Federated Learning for Sparse Bayesian Models with Applications to Electronic Health Records and Genomics.

Federated learning is becoming increasingly more popular as the concern of privacy breaches rises across disciplines including the biological and biomedical fields. The main idea is to train models locally on each server using data that are only available to that server and aggregate the model (not data) information at the global level. While federated learning has made significant advancements for machine learning methods such as deep neural networks, to the best of our knowledge, its development in sparse Bayesian models is still lacking. Sparse Bayesian models are highly interpretable with natural uncertain quantification, a desirable property for many scientific problems. However, without a federated learning algorithm, their applicability to sensitive biological/biomedical data from multiple sources is limited. Therefore, to fill this gap in the literature, we propose a new Bayesian federated learning framework that is capable of pooling information from different data sources without breaching privacy. The proposed method is conceptually simple to understand and implement, accommodates sampling heterogeneity (i.e., non-iid observations) across data sources, and allows for principled uncertainty quantification. We illustrate the proposed framework with three concrete sparse Bayesian models, namely, sparse regression, Markov random field, and directed graphical models. The application of these three models is demonstrated through three real data examples including a multi-hospital COVID-19 study, breast cancer protein-protein interaction networks, and gene regulatory networks.

A Bayesian nonparametric approach for evaluating the causal effect of treatment in randomized trials with semi-competing risks.

We develop a Bayesian nonparametric (BNP) approach to evaluate the causal effect of treatment in a randomized trial where a nonterminal event may be censored by a terminal event, but not vice versa (i.e., semi-competing risks). Based on the idea of principal stratification, we define a novel estimand for the causal effect of treatment on the nonterminal event. We introduce identification assumptions, indexed by a sensitivity parameter, and show how to draw inference using our BNP approach. We conduct simulation studies and illustrate our methodology using data from a brain cancer trial. The R code implementing our model and algorithm is available for download at https://github.com/YanxunXu/BaySemiCompeting.

Real-World Effectiveness of Remdesivir in Adults Hospitalized With Coronavirus Disease 2019 (COVID-19): A Retrospective, Multicenter Comparative Effectiveness Study.

BACKGROUND: There is an urgent need to understand the real-world effectiveness of remdesivir in the treatment of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). METHODS: This was a retrospective comparative effectiveness study. Individuals hospitalized in a large private healthcare network in the United States from 23 February 2020 through 11 February 2021 with a positive test for SARS-CoV-2 and ICD-10 diagnosis codes consistent with symptomatic coronavirus disease 2019 (COVID-19) were included. Remdesivir recipients were matched to controls using time-dependent propensity scores. The primary outcome was time to improvement with a secondary outcome of time to death. RESULTS: Of 96 859 COVID-19 patients, 42 473 (43.9%) received at least 1 remdesivir dose. The median age of remdesivir recipients was 65 years, 23 701 (55.8%) were male, and 22 819 (53.7%) were non-White. Matches were found for 18 328 patients (43.2%). Remdesivir recipients were significantly more likely to achieve clinical improvement by 28 days (adjusted hazard ratio [aHR] 1.19, 95% confidence interval [CI], 1.16-1.22). Remdesivir patients on no oxygen (aHR 1.30, 95% CI, 1.22-1.38) or low-flow oxygen (aHR 1.23, 95% CI, 1.19-1.27) were significantly more likely to achieve clinical improvement by 28 days. There was no significant impact on the likelihood of mortality overall (aHR 1.02, 95% CI, .97-1.08). Remdesivir recipients on low-flow oxygen were significantly less likely to die than controls (aHR 0.85, 95% CI, .77-.92; 28-day mortality 8.4% [865 deaths] for remdesivir patients, 12.5% [1334 deaths] for controls). CONCLUSIONS: These results support the use of remdesivir for hospitalized COVID-19 patients on no or low-flow oxygen. Routine initiation of remdesivir in more severely ill patients is unlikely to be beneficial.

Peripheral Plasma Cells Associated with Mortality Benefit in Severe COVID-19: A Marker of Disease Resolution.

BACKGROUND: Cytokines seen in severe coronavirus disease 2019 (COVID-19) are associated with proliferation, differentiation, and survival of plasma cells. Plasma cells are not routinely found in peripheral blood, though may produce virus-neutralizing antibodies in COVID-19 later in the course of an infection. METHODS: Using the Johns Hopkins COVID-19 Precision Medicine Analytics Platform Registry, we identified hospitalized adult patients with confirmed severe acute respiratory coronavirus 2 (SARS-CoV-2) infection and stratified by presence of plasma cells and World Health Organization (WHO) disease severity. To identify plasma cells, we employed a sensitive flow cytometric screening method for highly fluorescent lymphocytes and confirmed these microscopically. Cox regression models were used to evaluate time to death and time to clinical improvement by the presence of plasma cells in patients with severe disease. RESULTS: Of 2301 hospitalized patients with confirmed infection, 371 had plasma cells identified. Patients with plasma cells were more likely to have severe disease, though 86.6% developed plasma cells after onset of severe disease. In patients with severe disease, after adjusting for age, sex, body mass index, race, and other covariates associated with disease severity, patients with plasma cells had a reduced hazard of death (adjusted hazard ratio: 0.57; 95% confidence interval: 0.38-0.87; P value: .008). There was no significant association with the presence of plasma cells and time to clinical improvement. CONCLUSIONS: Patients with severe disease who have detectable plasma cells in the peripheral blood have improved mortality despite adjusting for known covariates associated with disease severity in COVID-19. Further investigation is warranted to understand the role of plasma cells in the immune response to COVID-19.

Comparison of Time to Clinical Improvement With vs Without Remdesivir Treatment in Hospitalized Patients With COVID-19.

Importance: Clinical effectiveness data on remdesivir are urgently needed, especially among diverse populations and in combination with other therapies. Objective: To examine whether remdesivir administered with or without corticosteroids for treatment of coronavirus disease 2019 (COVID-19) is associated with more rapid clinical improvement in a racially/ethnically diverse population. Design, Setting, and Participants: This retrospective comparative effectiveness research study was conducted from March 4 to August 29, 2020, in a 5-hospital health system in the Baltimore, Maryland, and Washington, DC, area. Of 2483 individuals with confirmed severe acute respiratory syndrome coronavirus 2 infection assessed by polymerase chain reaction, those who received remdesivir were matched to infected individuals who did not receive remdesivir using time-invariant covariates (age, sex, race/ethnicity, Charlson Comorbidity Index, body mass index, and do-not-resuscitate or do-not-intubate orders) and time-dependent covariates (ratio of peripheral blood oxygen saturation to fraction of inspired oxygen, blood pressure, pulse, temperature, respiratory rate, C-reactive protein level, complete white blood cell count, lymphocyte count, albumin level, alanine aminotransferase level, glomerular filtration rate, dimerized plasmin fragment D [D-dimer] level, and oxygen device). An individual in the remdesivir group with k days of treatment was matched to a control patient who stayed in the hospital at least k days (5 days maximum) beyond the matching day. Exposures: Remdesivir treatment with or without corticosteroid administration. Main Outcomes and Measures: The primary outcome was rate of clinical improvement (hospital discharge or decrease of 2 points on the World Health Organization severity score), and the secondary outcome, mortality at 28 days. An additional outcome was clinical improvement and time to death associated with combined remdesivir and corticosteroid treatment. Results: Of 2483 consecutive admissions, 342 individuals received remdesivir, 184 of whom also received corticosteroids and 158 of whom received remdesivir alone. For these 342 patients, the median age was 60 years (interquartile range, 46-69 years), 189 (55.3%) were men, and 276 (80.7%) self-identified as non-White race/ethnicity. Remdesivir recipients had a shorter time to clinical improvement than matched controls without remdesivir treatment (median, 5.0 days [interquartile range, 4.0-8.0 days] vs 7.0 days [interquartile range, 4.0-10.0 days]; adjusted hazard ratio, 1.47 [95% CI, 1.22-1.79]). Remdesivir recipients had a 28-day mortality rate of 7.7% (22 deaths) compared with 14.0% (40 deaths) among matched controls, but this difference was not statistically significant in the time-to-death analysis (adjusted hazard ratio, 0.70; 95% CI, 0.38-1.28). The addition of corticosteroids to remdesivir was not associated with a reduced hazard of death at 28 days (adjusted hazard ratio, 1.94; 95% CI, 0.67-5.57). Conclusions and Relevance: In this comparative effectiveness research study of adults hospitalized with COVID-19, receipt of remdesivir was associated with faster clinical improvement in a cohort of predominantly non-White patients. Remdesivir plus corticosteroid administration did not reduce the time to death compared with remdesivir administered alone.

Comprehensive molecular characterization of mitochondrial genomes in human cancers.

Mitochondria are essential cellular organelles that play critical roles in cancer. Here, as part of the International Cancer Genome Consortium/The Cancer Genome Atlas Pan-Cancer Analysis of Whole Genomes Consortium, which aggregated whole-genome sequencing data from 2,658 cancers across 38 tumor types, we performed a multidimensional, integrated characterization of mitochondrial genomes and related RNA sequencing data. Our analysis presents the most definitive mutational landscape of mitochondrial genomes and identifies several hypermutated cases. Truncating mutations are markedly enriched in kidney, colorectal and thyroid cancers, suggesting oncogenic effects with the activation of signaling pathways. We find frequent somatic nuclear transfers of mitochondrial DNA, some of which disrupt therapeutic target genes. Mitochondrial copy number varies greatly within and across cancers and correlates with clinical variables. Co-expression analysis highlights the function of mitochondrial genes in oxidative phosphorylation, DNA repair and the cell cycle, and shows their connections with clinically actionable genes. Our study lays a foundation for translating mitochondrial biology into clinical applications.

Sex Differences in Neurocognitive Function in Adults with HIV: Patterns, Predictors, and Mechanisms.

PURPOSE OF REVIEW: Sex differences in cognitive function are well documented yet few studies had adequate numbers of women and men living with HIV (WLWH; MLWH) to identify sex differences in neurocognitive impairment (NCI) and the factors contributing to NCI. Here, we review evidence that WLWH may be at greater risk for NCI. RECENT FINDINGS: We conducted a systematic review of recent studies of NCI in WLWH versus MLWH. A power analysis showed that few HIV studies have sufficient power to address male/female differences in NCI but studies with adequate power find evidence of greater NCI in WLWH, particularly in the domains of memory, speed of information processing, and motor function. Sex is an important determinant of NCI in HIV, and may relate to male/female differences in cognitive reserve, comorbidities (mental health and substance use disorders), and biological factors (e.g., inflammation, hormonal, genetic).

Bayesian non-parametric survival regression for optimizing precision dosing of intravenous busulfan in allogeneic stem cell transplantation.

Allogeneic stem cell transplantation (allo-SCT) is now part of standard of care for acute leukemia (AL). To reduce toxicity of the pre-transplant conditioning regimen, intravenous busulfan is usually used as a preparative regimen for AL patients undergoing allo-SCT. Systemic busulfan exposure, characterized by the area under the plasma concentration versus time curve (AUC), is strongly associated with clinical outcome. An AUC that is too high is associated with severe toxicities, while an AUC that is too low carries increased risks of disease recurrence and failure to engraft. Consequently, an optimal AUC interval needs to be determined for therapeutic use. To address the possibility that busulfan pharmacokinetics and pharmacodynamics vary significantly with patient characteristics, we propose a tailored approach to determine optimal covariate-specific AUC intervals. To estimate these personalized AUC intervals, we apply a flexible Bayesian nonparametric regression model based on a dependent Dirichlet process and Gaussian process, DDP-GP. Our analyses of a dataset of 151 patients identified optimal therapeutic intervals for AUC that varied substantively with age and whether the patient was in complete remission or had active disease at transplant. Extensive simulations to evaluate the DDP-GP model in similar settings showed that its performance compares favorably to alternative methods. We provide an R package, DDPGPSurv, that implements the DDP-GP model for a broad range of survival regression analyses.

A nonparametric Bayesian basket trial design.

Targeted therapies on the basis of genomic aberrations analysis of the tumor have shown promising results in cancer prognosis and treatment. Regardless of tumor type, trials that match patients to targeted therapies for their particular genomic aberrations have become a mainstream direction of therapeutic management of patients with cancer. Therefore, finding the subpopulation of patients who can most benefit from an aberration-specific targeted therapy across multiple cancer types is important. We propose an adaptive Bayesian clinical trial design for patient allocation and subpopulation identification. We start with a decision theoretic approach, including a utility function and a probability model across all possible subpopulation models. The main features of the proposed design and population finding methods are the use of a flexible nonparametric Bayesian survival regression based on a random covariate-dependent partition of patients, and decisions based on a flexible utility function that reflects the requirement of the clinicians appropriately and realistically, and the adaptive allocation of patients to their superior treatments. Through extensive simulation studies, the new method is demonstrated to achieve desirable operating characteristics and compares favorably against the alternatives.

Prediction of lung cancer risk at follow-up screening with low-dose CT: a training and validation study of a deep learning method.

Background: Current lung cancer screening guidelines use mean diameter, volume or density of the largest lung nodule in the prior computed tomography (CT) or appearance of new nodule to determine the timing of the next CT. We aimed at developing a more accurate screening protocol by estimating the 3-year lung cancer risk after two screening CTs using deep machine learning (ML) of radiologist CT reading and other universally available clinical information. Methods: A deep machine learning (ML) algorithm was developed from 25,097 participants who had received at least two CT screenings up to two years apart in the National Lung Screening Trial. Double-blinded validation was performed using 2,294 participants from the Pan-Canadian Early Detection of Lung Cancer Study (PanCan). Performance of ML score to inform lung cancer incidence was compared with Lung-RADS and volume doubling time using time-dependent ROC analysis. Exploratory analysis was performed to identify individuals with aggressive cancers and higher mortality rates. Findings: In the PanCan validation cohort, ML showed excellent discrimination with a 1-, 2- and 3-year time-dependent AUC values for cancer diagnosis of 0·968±0·013, 0·946±0·013 and 0·899±0·017. Although high ML score cohort included only 10% of the PanCan sample, it identified 94%, 85%, and 71% of incident and interval lung cancers diagnosed within 1, 2, and 3 years, respectively, after the second screening CT. Furthermore, individuals with high ML score had significantly higher mortality rates (HR=16·07, p<0·001) compared to those with lower risk. Interpretation: ML tool that recognizes patterns in both temporal and spatial changes as well as synergy among changes in nodule and non-nodule features may be used to accurately guide clinical management after the next scheduled repeat screening CT.

TCGA-assembler 2: software pipeline for retrieval and processing of TCGA/CPTAC data.

Motivation: The Cancer Genome Atlas (TCGA) program has produced huge amounts of cancer genomics data providing unprecedented opportunities for research. In 2014, we developed TCGA-Assembler, a software pipeline for retrieval and processing of public TCGA data. In 2016, TCGA data were transferred from the TCGA data portal to the Genomic Data Commons (GDCs), which is supported by a different set of data storage and retrieval mechanisms. In addition, new proteomics data of TCGA samples have been generated by the Clinical Proteomic Tumor Analysis Consortium (CPTAC) program, which were not available for downloading through TCGA-Assembler. It is desirable to acquire and integrate data from both GDC and CPTAC. Results: We develop TCGA-assembler 2 (TA2) to automatically download and integrate data from GDC and CPTAC. We make substantial improvement on the functionality of TA2 to enhance user experience and software performance. TA2 together with its previous version have helped more than 2000 researchers from 64 countries to access and utilize TCGA and CPTAC data in their research. Availability of TA2 will continue to allow existing and new users to conduct reproducible research based on TCGA and CPTAC data. Availability and implementation: http://www.compgenome.org/TCGA-Assembler/ or https://github.com/compgenome365/TCGA-Assembler-2. Contact: zhuyitan@gmail.com or koaeraser@gmail.com. Supplementary information: Supplementary data are available at Bioinformatics online.

A Decision-Theoretic Comparison of Treatments to Resolve Air Leaks After Lung Surgery Based on Nonparametric Modeling.

We propose a Bayesian nonparametric utility-based group sequential design for a randomized clinical trial to compare a gel sealant to standard care for resolving air leaks after pulmonary resection. Clinically, resolving air leaks in the days soon after surgery is highly important, since longer resolution time produces undesirable complications that require extended hospitalization. The problem of comparing treatments is complicated by the fact that the resolution time distributions are skewed and multi-modal, so using means is misleading. We address these challenges by assuming Bayesian nonparametric probability models for the resolution time distributions and basing the comparative test on weighted means. The weights are elicited as clinical utilities of the resolution times. The proposed design uses posterior expected utilities as group sequential test criteria. The procedure's frequentist properties are studied by extensive simulations.

Bayesian nonparametric statistics: A new toolkit for discovery in cancer research.

Many commonly used statistical methods for data analysis or clinical trial design rely on incorrect assumptions or assume an over-simplified framework that ignores important information. Such statistical practices may lead to incorrect conclusions about treatment effects or clinical trial designs that are impractical or that do not accurately reflect the investigator's goals. Bayesian nonparametric (BNP) models and methods are a very flexible new class of statistical tools that can overcome such limitations. This is because BNP models can accurately approximate any distribution or function and can accommodate a broad range of statistical problems, including density estimation, regression, survival analysis, graphical modeling, neural networks, classification, clustering, population models, forecasting and prediction, spatiotemporal models, and causal inference. This paper describes 3 illustrative applications of BNP methods, including a randomized clinical trial to compare treatments for intraoperative air leaks after pulmonary resection, estimating survival time with different multi-stage chemotherapy regimes for acute leukemia, and evaluating joint effects of targeted treatment and an intermediate biological outcome on progression-free survival time in prostate cancer.

Bayesian Nonparametric Estimation for Dynamic Treatment Regimes with Sequential Transition Times.

We analyze a dataset arising from a clinical trial involving multi-stage chemotherapy regimes for acute leukemia. The trial design was a 2 × 2 factorial for frontline therapies only. Motivated by the idea that subsequent salvage treatments affect survival time, we model therapy as a dynamic treatment regime (DTR), that is, an alternating sequence of adaptive treatments or other actions and transition times between disease states. These sequences may vary substantially between patients, depending on how the regime plays out. To evaluate the regimes, mean overall survival time is expressed as a weighted average of the means of all possible sums of successive transitions times. We assume a Bayesian nonparametric survival regression model for each transition time, with a dependent Dirichlet process prior and Gaussian process base measure (DDP-GP). Posterior simulation is implemented by Markov chain Monte Carlo (MCMC) sampling. We provide general guidelines for constructing a prior using empirical Bayes methods. The proposed approach is compared with inverse probability of treatment weighting, including a doubly robust augmented version of this approach, for both single-stage and multi-stage regimes with treatment assignment depending on baseline covariates. The simulations show that the proposed nonparametric Bayesian approach can substantially improve inference compared to existing methods. An R program for implementing the DDP-GP-based Bayesian nonparametric analysis is freely available at https://www.ma.utexas.edu/users/yxu/.

Subgroup-Based Adaptive (SUBA) Designs for Multi-Arm Biomarker Trials.

Targeted therapies based on biomarker profiling are becoming a mainstream direction of cancer research and treatment. Depending on the expression of specific prognostic biomarkers, targeted therapies assign different cancer drugs to subgroups of patients even if they are diagnosed with the same type of cancer by traditional means, such as tumor location. For example, Herceptin is only indicated for the subgroup of patients with HER2+ breast cancer, but not other types of breast cancer. However, subgroups like HER2+ breast cancer with effective targeted therapies are rare and most cancer drugs are still being applied to large patient populations that include many patients who might not respond or benefit. Also, the response to targeted agents in humans is usually unpredictable. To address these issues, we propose SUBA, subgroup-based adaptive designs that simultaneously search for prognostic subgroups and allocate patients adaptively to the best subgroup-specific treatments throughout the course of the trial. The main features of SUBA include the continuous reclassification of patient subgroups based on a random partition model and the adaptive allocation of patients to the best treatment arm based on posterior predictive probabilities. We compare the SUBA design with three alternative designs including equal randomization, outcome-adaptive randomization and a design based on a probit regression. In simulation studies we find that SUBA compares favorably against the alternatives.

Comprehensive Characterization of Molecular Differences in Cancer between Male and Female Patients.

An individual's sex has been long recognized as a key factor affecting cancer incidence, prognosis, and treatment responses. However, the molecular basis for sex disparities in cancer remains poorly understood. We performed a comprehensive analysis of molecular differences between male and female patients in 13 cancer types of The Cancer Genome Atlas and revealed two sex-effect groups associated with distinct incidence and mortality profiles. One group contains a small number of sex-affected genes, whereas the other shows much more extensive sex-biased molecular signatures. Importantly, 53% of clinically actionable genes (60/114) show sex-biased signatures. Our study provides a systematic molecular-level understanding of sex effects in diverse cancers and suggests a pressing need to develop sex-specific therapeutic strategies in certain cancer types.

Bayesian inference for latent biologic structure with determinantal point processes (DPP).

We discuss the use of the determinantal point process (DPP) as a prior for latent structure in biomedical applications, where inference often centers on the interpretation of latent features as biologically or clinically meaningful structure. Typical examples include mixture models, when the terms of the mixture are meant to represent clinically meaningful subpopulations (of patients, genes, etc.). Another class of examples are feature allocation models. We propose the DPP prior as a repulsive prior on latent mixture components in the first example, and as prior on feature-specific parameters in the second case. We argue that the DPP is in general an attractive prior model for latent structure when biologically relevant interpretation of such structure is desired. We illustrate the advantages of DPP prior in three case studies, including inference in mixture models for magnetic resonance images (MRI) and for protein expression, and a feature allocation model for gene expression using data from The Cancer Genome Atlas. An important part of our argument are efficient and straightforward posterior simulation methods. We implement a variation of reversible jump Markov chain Monte Carlo simulation for inference under the DPP prior, using a density with respect to the unit rate Poisson process.