age (as a continuous variable) and the presence of baseline cardiovascular disease risk factors. All tests for statistical significance were two-sided. Results The median follow-up time was 3.8 years (range = 0.1-11.3 years). Among the 1015 patients who received ADT, the median duration of ADT use was 4.1 months (range = 1.0-32.9 months). In a competing risks regression analysis that controlled for age and risk factors for cardiovascular disease, both ADT use (adjusted hazard ratio [HR] = 2.6; 95% confidence interval [CI] = 1.4 to 4.7; P = .002) and age (adjusted HR = 1.07; 95% CI = 1.02 to 1.1; P=.003) were associated with statistically significantly increased risks of death from cardiovascular causes in patients treated with radical prostatectomy. Among patients 65 years or older treated with radical prostatectomy, the 5-year cumulative incidence of cardiovascular death was 5.5% (95% CI = 1.2% to 9.8%) in those who received ADT and 2.0% (95% CI = 1.1% to 3.0%) in those who did not. Among patients 65 years or older treated with external beam radiation therapy, brachytherapy, or cryotherapy, ADT use was associated with a higher cumulative incidence of death from cardiovascular causes, but the difference did not reach statistical significance. Conclusions The use of ADT appears to be associated with an increased risk of death from cardiovascular causes in patients undergoing radical prostatectomy for localized prostate cancer. [ABSTRACT FROM AUTHOR]

In this paper, we carry out an in-depth theoretical investigation of Bayesian inference for the Cox regression model. We establish necessary and sufficient conditions for posterior propriety of the regression coefficient, β, in Cox's partial likelihood, which can be obtained as the limiting marginal posterior distribution of β through the specification of a gamma process prior for the cumulative baseline hazard and a uniform improper prior for β. We also examine necessary and sufficient conditions for posterior propriety of the regression coefficients, β, using full likelihood Bayesian approaches in which a gamma process prior is specified for the cumulative baseline hazard. We examine characterisation of posterior propriety under completely observed data settings as well as for settings involving missing covariates. Latent variables are introduced to facilitate a straightforward Gibbs sampling scheme in the Bayesian computation. A real dataset is presented to illustrate the proposed methodology. [ABSTRACT FROM PUBLISHER]

We propose Bayesian methods for estimating parameters in generalized linear models (GLMs) with nonignorably missing covariate data. We show that when improper uniform priors are used for the regression coefficients, φ, of the multinomial selection model for the missing data mechanism, the resulting joint posterior will always be improper if (i) all missing covariates are discrete and an intercept is included in the selection model for the missing data mechanism, or (ii) at least one of the covariates is continuous and unbounded. This impropriety will result regardless of whether proper or improper priors are specified for the regression parameters, β, of the GLM or the parameters, α, of the covariate distribution. To overcome this problem, we propose a novel class of proper priors for the regression coefficients, φ, in the selection model for the missing data mechanism. These priors are robust and computationally attractive in the sense that inferences about β are not sensitive to the choice of the hyperparameters of the prior for φ and they facilitate a Gibbs sampling scheme that leads to accelerated convergence. In addition, we extend the model assessment criterion of (2004a, Biometrika 91, 45–63), called the weighted L measure, to GLMs and missing data problems as well as extend the deviance information criterion (DIC) of (2002, Journal of the Royal Statistical Society B 64, 583–639) for assessing whether the missing data mechanism is ignorable or nonignorable. A novel Markov chain Monte Carlo sampling algorithm is also developed for carrying out posterior computation. Several simulations are given to investigate the performance of the proposed Bayesian criteria as well as the sensitivity of the prior specification. Real datasets from a melanoma cancer clinical trial and a liver cancer study are presented to further illustrate the proposed methods. [ABSTRACT FROM AUTHOR]

Background: Whether the prostate-specific antigen (PSA) response to salvage hormonal therapy can act as an intermediate end point for prostate cancer-specific mortality (PCSM) remains unclear. Therefore, we evaluated whether PSA response, defined as the absolute value of the ratio of the rate of PSA change after salvage hormonal therapy to the rate of PSA change before salvage therapy, is associated with the time to PCSM following salvage hormonal therapy. Methods: A single-institution and two pooled multiinstitution databases containing baseline, treatment, and follow-up information on men who received salvage hormonal therapy for PSA failure following surgery or radiation therapy from January 1, 1988, to January 1, 2002, formed the study (n = 199) and validation cohorts (n = 1255), respectively. The ability of PSA response and its constituents (i.e., pre-salvage hormonal therapy PSA slope and post-salvage hormonal therapy PSA slope) to predict time to PCSM following salvage hormonal therapy was assessed using Cox regression analysis. For illustrative purposes, PSA response was analyzed as a dichotomous variable with a breakpoint for the ratio of PSA response of 1. All statistical tests were two-sided. Results: PSA response was statistically significantly associated with time to PCSM following salvage hormonal therapy in both the study (PCox = .0014) and validation (PCox < .001) cohorts; however, its constituents were not (pre-salvage hormonal therapy PSA slope: PCox-study = .97, PCox-validation = .57; post-salvage hormonal therapy PSA slope: PCox-study = .27, PCox-validation = .31). Patients with a PSA response that was less than or equal to 1 had a statistically significantly shorter time to PCSM than patients with a PSA response of greater than 1 in both the study (hazard ratio [HR] = 3.6, 95% confidence interval [CI] = 1.3 to 10.3; PCox = .01) and validation (HR = 12.8, 95% CI = 6.2 to 26.3; PCox < .001) cohorts. Conclusion: The PSA response to salvage hormonal therapy can serve as an intermediate end point for PCSM in patients with a rising PSA level following surgery or radiation therapy. [ABSTRACT FROM AUTHOR]

We propose a general Bayesian criterion for model assessment for categorical data called the weighted L measure, which is constructed from the posterior predictive distribution of the data. The measure is based on weighting the observations according to the sampling variance of their future response vector. The weight component in the weighted L measure plays the role of a penalty term in the criterion, in which a greater weight assigned to covariate values implies a greater penalty term on the dimension of the model. A detailed justification is provided for such a weighting procedure and several theoretical properties of the weighted L measure are presented for a wide variety of discrete data models. For these models, we examine properties of the weighted L measure, and show that it can perform better than the unweighted L measure in a variety of settings. In addition, we show that the weighted quadratic loss L measure is more attractive than the unweighted L measure and the deviance loss L measure for categorical data. Moreover, a calibration for the weighted L measure is motivated and proposed, which allows us to compare formally the L measure values of competing models. A detailed simulation study is presented to examine the performance of the weighted L measure, and it is compared to other established model‐selection methods. Finally, the method is applied to a real dataset using a bivariate ordinal response model. [ABSTRACT FROM AUTHOR]

Background: The relationship between prostate-specific antigen (PSA)-defined recurrence and prostate cancer-specific mortality remains unclear. Therefore, we evaluated the hypothesis that a short post-treatment PSA doubling time (PSA-DT) after radiation therapy is a surrogate end point for prostate cancer-specific mortality by analyzing two multiinstitutional databases. Methods: Baseline, treatment, and follow-up information was compiled on a cohort of 8669 patients with prostate cancer treated with surgery (5918 men) or radiation (2751 men) from January 1, 1988, through January 1, 2002, for localized or locally advanced, non-metastatic prostate cancer. We used a Cox regression analysis to test whether the post-treatment PSA-DT was a prognostic factor that was independent of treatment received. All statistical tests were two-sided. Results: The post-treatment PSA-DT was statistically significantly associated with time to prostate cancer-specific mortality and with time to all-cause mortality (ali P[sub Cox]<.001). However, the treatment received was not statistically significantly associated with time to prostate cancer-specific mortality after PSA-defined disease recurrence for patients with a PSA-DT of less than 3 months (P[sub Cox] = .90) and for patients with a PSA-DT of 3 months or more (P[sub Cox] = .28) when controlling for the specific value of the PSA-DT. Furthermore, after a PSA-defined recurrence, a PSA-DT of less than 3 months was statistically significantly associated with time to prostate cancer-specific mortality (median time = 6 years; hazard ratio = 19.6, 95% confidence interval = 12.5 to 30.9). Conclusion: A post-treatment PSADT of less than 3 months and the specific value of the posttreatment PSA-DT when it is 3 months or more appear to be surrogate end points for prostate cancer-specific mortality after surgery or radiation therapy. We recommend that consideration be given to initiating androgen suppression therapy at the time of a PSA-defined recurrence when the PSA-DT is less than 3 months to delay the imminent onset of metastatic bone disease. [ABSTRACT FROM AUTHOR]

We propose a method for estimating parameters in the generalised linear mixed model with nonignorable missing response data and with nonmonotone patterns of missing data in the response variable. We develop a Monte Carlo EM algorithm for estimating the parameters in the model via the Gibbs sampler. For the normal random effects model, we derive a novel analytical form for the E‐ and M‐steps, which is facilitated by integrating out the random effects. This form leads to a computationally feasible and extremely efficient Monte Carlo EM algorithm for computing maximum likelihood estimates and standard errors. In addition, we propose a very general joint multinomial model for the missing data indicators, which can be specified via a sequence of one‐dimensional conditional distributions. This multinomial model allows for an arbitrary correlation structure between the missing data indicators, and has the potential of reducing the number of nuisance parameters. Real datasets from the International Breast Cancer Study Group and an environmental study involving dyspnoea in cotton workers are presented to illustrate the proposed methods. [ABSTRACT FROM PUBLISHER]

A critical issue in modelling binary response data is the choice of the links. We introduce a new link based on the generalized t-distribution. There are two parameters in the generalized t-link: one parameter purely controls the heaviness of the tails of the link and the second parameter controls the scale of the link. Two major advantages are offered by the generalized t-links. First, a symmetric generalized t-link with an unknown shape parameter is much more identifiable than a Student t-link with unknown degrees of freedom and a known scale parameter. Secondly, skewed generalized t-links with both unknown shape and scale parameters provide much more flexible and improved skewed link regression models than the existing skewed links. Various theoretical properties and attractive features of the proposed links are examined and explored in detail. An efficient Markov chain Monte Carlo algorithm is developed for sampling from the posterior distribution. The deviance information criterion measure is used for guiding the choice of links. The proposed methodology is motivated and illustrated by prostate cancer data. [ABSTRACT FROM AUTHOR]