Multiple models for outbreak decision support in the face of uncertainty.

Policymakers must make management decisions despite incomplete knowledge and conflicting model projections. Little guidance exists for the rapid, representative, and unbiased collection of policy-relevant scientific input from independent modeling teams. Integrating approaches from decision analysis, expert judgment, and model aggregation, we convened multiple modeling teams to evaluate COVID-19 reopening strategies for a mid-sized United States county early in the pandemic. Projections from seventeen distinct models were inconsistent in magnitude but highly consistent in ranking interventions. The 6-mo-ahead aggregate projections were well in line with observed outbreaks in mid-sized US counties. The aggregate results showed that up to half the population could be infected with full workplace reopening, while workplace restrictions reduced median cumulative infections by 82%. Rankings of interventions were consistent across public health objectives, but there was a strong trade-off between public health outcomes and duration of workplace closures, and no win-win intermediate reopening strategies were identified. Between-model variation was high; the aggregate results thus provide valuable risk quantification for decision making. This approach can be applied to the evaluation of management interventions in any setting where models are used to inform decision making. This case study demonstrated the utility of our approach and was one of several multimodel efforts that laid the groundwork for the COVID-19 Scenario Modeling Hub, which has provided multiple rounds of real-time scenario projections for situational awareness and decision making to the Centers for Disease Control and Prevention since December 2020.

Increasing aggressive prostate cancer.

INTRODUCTION: To compare prostate biopsy (Pbx) characteristics, before and after the 2012 United States Preventive Services Task Force (USPSTF) prostate cancer screening guidelines in our practice. MATERIALS AND METHODS: We completed a retrospective comparative analysis of 1703 sequential patients that had a Pbx in 2010 to 2012 (3 years) with 1006 patients biopsied in 2018, 2019 and 2021 (3 years). Data from a total of 2709 Pbx was collected on patient age, race, prostate-specific antigen (PSA), digital rectal examination (DRE) and Gleason sum score (GSS). The data was analyzed to determine whether the 2012 USPSTF screening recommendations against prostate cancer screening may have affected prostate cancer characteristics. Two study groups were defined as Group A and Group B. Group A represents Pbx prior to the 2012 USPSTF screening guidelines (2010-2012) and Group B represents Pbx in 2018-19 and 2021. The patient population consisted of 76% Black, 14% White and 11% other. RESULTS: The number of patients that had a Pbx in Groups A vs. B: 567 patients/year vs. 335 patients/year. The annual positive Pbx rate for Group A vs. B: 134/year vs. 175/year. High grade prostate cancer (GSS 7-10) in Groups A vs. B: 51.5% vs. 59%. The proportion of patients with a PSA 10 ng/mL or greater in Groups A vs. B: 25.4% vs. 31%. The PSA 10 ng/mL and over and GSS 7-10 was higher in Group B for all age groups. In 2021, GSS 7-10 was present in 64% of 70-80 year olds. In Group B, GSS 6 decreased by 7.5% while GSS 7-10 increased by 7.5% compared with Group A. CONCLUSIONS: Our data through the year 2021 shows that after the 2012 USPSTF recommendations against prostate cancer screening, Pbx decreased and prostate cancer diagnosis and high grade (GSS 7-10) prostate cancer increased. As our patient population consists of 76% Black patients and 33% of men age 70-80 years old, our results support annual prostate cancer screening for US men 50-80 years old and especially high-risk patients that include Black men, men with a family history of prostate cancer and healthy men age 70-80 years old. Annual DRE- and PSA- based prostate cancer screening will likely markedly decrease prostate cancer morbidity, mortality and the cost of prostate cancer management.

A mathematical model for persistent post-CSD vasoconstriction.

Cortical spreading depression (CSD) is the propagation of a relatively slow wave in cortical brain tissue that is linked to a number of pathological conditions such as stroke and migraine. Most of the existing literature investigates the dynamics of short term phenomena such as the depolarization and repolarization of membrane potentials or large ion shifts. Here, we focus on the clinically-relevant hour-long state of neurovascular malfunction in the wake of CSDs. This dysfunctional state involves widespread vasoconstriction and a general disruption of neurovascular coupling. We demonstrate, using a mathematical model, that dissolution of calcium that has aggregated within the mitochondria of vascular smooth muscle cells can drive an hour-long disruption. We model the rate of calcium clearance as well as the dynamical implications on overall blood flow. Based on reaction stoichiometry, we quantify a possible impact of calcium phosphate dissolution on the maintenance of F0F1-ATP synthase activity.

Reconstruction of Cell Focal Adhesions using Physical Constraints and Compressive Regularization.

We develop a method to reconstruct, from measured displacements of an underlying elastic substrate, the spatially dependent forces that cells or tissues impart on it. Given newly available high-resolution images of substrate displacements, it is desirable to be able to reconstruct small-scale, compactly supported focal adhesions that are often localized and exist only within the footprint of a cell. In addition to the standard quadratic data mismatch terms that define least-squares fitting, we motivate a regularization term in the objective function that penalizes vectorial invariants of the reconstructed surface stress while preserving boundaries. We solve this inverse problem by providing a numerical method for setting up a discretized inverse problem that is solvable by standard convex optimization techniques. By minimizing the objective function subject to a number of important physically motivated constraints, we are able to efficiently reconstruct stress fields with localized structure from simulated and experimental substrate displacements. Our method incorporates the exact solution for a stress tensor accurate to first-order finite differences and motivates the use of distance-based cutoffs for data inclusion and problem sparsification.

A transcription factor network specifying inhibitory versus excitatory neurons in the dorsal spinal cord.

The proper balance of excitatory and inhibitory neurons is crucial for normal processing of somatosensory information in the dorsal spinal cord. Two neural basic helix-loop-helix transcription factors (TFs), Ascl1 and Ptf1a, have contrasting functions in specifying these neurons. To understand how Ascl1 and Ptf1a function in this process, we identified their direct transcriptional targets genome-wide in the embryonic mouse neural tube using ChIP-Seq and RNA-Seq. We show that Ascl1 and Ptf1a directly regulate distinct homeodomain TFs that specify excitatory or inhibitory neuronal fates. In addition, Ascl1 directly regulates genes with roles in several steps of the neurogenic program, including Notch signaling, neuronal differentiation, axon guidance and synapse formation. By contrast, Ptf1a directly regulates genes encoding components of the neurotransmitter machinery in inhibitory neurons, and other later aspects of neural development distinct from those regulated by Ascl1. Moreover, Ptf1a represses the excitatory neuronal fate by directly repressing several targets of Ascl1. Ascl1 and Ptf1a bind sequences primarily enriched for a specific E-Box motif (CAGCTG) and for secondary motifs used by Sox, Rfx, Pou and homeodomain factors. Ptf1a also binds sequences uniquely enriched in the CAGATG E-box and in the binding motif for its co-factor Rbpj, providing two factors that influence the specificity of Ptf1a binding. The direct transcriptional targets identified for Ascl1 and Ptf1a provide a molecular understanding of how these DNA-binding proteins function in neuronal development, particularly as key regulators of homeodomain TFs required for neuronal subtype specification.

Regulatory inhibition of biological tissue mineralization by calcium phosphate through post-nucleation shielding by fetuin-A.

In vertebrates, insufficient availability of calcium and inorganic phosphate ions in extracellular fluids leads to loss of bone density and neuronal hyper-excitability. To counteract this problem, calcium ions are usually present at high concentrations throughout bodily fluids-at concentrations exceeding the saturation point. This condition leads to the opposite situation where unwanted mineral sedimentation may occur. Remarkably, ectopic or out-of-place sedimentation into soft tissues is rare, in spite of the thermodynamic driving factors. This fortunate fact is due to the presence of auto-regulatory proteins that are found in abundance in bodily fluids. Yet, many important inflammatory disorders such as atherosclerosis and osteoarthritis are associated with this undesired calcification. Hence, it is important to gain an understanding of the regulatory process and the conditions under which it can go awry. In this manuscript, we extend mean-field continuum classical nucleationtheory of the growth of clusters to encompass surface shielding. We use this formulation to study the regulation of sedimentation of calcium phosphate salts in biological tissues through the mechanism of post-nuclear shielding of nascent mineral particles by binding proteins. We develop a mathematical description of this phenomenon using a countable system of hyperbolic partial differential equations. A critical concentration of regulatory protein is identified as a function of the physical parameters that describe the system.

Bayesian Uncertainty Quantification for Bond Energies and Mobilities Using Path Integral Analysis.

Dynamic single-molecule force spectroscopy is often used to distort bonds. The resulting responses, in the form of rupture forces, work applied, and trajectories of displacements, are used to reconstruct bond potentials. Such approaches often rely on simple parameterizations of one-dimensional bond potentials, assumptions on equilibrium starting states, and/or large amounts of trajectory data. Parametric approaches typically fail at inferring complicated bond potentials with multiple minima, while piecewise estimation may not guarantee smooth results with the appropriate behavior at large distances. Existing techniques, particularly those based on work theorems, also do not address spatial variations in the diffusivity that may arise from spatially inhomogeneous coupling to other degrees of freedom in the macromolecule. To address these challenges, we develop a comprehensive empirical Bayesian approach that incorporates data and regularization terms directly into a path integral. All experimental and statistical parameters in our method are estimated directly from the data. Upon testing our method on simulated data, our regularized approach requires less data and allows simultaneous inference of both complex bond potentials and diffusivity profiles. Crucially, we show that the accuracy of the reconstructed bond potential is sensitive to the spatially varying diffusivity and accurate reconstruction can be expected only when both are simultaneously inferred. Moreover, after providing a means for self-consistently choosing regularization parameters from data, we derive posterior probability distributions, allowing for uncertainty quantification.

Gradient-flow adaptive importance sampling for Bayesian leave one out cross-validation for sigmoidal classification models.

We introduce a set of gradient-flow-guided adaptive importance sampling (IS) transformations to stabilize Monte-Carlo approximations of point-wise leave one out cross-validated (LOO) predictions for Bayesian classification models. One can leverage this methodology for assessing model generalizability by for instance computing a LOO analogue to the AIC or computing LOO ROC/PRC curves and derived metrics like the AUROC and AUPRC. By the calculus of variations and gradient flow, we derive two simple nonlinear single-step transformations that utilize gradient information to shift a model's pre-trained full-data posterior closer to the target LOO posterior predictive distributions. In doing so, the transformations stabilize importance weights. Because the transformations involve the gradient of the likelihood function, the resulting Monte Carlo integral depends on Jacobian determinants with respect to the model Hessian. We derive closed-form exact formulae for these Jacobian determinants in the cases of logistic regression and shallow ReLU-activated artificial neural networks, and provide a simple approximation that sidesteps the need to compute full Hessian matrices and their spectra. We test the methodology on an n≪p dataset that is known to produce unstable LOO IS weights.

Interpretable (not just posthoc-explainable) medical claims modeling for discharge placement to reduce preventable all-cause readmissions or death.

We developed an inherently interpretable multilevel Bayesian framework for representing variation in regression coefficients that mimics the piecewise linearity of ReLU-activated deep neural networks. We used the framework to formulate a survival model for using medical claims to predict hospital readmission and death that focuses on discharge placement, adjusting for confounding in estimating causal local average treatment effects. We trained the model on a 5% sample of Medicare beneficiaries from 2008 and 2011, based on their 2009-2011 inpatient episodes (approximately 1.2 million), and then tested the model on 2012 episodes (approximately 400 thousand). The model scored an out-of-sample AUROC of approximately 0.75 on predicting all-cause readmissions-defined using official Centers for Medicare and Medicaid Services (CMS) methodology-or death within 30-days of discharge, being competitive against XGBoost and a Bayesian deep neural network, demonstrating that one need-not sacrifice interpretability for accuracy. Crucially, as a regression model, it provides what blackboxes cannot-its exact gold-standard global interpretation, explicitly defining how the model performs its internal "reasoning" for mapping the input data features to predictions. In doing so, we identify relative risk factors and quantify the effect of discharge placement. We also show that the posthoc explainer SHAP provides explanations that are inconsistent with the ground truth model reasoning that our model readily admits.

AAV8 gene therapy reverses cardiac pathology and prevents early mortality in a mouse model of Friedreich's ataxia.

Friedreich's ataxia (FRDA) is an autosomal-recessive disorder primarily attributed to biallelic GAA repeat expansions that reduce expression of the mitochondrial protein frataxin (FXN). FRDA is characterized by progressive neurodegeneration, with many patients developing cardiomyopathy that progresses to heart failure and death. The potential to reverse or prevent progression of the cardiac phenotype of FRDA was investigated in a mouse model of FRDA, using an adeno-associated viral vector (AAV8) containing the coding sequence of the FXN gene. The Fxnflox/null::MCK-Cre conditional knockout mouse (FXN-MCK) has an FXN gene ablation that prevents FXN expression in cardiac and skeletal muscle, leading to cardiac insufficiency, weight loss, and morbidity. FXN-MCK mice received a single intravenous injection of an AAV8 vector containing human (hFXN) or mouse (mFXN) FXN genes under the control of a phosphoglycerate kinase promoter. Compared to vehicle-treated FXN-MCK control mice, AAV-treated FXN-MCK mice displayed increases in body weight, reversal of cardiac deficits, and increases in survival without apparent toxicity in the heart or liver for up to 12 weeks postdose. FXN protein expression in heart tissue was detected in a dose-dependent manner, exhibiting wide distribution throughout the heart similar to wild type, but more speckled. These results support an AAV8-based approach to treat FRDA-associated cardiomyopathy.

Accuracy of Medical Student Measurements of CT Right-to-Left Ventricular Diameter in Patients with Acute Pulmonary Embolism.

Objectives: Acute pulmonary embolism (PE) is a common disease, necessitating risk stratification to determine management. A right ventricle (RV) to left ventricle (LV) diameter ratio ≥1.0 on computed tomography pulmonary angiography (CTPA) suggests RV strain, which may indicate a worse prognosis. Two prior studies showed that residents with brief training by a radiologist could accurately measure RV/LV ratio. We assessed whether medical students could accurately measure RV dilatation. Methods: We conducted a post hoc analysis of a retrospective cohort study of adults undergoing management for acute PE at 21 community emergency departments across Kaiser Permanente Northern California from 2013 to 2015. We created a sample, stratified to contain an equal number of patients from each of the 5 PE Severity Index classes. Four medical students measured RV and LV diameter on CTPA after training from an emergency medicine physician and an interventional radiologist. We used Cohen's kappa statistics, Bland-Altman plots, and Pearson correlation coefficients to assess interrater reliability. Results: Of the 108 CTPAs reviewed, 79 (73%) showed RV dilatation and 29 (27%) did not. The kappa statistic for the presence of RV dilatation of the medical students compared to the radiologist showed moderate agreement for 3 medical students (kappa (95% CI): 0.46 (0.21-0.70), 0.49 (0.31-0.68), 0.50 (0.32-0.68)) and fair agreement for 1 medical student (kappa (95% CI): 0.29 (0.10-0.47)). The average interrater differences in RV/LV ratio between a radiologist and each of the 4 medical students were -0.04, -0.05, 0.04, and 0.24. Pearson correlation coefficients were 0.87, 0.80, 0.74, and 0.78, respectively, indicating moderate correlation (P < .001 for all). Conclusion: Medical students were able to identify RV dilatation on CTPA in moderate agreement with that of a radiologist. Further study is needed to determine whether medical student accuracy could improve with additional training.

Regularized Bayesian calibration and scoring of the WD-FAB IRT model improves predictive performance over marginal maximum likelihood.

Item response theory (IRT) is the statistical paradigm underlying a dominant family of generative probabilistic models for test responses, used to quantify traits in individuals relative to target populations. The graded response model (GRM) is a particular IRT model that is used for ordered polytomous test responses. Both the development and the application of the GRM and other IRT models require statistical decisions. For formulating these models (calibration), one needs to decide on methodologies for item selection, inference, and regularization. For applying these models (test scoring), one needs to make similar decisions, often prioritizing computational tractability and/or interpretability. In many applications, such as in the Work Disability Functional Assessment Battery (WD-FAB), tractability implies approximating an individual's score distribution using estimates of mean and variance, and obtaining that score conditional on only point estimates of the calibrated model. In this manuscript, we evaluate the calibration and scoring of models under this common use-case using Bayesian cross-validation. Applied to the WD-FAB responses collected for the National Institutes of Health, we assess the predictive power of implementations of the GRM based on their ability to yield, on validation sets of respondents, ability estimates that are most predictive of patterns of item responses. Our main finding indicates that regularized Bayesian calibration of the GRM outperforms the regularization-free empirical Bayesian procedure of marginal maximum likelihood. We also motivate the use of compactly supported priors in test scoring.

Cerebellar Pathology in an Inducible Mouse Model of Friedreich Ataxia.

Friedreich ataxia (FRDA) is an autosomal recessive neurodegenerative disorder caused by deficiency of the mitochondrial protein frataxin. Lack of frataxin causes neuronal loss in various areas of the CNS and PNS. In particular, cerebellar neuropathology in FRDA patients includes loss of large principal neurons and synaptic terminals in the dentate nucleus (DN), and previous studies have demonstrated early synaptic deficits in the Knockin-Knockout mouse model of FRDA. However, the exact correlation of frataxin deficiency with cerebellar neuropathology remains unclear. Here we report that doxycycline-induced frataxin knockdown in a mouse model of FRDA (FRDAkd) leads to synaptic cerebellar degeneration that can be partially reversed by AAV8-mediated frataxin restoration. Loss of cerebellar Purkinje neurons and large DN principal neurons are observed in the FRDAkd mouse cerebellum. Levels of the climbing fiber-specific glutamatergic synaptic marker VGLUT2 decline starting at 4 weeks after dox induction, whereas levels of the parallel fiber-specific synaptic marker VGLUT1 are reduced by 18-weeks. These findings suggest initial selective degeneration of climbing fiber synapses followed by loss of parallel fiber synapses. The GABAergic synaptic marker GAD65 progressively declined during dox induction in FRDAkd mice, while GAD67 levels remained unaltered, suggesting specific roles for frataxin in maintaining cerebellar synaptic integrity and function during adulthood. Expression of frataxin following AAV8-mediated gene transfer partially restored VGLUT1/2 levels. Taken together, our findings show that frataxin knockdown leads to cerebellar degeneration in the FRDAkd mouse model, suggesting that frataxin helps maintain cerebellar structure and function.

Biphasic direct current shift, haemoglobin desaturation and neurovascular uncoupling in cortical spreading depression.

Cortical spreading depression is a propagating wave of depolarization that plays important roles in migraine, stroke, subarachnoid haemorrhage and brain injury. Cortical spreading depression is associated with profound vascular changes that may be a significant factor in the clinical response to cortical spreading depression events. We used a combination of optical intrinsic signal imaging, electro-physiology, potassium sensitive electrodes and spectroscopy to investigate neurovascular changes associated with cortical spreading depression in the mouse. We identified two distinct phases of altered neurovascular function, one during the propagating cortical spreading depression wave and a second much longer phase after passage of the wave. The direct current shift associated with the cortical spreading depression wave was accompanied by marked arterial constriction and desaturation of cortical haemoglobin. After recovery from the initial cortical spreading depression wave, we observed a second phase of prolonged, negative direct current shift, arterial constriction and haemoglobin desaturation, lasting at least an hour. Persistent disruption of neurovascular coupling was demonstrated by a loss of coherence between electro-physiological activity and perfusion. Extracellular potassium concentration increased during the cortical spreading depression wave, but recovered and remained at baseline after passage of the wave, consistent with different mechanisms underlying the first and second phases of neurovascular dysfunction. These findings indicate that cortical spreading depression is associated with a multiphasic alteration in neurovascular function, including a novel second direct current shift accompanied by arterial constriction and decrease in tissue oxygen supply, that is temporally and mechanistically distinct from the initial propagated cortical spreading depression wave. Vascular/metabolic uncoupling with cortical spreading depression may have important clinical consequences, and the different phases of dysfunction may represent separate therapeutic targets in the disorders where cortical spreading depression occurs.

Global prediction of unreported SARS-CoV2 infection from observed COVID-19 cases.

Estimation of infectiousness and fatality of the SARS-CoV-2 virus in the COVID-19 global pandemic is complicated by ascertainment bias resulting from incomplete and non-representative samples of infected individuals. We developed a strategy for overcoming this bias to obtain more plausible estimates of the true values of key epidemiological variables. We fit mechanistic Bayesian latent-variable SIR models to confirmed COVID-19 cases, deaths, and recoveries, for all regions (countries and US states) independently. Bayesian averaging over models, we find that the raw infection incidence rate underestimates the true rate by a factor, the case ascertainment ratio CARt that depends upon region and time. At the regional onset of COVID-19, the predicted global median was 13 infections unreported for each case confirmed (CARt = 0.07 C.I. (0.02, 0.4)). As the infection spread, the median CARt rose to 9 unreported cases for every one diagnosed as of April 15, 2020 (CARt = 0.1 C.I. (0.02, 0.5)). We also estimate that the median global initial reproduction number R0 is 3.3 (C.I (1.5, 8.3)) and the total infection fatality rate near the onset is 0.17% (C.I. (0.05%, 0.9%)). However the time-dependent reproduction number Rt and infection fatality rate as of April 15 were 1.2 (C.I. (0.6, 2.5)) and 0.8% (C.I. (0.2%,4%)), respectively. We find that there is great variability between country- and state-level values. Our estimates are consistent with recent serological estimates of cumulative infections for the state of New York, but inconsistent with claims that very large fractions of the population have already been infected in most other regions. For most regions, our estimates imply a great deal of uncertainty about the current state and trajectory of the epidemic.

Inorganic polyphosphate is required for sustained free mitochondrial calcium elevation, following calcium uptake.

Mitochondrial free calcium is critically linked to the regulation of cellular metabolism. Free ionic calcium concentration within these organelles is determined by the interplay between two processes: exchange across the mitochondrial inner membrane and calcium-buffering within the matrix. During stimulated calcium uptake, calcium is primarily buffered by orthophosphate, preventing calcium toxicity while allowing for well-regulated yet elevated calcium loads. However, if limited to orthophosphates only, this buffering system is expected to lead to the irreversible formation of insoluble precipitates, which are not observed in living cells, under physiological conditions. Here, we demonstrate that the regulation of free mitochondrial calcium requires the presence of free inorganic polyphosphate (polyP) within the organelle. We found that the overexpression of a mitochondrial-targeted enzyme hydrolyzing polyP leads to the loss of the cellular ability to maintain elevated calcium concentrations within the organelle, following stimulated cytoplasmic signal. We hypothesize that the presence of polyP prevents the formation of calcium-phosphate insoluble clusters, allowing for the maintenance of elevated free calcium levels, during stimulated calcium uptake.

COVID-19 reopening strategies at the county level in the face of uncertainty: Multiple Models for Outbreak Decision Support.

Policymakers make decisions about COVID-19 management in the face of considerable uncertainty. We convened multiple modeling teams to evaluate reopening strategies for a mid-sized county in the United States, in a novel process designed to fully express scientific uncertainty while reducing linguistic uncertainty and cognitive biases. For the scenarios considered, the consensus from 17 distinct models was that a second outbreak will occur within 6 months of reopening, unless schools and non-essential workplaces remain closed. Up to half the population could be infected with full workplace reopening; non-essential business closures reduced median cumulative infections by 82%. Intermediate reopening interventions identified no win-win situations; there was a trade-off between public health outcomes and duration of workplace closures. Aggregate results captured twice the uncertainty of individual models, providing a more complete expression of risk for decision-making purposes.

Predictive Bayesian selection of multistep Markov chains, applied to the detection of the hot hand and other statistical dependencies in free throws.

Consider the problem of modelling memory effects in discrete-state random walks using higher-order Markov chains. This paper explores cross-validation and information criteria as proxies for a model's predictive accuracy. Our objective is to select, from data, the number of prior states of recent history upon which a trajectory is statistically dependent. Through simulations, I evaluate these criteria in the case where data are drawn from systems with fixed orders of history, noting trends in the relative performance of the criteria. As a real-world illustrative example of these methods, this manuscript evaluates the problem of detecting statistical dependencies in shot outcomes in free throw shooting. Over three National Basketball Association (NBA) seasons analysed, several players exhibited statistical dependencies in free throw hitting probability of various types-hot handedness, cold handedness and error correction. For the 2013-2014 to 2015-2016 NBA seasons, I detected statistical dependencies in 23% of all player-seasons. Focusing on a single player, in two of these three seasons, LeBron James shot a better percentage after an immediate miss than otherwise. Conditioning on the previous outcome makes for a more-predictive model than treating free throw makes as independent. When extended specifically to LeBron James' 2016-2017 season, a model depending on the previous shot (single-step Markovian) does not clearly beat a model with independent outcomes. An error-correcting variable length model of two parameters, where James shoots a higher percentage after a missed free throw than otherwise, is more predictive than either model.

Adaptive adipose tissue stromal plasticity in response to cold stress and antibody-based metabolic therapy.

In response to environmental and nutrient stress, adipose tissues must establish a new homeostatic state. Here we show that cold exposure of obese mice triggers an adaptive tissue remodeling in visceral adipose tissue (VAT) that involves extracellular matrix deposition, angiogenesis, sympathetic innervation, and adipose tissue browning. Obese VAT is predominated by pro-inflammatory M1 macrophages; cold exposure induces an M1-to-M2 shift in macrophage composition and dramatic changes in macrophage gene expression in both M1 and M2 macrophages. Antibody-mediated CSF1R blocking prevented the cold-induced recruitment of adipose tissue M2 macrophages, suggesting the role of CSF1R signaling in the process. These cold-induced effects in obese VAT are phenocopied by an administration of the FGF21-mimetic antibody, consistent with its action to stimulate sympathetic nerves. Collectively, these studies illuminate adaptive visceral adipose tissue plasticity in obese mice in response to cold stress and antibody-based metabolic therapy.