Efficient alternatives for Bayesian hypothesis tests in psychology

Bayesian hypothesis testing procedures have gained increased acceptance in recent years. A key advantage that Bayesian tests have over classical testing procedures is their potential to quantify information in support of true null hypotheses. Ironically, default implementations of Bayesian tests prevent the accumulation of strong evidence in favor of true null hypotheses because associated default alternative hypotheses assign a high probability to data that are most consistent with a null effect. We propose the use of "nonlocal" alternative hypotheses to resolve this paradox. The resulting class of Bayesian hypothesis tests permits more rapid accumulation of evidence in favor of both true null hypotheses and alternative hypotheses that are compatible with standardized effect sizes of most interest in psychology. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Global Estimate of Human Brucellosis Incidence

Brucellosis is a major public health concern worldwide, especially for persons living in resource-limited settings. Historically, an evidence-based estimate of the global annual incidence of human cases has been elusive. We used international public health data to fill this information gap through application of risk metrics to worldwide and regional at-risk populations. We performed estimations using 3 statistical models (weighted average interpolation, bootstrap resampling, and Bayesian inference) and considered missing information. An evidence-based conservative estimate of the annual global incidence is 2.1 million, significantly higher than was previously assumed. Our models indicate Africa and Asia sustain most of the global risk and cases, although areas within the Americas and Europe remain of concern. This study reveals that disease risk and incidence are higher than previously suggested and lie mainly within resource-limited settings. Clarification of both misdiagnosis and underdiagnosis is required because those factors will amplify case estimates.

A Hyperparameter-Free, Fast and Efficient Framework to Detect Clusters From Limited Samples Based on Ultra High-Dimensional Features

Clustering is a challenging problem in machine learning in which one attempts to group N objects into K₀ groups based on P features measured on each object. In this article, we examine the case where N ≪ P and K₀ is not known. Clustering in such high dimensional, small sample size settings has numerous applications in biology, medicine, the social sciences, clinical trials, and other scientific and experimental fields. Whereas most existing clustering algorithms either require the number of clusters to be known a priori or are sensitive to the choice of tuning parameters, our method does not require the prior specification of K₀ or any tuning parameters. This represents an important advantage for our method because training data are not available in the applications we consider (i.e., in unsupervised learning problems). Without training data, estimating K₀ and other hyperparameters-and thus applying alternative clustering algorithms-can be difficult and lead to inaccurate results. Our method is based on a simple transformation of the Gram matrix and application of the strong law of large numbers to the transformed matrix. If the correlation between features decays as the number of features grows, we show that the transformed feature vectors concentrate tightly around their respective cluster expectations in a low-dimensional space. This result simplifies the detection and visualization of the unknown cluster configuration. We illustrate the algorithm by applying it to 32 benchmarked microarray datasets, each containing thousands of genomic features measured on a relatively small number of tissue samples. Compared to 21 other commonly used clustering methods, we find that the proposed algorithm is faster and twice as accurate in determining the "best" cluster configuration.

A Modified Sequential Probability Ratio Test

We describe a modified sequential probability ratio test that can be used to reduce the average sample size required to perform statistical hypothesis tests at specified levels of significance and power. Examples are provided for z tests, t tests, and tests of binomial success probabilities. A description of a software package to implement the test designs is provided. We compare the sample sizes required in fixed design tests conducted at 5% significance levels to the average sample sizes required in sequential tests conducted at 0.5% significance levels, and we find that the two sample sizes are approximately equal.

On the Existence of Uniformly Most Powerful Bayesian Tests With Application to Non-Central Chi-Squared Tests

Uniformly most powerful Bayesian tests (UMPBT's) are an objective class of Bayesian hypothesis tests that can be considered the Bayesian counterpart of classical uniformly most powerful tests. Because the rejection regions of UMPBT's can be matched to the rejection regions of classical uniformly most powerful tests (UMPTs), UMPBT's provide a mechanism for calibrating Bayesian evidence thresholds, Bayes factors, classical significance levels and p-values. The purpose of this article is to expand the application of UMPBT's outside the class of exponential family models. Specifically, we introduce sufficient conditions for the existence of UMPBT's and propose a unified approach for their derivation. An important application of our methodology is the extension of UMPBT's to testing whether the non-centrality parameter of a chi-squared distribution is zero. The resulting tests have broad applicability, providing default alternative hypotheses to compute Bayes factors in, for example, Pearson's chi-squared test for goodness-of-fit, tests of independence in contingency tables, and likelihood ratio, score and Wald tests.

A pedigree-based prediction model identifies carriers of deleterious de novo mutations in families with Li-Fraumeni syndrome

De novo mutations (DNMs) are increasingly recognized as rare disease causal factors. Identifying DNM carriers will allow researchers to study the likely distinct molecular mechanisms of DNMs. We developed Famdenovo to predict DNM status (DNM or familial mutation [FM]) of deleterious autosomal dominant germline mutations for any syndrome. We introduce Famdenovo.TP53 for Li-Fraumeni syndrome (LFS) and analyze 324 LFS family pedigrees from four US cohorts: a validation set of 186 pedigrees and a discovery set of 138 pedigrees. The concordance index for Famdenovo.TP53 prediction was 0.95 (95% CI: [0.92, 0.98]). Forty individuals (95% CI: [30, 50]) were predicted as DNM carriers, increasing the total number from 42 to 82. We compared clinical and biological features of FM versus DNM carriers: (1) cancer and mutation spectra along with parental ages were similarly distributed; (2) ascertainment criteria like early-onset breast cancer (age 20–35 yr) provides a condition for an unbiased estimate of the DNM rate: 48% (23 DNMs vs. 25 FMs); and (3) hotspot mutation R248W was not observed in DNMs, although it was as prevalent as hotspot mutation R248Q in FMs. Furthermore, we introduce Famdenovo.BRCA for hereditary breast and ovarian cancer syndrome and apply it to a small set of family data from the Cancer Genetics Network. In summary, we introduce a novel statistical approach to systematically evaluate deleterious DNMs in inherited cancer syndromes. Our approach may serve as a foundation for future studies evaluating how new deleterious mutations can be established in the germline, such as those in TP53.

BAYESIAN VARIABLE SELECTION FOR SURVIVAL DATA USING INVERSE MOMENT PRIORS

Efficient variable selection in high dimensional cancer genomic studies is critical for discovering genes associated with specific cancer types and for predicting response to treatment. Censored survival data is prevalent in such studies. In this article we introduce a Bayesian variable selection procedure that uses a mixture prior composed of a point mass at zero and an inverse moment prior in conjunction with the partial likelihood defined by the Cox proportional hazard model. The procedure is implemented in the R package BVSNLP, which supports parallel computing and uses a stochastic search method to explore the model space. Bayesian model averaging is used for prediction. The proposed algorithm provides better performance than other variable selection procedures in simulation studies, and appears to provide more consistent variable selection when applied to actual genomic datasets.

GWASinlps: non-local prior based iterative SNP selection tool for genome-wide association studies

Motivation

Multiple marker analysis of the genome-wide association study (GWAS) data has gained ample attention in recent years. However, because of the ultra high-dimensionality of GWAS data, such analysis is challenging. Frequently used penalized regression methods often lead to large number of false positives, whereas Bayesian methods are computationally very expensive. Motivated to ameliorate these issues simultaneously, we consider the novel approach of using non-local priors in an iterative variable selection framework.

Results

We develop a variable selection method, named, iterative non-local prior based selection for GWAS, or GWASinlps, that combines, in an iterative variable selection framework, the computational efficiency of the screen-and-select approach based on some association learning and the parsimonious uncertainty quantification provided by the use of non-local priors. The hallmark of our method is the introduction of 'structured screen-and-select' strategy, that considers hierarchical screening, which is not only based on response-predictor associations, but also based on response-response associations and concatenates variable selection within that hierarchy. Extensive simulation studies with single nucleotide polymorphisms having realistic linkage disequilibrium structures demonstrate the advantages of our computationally efficient method compared to several frequentist and Bayesian variable selection methods, in terms of true positive rate, false discovery rate, mean squared error and effect size estimation error. Further, we provide empirical power analysis useful for study design. Finally, a real GWAS data application was considered with human height as phenotype.

Availability and implementation

An R-package for implementing the GWASinlps method is available at https://cran.r-project.org/web/packages/GWASinlps/index.html.

Supplementary information

Supplementary data are available at Bioinformatics online.

Functional Horseshoe Priors for Subspace Shrinkage

We introduce a new shrinkage prior on function spaces, called the functional horseshoe prior (fHS), that encourages shrinkage towards parametric classes of functions. Unlike other shrinkage priors for parametric models, the fHS shrinkage acts on the shape of the function rather than inducing sparsity on model parameters. We study the efficacy of the proposed approach by showing an adaptive posterior concentration property on the function. We also demonstrate consistency of the model selection procedure that thresholds the shrinkage parameter of the functional horseshoe prior. We apply the fHS prior to nonparametric additive models and compare its performance with procedures based on the standard horseshoe prior and several penalized likelihood approaches. We find that the new procedure achieves smaller estimation error and more accurate model selection than other procedures in several simulated and real examples. The supplementary material for this article, which contains additional simulated and real data examples, MCMC diagnostics, and proofs of the theoretical results, is available online.

Abstract 2406: A general probabilistic algorithm to predict de novo mutations in familial diseases as demonstrated in Li-Fraumeni Syndrome

Purpose:

Germline variants (mutations) occurring in the carrier but not in the carrier's parents are defined as de novo mutations (DNMs). DNMs have been increasingly recognized as causal factors for rare diseases. Given this important role, the ability to identify DNM carriers, scarcely distributed among all mutation carriers, will allow researchers to study the likely distinct molecular mechanisms of DNMs in these diseases. However, testing the mutation status of the parents is often impractical in retrospective studies due to lack of blood samples.

Methods:

To fill this need, using Li-Fraumeni syndrome (LFS) as a representative of a rare inherited syndrome, we developed a method called Famdenovo to predict the de novo status (de novo or familial) of germline mutations, such as in TP53. We have collected a total 324 LFS family pedigrees with confirmed TP53 germline mutations from four medical centers in the US, with 186 pedigrees having TP53 genetic testing results available in at least one full trio in the pedigree, serving as a validation set. We further apply Famdenovo to the remaining 138 families (discovery set) to identify individuals who may carry a de novo mutation in TP53.

Results:

In the validation set, the area under the ROC curve (AUC) of Famdenovo is 0.95 (95% CI: [0.92, 0.98]), suggesting excellent ability for discrimination, and the ratio of the observed to expected (OE) is 1.32 suggesting good concordance. In the discovery set, we predict that an additional 40 individuals (95% CI: [30, 50]) are DNM carriers, which increases the total number of DNM carriers to 82. The corresponding predicted DNM rate is 28.9%, also consistent with what is observed in the validation set. Across the validation and discovery sets, we observed similar distributions of ages-of-onset for DNM carriers at specific cancer sites: breast, brain, leukemia, osteosarcoma, soft tissue sarcoma, lung, and adrenal cortical carcinoma. Interestingly, Lung cancer only occurred in female DNM carriers. Among TP53 hotspot mutations, R248Q is most enriched in DNMs.

Conclusions:

Our new statistical method Famdenovo provides the probability of a person carrying a de novo germline mutation in rare inherited syndromes when the mutation status of both parents is not available. Famdenovo is freely available as an R package from http://bioinformatics.mdanderson.org/main/Famdenovo. The LFS family cohorts not only served as a validation set for the accuracy of Famdenovo in predicting DNMs, but also benefited from using Famdenovo to discover additional individuals carrying DNMs in TP53. The computer-based identification of DNM carriers in TP53, who are otherwise hidden in a wide population, enabled epidemiological interpretation of the cancer outcomes for DNM carriers. Famdenovo is a general tool and can be applied to other cancer genes where there is a good understanding of the penetrance of the associated disease phenotype.

Citation Format: Xuedong Pan, Fan Gao, Elissa B. Dodd, Jasmina Bojadzieva, Phuong L. Mai, Valen E. Johnson, Kristin Zelley, Kim E. Nichols, Judy E. Garber, Sharon A. Savage, Louise C. Strong, Wenyi Wang. A general probabilistic algorithm to predict de novo mutations in familial diseases as demonstrated in Li-Fraumeni Syndrome [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2406.

Evidence from marginally significant t statistics

This article examines the evidence contained in t statistics that are marginally significant in 5% tests. The bases for evaluating evidence are likelihood ratios and integrated likelihood ratios, computed under a variety of assumptions regarding the alternative hypotheses in null hypothesis significance tests. Likelihood ratios and integrated likelihood ratios provide a useful measure of the evidence in favor of competing hypotheses because they can be interpreted as representing the ratio of the probabilities that each hypothesis assigns to observed data. When they are either very large or very small, they suggest that one hypothesis is much better than the other in predicting observed data. If they are close to 1.0, then both hypotheses provide approximately equally valid explanations for observed data. I find that p-values that are close to 0.05 (i.e., that are "marginally significant") correspond to integrated likelihood ratios that are bounded by approximately 7 in two-sided tests, and by approximately 4 in one-sided tests. The modest magnitude of integrated likelihood ratios corresponding to p-values close to 0.05 clearly suggests that higher standards of evidence are needed to support claims of novel discoveries and new effects.

Scalable Bayesian Variable Selection Using Nonlocal Prior Densities in Ultrahigh-dimensional Settings

Bayesian model selection procedures based on nonlocal alternative prior densities are extended to ultrahigh dimensional settings and compared to other variable selection procedures using precision-recall curves. Variable selection procedures included in these comparisons include methods based on g-priors, reciprocal lasso, adaptive lasso, scad, and minimax concave penalty criteria. The use of precision-recall curves eliminates the sensitivity of our conclusions to the choice of tuning parameters. We find that Bayesian variable selection procedures based on nonlocal priors are competitive to all other procedures in a range of simulation scenarios, and we subsequently explain this favorable performance through a theoretical examination of their consistency properties. When certain regularity conditions apply, we demonstrate that the nonlocal procedures are consistent for linear models even when the number of covariates p increases sub-exponentially with the sample size n. A model selection procedure based on Zellner's g-prior is also found to be competitive with penalized likelihood methods in identifying the true model, but the posterior distribution on the model space induced by this method is much more dispersed than the posterior distribution induced on the model space by the nonlocal prior methods. We investigate the asymptotic form of the marginal likelihood based on the nonlocal priors and show that it attains a unique term that cannot be derived from the other Bayesian model selection procedures. We also propose a scalable and efficient algorithm called Simplified Shotgun Stochastic Search with Screening (S5) to explore the enormous model space, and we show that S5 dramatically reduces the computing time without losing the capacity to search the interesting region in the model space, at least in the simulation settings considered. The S5 algorithm is available in an R package BayesS5 on CRAN.

On the Reproducibility of Psychological Science

Investigators from a large consortium of scientists recently performed a multi-year study in which they replicated 100 psychology experiments. Although statistically significant results were reported in 97% of the original studies, statistical significance was achieved in only 36% of the replicated studies. This article presents a reanalysis of these data based on a formal statistical model that accounts for publication bias by treating outcomes from unpublished studies as missing data, while simultaneously estimating the distribution of effect sizes for those studies that tested nonnull effects. The resulting model suggests that more than 90% of tests performed in eligible psychology experiments tested negligible effects, and that publication biases based on p-values caused the observed rates of nonreproducibility. The results of this reanalysis provide a compelling argument for both increasing the threshold required for declaring scientific discoveries and for adopting statistical summaries of evidence that account for the high proportion of tested hypotheses that are false. Supplementary materials for this article are available online.

TOX expression and role in CTCL

BACKGROUND

Cutaneous T-cell lymphomas (CTCL) are skin malignancies including mycosis fungoides (MF) and CD30(+) lymphoproliferative disorders (LPD). In early disease, CTCL can be difficult to diagnose, especially in MF for which there is no reliable diagnostic marker. MF/CTCL have increased expression of thymocyte selection-associated HMG box protein (TOX). Although TOX has been proposed to be a diagnostic marker for MF, further validation studies are needed. Moreover, it is unclear what drives TOX expression or its role in MF/CTCL.

OBJECTIVE

We hypothesize evaluation of TOX levels across a spectrum of CTCL, including MF precursor (large plaque parapsoriasis, LPP), will help elucidate the implications of altered TOX expression.

MATERIALS AND METHODS

TOX staining was performed in MF, CD30(+) LPD, LPP as well as benign inflammatory dermatoses (BID) and normal skin (NS). CTCL cell lines were utilized to evaluate the regulation of TOX.

RESULTS

Positive TOX expression was identified in 73.6% of MF cases and in 31.6% of BID/NS. TOX had a positive predictive value (PPV) for MF of 86.7% and a negative predictive value (NPV) of 48.1%. TOX expression in MF was detected more commonly in Black patients (P = 0.015) and less commonly in transformed MF (P = 0.045). LPP had positive TOX staining in 70.0%. In CTCL cells, GATA3 knockdown decreased TOX mRNA and protein expression. TOX expression also decreased in the presence of CTCL therapeutics.

CONCLUSION

Our data indicate that TOX is useful as a diagnostic marker in MF. Moreover, TOX expression was evident in LPP, indicating it may have a previously unappreciated role in the development of MF. Finally, our data suggest that GATA3 regulates TOX, revealing insight into TOX regulation.

Bayesian variable selection for binary outcomes in high-dimensional genomic studies using non-local priors

Motivation: The advent of new genomic technologies has resulted in the production of massive data sets. Analyses of these data require new statistical and computational methods. In this article, we propose one such method that is useful in selecting explanatory variables for prediction of a binary response. Although this problem has recently been addressed using penalized likelihood methods, we adopt a Bayesian approach that utilizes a mixture of non-local prior densities and point masses on the binary regression coefficient vectors.

Results: The resulting method, which we call iMOMLogit, provides improved performance in identifying true models and reducing estimation and prediction error in a number of simulation studies. More importantly, its application to several genomic datasets produces predictions that have high accuracy using far fewer explanatory variables than competing methods. We also describe a novel approach for setting prior hyperparameters by examining the total variation distance between the prior distributions on the regression parameters and the distribution of the maximum likelihood estimator under the null distribution. Finally, we describe a computational algorithm that can be used to implement iMOMLogit in ultrahigh-dimensional settings (p>>n) and provide diagnostics to assess the probability that this algorithm has identified the highest posterior probability model.

Availability and implementation: Software to implement this method can be downloaded at: http://www.stat.tamu.edu/∼amir/code.html.

Contact:wwang7@mdanderson.org or vjohnson@stat.tamu.edu

Supplementary information:Supplementary data are available at Bioinformatics online.

A robust Bayesian dose-finding design for phase I/II clinical trials

We propose a Bayesian phase I/II dose-finding trial design that simultaneously accounts for toxicity and efficacy. We model the toxicity and efficacy of investigational doses using a flexible Bayesian dynamic model, which borrows information across doses without imposing stringent parametric assumptions on the shape of the dose-toxicity and dose-efficacy curves. An intuitive utility function that reflects the desirability trade-offs between efficacy and toxicity is used to guide the dose assignment and selection. We also discuss the extension of this design to handle delayed toxicity and efficacy. We conduct extensive simulation studies to examine the operating characteristics of the proposed method under various practical scenarios. The results show that the proposed design possesses good operating characteristics and is robust to the shape of the dose-toxicity and dose-efficacy curves.

Evaluation of image registration spatial accuracy using a Bayesian hierarchical model

To evaluate the utility of automated deformable image registration (DIR) algorithms, it is necessary to evaluate both the registration accuracy of the DIR algorithm itself, as well as the registration accuracy of the human readers from whom the "gold standard" is obtained. We propose a Bayesian hierarchical model to evaluate the spatial accuracy of human readers and automatic DIR methods based on multiple image registration data generated by human readers and automatic DIR methods. To fully account for the locations of landmarks in all images, we treat the true locations of landmarks as latent variables and impose a hierarchical structure on the magnitude of registration errors observed across image pairs. DIR registration errors are modeled using Gaussian processes with reference prior densities on prior parameters that determine the associated covariance matrices. We develop a Gibbs sampling algorithm to efficiently fit our models to high-dimensional data, and apply the proposed method to analyze an image dataset obtained from a 4D thoracic CT study.

Inflammatory markers and development of symptom burden in patients with multiple myeloma during autologous stem cell transplantation

PURPOSE

Increasing research suggests that inflammation mediates symptom development. In this longitudinal study, we examined inflammatory factors related to the development of high symptom burden during autologous hematopoietic stem cell transplant (AuSCT) for multiple myeloma.

EXPERIMENTAL DESIGN

Patients (n = 63) repeatedly reported symptom severity on the MD Anderson Symptom Inventory multiple myeloma module (MDASI-MM) and contributed blood samples periodically for up to 100 days after AuSCT for inflammatory marker assays. The temporal associations between serum inflammatory marker concentrations and symptom severity outcomes were examined by nonlinear mixed-effect modeling.

RESULTS

Fatigue, pain, disturbed sleep, lack of appetite, and drowsiness were consistently the most severe MDASI-MM symptoms during the study. Peak symptom severity occurred on day 8 after AuSCT, during white blood cell count nadir. Patterns of serum interleukin (IL)-6 (peak on day 9) and soluble IL-6 receptor (sIL-6R; nadir on day 8) expression paralleled symptom development over time (both P < 0.0001). By univariate analysis, serum IL-6, sIL-6R, IL-10, C-reactive protein, macrophage inflammatory protein (MIP)-1α, sIL-1R2, sIL-1RA, and soluble tumor necrosis factor receptor 1 were significantly related to the most severe symptoms during the first 30 days after AuSCT (all P < 0.05). By multivariate analysis, IL-6 (estimate = 0.170; P = 0.004) and MIP-1α (estimate = -0.172; P = 0.006) were temporally associated with the severity of the component symptom score.

CONCLUSIONS

Systemic inflammatory response was associated with high symptom burden during the acute phase of AuSCT. Additional research is needed to understand how the inflammatory response is mechanistically associated with symptom expression and whether suppression of this response can reduce symptoms without compromising tumor control.

On Numerical Aspects of Bayesian Model Selection in High and Ultrahigh-dimensional Settings

This article examines the convergence properties of a Bayesian model selection procedure based on a non-local prior density in ultrahigh-dimensional settings. The performance of the model selection procedure is also compared to popular penalized likelihood methods. Coupling diagnostics are used to bound the total variation distance between iterates in an Markov chain Monte Carlo (MCMC) algorithm and the posterior distribution on the model space. In several simulation scenarios in which the number of observations exceeds 100, rapid convergence and high accuracy of the Bayesian procedure is demonstrated. Conversely, the coupling diagnostics are successful in diagnosing lack of convergence in several scenarios for which the number of observations is less than 100. The accuracy of the Bayesian model selection procedure in identifying high probability models is shown to be comparable to commonly used penalized likelihood methods, including extensions of smoothly clipped absolute deviations (SCAD) and least absolute shrinkage and selection operator (LASSO) procedures.

Genetic markers associated with progression in early mycosis fungoides

BACKGROUND

Mycosis fungoides (MF) is a rare, but potentially devastating malignancy. It classically presents with cutaneous patches and plaques and can progress to tumours on the skin with lymph node, blood and visceral involvement. While most patients with MF have a relatively benign disease course, a subset of patients will develop progressive disease that is often fatal.

OBJECTIVE

The aim of this study was to identify genetic markers in early MF limited to the skin (stages IA-IIA) that distinguish those patients who will have progressive disease from those who will not, so that early appropriate treatment may be instituted.

METHODS

The study includes 18 patients who were diagnosed with early stage MF at the time of biopsy and had follow-up to determine which patients developed progressive disease. RNA was extracted from skin biopsy specimens and analysed for expression of CD3, FOXP3, IFNγ, Interleukin (IL)-4, IL-13, KIR3DL2, MICB, PLS3 and STAT4 by quantitative real-time polymerase chain reaction.

RESULTS/CONCLUSIONS

Reduced expression of FOXP3 and STAT4 and increased expression of IL-4 relative to CD3 expression levels were significantly associated with MF progression. Further studies will be needed to fully assess the usefulness of these genetic markers to predict disease progression and guide treatment options in patients diagnosed with early MF.

Prognostic value of patient-reported symptom interference in patients with late-stage lung cancer

PURPOSE

Patient-reported outcomes (PROs) have been found to be significant predictors of clinical outcomes such as overall survival (OS), but the effect of demographic and clinical factors on the prognostic ability of PROs is less understood. Several PROs derived from the 12-item Short-Form Health Survey (SF-12) and M. D. Anderson Symptom Inventory (MDASI) were investigated for association with OS, with adjustments for other factors, including performance status.

METHODS

A retrospective analysis was performed on data from 90 patients with stage IV non-small cell lung cancer. Several baseline PROs were added to a base Cox proportional hazards model to examine the marginal significance and improvement in model fit attributable to the PRO: mean MDASI symptom interference level; mean MDASI symptom severity level for five selected symptoms; SF-12 physical and mental component summaries; and the SF-12 general health item. Bootstrap resampling was used to assess the robustness of the findings.

RESULTS

The MDASI mean interference level had a significant effect on OS (p = 0.007) when the model was not adjusted for interactions with other prognostic factors. Further exploration suggested the significance was due to an interaction with performance status (p = 0.001). The MDASI mean symptom severity level and the SF-12 physical component summary, mental component summary, and general health item did not have a significant effect on OS.

CONCLUSIONS

Symptom interference adds prognostic information for OS in advanced lung cancer patients with poor performance status, even when demographic and clinical prognostic factors are accounted for.

UNIFORMLY MOST POWERFUL BAYESIAN TESTS

Uniformly most powerful tests are statistical hypothesis tests that provide the greatest power against a fixed null hypothesis among all tests of a given size. In this article, the notion of uniformly most powerful tests is extended to the Bayesian setting by defining uniformly most powerful Bayesian tests to be tests that maximize the probability that the Bayes factor, in favor of the alternative hypothesis, exceeds a specified threshold. Like their classical counterpart, uniformly most powerful Bayesian tests are most easily defined in one-parameter exponential family models, although extensions outside of this class are possible. The connection between uniformly most powerful tests and uniformly most powerful Bayesian tests can be used to provide an approximate calibration between p-values and Bayes factors. Finally, issues regarding the strong dependence of resulting Bayes factors and p-values on sample size are discussed.

Validation of the M. D. Anderson Symptom Inventory multiple myeloma module

Background

The symptom burden associated with multiple myeloma (MM) is often severe. Presently, no instrument comprehensively assesses disease-related and treatment-related symptoms in patients with MM. We sought to validate a module of the M. D. Anderson Symptom Inventory (MDASI) developed specifically for patients with MM (MDASI-MM).

Methods

The MDASI-MM was developed with clinician input, cognitive debriefing, and literature review, and administered to 132 patients undergoing induction chemotherapy or stem cell transplantation. We demonstrated the MDASI-MM’s reliability (Cronbach α values); criterion validity (item and subscale correlations between the MDASI-MM and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) and the EORTC MM module (QLQ-MY20)), and construct validity (differences between groups by performance status). Ratings from transplant patients were examined to demonstrate the MDASI-MM’s sensitivity in detecting the acute worsening of symptoms post-transplantation.

Results

The MDASI-MM demonstrated excellent correlations with subscales of the 2 EORTC instruments, strong ability to distinguish clinically different patient groups, high sensitivity in detecting change in patients’ performance status, and high reliability. Cognitive debriefing confirmed that the MDASI-MM encompasses the breadth of symptoms relevant to patients with MM.

Conclusion

The MDASI-MM is a valid, reliable, comprehensive-yet-concise tool that is recommended as a uniform symptom assessment instrument for patients with MM.

Bayesian adaptive phase II screening design for combination trials

BACKGROUND

Trials of combination therapies for the treatment of cancer are playing an increasingly important role in the battle against this disease. To more efficiently handle the large number of combination therapies that must be tested, we propose a novel Bayesian phase II adaptive screening design to simultaneously select among possible treatment combinations involving multiple agents.

METHODS

Our design is based on formulating the selection procedure as a Bayesian hypothesis testing problem in which the superiority of each treatment combination is equated to a single hypothesis. During the trial conduct, we use the current values of the posterior probabilities of all hypotheses to adaptively allocate patients to treatment combinations.

RESULTS

Simulation studies show that the proposed design substantially outperforms the conventional multiarm balanced factorial trial design. The proposed design yields a significantly higher probability for selecting the best treatment while allocating substantially more patients to efficacious treatments.

LIMITATIONS

The proposed design is most appropriate for the trials combining multiple agents and screening out the efficacious combination to be further investigated.

CONCLUSIONS

The proposed Bayesian adaptive phase II screening design substantially outperformed the conventional complete factorial design. Our design allocates more patients to better treatments while providing higher power to identify the best treatment at the end of the trial.

Specific lymphocyte subsets predict response to adoptive cell therapy using expanded autologous tumor-infiltrating lymphocytes in metastatic melanoma patients

PURPOSE

Adoptive cell therapy (ACT) using autologous tumor-infiltrating lymphocytes (TIL) is a promising treatment for metastatic melanoma unresponsive to conventional therapies. We report here on the results of an ongoing phase II clinical trial testing the efficacy of ACT using TIL in patients with metastatic melanoma and the association of specific patient clinical characteristics and the phenotypic attributes of the infused TIL with clinical response.

EXPERIMENTAL DESIGN

Altogether, 31 transiently lymphodepleted patients were treated with their expanded TIL, followed by two cycles of high-dose interleukin (IL)-2 therapy. The effects of patient clinical features and the phenotypes of the T cells infused on the clinical response were determined.

RESULTS

Overall, 15 of 31 (48.4%) patients had an objective clinical response using immune-related response criteria (irRC) with 2 patients (6.5%) having a complete response. Progression-free survival of more than 12 months was observed for 9 of 15 (60%) of the responding patients. Factors significantly associated with the objective tumor regression included a higher number of TIL infused, a higher proportion of CD8(+) T cells in the infusion product, a more differentiated effector phenotype of the CD8(+) population, and a higher frequency of CD8(+) T cells coexpressing the negative costimulation molecule "B- and T-lymphocyte attenuator" (BTLA). No significant difference in the telomere lengths of TIL between responders and nonresponders was identified.

CONCLUSION

These results indicate that the immunotherapy with expanded autologous TIL is capable of achieving durable clinical responses in patients with metastatic melanoma and that CD8(+) T cells in the infused TIL, particularly differentiated effectors cells and cells expressing BTLA, are associated with tumor regression.

Reproducibility and comparison of DCE-MRI and DCE-CT perfusion parameters in a rat tumor model

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and computed tomography (DCE-CT) provide independent measures of biomarkers related to tumor perfusion. We compared the reproducibilities and absolute values of DCE-MRI and DCE-CT biomarkers in the same tumors in an animal model, to investigate the physiologic validity of both approaches. DCE-MRI and DCE-CT were each performed sequentially on three consecutive days in each of twelve rats bearing C6 glioma xenografts. DCE-MRI yielded endothelial transfer constant (K(trans)), extracellular, extravascular space volume fraction (v(e)), and contrast agent reflux rate constant (k(ep)); and DCE-CT, blood flow (BF), blood volume (BV), mean transit time (MTT), and permeability-surface area product (PS) using Tofts and deconvolution physiological models, with 6.6 and 0.4 seconds temporal resolutions, respectively. Variability in DCE-CT and DCE-MRI were evaluated by variance components analysis. Intra-rat coefficients of variation for DCE-CT parameters BF, BV, MTT and PS were 25%, 22%, 18% and 23%; and for DCE-MRI parameters K(trans), k(ep) and v(e) were 23%, 16% and 20%, respectively. Mean (±SD) BF, BV, MTT and PS were: 44.6 (±13.7) ml min(-1) 100 g(-1), 5.7 (±1.5) ml 100 g(-1), 10.8 (±2.3) seconds, and 14.6 (±4.7) ml min(-1) 100 g(-1), respectively. Mean (±SD) K(trans), k(ep) and v(e) were: 0.21 (±0.05) min(-1), 0.68 (±0.14) min(-1), and 0.29 (±0.06), respectively. Permeability estimates from DCE-MRI (K(trans)) were 44% higher than from DCE-CT (PS), despite application of appropriate corrections. DCE-MRI and DCE-CT biomarkers of tumor perfusion have similar reproducibilities suggesting that they may have comparable utility, but their derived parameter values are not equivalent.

Bayesian Model Selection in High-Dimensional Settings

Standard assumptions incorporated into Bayesian model selection procedures result in procedures that are not competitive with commonly used penalized likelihood methods. We propose modifications of these methods by imposing nonlocal prior densities on model parameters. We show that the resulting model selection procedures are consistent in linear model settings when the number of possible covariates p is bounded by the number of observations n, a property that has not been extended to other model selection procedures. In addition to consistently identifying the true model, the proposed procedures provide accurate estimates of the posterior probability that each identified model is correct. Through simulation studies, we demonstrate that these model selection procedures perform as well or better than commonly used penalized likelihood methods in a range of simulation settings. Proofs of the primary theorems are provided in the Supplementary Material that is available online.

Longitudinal relationship between inflammatory markers and patient-reported symptom severity during induction therapy for multiple myeloma

9083

Background: Multiple myeloma (MM) patients undergoing induction therapy experience both disease- and therapy-related symptoms. We investigated the association between the trajectory of symptom severity and changes in levels of inflammatory markers. Methods: MM patients (N=62) rated symptoms via the M. D. Anderson Symptom Inventory (MDASI) twice a week during induction therapy. Patients contributed serum samples before the start of every chemotherapy cycle. A panel of pro- and anti-inflammatory cytokines and markers was evaluated by Luminex. Ordinal regression analyses were used to describe the relationship between cytokines and symptom outcomes across time. These analyses were adjusted for patient and clinical factors (age, sex, diabetes diagnosis, anemia, BMI, comorbidity, staging, ECOG PS, prior treatment status, tumor response, opioid use, and chemo regimen). Results: Bortezomib-based induction therapy was received by 89% of the sample. Fatigue was persistently the most severe symptom during induction therapy, followed by disturbed sleep, muscle weakness, pain, drowsiness, and bone aches. Numbness, which is representative of chemotherapy-induced peripheral neuropathy, significantly worsened from baseline (p=0.01). We observed significant longitudinal associations between sIL-1R1 and distress and sadness (both p=0.02); between sIL-6R and disturbed sleep (p=0.001), poor appetite (p=0.04), and sore mouth (p=0.006). IL-6 was significantly associated with pain, fatigue, nausea, and sore mouth (all p<.05). A negative association between sTNF-RII and pain, sleep, distress, remembering, poor appetite, and nausea (all p<0.05) was also observed. MCP-1 was positively associated with numbness (p=0.04); while MIP-1α was negatively associated with sleep, numbness, constipation, poor attention (all p<0.01), and bone aches (p=0.0006). IL-10 was negatively associated with mood interference (p=0.04). Conclusions: Frequent assessment can document the longitudinal course of multiple symptoms during induction and provides opportunity to evaluate systemic inflammation as a potential source of symptom burden during induction therapy for MM.

Predictive value of baseline serum MIP-1α and CRP on symptom burden and tumor response to induction therapy in patients with multiple myeloma

8091

Background: Macrophage inflammatory protein-1α(MIP-1α) is a growth factor for human multiple myeloma (MM) cells. As an osteoclastic factor, its presence provides pathophysiological evidence of the development of lytic bone lesions in MM. C-reactive protein (CRP) indicates systemic inflammation. The relationship between disease-driven inflammatory markers and both patient-experienced symptoms and tumor response to induction therapy is unknown. Methods: MM patients (N=39) who were either newly diagnosed or had received fewer than 2 cycles of chemotherapy, and who also were to receive induction therapy, were enrolled. To test concentrations of MIP-1α and CRP, serum samples were collected before and after induction and assayed by Luminex. Multiple symptoms were measured twice a week via the M. D. Anderson Symptom Inventory MM module (MDASI-MM) from -8 to +112 days of induction. The MM-specific items of the MDASI-MM are bone aches, constipation, muscle weakness, diarrhea, sore mouth or throat, rash, and difficulty paying attention. Correlation between symptom severity and inflammatory markers at baseline was examined by linear regression modeling. Kruskal-Wallis significance test and Wilcoxon test were used to examine association between MIP-1α and tumor response. Results: Patients received either bortezomib-based (89%) or lenalidomide-based induction therapy. Baseline MIP-1α and CRP were significantly inversely related to the mean severity component score of the 5 most-severe symptoms (fatigue, pain, bone aches, poor sleep, drowsiness) (p=.03; p=.02), and significantly inversely related to the severity of a component score of the module-specific symptoms (p=0.04; p=.002). Change over time in MIP-1α differed significantly by tumor response category (p=.04), with the partial response group having a higher median score than the complete response group (1.483 vs. 0.016, p=.01). Conclusions: Our data suggest that higher baseline levels of serum MIP-1α and CRP predict effective chemotherapy-induced reduction of disease-related symptoms in MM patients. Higher serum MIP-1α expression after induction therapy was related to less-ideal tumor response.

Inflammatory markers and symptom burden in patients with multiple myeloma undergoing autologous stem cell transplantation

9081

Background: We explored the association between activation of inflammatory networks and development of severe symptoms during mobilization and the acute phase of autologous hematopoietic stem cell transplantation (ASCT) for multiple myeloma (MM). Methods: From mobilization through the first 3 months after ASCT, multiple symptoms were measured repeatedly via the myeloma module of the M. D. Anderson Symptom Inventory (MDASI-MM). Serum levels of interleukin (IL)-6, IL-10, tumor necrosis factor (TNF)-a, soluble TNF receptors 1 and 2 (sTNF-R1, sTNF-R2), IL-1 receptor antagonist (IL-1RA), vascular endothelial growth factor (VEGF), macrophage inflammatory protein-1 (MIP-1a), monocyte chemoattractant protein (MCP)-1 and C-reactive protein (CRP) were collected up to 11 times per patient and assayed by Luminex. Ordinal regression modeling was used to analyze repeated-measures data. Results: Among50 MM patients, the most severe symptoms reported by werefatigue, pain, muscle weakness, disturbed sleep, drowsiness, numbness, poor appetite, and bone aches. Symptoms worsened rapidly from conditioning therapy and peaked at WBC nadir. Over time, controlling for age, sex, and MM stage, increased serum IL-6 (P=.0007) and MCP-1 (P=.0005) were significantly associated with worsening of the most severe symptoms; MIP-1a (P=.005) and VEGF (P=.002) were inversely associated with these symptoms. Increased CRP was significantly associated with worsening pain (P=.01), fatigue (P=.007), and bone aches (P=.006). Conclusions: This longitudinal study indicates a strong association between dynamic changes in circulating inflammatory molecules and the most severe symptoms in MM patients during the acute phase of ASCT. This observation, similar to IL-6-related multisymptom development around WBC nadir of allogeneic SCT (Wang et al, 2008), provides evidence of potential inflammation-induced behavioral changes in humans. Testing of anti-inflammatory interventions is warranted to further confirm the role of inflammation in the development of symptoms and identify effective mechanism-driven symptom management during ASCT.

PET/CT in the management of patients with stage IIIC and IV metastatic melanoma considered candidates for surgery: evaluation of the additive value after conventional imaging

OBJECTIVE

The purpose of this article is to determine how often unexpected (18)F-FDG PET/CT findings result in a change in management for patients with stage IV and clinically evident stage III melanoma with resectable disease according to conventional imaging.

SUBJECTS AND METHODS

Thirty-two patients with oligometastatic stage IV and clinically evident stage III melanoma were identified by surgical oncologists according to the results of conventional imaging, which included contrast-enhanced CT of the chest, abdomen, and pelvis and MRI of the brain. The surgical plan included resection of known metastases or isolated limb perfusion with chemotherapy. Thirty-three FDG PET/CT scans were performed within 36 days of their contrast-enhanced CT. The impact of PET/CT was defined as the percentage of cases in which a change in the surgical plan resulted from the unanticipated PET/CT findings.

RESULTS

PET/CT revealed unexpected melanoma metastases in 12% of scans (4/33). As a result, the surgery was canceled for two patients, and the planned approach was altered for another two patients to address the unexpected sites. In 6% of scans (2/33), the unexpected metastases were detected in the extremities, which were not included in conventional imaging. Three scans (9%) showed false-positive FDG-avid findings that proved to be benign by subsequent stability or resolution with no therapy.

CONCLUSION

In patients with surgically treatable metastatic melanoma, FDG PET/CT can detect unexpected metastases that are missed or not imaged with conventional imaging, and can be considered as part of preoperative workup.

Goodness-of-fit diagnostics for Bayesian hierarchical models

This article proposes methodology for assessing goodness of fit in Bayesian hierarchical models. The methodology is based on comparing values of pivotal discrepancy measures (PDMs), computed using parameter values drawn from the posterior distribution, to known reference distributions. Because the resulting diagnostics can be calculated from standard output of Markov chain Monte Carlo algorithms, their computational costs are minimal. Several simulation studies are provided, each of which suggests that diagnostics based on PDMs have higher statistical power than comparable posterior-predictive diagnostic checks in detecting model departures. The proposed methodology is illustrated in a clinical application; an application to discrete data is described in supplementary material.

Perfusion CT assessment of tissue hemodynamics following hepatic arterial infusion of increasing doses of angiotensin II in a rabbit liver tumor model

PURPOSE

To investigate the effects of increasing doses of angiotensin II on hepatic hemodynamics in the normal rabbit liver and in hepatic VX2 tumors by using dynamic contrast material-enhanced perfusion computed tomography (CT).

MATERIALS AND METHODS

This study was approved by the institutional animal care and use committee. Solitary hepatic VX2 tumors were implanted into 12 rabbits. In each animal, perfusion CT of the liver was performed before (at baseline) and after hepatic arterial infusion of varying doses (0.1-50.0 μg/mL) of angiotensin II. Images were acquired continuously for 80 seconds after the start of the intravenous contrast material administration. Blood flow (BF), blood volume (BV), mean transit time (MTT), and capillary permeability-surface area product were calculated for the tumor and the adjacent and distant normal liver tissue. Generalized linear mixed models were used to estimate the effects of angiotensin II dose on outcome measures.

RESULTS

Angiotensin II infusion increased contrast enhancement of the tumor and distal liver vessels. Tumor BF increased in a dose-dependent manner after administration of 0.5-25.0 μg/mL angiotensin II, but only the 2.5 μg/mL dose induced a significant increase in tumor BF compared with BF in the adjacent (68.0 vs 26.3 mL/min/100 g, P < .0001) and distant (68.0 vs 28.3 mL/min/100 g, P = .02) normal liver tissue. Tumor BV varied with angiotensin II dose but was greater than the BV of the adjacent and distant liver tissue at only the 2.5 μg/mL (4.8 vs 3.5 mL/100 g for adjacent liver [P < .0001], 4.8 vs 3.3 mL/100 g for distant liver [P = .0006]) and 10.0 μg/mL (4.9 vs 4.4 mL/100 g for adjacent liver [P = .007], 4.9 vs 4.3 mL/100 g for distant liver [P = .04]) doses. Tumor MTT was significantly shorter than the adjacent liver tissue MTT at angiotensin II doses of 2.5 μg/mL (9.7 vs 15.8 sec, P = .001) and 10.0 μg/mL (5.1 vs 13.2 sec, P = .007) and significantly shorter than the distant liver tissue MTT at 2.5 μg/mL only (9.7 vs 15.3 sec, P = .0006). The capillary permeability-surface area product for the tumor was higher than that for the adjacent liver tissue at the 2.5 μg/mL angiotensin II dose only (11.5 vs 8.1 mL/min/100 g, P = .01).

CONCLUSION

Perfusion CT enables a mechanistic understanding of angiotensin II infusion in the liver and derivation of the optimal effective dose. The 2.5 μg/mL angiotensin II dose increases perfusion in hepatic VX2 tumors versus that in adjacent and distant normal liver tissue primarily by constricting normal distal liver vessels and in turn increasing tumor BF and BV.

Chyluria after radiofrequency ablation of renal tumors

PURPOSE

To assess the incidence of chyluria after radiofrequency (RF) ablation of renal tumors and attempt to identify predictors of this phenomenon.

MATERIALS AND METHODS

Over a 3-year period, 62 consecutive patients with renal tumors were treated by percutaneous computed tomography (CT)-guided or laparoscopic RF ablation, of which 41 underwent at least three posttreatment CT studies and were evaluated in this study. Three radiologists reviewed the pretreatment and posttreatment CT images for the presence or absence of fat-fluid levels in the bladder, the location of the tumor, and the size of the postablation defect. A logistic regression model was used to assess whether ablation defect size or tumor location predicted chyluria.

RESULTS

Chyluria was detected at a mean time of 44.5 weeks in 17 (41%) of 41 patients with renal tumors treated by RF ablation. A pretreatment biopsy specimen showed renal cell carcinoma in 74%. Mean tumor size was 2.77 cm, and mean initial ablation size was 4.2 cm. Chyluria persisted in seven patients. Zone of ablation defect size and tumor location were not significant predictors of chyluria (P = .64 and P = .42). Mean follow-up was 77 weeks.

CONCLUSIONS

Chyluria is a common and asymptomatic finding in a significant proportion of patients undergoing RF ablation for renal tumors. Tumor location and zone of ablation defect size were not predictors of chyluria. The presence of a fat-fluid level should not be mistaken for an air-fluid level.

Semiautomated motion correction of tumors in lung CT-perfusion studies

RATIONALE AND OBJECTIVES

To compare the relative performance of one-dimensional (1D) manual, rigid-translational, and nonrigid registration techniques to correct misalignment of lung tumor anatomy acquired from computed tomography perfusion (CTp) datasets.

MATERIALS AND METHODS

Twenty-five datasets in patients with lung tumors who had undergone a CTp protocol were evaluated. Each dataset consisted of one reference CT image from an initial cine slab and six subsequent breathhold helical volumes (16-row multi-detector CT), acquired during intravenous contrast administration. Each helical volume was registered to the reference image using two semiautomated intensity-based registration methods (rigid-translational and nonrigid), and 1D manual registration (the only registration method available in the relevant application software). The performance of each technique to align tumor regions was assessed quantitatively (percent overlap and distance of center of mass), and by a visual validation study (using a 5-point scale). The registration methods were statistically compared using linear mixed and ordinal probit regression models.

RESULTS

Quantitatively, tumor alignment with the nonrigid method compared to rigid-translation was borderline significant, which in turn was significantly better than the 1D manual method: average (± SD) percent overlap, 91.8 ± 2.3%, 87.7 ± 5.5%, and 77.6 ± 5.9%, respectively; and average (± SD) DCOM, 0.41 ± 0.16 mm, 1.08 ± 1.13 mm, and 2.99 ± 2.93 mm, respectively (all P < .0001). Visual validation confirmed these findings.

CONCLUSION

Semiautomated registration methods achieved superior alignment of lung tumors compared to the 1D manual method. This will hopefully translate into more reliable CTp analyses.

Impact of cultural and linguistic factors on symptom reporting by patients with cancer

BACKGROUND

Patient reporting of the severity and impact of symptoms is an essential component of cancer symptom management and cancer treatment clinical trials. In multinational clinical trials, cultural and linguistic variations in patient-reported outcomes instruments could confound the interpretation of study results.

METHODS

The severity and interference of multiple symptoms in 1433 cancer patients with mixed diagnoses and treatment status from the United States, China, Japan, Russia, and Korea were measured with psychometrically validated language versions of the M. D. Anderson Symptom Inventory (MDASI). Mixed-effect ordinal probit regression models were fitted to the pooled data to compare the magnitude of the effect of "country" (nation and linguistic factors) with between-subjects effects on symptom reporting, adjusted for patient and clinical factors (age, sex, performance status, and chemotherapy status).

RESULTS

For the pooled sample, fatigue, disturbed sleep, distress, pain, and lack of appetite were the most severe patient-reported MDASI symptoms. The magnitude of the variance of the country random effects was only one-fourth to one-half of the interpatient variation (sigma(2) = 0.23-0.46) for all symptoms, except nausea and vomiting.

CONCLUSIONS

Cultural and linguistic variations in symptom reporting among the five language versions of the validated MDASI were limited. Ordinal probit modeling provided a simple mechanism for accounting for cultural and linguistic differences in patient populations. The equivalence among MDASI translations in this study suggests that symptom ratings collected from various cultural and language groups using the MDASI can be interpreted in a similar way in oncology practice, clinical trials, and clinical research.

Prognostic value of symptom burden for overall survival in patients receiving chemotherapy for advanced nonsmall cell lung cancer

BACKGROUND

Patient-reported outcomes have shown independent prognostic value for patients with nonsmall cell lung cancer (NSCLC). However, translating patient-reported outcomes into useful prognostic information for individual patients has been problematic.

METHODS

A total of 94 patients with advanced NSCLC and an Eastern Cooperative Oncology Group performance status (PS) of 0 to 2 who qualified for chemotherapy rated symptom severity using the M. D. Anderson Symptom Inventory before and after their first chemotherapy cycle. Prognostic values of baseline symptoms and changes in symptom severity were examined by Cox proportional hazards models.

RESULTS

In multivariate analysis, controlled for demographic and other factors, baseline coughing rated > or =4 independently predicted significantly higher risk for shorter survival (hazards ratio [HR], 8.69; P < .0001). Patients with coughing > or =4 and a PS of 2 were more likely to have shorter survival (HR, 20.6; P < .0001) than patients with coughing <4 and a PS of 0 to 1. A 1-point or greater increase in severity of fatigue (P < .05), shortness of breath, or poor appetite (P < .01) from baseline to the end of the first chemotherapy cycle was also found to be independently associated with higher risk for poor survival.

CONCLUSIONS

An increased risk for shorter survival was indicated by moderate to severe coughing at baseline or by increased fatigue or shortness of breath during the first chemotherapy cycle in patients with advanced NSCLC. Although cross-validation is needed, these data suggest that an individual patient's symptom severity scores, quickly obtainable in the clinic, might contribute clinically useful information for treatment planning for that patient. Society.

Bayesian design of single-arm phase II clinical trials with continuous monitoring

BACKGROUND

Bayesian designs are increasingly used to conduct phase II clinical trials. However, stopping boundaries in most Bayesian designs are defined from posterior credible intervals. The use of designs based on posterior credible intervals results in a loss of efficiency when compared to formal stopping rules based on Bayesian hypothesis tests. Such designs also introduce an unnecessary element of subjectivity in the interpretation of trial results.

METHODS

We derive a new class of Bayesian designs based on formal hypothesis tests. The prior densities used to define the alternative hypotheses in these tests assign no mass to parameter values that are consistent with the null hypotheses and are called nonlocal alternative prior densities.

RESULTS

We show that Bayesian designs based on hypothesis tests and nonlocal alternative prior densities are more efficient than common Bayesian designs based on posterior credible intervals and common frequentist designs. In contrast to trial designs based on Bayesian credible intervals, we demonstrate that the mis-specification of the prior densities used to describe the anticipated effect of the experimental treatment in designs based on hypothesis tests cannot increase the expected weight of evidence in favor of the trial agent.

LIMITATIONS

Extension of test-based designs to phase I-II designs and randomized phase II designs remains an open research question.

CONCLUSIONS

Phase II single-arm trials designed using Bayesian hypothesis tests with nonlocal alternatives provide better operating characteristics, use fewer patients per correct decision, and provide more directly interpretable results than other commonly used Bayesian and frequentist designs. Because the mis-specification of the prior density on the effect of the experimental agent decreases the expected weight of evidence that is collected in favor of the experimental treatment, the use of Bayesian hypothesis tests to design clinical trials also eliminates a potential source of bias often associated with trials designed using posterior credible intervals.

A Bayesian chi-squared goodness-of-fit test for censored data models

We propose a Bayesian chi-squared model diagnostic for analysis of data subject to censoring. The test statistic has the form of Pearson's chi-squared test statistic and is easy to calculate from standard output of Markov chain Monte Carlo algorithms. The key innovation of this diagnostic is that it is based only on observed failure times. Because it does not rely on the imputation of failure times for observations that have been censored, we show that under heavy censoring it can have higher power for detecting model departures than a comparable test based on the complete data. In a simulation study, we show that tests based on this diagnostic exhibit comparable power and better nominal Type I error rates than a commonly used alternative test proposed by Akritas (1988, Journal of the American Statistical Association 83, 222-230). An important advantage of the proposed diagnostic is that it can be applied to a broad class of censored data models, including generalized linear models and other models with nonidentically distributed and nonadditive error structures. We illustrate the proposed model diagnostic for testing the adequacy of two parametric survival models for Space Shuttle main engine failures.

General intelligence in another primate: individual differences across cognitive task performance in a New World monkey (Saguinus oedipus)

BACKGROUND

Individual differences in human cognitive abilities show consistently positive correlations across diverse domains, providing the basis for the trait of "general intelligence" (g). At present, little is known about the evolution of g, in part because most comparative studies focus on rodents or on differences across higher-level taxa. What is needed, therefore, are experiments targeting nonhuman primates, focusing on individual differences within a single species, using a broad battery of tasks. To this end, we administered a large battery of tasks, representing a broad range of cognitive domains, to a population of captive cotton-top tamarin monkeys (Saguinus oedipus).

METHODOLOGY AND RESULTS

Using a Bayesian latent variable model, we show that the pattern of correlations among tasks is consistent with the existence of a general factor accounting for a small but significant proportion of the variance in each task (the lower bounds of 95% Bayesian credibility intervals for correlations between g and task performance all exceed 0.12).

CONCLUSION

Individual differences in cognitive abilities within at least one other primate species can be characterized by a general intelligence factor, supporting the hypothesis that important aspects of human cognitive function most likely evolved from ancient neural substrates.

A framework for evaluation of deformable image registration spatial accuracy using large landmark point sets

Expert landmark correspondences are widely reported for evaluating deformable image registration (DIR) spatial accuracy. In this report, we present a framework for objective evaluation of DIR spatial accuracy using large sets of expert-determined landmark point pairs. Large samples (>1100) of pulmonary landmark point pairs were manually generated for five cases. Estimates of inter- and intra-observer variation were determined from repeated registration. Comparative evaluation of DIR spatial accuracy was performed for two algorithms, a gradient-based optical flow algorithm and a landmark-based moving least-squares algorithm. The uncertainty of spatial error estimates was found to be inversely proportional to the square root of the number of landmark point pairs and directly proportional to the standard deviation of the spatial errors. Using the statistical properties of this data, we performed sample size calculations to estimate the average spatial accuracy of each algorithm with 95% confidence intervals within a 0.5 mm range. For the optical flow and moving least-squares algorithms, the required sample sizes were 1050 and 36, respectively. Comparative evaluation based on fewer than the required validation landmarks results in misrepresentation of the relative spatial accuracy. This study demonstrates that landmark pairs can be used to assess DIR spatial accuracy within a narrow uncertainty range.

Bayesian model selection using test statistics

Existing Bayesian model selection procedures require the specification of prior distributions on the parameters appearing in every model in the selection set. In practice, this requirement limits the application of Bayesian model selection methodology. To overcome this limitation, we propose a new approach towards Bayesian model selection that uses classical test statistics to compute Bayes factors between possible models. In several test cases, our approach produces results that are similar to previously proposed Bayesian model selection and model averaging techniques in which prior distributions were carefully chosen. In addition to eliminating the requirement to specify complicated prior distributions, this method offers important computational and algorithmic advantages over existing simulation-based methods. Because it is easy to evaluate the operating characteristics of this procedure for a given sample size and specified number of covariates, our method facilitates the selection of hyperparameter values through prior-predictive simulation.

Statistical analysis of the National Institutes of Health peer review system

A statistical model is proposed for the analysis of peer-review ratings of R01 grant applications submitted to the National Institutes of Health. Innovations of this model include parameters that reflect differences in reviewer scoring patterns, a mechanism to account for the transfer of information from an application's preliminary ratings and group discussion to final ratings provided by all panel members and posterior estimates of the uncertainty associated with proposal ratings. Application of this model to recent R01 rating data suggests that statistical adjustments to panel rating data would lead to a 25% change in the pool of funded proposals. Viewed more broadly, the methodology proposed in this article provides a general framework for the analysis of data collected interactively from expert panels through the use of the Delphi method and related procedures.

Radiation pneumonitis: correlation of toxicity with pulmonary metabolic radiation response

PURPOSE

To characterize the relationship between radiation pneumonitis (RP) clinical symptoms and pulmonary metabolic activity on post-treatment [(18)F]-fluorodeoxyglucose positron emission tomography (FDG-PET).

PATIENTS AND METHODS

We retrospectively studied 101 esophageal cancer patients who underwent restaging FDG-PET/computed tomography imaging 3-12 weeks after completing thoracic radiotherapy. The National Institutes of Health Common Toxicity Criteria, version 3, was used to score the RP clinical symptoms. Linear regression was applied to the FDG-PET/computed tomography images to determine the normalized FDG uptake vs. radiation dose. The pulmonary metabolic radiation response (PMRR) was quantified as this slope. Modeling was performed to determine the interaction of PMRR, mean lung dose (MLD), and the percentage of lung receiving >20 Gy with RP outcomes.

RESULTS

Of the 101 patients, 25 had Grade 0, 10 had Grade 1, 60 had Grade 2, 5 had Grade 3, and 1 had Grade 5 RP symptoms. Logistic regression analysis demonstrated that increased values of both MLD and PMRR were associated with a greater probability of RP clinical symptoms (p = 0.032 and p = 0.033, respectively). Spearman's rank correlation found no association between the PMRR and the dosimetric parameters (planning target volume, MLD, percentage of lung receiving >5-30 Gy). Twofold cross-validation demonstrated that the combination of MLD and PMRR was superior to either alone for assessing the development of clinical RP symptoms. The combined MLD (or percentage of lung receiving >20 Gy) and PMRR had a greater sensitivity and accuracy (53.3% and 62.5%, respectively) than either alone.

CONCLUSION

The results of this study have demonstrated a significant correlation between RP clinical symptoms and the PMRR measured by FDG-PET/computed tomography after thoracic radiotherapy.