Simultaneous Inference of Multiple Binary Endpoints in Biomedical Research: Small Sample Properties of Multiple Marginal Models and a Resampling Approach.

In biomedical research, the simultaneous inference of multiple binary endpoints may be of interest. In such cases, an appropriate multiplicity adjustment is required that controls the family-wise error rate, which represents the probability of making incorrect test decisions. In this paper, we investigate two approaches that perform single-step p $p$ -value adjustments that also take into account the possible correlation between endpoints. A rather novel and flexible approach known as multiple marginal models is considered, which is based on stacking of the parameter estimates of the marginal models and deriving their joint asymptotic distribution. We also investigate a nonparametric vector-based resampling approach, and we compare both approaches with the Bonferroni method by examining the family-wise error rate and power for different parameter settings, including low proportions and small sample sizes. The results show that the resampling-based approach consistently outperforms the other methods in terms of power, while still controlling the family-wise error rate. The multiple marginal models approach, on the other hand, shows a more conservative behavior. However, it offers more versatility in application, allowing for more complex models or straightforward computation of simultaneous confidence intervals. The practical application of the methods is demonstrated using a toxicological dataset from the National Toxicology Program.

Discordant Health Implications and Molecular Mechanisms of Vitamin D in Clinical and Preclinical Studies of Prostate Cancer: A Critical Appraisal of the Literature Data.

Clinical and preclinical studies have provided conflicting data on the postulated beneficial effects of vitamin D in patients with prostate cancer. In this opinion piece, we discuss reasons for discrepancies between preclinical and clinical vitamin D studies. Different criteria have been used as evidence for the key roles of vitamin D. Clinical studies report integrative cancer outcome criteria such as incidence and mortality in relation to vitamin D status over time. In contrast, preclinical vitamin D studies report molecular and cellular changes resulting from treatment with the biologically active vitamin D metabolite, 1,25-dihydroxyvitamin D3 (calcitriol) in tissues. However, these reported changes in preclinical in vitro studies are often the result of treatment with biologically irrelevant high calcitriol concentrations. In typical experiments, the used calcitriol concentrations exceed the calcitriol concentrations in normal and malignant prostate tissue by 100 to 1000 times. This raises reasonable concerns regarding the postulated biological effects and mechanisms of these preclinical vitamin D approaches in relation to clinical relevance. This is not restricted to prostate cancer, as detailed data regarding the tissue-specific concentrations of vitamin D metabolites are currently lacking. The application of unnaturally high concentrations of calcitriol in preclinical studies appears to be a major reason why the results of preclinical in vitro studies hardly match up with outcomes of vitamin D-related clinical studies. Regarding future studies addressing these concerns, we suggest establishing reference ranges of tissue-specific vitamin D metabolites within various cancer entities, carrying out model studies on human cancer cells and patient-derived organoids with biologically relevant calcitriol concentrations, and lastly improving the design of vitamin D clinical trials where results from preclinical studies guide the protocols and endpoints within these trials.

bootGSEA: a bootstrap and rank aggregation pipeline for multi-study and multi-omics enrichment analyses.

Introduction: Gene set enrichment analysis (GSEA) subsequent to differential expression analysis is a standard step in transcriptomics and proteomics data analysis. Although many tools for this step are available, the results are often difficult to reproduce because set annotations can change in the databases, that is, new features can be added or existing features can be removed. Finally, such changes in set compositions can have an impact on biological interpretation. Methods: We present bootGSEA, a novel computational pipeline, to study the robustness of GSEA. By repeating GSEA based on bootstrap samples, the variability and robustness of results can be studied. In our pipeline, not all genes or proteins are involved in the different bootstrap replicates of the analyses. Finally, we aggregate the ranks from the bootstrap replicates to obtain a score per gene set that shows whether it gains or loses evidence compared to the ranking of the standard GSEA. Rank aggregation is also used to combine GSEA results from different omics levels or from multiple independent studies at the same omics level. Results: By applying our approach to six independent cancer transcriptomics datasets, we showed that bootstrap GSEA can aid in the selection of more robust enriched gene sets. Additionally, we applied our approach to paired transcriptomics and proteomics data obtained from a mouse model of spinal muscular atrophy (SMA), a neurodegenerative and neurodevelopmental disease associated with multi-system involvement. After obtaining a robust ranking at both omics levels, both ranking lists were combined to aggregate the findings from the transcriptomics and proteomics results. Furthermore, we constructed the new R-package "bootGSEA," which implements the proposed methods and provides graphical views of the findings. Bootstrap-based GSEA was able in the example datasets to identify gene or protein sets that were less robust when the set composition changed during bootstrap analysis. Discussion: The rank aggregation step was useful for combining bootstrap results and making them comparable to the original findings on the single-omics level or for combining findings from multiple different omics levels.

Dimension reduction and outlier detection of 3-D shapes derived from multi-organ CT images.

BACKGROUND: Unsupervised clustering and outlier detection are important in medical research to understand the distributional composition of a collective of patients. A number of clustering methods exist, also for high-dimensional data after dimension reduction. Clustering and outlier detection may, however, become less robust or contradictory if multiple high-dimensional data sets per patient exist. Such a scenario is given when the focus is on 3-D data of multiple organs per patient, and a high-dimensional feature matrix per organ is extracted. METHODS: We use principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE) and multiple co-inertia analysis (MCIA) combined with bagplots to study the distribution of multi-organ 3-D data taken by computed tomography scans. After point-set registration of multiple organs from two public data sets, multiple hundred shape features are extracted per organ. While PCA and t-SNE can only be applied to each organ individually, MCIA can project the data of all organs into the same low-dimensional space. RESULTS: MCIA is the only approach, here, with which data of all organs can be projected into the same low-dimensional space. We studied how frequently (i.e., by how many organs) a patient was classified to belong to the inner or outer 50% of the population, or as an outlier. Outliers could only be detected with MCIA and PCA. MCIA and t-SNE were more robust in judging the distributional location of a patient in contrast to PCA. CONCLUSIONS: MCIA is more appropriate and robust in judging the distributional location of a patient in the case of multiple high-dimensional data sets per patient. It is still recommendable to apply PCA or t-SNE in parallel to MCIA to study the location of individual organs.

Clinical implications of AGR2 in primary prostate cancer: Results from a large-scale study.

Human anterior gradient-2 (AGR2) has been implicated in carcinogenesis of various solid tumours, but the expression data in prostate cancer are contradictory regarding its prognostic value. The objective of this study is to evaluate the expression of AGR2 in a large prostate cancer cohort and to correlate it with clinicopathological data. AGR2 protein expression was analysed immunohistochemically in 1023 well-characterized prostate cancer samples with a validated antibody. AGR2 expression levels in carcinomas were compared with matched tissue samples of adjacent normal glands. AGR2 expression levels were dichotomized and tested for statistical significance. Increased AGR2 expression was found in 93.5% of prostate cancer cases. AGR2 levels were significantly higher in prostate cancer compared with normal prostate tissue. A gradual loss of AGR2 expression was associated with increasing tumour grade (ISUP), and AGR2 expression is inversely related to patient survival, however, multivariable significance is not achieved. AGR2 is clearly upregulated in the majority of prostate cancer cases, yet a true diagnostic value appears unlikely. In spite of the negative correlation of AGR2 expression with increasing tumour grade, no independent prognostic significance was found in this large-scale study.

Identifying biomarkers deciphering sepsis from trauma-induced sterile inflammation and trauma-induced sepsis.

Objective: The purpose of this study was to identify a panel of biomarkers for distinguishing early stage sepsis patients from non-infected trauma patients. Background: Accurate differentiation between trauma-induced sterile inflammation and real infective sepsis poses a complex life-threatening medical challenge because of their common symptoms albeit diverging clinical implications, namely different therapies. The timely and accurate identification of sepsis in trauma patients is therefore vital to ensure prompt and tailored medical interventions (provision of adequate antimicrobial agents and if possible eradication of infective foci) that can ultimately lead to improved therapeutic management and patient outcome. The adequate withholding of antimicrobials in trauma patients without sepsis is also important in aspects of both patient and environmental perspective. Methods: In this proof-of-concept study, we employed advanced technologies, including Matrix-Assisted Laser Desorption/Ionization (MALDI) and multiplex antibody arrays (MAA) to identify a panel of biomarkers distinguishing actual sepsis from trauma-induced sterile inflammation. Results: By comparing patient groups (controls, infected and non-infected trauma and septic shock patients under mechanical ventilation) at different time points, we uncovered distinct protein patterns associated with early trauma-induced sterile inflammation on the one hand and sepsis on the other hand. SYT13 and IL1F10 emerged as potential early sepsis biomarkers, while reduced levels of A2M were indicative of both trauma-induced inflammation and sepsis conditions. Additionally, higher levels of TREM1 were associated at a later stage in trauma patients. Furthermore, enrichment analyses revealed differences in the inflammatory response between trauma-induced inflammation and sepsis, with proteins related to complement and coagulation cascades being elevated whereas proteins relevant to focal adhesion were diminished in sepsis. Conclusions: Our findings, therefore, suggest that a combination of biomarkers is needed for the development of novel diagnostic approaches deciphering trauma-induced sterile inflammation from actual infective sepsis.