Nonparametric analysis of delayed treatment effects using single-crossing constraints.

Clinical trials involving novel immuno-oncology therapies frequently exhibit survival profiles which violate the proportional hazards assumption due to a delay in treatment effect, and, in such settings, the survival curves in the two treatment arms may have a crossing before the two curves eventually separate. To flexibly model such scenarios, we describe a nonparametric approach for estimating the treatment arm-specific survival functions which constrains these two survival functions to cross at most once without making any additional assumptions about how the survival curves are related. A main advantage of our approach is that it provides an estimate of a crossing time if such a crossing exists, and, moreover, our method generates interpretable measures of treatment benefit including crossing-conditional survival probabilities and crossing-conditional estimates of restricted residual mean life. Our estimates of these measures may be used together with efficacy measures from a primary analysis to provide further insight into differences in survival across treatment arms. We demonstrate the use and effectiveness of our approach with a large simulation study and an analysis of reconstructed outcomes from a recent combination therapy trial.

Clinical implications of AR alterations in advanced prostate cancer: a multi-institutional collaboration.

BACKGROUND: AR gene alterations can develop in response to pressure of testosterone suppression and androgen receptor targeting agents (ARTA). Despite this, the relevance of these gene alterations in the context of ARTA treatment and clinical outcomes remains unclear. METHODS: Patients with castration-resistant prostate cancer (CRPC) who had undergone genomic testing and received ARTA treatment were identified in the Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE) database. Patients were stratified according to the timing of genomic testing relative to the first ARTA treatment (pre-/post-ARTA). Clinical outcomes such as time to progression, PSA response, and overall survival were compared based on alteration types. RESULTS: In total, 540 CRPC patients who received ARTA and had tissue-based (n = 321) and/or blood-based (n = 244) genomic sequencing were identified. Median age was 62 years (range 39-90) at the time of the diagnosis. Majority were White (72.2%) and had metastatic disease (92.6%) at the time of the first ARTA treatment. Pre-ARTA genomic testing was available in 24.8% of the patients, and AR mutations and amplifications were observed in 8.2% and 13.1% of the patients, respectively. Further, time to progression was longer in patients with AR amplifications (25.7 months) compared to those without an AR alteration (9.6 months; p = 0.03). In the post-ARTA group (n = 406), AR mutations and AR amplifications were observed in 18.5% and 35.7% of the patients, respectively. The most common mutation in post-ARTA group was L702H (9.9%). CONCLUSION: In this real-world clinicogenomics database-driven study we explored the development of AR alterations and their association with ARTA treatment outcomes. Our study showed that AR amplifications are associated with longer time to progression on first ARTA treatment. Further prospective studies are needed to optimize therapeutic strategies for patients with AR alterations.

Biomarker-Directed Therapy in Black and White Men With Metastatic Castration-Resistant Prostate Cancer.

Importance: Black men have higher incidence and mortality from prostate cancer. Whether precision oncology disparities affect Black men with metastatic castration-resistant prostate cancer (mCRPC) is unknown. Objective: To compare precision medicine data and outcomes between Black and White men with mCRPC. Design, Setting, and Participants: This retrospective cohort study used data collected by the Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE) consortium, a multi-institutional registry with linked clinicogenomic data, from April 2020 to December 2021. Participants included Black and White patients with mCRPC with molecular data. Data were analyzed from December 2021 to May 2023. Exposures: Database-reported race and ethnicity. Main Outcomes and Measures: The primary outcome was the frequency of actionable molecular data, defined as the presence of mismatch repair deficiency (MMRD) or high microsatellite instability (MSI-H), homologous recombination repair deficiency, or tumor mutational burden of 10 mutations per megabase or greater. Secondary outcomes included the frequency of other alterations, the type and timing of genomic testing performed, and use of targeted therapy. Efficacy outcomes were prostate-specific antigen response rate, site-reported radiographic response, and overall survival. Results: A total of 962 eligible patients with mCRPC were identified, including 204 Black patients (21.2%; median [IQR] age at diagnosis, 61 [55-67] years; 131 patients [64.2%] with Gleason scores 8-10; 92 patients [45.1%] with de novo metastatic disease) and 758 White patients (78.8%; median [IQR] age, 63 [57-69] years; 445 patients [58.7%] with Gleason scores 8-10; 310 patients [40.9%] with de novo metastatic disease). Median (IQR) follow-up from mCRPC was 26.6 (14.2-44.7) months. Blood-based molecular testing was more common in Black men (111 men [48.7%]) than White men (317 men [36.4%]; P < .001). Rates of actionable alterations were similar between groups (65 Black men [32.8%]; 215 White men [29.1%]; P = .35), but MMRD or MSI-H was more common in Black men (18 men [9.1]) than White men (36 men [4.9%]; P = .04). PTEN alterations were less frequent in Black men than White men (31 men [15.7%] vs 194 men [26.3%]; P = .003), as were TMPRSS alterations (14 men [7.1%] vs 155 men [21.0%]; P < .001). No other differences were seen in the 15 most frequently altered genes, including TP53, AR, CDK12, RB1, and PIK3CA. Matched targeted therapy was given less frequently in Black men than White men (22 men [33.5%] vs 115 men [53.5%]; P = .008). There were no differences in response to targeted therapy or survival between the two cohorts. Conclusions and Relevance: This cohort study of men with mCRPC found higher frequency of MMRD or MSI-H and lower frequency of PTEN and TMPRSS alterations in Black men compared with White men. Although Black men received targeted therapy less frequently than White men, no differences were observed in clinical outcomes.

Clinical implications of AR alterations in advanced prostate cancer: A multi-institutional collaboration.

Background: AR gene alterations can develop in response to pressure of testosterone suppression and androgen receptor targeting agents (ARTA). Despite this, the relevance of these gene alterations in the context of ARTA treatment and clinical outcomes remains unclear. Methods: Patients with castration-resistant prostate cancer (CRPC) who had undergone genomic testing and received ARTA treatment were identified in the Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE) database. Patients were stratified according to the timing of genomic testing relative to the first ARTA treatment (pre-/post-ARTA). Clinical outcomes such as time to progression, PSA response, and overall survival were compared based on alteration types. Results: In total, 540 CRPC patients who received ARTA and had tissue-based (n=321) and/or blood-based (n=244) genomic sequencing were identified. Median age was 62 years (range 39-90) at the time of the diagnosis. Majority were White (72.2%) and had metastatic disease (92.6%) at the time of the first ARTA treatment. Pre-ARTA genomic testing was available in 24.8% of the patients, and AR mutations and amplifications were observed in 8.2% and 13.1% of the patients, respectively. Further, time to progression was longer in patients with AR amplifications (25.7 months) compared to those without an AR alteration (9.6 months; p=0.03). In the post-ARTA group (n=406), AR mutations and AR amplifications were observed in 18.5% and 35.7% of the patients, respectively. The most common mutation in post-ARTA group was L702H (9.9%). Conclusion: To our knowledge, this is the largest real-world clinicogenomics database-driven study exploring the development of ARalterations and their association with ARTA treatment outcomes. Our study showed that AR amplifications are associated with longer time to progression on first ARTA treatment. Further prospective studies are needed to optimize therapeutic strategies for patients with AR alterations.

Effects of CYP3A4 and CYP2C9 genotype on systemic anastrozole and fulvestrant concentrations in SWOG S0226.

Objective & methods: This study tested associations of genotype-predicted activity of CYP3A4, other pharmacogenes, SLC28A7 (rs11648166) and ALPPL2 (rs28845026) with systemic concentrations of the endocrine therapies anastrozole and fulvestrant in SWOG S0226 trial participants. Results: Participants in the anastrozole-only arm with low CYP3A4 activity (i.e. CYP3A4*22 carriers) had higher systemic anastrozole concentrations than patients with high CYP3A4 activity (ß-coefficient = 10.03; 95% CI: 1.42, 18.6; p = 0.025). In an exploratory analysis, participants with low CYP2C9 activity had lower anastrozole concentrations and higher fulvestrant concentrations than participants with high CYP2C9 activity. Conclusion: Inherited genetic variation in CYP3A4 and CYP2C9 may affect concentrations of endocrine therapy and may be useful to personalize dosing and improve treatment outcomes.

Impaired in vitro fertilization outcomes following testosterone treatment improve with washout in a mouse model of gender-affirming hormone treatment.

BACKGROUND: The impact of gender-affirming testosterone on fertility is poorly understood, with ovarian histopathologic studies showing variable results, some with a detrimental effect on reproductive capacity and uncertain reversibility. Assisted reproductive outcome data are restricted to small case series that lack the ability to inform clinical practice guidelines and limit fertility preservation counseling for transgender and nonbinary individuals. OBJECTIVE: This study aimed to determine the impact of current testosterone and testosterone washout on in vitro fertilization outcomes in a mouse model for gender-affirming hormone treatment. We hypothesized that current or previous testosterone treatment would not affect in vitro fertilization outcomes. STUDY DESIGN: C57BL/6N female mice (n=120) were assigned to 4 treatment groups: (1) current control, (2) current testosterone, (3) control washout, and (4) testosterone washout. Testosterone implants remained in situ for 6 or 12 weeks, representing the short- and long-term treatment arms, respectively. Current treatment groups underwent ovarian stimulation with implants in place, and washout treatment groups were explanted and had ovarian stimulation after 2 weeks. Oocytes were collected, fertilized, and cultured in vitro, with one arm continuing to the blastocyst stage and the other having transfer of cleavage-stage embryos. Statistical analysis was performed using GraphPad Prism, version 9.0 and R statistical software, version 4.1.2, with statistical significance defined by P<.05. RESULTS: Current long-term testosterone treatment impaired in vitro fertilization outcomes, with fewer mature oocytes retrieved (13.7±5.1 [standard deviation] vs 28.6±7.8 [standard deviation]; P<.0001) leading to fewer cleavage-stage embryos (12.1±5.1 vs 26.5±8.2; P<.0001) and blastocysts (10.0±3.2 vs 25.0±6.5; P<.0001). There was recovery of in vitro fertilization outcomes following washout in the short-term treatment cohort, with incomplete reversibility in the long-term cohort. Testosterone did not negatively affect maturity, fertilization, or blastulation rates. CONCLUSION: In a mouse model of gender-affirming hormone treatment, testosterone negatively affected oocyte yield without affecting oocyte quality. Our findings suggest that testosterone reversibility is duration-dependent. These results demonstrate the feasibility of in vitro fertilization without testosterone discontinuation while supporting a washout period for optimization of mature oocyte yield.

Computational and Experimental Evidence for Templated Macrocyclization: The Role of a Hydrogen Bond Network in the Quantitative Dimerization of 24-Atom Macrocycles.

In the absence of preorganization, macrocyclization reactions are often plagued by oligomeric and polymeric side products. Here, a network of hydrogen bonds was identified as the basis for quantitative yields of macrocycles derived from the dimerization of monomers. Oligomers and polymers were not observed. Macrocyclization, the result of the formation of two hydrazones, was hypothesized to proceed in two steps. After condensation to yield the monohydrazone, a network of hydrogen bonds formed to preorganize the terminal acetal and hydrazine groups for cyclization. Experimental evidence for preorganization derived from macrocycles and acyclic models. Solution NMR spectroscopy and single-crystal X-ray diffraction revealed that the macrocycles isolated from the cyclization reaction were protonated twice. These protons contributed to an intramolecular network of hydrogen bonds that engaged distant carbonyl groups to realize a long-range order. DFT calculations showed that this network of hydrogen bonds contributed 8.7 kcal/mol to stability. Acyclic models recapitulated this network in solution. Condensation of an acetal and a triazinyl hydrazine, which adopted a number of conformational isomers, yielded a hydrazone that adopted a favored rotamer conformation in solution. The critical hydrogen-bonded proton was also evident. DFT calculations of acyclic models showed that the rotamers were isoenergetic when deprotonated. Upon protonation, however, energies diverged with one low-energy rotamer adopting the conformation observed in the macrocycle. This conformation anchored the network of hydrogen bonds of the intermediate. Computation revealed that the hydrogen-bonded network in the acyclic intermediate contributed up to 14 kcal/mol of stability and preorganized the acetal and hydrazine for cyclization.

Single-cell analysis of Sézary syndrome reveals novel markers and shifting gene profiles associated with treatment.

Cutaneous T-cell lymphomas (CTCLs) are a spectrum of diseases with varied clinical courses caused by malignant clonal proliferation of skin-tropic T cells. Most patients have an indolent disease course managed with skin-directed therapies. In contrast, others, especially in advanced stages of disease or with specific forms, have aggressive progression and poor median survival. Sézary syndrome (SS), a leukemic variant of CTCL, lacks highly consistent phenotypic and genetic markers that may be leveraged to prevent the delay in diagnosis experienced by most patients with CTCL and could be useful for optimal treatment selection. Using single-cell mRNA and T-cell receptor sequencing of peripheral blood immune cells in SS, we extensively mapped the transcriptomic variations of nearly 50 000 T cells of both malignant and nonmalignant origins. We identified potential diverging SS cell populations, including quiescent and proliferative populations shared across multiple patients. In particular, the expression of AIRE was the most highly upregulated gene in our analysis, and AIRE protein expression could be observed over a variety of CTCLs. Furthermore, within a single patient, we were able to characterize differences in cell populations by comparing malignant T cells over the course of treatment with histone deacetylase inhibition and photopheresis. New cellular clusters after progression of the therapy notably exhibited increased expression of the transcriptional factor FOXP3, a master regulator of regulatory T-cell function, raising the potential implication of an evolving mechanism of immune evasion.

BaySyn: Bayesian Evidence Synthesis for Multi-system Multiomic Integration.

The discovery of cancer drivers and drug targets are often limited to the biological systems - from cancer model systems to patients. While multiomic patient databases have sparse drug response data, cancer model systems databases, despite covering a broad range of pharmacogenomic platforms, provide lower lineage-specific sample sizes, resulting in reduced statistical power to detect both functional driver genes and their associations with drug sensitivity profiles. Hence, integrating evidence across model systems, taking into account the pros and cons of each system, in addition to multiomic integration, can more efficiently deconvolve cellular mechanisms of cancer as well as learn therapeutic associations. To this end, we propose BaySyn - a hierarchical Bayesian evidence synthesis framework for multi-system multiomic integration. BaySyn detects functionally relevant driver genes based on their associations with upstream regulators using additive Gaussian process models and uses this evidence to calibrate Bayesian variable selection models in the (drug) outcome layer. We apply BaySyn to multiomic cancer cell line and patient datasets from the Cancer Cell Line Encyclopedia and The Cancer Genome Atlas, respectively, across pan-gynecological cancers. Our mechanistic models implicate several relevant functional genes across cancers such as PTPN6 and ERBB2 in the KEGG adherens junction gene set. Furthermore, our outcome model is able to make higher number of discoveries in drug response models than its uncalibrated counterparts under the same thresholds of Type I error control, including detection of known lineage-specific biomarker associations such as BCL11A in breast and FGFRL1 in ovarian cancers. All our results and implementation codes are freely available via an interactive R Shiny dashboard at tinyurl.com/BaySynApp. The supplementary materials are available online at tinyurl.com/BaySynSup.

Antenna movements as a function of odorants' biological value in honeybees (Apis mellifera L.).

In honeybees, the antennae are highly mobile sensory organs that express scanning movements in various behavioral contexts and toward many stimuli, especially odorants. The rules underlying these movements are still unclear. Using a motion-capture system, we analyzed bees' antennal responses to a panel of pheromonal and other biologically relevant odorants. We observed clear differences in bees' antennal responses, with opposite movements to stimuli related to opposite contexts: slow backward movements were expressed in response to alarm pheromones, while fast forward movements were elicited by food related cues as well as brood and queen related pheromones. These responses are reproducible, as a similar pattern of odor-specific responses was observed in bees from different colonies, on different years. We then tested whether odorants' attractiveness for bees, measured using an original olfactory orientation setup, may predict antenna movements. This simple measure of odorants' valence did however not correlate with either antennal position or velocity measures, showing that more complex rules than simple hedonics underlie bees' antennal responses to odorants. Lastly, we show that newly-emerged bees express only limited antennal responses compared to older bees, suggesting that a significant part of the observed responses are acquired during bees' behavioral development.

Efficacy of enfortumab vedotin in advanced urothelial cancer: Analysis from the Urothelial Cancer Network to Investigate Therapeutic Experiences (UNITE) study.

BACKGROUND: Enfortumab vedotin (EV) is a novel antibody-drug conjugate approved for advanced urothelial cancer (aUC) refractory to prior therapy. In the Urothelial Cancer Network to Investigate Therapeutic Experiences (UNITE) study, the authors looked at the experience with EV in patient subsets of interest for which activity had not been well defined in clinical trials. METHODS: UNITE was a retrospective study of patients with aUC treated with recently approved agents. This initial analysis focused on patients treated with EV. Patient data were abstracted from chart reviews by investigators at each site. The observed response rate (ORR) was investigator-assessed for patients with at least 1 post-baseline scan or clear evidence of clinical progression. ORRs were compared across subsets of interest for patients treated with EV monotherapy. RESULTS: The initial UNITE analysis included 304 patients from 16 institutions; 260 of these patients were treated with EV monotherapy and included in the analyses. In the monotherapy cohort, the ORR was 52%, and it was >40% in all reported subsets of interest, including patients with comorbidities previously excluded from clinical trials (baseline renal impairment, diabetes, and neuropathy) and patients with fibroblast growth factor receptor 3 (FGFR3) alterations. Progression-free survival and overall survival were 6.8 and 14.4 months, respectively. Patients with a pure urothelial histology had a higher ORR than patients with a variant histology component (58% vs 42%; P = .06). CONCLUSIONS: In a large retrospective cohort, responses to EV monotherapy were consistent with data previously reported in clinical trials and were also observed in various patient subsets, including patients with variant histology, patients with FGFR3 alterations, and patients previously excluded from clinical trials with an estimated glomerular filtration rate < 30 mL/min and significant comorbidities. LAY SUMMARY: Enfortumab vedotin, approved by the Food and Drug Administration in 2019, is an important new drug for the treatment of patients with advanced bladder cancer. This study looks at the effectiveness of enfortumab vedotin as it has been used at multiple centers since approval, and focuses on important patient populations previously excluded from clinical trials. These populations include patients with decreased kidney function, diabetes, and important mutations. Enfortumab vedotin is effective for treating these patients. Previously reported clinical trial data have been replicated in this real-world setting, and support the use of this drug in broader patient populations.

Exploratory polarization facilitates mating partner selection in Saccharomyces cerevisiae.

Yeast decode pheromone gradients to locate mating partners, providing a model for chemotropism. How yeast polarize toward a single partner in crowded environments is unclear. Initially, cells often polarize in unproductive directions, but then they relocate the polarity site until two partners' polarity sites align, whereupon the cells "commit" to each other by stabilizing polarity to promote fusion. Here we address the role of the early mobile polarity sites. We found that commitment by either partner failed if just one partner was defective in generating, orienting, or stabilizing its mobile polarity sites. Mobile polarity sites were enriched for pheromone receptors and G proteins, and we suggest that such sites engage in an exploratory search of the local pheromone landscape, stabilizing only when they detect elevated pheromone levels. Mobile polarity sites were also enriched for pheromone secretion factors, and simulations suggest that only focal secretion at polarity sites would produce high pheromone concentrations at the partner's polarity site, triggering commitment.

A transcriptomic map of murine and human alopecia areata.

Alopecia areata (AA) is a common autoimmune condition, presenting initially with loss of hair without other overt skin changes. The unremarkable appearance of the skin surface contrasts with the complex immune activity occurring at the hair follicle. AA pathogenesis is due to the loss of immune privilege of the hair follicle, leading to autoimmune attack. Although the literature has focused on CD8+ T cells, vital roles for CD4+ T cells and antigen-presenting cells have been suggested. Here, we use single-cell sequencing to reveal distinct expression profiles of immune cells in murine AA. We found clonal expansions of both CD4+ and CD8+ T cells, with shared clonotypes across varied transcriptional states. The murine AA data were used to generate highly predictive models of human AA disease. Finally, single-cell sequencing of T cells in human AA recapitulated the clonotypic findings and the gene expression of the predictive models.

Individualized treatment effects with censored data via fully nonparametric Bayesian accelerated failure time models.

Individuals often respond differently to identical treatments, and characterizing such variability in treatment response is an important aim in the practice of personalized medicine. In this article, we describe a nonparametric accelerated failure time model that can be used to analyze heterogeneous treatment effects (HTE) when patient outcomes are time-to-event. By utilizing Bayesian additive regression trees and a mean-constrained Dirichlet process mixture model, our approach offers a flexible model for the regression function while placing few restrictions on the baseline hazard. Our nonparametric method leads to natural estimates of individual treatment effect and has the flexibility to address many major goals of HTE assessment. Moreover, our method requires little user input in terms of model specification for treatment covariate interactions or for tuning parameter selection. Our procedure shows strong predictive performance while also exhibiting good frequentist properties in terms of parameter coverage and mitigation of spurious findings of HTE. We illustrate the merits of our proposed approach with a detailed analysis of two large clinical trials (N = 6769) for the prevention and treatment of congestive heart failure using an angiotensin-converting enzyme inhibitor. The analysis revealed considerable evidence for the presence of HTE in both trials as demonstrated by substantial estimated variation in treatment effect and by high proportions of patients exhibiting strong evidence of having treatment effects which differ from the overall treatment effect.

Ratiometric GPCR signaling enables directional sensing in yeast.

Accurate detection of extracellular chemical gradients is essential for many cellular behaviors. Gradient sensing is challenging for small cells, which can experience little difference in ligand concentrations on the up-gradient and down-gradient sides of the cell. Nevertheless, the tiny cells of the yeast Saccharomyces cerevisiae reliably decode gradients of extracellular pheromones to find their mates. By imaging the behavior of polarity factors and pheromone receptors, we quantified the accuracy of initial polarization during mating encounters. We found that cells bias the orientation of initial polarity up-gradient, even though they have unevenly distributed receptors. Uneven receptor density means that the gradient of ligand-bound receptors does not accurately reflect the external pheromone gradient. Nevertheless, yeast cells appear to avoid being misled by responding to the fraction of occupied receptors rather than simply the concentration of ligand-bound receptors. Such ratiometric sensing also serves to amplify the gradient of active G protein. However, this process is quite error-prone, and initial errors are corrected during a subsequent indecisive phase in which polarity clusters exhibit erratic mobile behavior.

AR(1) latent class models for longitudinal count data.

In a variety of applications involving longitudinal or repeated-measurements data, it is desired to uncover natural groupings or clusters that exist among study subjects. Motivated by the need to recover clusters of longitudinal trajectories of conduct problems in the field of developmental psychopathology, we propose a method to address this goal when the response data in question are counts. We assume the subject-specific observations are generated from a first-order autoregressive process that is appropriate for count data. A key advantage of our approach is that the class-specific likelihood function arising from each subject's data can be expressed in closed form, circumventing common computational issues associated with random effects models. To further improve computational efficiency, we propose an approximate EM procedure for estimating the model parameters where, within each EM iteration, the maximization step is approximated by solving an appropriately chosen set of estimating equations. We explore the effectiveness of our procedures through simulations based on a four-class model, placing a special emphasis on recovery of the latent trajectories. Finally, we analyze data and recover trajectories of conduct problems in an important nationally representative sample. The methods discussed here are implemented in the R package inarmix, which is available from the Comprehensive R Archive Network (http://cran.r-project.org).

Considerations when assessing heterogeneity of treatment effect in patient-centered outcomes research.

When baseline risk of an outcome varies within a population, the effect of a treatment on that outcome will vary on at least one scale (e.g., additive, multiplicative). This treatment effect heterogeneity is of interest in patient-centered outcomes research. Based on a literature review and solicited expert opinion, we assert the following: (1) Treatment effect heterogeneity on the additive scale is most interpretable to health-care providers and patients using effect estimates to guide treatment decision-making; heterogeneity reported on the multiplicative scale may be misleading as to the magnitude or direction of a substantively important interaction. (2) The additive scale may give clues about sufficient-cause interaction, although such interaction is typically not relevant to patients' treatment choices. (3) Statistical modeling need not be conducted on the same scale as results are communicated. (4) Statistical testing is one tool for investigations, provided important subgroups are identified a priori, but test results should be interpreted cautiously given nonequivalence of statistical and clinical significance. (5) Qualitative interactions should be evaluated in a prespecified manner for important subgroups. Principled analytic plans that take into account the purpose of investigation of treatment effect heterogeneity are likely to yield more useful results for guiding treatment decisions.

Logistic Regression Likelihood Ratio Test Analysis for Detecting Signals of Adverse Events in Post-market Safety Surveillance.

The Vaccine Adverse Event Reporting System (VAERS) and other product surveillance systems compile reports of product-associated adverse events (AEs), and these reports may include a wide range of information including age, gender, and concomitant vaccines. Controlling for possible confounding variables such as these is an important task when utilizing surveillance systems to monitor post-market product safety. A common method for handling possible confounders is to compare observed product-AE combinations with adjusted baseline frequencies where the adjustments are made by stratifying on observable characteristics. Though approaches such as these have proven to be useful, in this article we propose a more flexible logistic regression approach which allows for covariates of all types rather than relying solely on stratification. Indeed, a main advantage of our approach is that the general regression framework provides flexibility to incorporate additional information such as demographic factors and concomitant vaccines. As part of our covariate-adjusted method, we outline a procedure for signal detection that accounts for multiple comparisons and controls the overall Type 1 error rate. To demonstrate the effectiveness of our approach, we illustrate our method with an example involving febrile convulsion, and we further evaluate its performance in a series of simulation studies.

Bayesian analysis of heterogeneous treatment effects for patient-centered outcomes research.

Evaluation of heterogeneity of treatment effect (HTE) is an essential aspect of personalized medicine and patient-centered outcomes research. Our goal in this article is to promote the use of Bayesian methods for subgroup analysis and to lower the barriers to their implementation by describing the ways in which the companion software beanz can facilitate these types of analyses. To advance this goal, we describe several key Bayesian models for investigating HTE and outline the ways in which they are well-suited to address many of the commonly cited challenges in the study of HTE. Topics highlighted include shrinkage estimation, model choice, sensitivity analysis, and posterior predictive checking. A case study is presented in which we demonstrate the use of the methods discussed.

Making the cut: improved ranking and selection for large-scale inference.

Identifying leading measurement units from a large collection is a common inference task in various domains of large-scale inference. Testing approaches, which measure evidence against a null hypothesis rather than effect magnitude, tend to overpopulate lists of leading units with those associated with low measurement error. By contrast, local maximum likelihood (ML) approaches tend to favor units with high measurement error. Available Bayesian and empirical Bayesian approaches rely on specialized loss functions that result in similar deficiencies. We describe and evaluate a generic empirical Bayesian ranking procedure that populates the list of top units in a way that maximizes the expected overlap between the true and reported top lists for all list sizes. The procedure relates unit-specific posterior upper tail probabilities with their empirical distribution to yield a ranking variable. It discounts high-variance units less than popular non-ML methods and thus achieves improved operating characteristics in the models considered.