Search

OBJECTIVE Although seizures are a relatively common phenomenon in the setting of brain metastases (BMs), there are no discrete recommendations regarding the use of antiepileptic drugs (AEDs) in this population, either in general or in the context of treatment. The authors’ aim was to better understand the underlying pathological factors as well as the therapeutic techniques that may lead to seizures following the radiosurgical treatment of BMs with the goal of guiding appropriate AED prophylaxis. METHODS Adult patients with BMs diagnosed from 2013 to 2020 at a single academic institution and treated with radiation therapy were included in this study. The authors evaluated factors associated with the incidence of seizures throughout the disease course, with a focus on seizures in the 90-day period following stereotactic radiosurgery (SRS). RESULTS Four hundred forty-four patients with newly diagnosed BMs were identified, 10% of whom had seizures at the time of presentation and 28% of whom had a seizure at any point during the study period. Tumor histology was significantly associated with initial seizure risk. AED use was highly variable. In the 90-day post-SRS period, the summed total planning target volume (PTV) was independently predictive of post-SRS seizures, regardless of the fractionation scheme (single fraction vs hypofractionated) and other clinical factors. The number of supratentorial BMs was not predictive of post-SRS seizures. CONCLUSIONS PTV is a superior predictor of post-SRS seizures relative to the number of supratentorial BMs, as it serves as a volumetric proxy for intracranial disease burden. A larger PTV, alongside tumor histology and prior seizure history, should be considered in the decision-making process for AED use following radiosurgery.

Bayesian methods provide direct uncertainty quantification in functional data analysis applications without reliance on bootstrap techniques. A major tool in functional data applications is the functional principal component analysis which decomposes the data around a common mean function and identifies leading directions of variation. Bayesian functional principal components analysis (BFPCA) provides uncertainty quantification on the estimated functional model components via the posterior samples obtained. We propose central posterior envelopes (CPEs) for BFPCA based on functional depth as a descriptive visualization tool to summarize variation in the posterior samples of the estimated functional model components, contributing to uncertainty quantification in BFPCA. The proposed BFPCA relies on a latent factor model and targets model parameters within a hierarchical modeling framework using modified multiplicative gamma process shrinkage priors on the variance components. Functional depth provides a center-outward order to a sample of functions. We utilize modified band depth and modified volume depth for ordering of a sample of functions and surfaces, respectively, to derive at CPEs of the mean and eigenfunctions within the BFPCA framework. The proposed CPEs are showcased in extensive simulations. Finally, the proposed CPEs are applied to the analysis of a sample of power spectral densities from resting state electroencephalography where they lead to novel insights on diagnostic group differences among children diagnosed with autism spectrum disorder and their typically developing peers across age.

Supplemental Figure 5. Gene ontology terms enriched in gene expression microarray analysis following β-catenin knockdown.

Figure S1 - CONSORT diagram.

Purpose:In a single-institution phase II study, we evaluated the safety of a 5-day dose-equivalent neoadjuvant radiotherapy (RT) regimen for high-risk primary soft tissue sarcoma.Patients and Methods:Patients received neoadjuvant RT alone (30 Gy in five fractions) to the primary tumor with standard margins. The primary endpoint was grade ≥2 late-radiation toxicity. Major wound complications, local recurrences, and distant metastases were also examined. In exploratory analysis, we evaluated germline biomarkers for wound toxicity and the effects of the study on treatment utilization.Results:Over 2 years, 52 patients were enrolled with median follow-up of 29 months. Seven of 44 evaluable patients (16%) developed grade ≥2 late toxicity. Major wound complications occurred in 16 of 50 patients (32%); a signature defined by 19 germline SNPs in miRNA-binding sites of immune and DNA damage response genes, in addition to lower extremity tumor location, demonstrated strong predictive performance for major wound complications. Compared with the preceding 2-year period, the number of patients treated with neoadjuvant RT alone at our institution increased 3-fold, with a concomitant increase in the catchment area.Conclusions:A shorter 5-day neoadjuvant RT regimen results in favorable rates of wound complications and grade ≥2 toxicity after 2-year follow-up. Five-day RT significantly increased utilization of neoadjuvant RT at our high-volume sarcoma center. With further validation, a putative germline biomarker for wound complications may guide safer RT utilization.

Supplementary Tables S1-S5

Supplementary Figure Legends

Supplementary Methods

BACKGROUND AND PURPOSE: The purpose of this study was to determine whether single nucleotide polymorphisms disrupting microRNA targets (mirSNPs) can serve as predictive biomarkers for toxicity after radiotherapy for prostate cancer and whether these may be differentially predictive depending on radiation fractionation. MATERIALS AND METHODS: We identified 201 men treated with two forms of definitive radiotherapy for prostate cancer at two institutions: 108 men received conventionally-fractionated radiotherapy (CF-RT) and 93 received stereotactic body radiotherapy (SBRT). Germline DNA was evaluated for the presence of functional mirSNPs. Random forest, boosted trees and elastic net models were developed to predict late grade ≥2 GU toxicity by the RTOG scale. RESULTS: The crude incidence of late grade ≥2 GU toxicity was 16% after CF-RT and 15% after SBRT. An elastic net model based on 22 mirSNPs differentiated CF-RT patients at high risk (71.5%) versus low risk (7.5%) for toxicity, with an area under the curve (AUC) values of 0.76–0.81. An elastic net model based on 32 mirSNPs differentiated SBRT patients at high risk (64.7%) versus low risk (3.9%) for toxicity, with an area under the curve (AUC) values of 0.81–0.87. These models were specific to treatment type delivered. Prospective studies are warranted to further validate these results. CONCLUSION: Predictive models using germline mirSNPs have high accuracy for predicting late grade ≥2 GU toxicity after either CF-RT or SBRT, and are unique for each treatment, suggesting that germline predictors of late radiation sensitivity are fractionation-dependent. Prospective studies are warranted to further validate these results.

Electroencephalography experiments produce region-referenced functional data representing brain signals in the time or the frequency domain collected across the scalp. The data typically also have a multilevel structure with high-dimensional observations collected across multiple experimental conditions or visits. Common analysis approaches reduce the data complexity by collapsing the functional and regional dimensions, where event-related potential (ERP) features or band power are targeted in a pre-specified scalp region. This practice can fail to portray more comprehensive differences in the entire ERP signal or the power spectral density (PSD) across the scalp. Building on the weak separability of the high-dimensional covariance process, the proposed multilevel hybrid principal components analysis (M-HPCA) utilizes dimension reduction tools from both vector and functional principal components analysis to decompose the total variation into between- and within-subject variance. The resulting model components are estimated in a mixed effects modeling framework via a computationally efficient minorization-maximization algorithm coupled with bootstrap. The diverse array of applications of M-HPCA is showcased with two studies of individuals with autism. While ERP responses to match vs mismatch conditions are compared in an audio odd-ball paradigm in the first study, short-term reliability of the PSD across visits is compared in the second. Finite sample properties of the proposed methodology are studied in extensive simulations.

Purpose: In a single-institution phase II study, we evaluated the safety of a 5-day dose-equivalent neoadjuvant radiotherapy (RT) regimen for high-risk primary soft tissue sarcoma. Patients and Methods: Patients received neoadjuvant RT alone (30 Gy in five fractions) to the primary tumor with standard margins. The primary endpoint was grade ≥2 late-radiation toxicity. Major wound complications, local recurrences, and distant metastases were also examined. In exploratory analysis, we evaluated germline biomarkers for wound toxicity and the effects of the study on treatment utilization. Results: Over 2 years, 52 patients were enrolled with median follow-up of 29 months. Seven of 44 evaluable patients (16%) developed grade ≥2 late toxicity. Major wound complications occurred in 16 of 50 patients (32%); a signature defined by 19 germline SNPs in miRNA-binding sites of immune and DNA damage response genes, in addition to lower extremity tumor location, demonstrated strong predictive performance for major wound complications. Compared with the preceding 2-year period, the number of patients treated with neoadjuvant RT alone at our institution increased 3-fold, with a concomitant increase in the catchment area. Conclusions: A shorter 5-day neoadjuvant RT regimen results in favorable rates of wound complications and grade ≥2 toxicity after 2-year follow-up. Five-day RT significantly increased utilization of neoadjuvant RT at our high-volume sarcoma center. With further validation, a putative germline biomarker for wound complications may guide safer RT utilization.

Background Containing coronavirus disease 2019 (COVID-19) has been difficult, due to both the large number of asymptomatic infected individuals and the long duration of infection. Managing these challenges requires understanding of the differences between asymptomatic vs symptomatic patients and those with a longer duration of infectivity. Methods Individuals from Los Angeles were tested for COVID-19, and a group positive for COVID-19 chose to have follow-up testing. Associations between symptoms and demographic factors, viral burden measured by cycle threshold (CT) value, and duration of polymerase chain reaction (PCR) positivity were analyzed. Results Eighteen point eight percent of patients were positive for COVID-19. Asymptomatic COVID-19-positive patients were significantly younger than symptomatic patients (2.6 years; P Conclusions We found important differences and similarities between asymptomatic and symptomatic COVID-19-positive patients, the most meaningful being a similar level of virus as measured by PCR, but a shorter duration of PCR positivity for asymptomatic patients. These findings suggest that asymptomatic patients may have more efficient clearance of virus, which may be relevant for management and screening.

Objective It is unclear whether relatively low glycated haemoglobin (HbA1c) levels are beneficial or harmful for the long-term health outcomes among people without diabetes. We aimed to investigate the association between low HbA1c levels and mortality among the US general population. Methods This study includes a nationally representative sample of 39 453 US adults from the National Health and Nutrition Examination Surveys 1999–2014, linked to mortality data through 2015. We employed the parametric g-formula with pooled logistic regression models and the ensemble machine learning algorithms to estimate the time-varying risk of all-cause and cardiovascular mortality by HbA1c categories (low, 4.0 to Results Over a median follow-up of 7.5 years, 5118 (13%) all-cause deaths, and 1116 (3%) cardiovascular deaths were observed. Logistic regression models and machine learning algorithms showed nearly identical predictive performance of death and risk estimates. Compared with mid-level HbA1c, low HbA1c was associated with a 30% (95% CI, 16 to 48) and a 12% (95% CI, 3 to 22) increased risk of all-cause mortality at 5 years and 10 years of follow-up, respectively. We found no evidence that low HbA1c levels were associated with cardiovascular mortality risk. The diabetes group, but not the prediabetes group, also showed an increased risk of all-cause mortality. Conclusions Using the US national database and adjusting for an extensive set of potential confounders with flexible modelling, we found that adults with low HbA1c were at increased risk of all-cause mortality. Further evaluation and careful monitoring of low HbA1c levels need to be considered.

Wildfires have been increasing in frequency in the western United States (US) with the 2017 and 2018 fire seasons experiencing some of the worst wildfires in terms of suppression costs and air pollution that the western US has seen. Although growing evidence suggests respiratory exacerbations from elevated fine particulate matter (PM2.5) during wildfires, significantly less is known about the impacts on human health of ozone (O3) that may also be increased due to wildfires. Using machine learning, we created daily surface concentration maps for PM2.5 and O3 during an intense wildfire in California in 2008. We then linked these daily exposures to counts of respiratory hospitalizations and emergency department visits at the ZIP code level. We calculated relative risks of respiratory health outcomes using Poisson generalized estimating equations models for each exposure in separate and mutually-adjusted models, additionally adjusted for pertinent covariates. During the active fire periods, PM2.5 was significantly associated with exacerbations of asthma and chronic obstructive pulmonary disease (COPD) and these effects remained after controlling for O3. Effect estimates of O3 during the fire period were non-significant for respiratory hospitalizations but were significant for ED visits for asthma (RR = 1.05 and 95% CI = (1.022, 1.078) for a 10 ppb increase in O3). In mutually-adjusted models, the significant findings for PM2.5 remained whereas the associations with O3 were confounded. Adjusted for O3, the RR for asthma ED visits associated with a 10 μg/m3 increase in PM2.5 was 1.112 and 95% CI = (1.087, 1.138). The significant findings for PM2.5 but not for O3 in mutually-adjusted models is likely due to the fact that PM2.5 levels during these fires exceeded the 24-hour National Ambient Air Quality Standard (NAAQS) of 35 μg/m3 for 4976 ZIP-code days and reached levels up to 6.073 times the NAAQS, whereas our estimated O3 levels during the fire period only occasionally exceeded the NAAQS of 70 ppb with low exceedance levels. Future studies should continue to investigate the combined role of O3 and PM2.5 during wildfires to get a more comprehensive assessment of the cumulative burden on health from wildfire smoke. Keywords: Wildfires, Ozone, Particulate matter, Respiratory disease

Functional brain imaging through electroencephalography (EEG) relies upon the analysis and interpretation of high-dimensional, spatially organized time series. We propose to represent time-localized frequency domain characterizations of EEG data as region-referenced functional data. This representation is coupled with a hierarchical regression modeling approach to multivariate functional observations. Within this familiar setting we discuss how several prior models relate to structural assumptions about multivariate covariance operators. An overarching modeling framework, based on infinite factorial decompositions, is finally proposed to balance flexibility and efficiency in estimation. The motivating application stems from a study of implicit auditory learning, in which typically developing (TD) children, and children with autism spectrum disorder (ASD) were exposed to a continuous speech stream. Using the proposed model, we examine differential band power dynamics as brain function is interrogated throughout the duration of a computer-controlled experiment. Our work offers a novel look at previous findings in psychiatry and provides further insights into the understanding of ASD. Our approach to inference is fully Bayesian and implemented in a highly optimized Rcpp package.

BackgroundThere is great interest in finding ways to identify patients who will develop toxicity to cancer therapies. This has become especially pressing in the era of immune therapy, where toxicity can be long-lasting and life-altering, and primarily comes in the form of immune-related adverse effects (irAEs). Treatment with the first drugs in this class, anti-programmed death 1 (anti-PD1)/programmed death-ligand 1 (PDL1) checkpoint therapies, results in grade 2 or higher irAEs in up to 25%–30% of patients, which occur most commonly within the first 6 months of treatment and can include arthralgias, rash, pruritus, pneumonitis, diarrhea and/or colitis, hepatitis, and endocrinopathies. We tested the hypothesis that germline microRNA pathway functional variants, known to predict altered systemic stress responses to cancer therapies, would predict irAEs in patients across cancer types.MethodsMicroRNA pathway variants were evaluated for an association with grade 2 or higher toxicity using four classifiers on 62 patients with melanoma, and then the panel’s performance was validated on 99 patients with other cancer types. Trained classifiers included classification trees, LASSO-regularized logistic regression, boosted trees, and random forests. Final performance measures were reported on the training set using leave-one-out cross validation and validated on held-out samples. The predicted probability of toxicity was evaluated for its association, if any, with response categories to anti-PD1/PDL1 therapy in the melanoma cohort.ResultsA biomarker panel was identified that predicts toxicity with 80% accuracy (F1=0.76, area under the curve (AUC)=0.82) in the melanoma training cohort and 77.6% accuracy (F1=0.621, AUC=0.778) in the pan-cancer validation cohort. In the melanoma cohort, the predictive probability of toxicity was not associated with response categories to anti-PD1/PDL1 therapy (p=0.70). In the same cohort, the most significant biomarker of toxicity in RAC1, predicting a greater than ninefold increased risk of toxicity (pConclusionsA germline microRNA-based biomarker signature predicts grade 2 and higher irAEs to anti-PD1/PDL1 therapy, regardless of tumor type, in a pan-cancer manner. These findings represent an important step toward personalizing checkpoint therapy, the use of which is growing rapidly.

We developed a custom-designed liposome carrier for codelivery of a potent immunogenic cell death (ICD) stimulus plus an inhibitor of the indoleamine 2,3-dioxygenase (IDO-1) pathway to establish a chemo-immunotherapy approach for solid tumors in syngeneic mice. The carrier was constructed by remote import of the anthraquinone chemotherapeutic agent, mitoxantrone (MTO), into the liposomes, which were further endowed with a cholesterol-conjugated indoximod (IND) prodrug in the lipid bilayer. For proof-of-principle testing, we used IV injection of the MTO/IND liposome in a CT26 colon cancer model to demonstrate the generation of a robust immune response, characterized by the appearance of ICD markers (CRT and HMGB-1) as well as evidence of cytotoxic cancer cell death, mediated by perforin and granzyme B. Noteworthy, the cytotoxic effects involved natural killer (NK) cell, which suggests a different type of ICD response. The immunotherapy response was significantly augmented by codelivery of the IND prodrug, which induced additional CRT expression, reduced number of Foxp3+ Treg, and increased perforin release, in addition to extending animal survival beyond the effect of an MTO-only liposome. The outcome reflects the improved pharmacokinetics of MTO delivery to the cancer site by the carrier. In light of the success in the CT26 model, we also assessed the platform efficacy in further breast cancer (EMT6 and 4T1) and renal cancer (RENCA) models, which overexpress IDO-1. Encapsulated MTO delivery was highly effective for inducing chemo-immunotherapy responses, with NK participation, in all tumor models. Moreover, the growth inhibitory effect of MTO was enhanced by IND codelivery in EMT6 and 4T1 tumors. All considered, our data support the use of encapsulated MTO delivery for chemo-immunotherapy, with the possibility to boost the immune response by codelivery of an IDO-1 pathway inhibitor.

Electroencephalography (EEG) studies produce region-referenced functional data in the form of EEG signals recorded across electrodes on the scalp. It is of clinical interest to relate the highly structured EEG data to scalar outcomes such as diagnostic status. In our motivating study, resting state EEG is collected on both typically developing (TD) children and children with Autism Spectrum Disorder (ASD) aged two to twelve years old. The peak alpha frequency (PAF), defined as the location of a prominent peak in the alpha frequency band of the spectral density, is an important biomarker linked to neurodevelopment and is known to shift from lower to higher frequencies as children age. To retain the most amount of information from the data, we consider the oscillations in the spectral density within the alpha band, rather than just the peak location, as a functional predictor of diagnostic status (TD vs. ASD), adjusted for chronological age. A covariate-adjusted region-referenced generalized functional linear model (CARR-GFLM) is proposed for modeling scalar outcomes from region-referenced functional predictors, which utilizes a tensor basis formed from one-dimensional discrete and continuous bases to estimate functional effects across a discrete regional domain while simultaneously adjusting for additional non-functional covariates, such as age. The proposed methodology provides novel insights into differences in neural development of TD and ASD children. The efficacy of the proposed methodology is investigated through extensive simulation studies.

Summary Electroencephalography (EEG) data possess a complex structure that includes regional, functional, and longitudinal dimensions. Our motivating example is a word segmentation paradigm in which typically developing (TD) children, and children with autism spectrum disorder (ASD) were exposed to a continuous speech stream. For each subject, continuous EEG signals recorded at each electrode were divided into one-second segments and projected into the frequency domain via fast Fourier transform. Following a spectral principal components analysis, the resulting data consist of region-referenced principal power indexed regionally by scalp location, functionally across frequencies, and longitudinally by one-second segments. Standard EEG power analyses often collapse information across the longitudinal and functional dimensions by averaging power across segments and concentrating on specific frequency bands. We propose a hybrid principal components analysis for region-referenced longitudinal functional EEG data, which utilizes both vector and functional principal components analyses and does not collapse information along any of the three dimensions of the data. The proposed decomposition only assumes weak separability of the higher-dimensional covariance process and utilizes a product of one dimensional eigenvectors and eigenfunctions, obtained from the regional, functional, and longitudinal marginal covariances, to represent the observed data, providing a computationally feasible non-parametric approach. A mixed effects framework is proposed to estimate the model components coupled with a bootstrap test for group level inference, both geared towards sparse data applications. Analysis of the data from the word segmentation paradigm leads to valuable insights about group-region differences among the TD and verbal and minimally verbal children with ASD. Finite sample properties of the proposed estimation framework and bootstrap inference procedure are further studied via extensive simulations.

Background Chronic health effects of traffic-related air pollution, like nitrogen dioxide (NO2), are well-documented. Animal models suggested that NO2 exposures dysregulate cortisol function. Objectives We evaluated the association between traffic-related NO2 exposure and adolescent human cortisol concentrations, utilizing measures of the cortisol diurnal slope. Methods 140 adolescents provided repeated salivary cortisol samples throughout one day. We built a land use regression model to estimate chronic NO2 exposures based on home and school addresses. We then generated model-based estimates of the association between cortisol and NO2 exposure one year prior to cortisol sampling, examining changes in cortisol diurnal slope. The final model was adjusted other criteria pollutants, measures of psychosocial stress, anthropometry, and other demographic and covariates. Results We observed a decrease in diurnal slope in cortisol for adolescents exposed to the estimated 75th percentile of ambient NO2 (high exposure) relative to those exposed at the 25th percentile (low exposure). For a highly exposed adolescent, the log cortisol was lower by 0.06 µg/dl at waking (95% CI: −0.15, 0.02), 0.07 µg/dl at 30 min post waking (95% CI: −0.15, 0.02), and higher by 0.05 µg/dl at bedtime (95% CI: 0.05, 0.15), compared to a low exposed adolescent. For an additional interquartile range of exposure, the model-based predicted diurnal slope significantly decreased by 0.12 (95% CI: −0.23, −0.01). Conclusions In adolescents, we found that increased, chronic exposure to NO2 and the mixture of pollutants from traffic sources was associated with a flattened diurnal slope of cortisol, a marker of an abnormal cortisol response which we hypothesize may be a mechanism through which air pollution may affect respiratory function and asthma in adolescents.

Latino Americans are a rapidly growing ethnic group in the United States but studies of glioblastoma in this population are limited. We have evaluated characteristics of 21,184 glioblastoma patients from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute. This SEER data from 2001 to 2011 draws from 28% of the U.S.PopulationLatinos have a lower incidence of GBM and present slightly younger than non-Latino Whites. Cubans present at an older age than other Latino sub-populations. Latinos have a higher incidence of giant cell glioblastoma than non-Latino Whites while the incidence of gliosarcoma is similar. Despite lower rates of radiation therapy and greater rates of sub-total resection than non-Latino Whites, Latinos have better 1 and 5year survival rates. SEER does not record chemotherapy data. Survivals of Latino sub-populations are similar with each other. Age, extent of resection, and the use of radiation therapy are associated with improved survival but none of these variables are sufficient in a multivariate analysis to explain the improved survival of Latinos relative to non-Latino Whites. As molecular data is not available in SEER records, we studied the MGMT and IDH status of 571 patients from a UCLA database. MGMT methylation and IDH1 mutation rates are not statistically significantly different between non-Latino Whites and Latinos. For UCLA patients with available information, chemotherapy and radiation rates are similar for non-Latino White and Latino patients, but the latter have lower rates of gross total resection and present at a younger age.

The electroencephalography (EEG) data created in event-related potential (ERP) experiments have a complex high-dimensional structure. Each stimulus presentation, or trial, generates an ERP waveform which is an instance of functional data. The experiments are made up of sequences of multiple trials, resulting in longitudinal functional data and moreover, responses are recorded at multiple electrodes on the scalp, adding an electrode dimension. Traditional EEG analyses involve multiple simplifications of this structure to increase the signal-to-noise ratio, effectively collapsing the functional and longitudinal components by identifying key features of the ERPs and averaging them across trials. Motivated by an implicit learning paradigm used in autism research in which the functional, longitudinal, and electrode components all have critical interpretations, we propose a multidimensional functional principal components analysis (MD-FPCA) technique which does not collapse any of the dimensions of the ERP data. The proposed decomposition is based on separation of the total variation into subject and subunit level variation which are further decomposed in a two-stage functional principal components analysis. The proposed methodology is shown to be useful for modeling longitudinal trends in the ERP functions, leading to novel insights into the learning patterns of children with Autism Spectrum Disorder (ASD) and their typically developing peers as well as comparisons between the two groups. Finite sample properties of MD-FPCA are further studied via extensive simulations.

Summary Multi-dimensional functional data arises in numerous modern scientific experimental and observational studies. In this article, we focus on longitudinal functional data, a structured form of multidimensional functional data. Operating within a longitudinal functional framework we aim to capture low dimensional interpretable features. We propose a computationally efficient nonparametric Bayesian method to simultaneously smooth observed data, estimate conditional functional means and functional covariance surfaces. Statistical inference is based on Monte Carlo samples from the posterior measure through adaptive blocked Gibbs sampling. Several operative characteristics associated with the proposed modeling framework are assessed comparatively in a simulated environment. We illustrate the application of our work in two case studies. The first case study involves age-specific fertility collected over time for various countries. The second case study is an implicit learning experiment in children with autism spectrum disorder.

Epidemiologists use prediction models to downscale (i.e., interpolate) air pollution exposure where monitoring data is insufficient. This study compares machine learning prediction models for ground-level ozone during wildfires, evaluating the predictive accuracy of ten algorithms on the daily 8-hour maximum average ozone during a 2008 wildfire event in northern California. Models were evaluated using a leave-one-location-out cross-validation (LOLO CV) procedure to account for the spatial and temporal dependence of the data and produce more realistic estimates of prediction error. LOLO CV avoids both the well-known overly optimistic bias of k-fold cross-validation on dependent data and the conservative bias of evaluating prediction error over a coarser spatial resolution via leave-k-locations-out CV. Gradient boosting was the most accurate of the ten machine learning algorithms with the lowest LOLO CV estimated root mean square error (0.228) and the highest LOLO CV R ˆ 2 (0.677). Random forest was the second best performing algorithm with an LOLO CV R ˆ 2 of 0.661. The LOLO CV estimates of predictive accuracy were less optimistic than 10-fold CV estimates for all ten models. The difference in estimated accuracy between the 10-fold CV and LOLO CV was greater for more flexible models like gradient boosting and random forest. The order of estimated model accuracy depended on the choice of evaluation metric, indicating that 10-fold CV and LOLO CV may select different models or sets of covariates as optimal, which calls into question the reliability of 10-fold CV for model (or variable) selection. These prediction models are designed for interpolating ozone exposure, and are not suited to inferring the effect of wildfires on ozone or extrapolating to predict ozone in other spatial or temporal domains. This is demonstrated by the inability of the best performing models to accurately predict ozone during 2007 southern California wildfires.

Summary 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.

Motivated by a study on visual implicit learning in young children with Autism Spectrum Disorder (ASD), we propose a robust functional clustering (RFC) algorithm to identify subgroups within electroencephalography (EEG) data. The proposed RFC is an iterative algorithm based on functional principal component analysis, where cluster membership is updated via predictions of the functional trajectories obtained through a non-parametric random effects model. We consider functional data resulting from event-related potential (ERP) waveforms representing EEG time-locked to stimuli over the course of an implicit learning experiment, after applying a previously proposed meta-preprocessing step. This meta-preprocessing is designed to increase the low signal-to-noise ratio in the raw data and to mitigate the longitudinal changes in the ERP waveforms which characterize the nature and speed of learning. The resulting functional ERP components (peak amplitudes and latencies) inherently exhibit covariance heterogeneity due to low data quality over some stimuli inducing the averaging of different numbers of waveforms in sliding windows of the meta-preprocessing step. The proposed RFC algorithm incorporates this known covariance heterogeneity into the clustering algorithm, improving cluster quality, as illustrated in the data application and extensive simulation studies. ASD is a heterogeneous syndrome and identifying subgroups within ASD children is of interest for understanding the diverse nature of this complex disorder. Applications to the implicit learning paradigm identify subgroups within ASD and typically developing children with diverse learning patterns over the course of the experiment, which may inform clinical stratification of ASD.

Computational approaches have evolved as efficient alternatives to understand the adverse effects of nanoparticles on human health and the environment. The potential of using Quantitative Structure–Activity Relationship (QSAR) modeling to establish statistically significant models for predicting the cytotoxicity of various metal oxide (MeOx) nanoparticles (NPs) has been investigated. A novel kind of nanospecific theoretical descriptor was proposed by integrating codes of certain physicochemical features into SMILES-based optimal descriptors to characterize the nanostructure information of NPs. The new descriptors were then applied to model MeOx NP cytotoxicity to both Escherichia coli bacteria and HaCaT cells for comparison purposes. The effects of size variation on the cytotoxicity to both types of cells were also investigated. The four resulting QSAR models were then rigorously validated, and extensively compared to other previously published models. The results demonstrated the robustness, validity and predictivity of these models. Predominant nanostructure factors responsible for MeOx NP cytotoxicity were identified through model interpretation. The results verified different mechanisms of nanotoxicity for these two types of cells. The proposed models can be expected to reliably predict the cytotoxicity of novel NPs solely from the newly developed descriptors, and provide guidance for prioritizing the design and manufacture of safer nanomaterials with desired properties.

163 Background: Intermediate and high-risk prostate cancer can be cured with radiation (RT) to the prostate and pelvic lymph nodes with androgen deprivation therapy (ADT), but both acute and late toxicity of the GU and GI systems are common. There are no biomarkers predicting radiation outcomes, limiting the opportunity to best personalize prostate radiation therapy. Methods: A prospectively enrolled single arm trial for locally advanced prostate cancer patients (T1-T4N0-N1M0) treated with definitive RT (74Gy IMRT) plus ADT was studied. Biologic samples were available in 108 of 138 patients. Toxicity was recorded using the RTOG morbidity grading system. We applied a panel of microRNA-based germline mutations shown to predict cancer therapy endpoints. Machine learning techniques were used to simultaneously identify prognostic features and perform classification of the biomarkers. Upsampling nested LOO-CV was used to assess performance and generality. Independent Fisher’s exact tests were performed to identify statistically significant marginal associations. Three classifiers were studied: logistic regression with elastic net regularization (EN-LR), classification trees (CT), and random forests (RF), with corresponding hyper-parameters of regularization weights (EN-LR), minimum split and bucket level sample size (CT), number of trees and mtry (RF). Normalized on the simplex, feature importance was defined as absolute value of regression weights for EN-LR, and cumulative decrease in Gini impurity for primary and surrogate splits at each node/splits for CT and RF. Results: Grade 2 or higher toxicity included acute GI (11%), acute GU (34%), late GI (3%) and late GU (16%). GI and GU toxicity and acute and late toxicity had unique predictive biomarkers. The top three marginal genetic associations for late GU toxicity were microRNA site variants in CD6 and CD274 (PDL1)(p.val < 0.01) and BRCA2 (p.val = 0.014). Using RF, CT and EN_LR we could predict late GU toxicity with up to 70% sensitivity, 96% specificity, and 90% accuracy. Conclusions: We have identified microRNA-based biomarkers that can predict late GU toxicity. Work incorporating patient reported outcomes and to identify biomarkers for additional endpoints is ongoing.

BACKGROUND: High-cost patients are a frequent focus of improvement projects based on primary care and other settings. Efforts to characterize high-cost, high-need patients are needed to inform care planning, but such efforts often rely on a priori assumptions, masking underlying complexities of a heterogenous population. OBJECTIVE: To define recognizable subgroups of patients among high-cost adults based on clinical conditions, and describe their survival and future spending. DESIGN: Retrospective observational cohort study. PARTICIPANTS: Within a large integrated delivery system with 2.7 million adult members, we selected the top 1% of continuously enrolled adults with respect to total healthcare expenditures during 2010. MAIN MEASURES: We used latent class analysis to identify clusters of alike patients based on 53 hierarchical condition categories. Prognosis as measured by healthcare spending and survival was assessed through 2014 for the resulting classes of patients. RESULTS: Among 21,183 high-cost adults, seven clinically distinctive subgroups of patients emerged. Classes included end-stage renal disease (12% of high-cost population), cardiopulmonary conditions (17%), diabetes with multiple comorbidities (8%), acute illness superimposed on chronic conditions (11%), conditions requiring highly specialized care (14%), neurologic and catastrophic conditions (5%), and patients with few comorbidities (the largest class, 33%). Over 4 years of follow-up, 6566 (31%) patients died, and survival in the classes ranged from 43 to 88%. Spending regressed to the mean in all classes except the ESRD and diabetes with multiple comorbidities groups. CONCLUSIONS: Data-driven characterization of high-cost adults yielded clinically intuitive classes that were associated with survival and reflected markedly different healthcare needs. Relatively few high-cost patients remain persistently high cost over 4 years. Our results suggest that high-cost patients, while not a monolithic group, can be segmented into few subgroups. These subgroups may be the focus of future work to understand appropriateness of care and design interventions accordingly. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s11606-018-4626-0) contains supplementary material, which is available to authorized users.

Differential brain response to sensory stimuli is very small (a few microvolts) compared to the overall magnitude of spontaneous electroencephalogram (EEG), yielding a low signal-to-noise ratio (SNR) in studies of event-related potentials (ERP). To cope with this phenomenon, stimuli are applied repeatedly and the ERP signals arising from the individual trials are averaged at the subject level. This results in loss of information about potentially important changes in the magnitude and form of ERP signals over the course of the experiment. In this article, we develop a meta-preprocessing step utilizing a moving average of ERP across sliding trial windows, to capture such longitudinal trends. We embed this procedure in a weighted linear mixed effects model to describe longitudinal trends in features such as ERP peak amplitude and latency across trials while adjusting for the inherent heteroskedasticity created at the meta-preprocessing step. The proposed unified framework, including the meta-processing and the weighted linear mixed effects modeling steps, is referred to as MAP-ERP (moving-averaged-processed ERP). We perform simulation studies to assess the performance of MAP-ERP in reconstructing existing longitudinal trends and apply MAP-ERP to data from young children with autism spectrum disorder (ASD) and their typically developing counterparts to examine differences in patterns of implicit learning, providing novel insights about the mechanisms underlying social and/or cognitive deficits in this disorder.

Background: Ten years ago, leaders in the field of toxicology called for a transformation of the discipline and a shift from primarily relying on traditional animal testing to incorporating advances in biotechnology and predictive methodologies into alternative testing strategies (ATS). Governmental agencies and academic and industry partners initiated programs to support such a transformation, but a decade later, the outcomes of these efforts are not well understood. Objectives: We aimed to assess the use of ATS and the perceived barriers and drivers to their adoption by toxicologists and by others working in, or closely linked with, the field of toxicology. Methods: We surveyed 1,381 toxicologists and experts in associated fields regarding the viability and use of ATS and the perceived barriers and drivers of ATS for a range of applications. We performed ranking, hierarchical clustering, and correlation analyses of the survey data. Results: Many respondents indicated that they were already using ATS, or believed that ATS were already viable approaches, for toxicological assessment of one or more end points in their primary area of interest or concern (26–86%, depending on the specific ATS/application pair). However, the proportions of respondents reporting use of ATS in the previous 12 mo were smaller (4.5–41%). Concern about regulatory acceptance was the most commonly cited factor inhibiting the adoption of ATS, and a variety of technical concerns were also cited as significant barriers to ATS viability. The factors most often cited as playing a significant role (currently or in the future) in driving the adoption of ATS were the need for expedited toxicology information, the need for reduced toxicity testing costs, demand by regulatory agencies, and ethical or moral concerns. Conclusions: Our findings indicate that the transformation of the field of toxicology is partly implemented, but significant barriers to acceptance and adoption remain. https://doi.org/10.1289/EHP1435

Jointly achieving parsimony and good predictive power in high dimensions is a main challenge in statistics. Non-local priors (NLPs) possess appealing properties for model choice, but their use for estimation has not been studied in detail. We show that for regular models NLP-based Bayesian model averaging (BMA) shrink spurious parameters either at fast polynomial or quasi-exponential rates as the sample size n increases, while non-spurious parameter estimates are not shrunk. We extend some results to linear models with dimension p growing with n. Coupled with our theoretical investigations, we outline the constructive representation of NLPs as mixtures of truncated distributions that enables simple posterior sampling and extending NLPs beyond previous proposals. Our results show notable high-dimensional estimation for linear models with p > >n at low computational cost. NLPs provided lower estimation error than benchmark and hyper-g priors, SCAD and LASSO in simulations, and in gene expression data achieved higher cross-validated R2 with less predictors. Remarkably, these results were obtained without pre-screening variables. Our findings contribute to the debate of whether different priors should be used for estimation and model selection, showing that selection priors may actually be desirable for high-dimensional estimation.\ud

Studies of human development require longitudinal data analysis methods that describe within- and between-individual variation in developmental and behavioral trajectories. This article reviews life-course data analysis methods for modeling these trajectories, as well as their application in studies of antisocial behavior and of crime in childhood, in adolescence, and throughout life. We set the stage by introducing growth curve (hierarchical linear) models. We focus our review on finite mixture models for life-course data, known as group-based trajectory and growth mixture models. We then discuss how these models are applied within criminology and developmental psychology, recent controversies over their substantive use and interpretation, and important issues of statistical practice and the challenges they raise. Building on the critical literature, we offer several recommendations for the applied users of the models. Finally, we present the most recent method of examining behavioral trajectories in criminology, the unimodal curve registration (UCR) approach. We briefly contrast the UCR model with growth curve and finite mixture models for life-course data analysis.

96 Background: Treatment with anti-PD1/anti-PDL1 agents is associated with toxicity termed immune related adverse events (iRAEs). While the prevalence of Grade 2 and higher iRAEs is approximately 25-30%, biomarkers have not been previously identified. We tested the hypothesis that functional, germ-line mutations would predict iRAEs. Methods: Four classifiers were trained on a set of 61 melanoma patients evaluated for toxicity and response. Subjects were classified as experiencing high toxicity (≥ Grade 2) vs low toxicity (< Grade 2). Performance of the classifiers was tested on a validation set of 89 cancer patients with a variety of cancer types, with the most common being GU and NSCLC. Classifiers were built for each treatment of marker data including classification trees, LASSO-regularized logistic regression, boosted trees (BT), and random forests. The final performance measures, accuracy, specificity, sensitivity, negative predictive value, positive predictive value, area under the curve (AUC), and F1 score, were reported on the categorical treatment of the training data using leave-one-out cross validation on the validation data. We also evaluated the association between our most significant toxicity biomarker and response to anti-PD1/PDL1 therapy. Results: Within the melanoma training sample, we found a biomarker signature where toxicity is predicted with 79.0% accuracy (F1 = .714, AUC = .827) using BT. The same biomarker panel also accurately predicted toxicity in the validation cohort with 85.6% accuracy (F1 = .760, AUC = .883). Of the most predictive biomarkers, three were in microRNA binding sites in RAC1, CD274, and KRAS, two in immune related genes IL2RA and FCGR2A, and one in the DNA repair gene ATM. Our most significant biomarker in RAC1 did not predict response to anti-PD1/PDL1 treatment (p=0.91). Conclusions: We have identified a germ-line biomarker signature which predicts Grade 2 or higher iRAEs for patients treated with anti-PD1/anti-PDL1 therapy, regardless of cancer type, and does not predict an increased likelihood of response to these therapies. These findings are an important step in defining how to better safely personalize immune therapy, whose use is growing rapidly.

A fundamental goal in nano-toxicology is that of identifying particle physical and chemical properties, which are likely to explain biological hazard. The first line of screening for potentially adverse outcomes often consists of exposure escalation experiments, involving the exposure of micro-organisms or cell lines to a library of nanomaterials. We discuss a modeling strategy, that relates the outcome of an exposure escalation experiment to nanoparticle properties. Our approach makes use of a hierarchical decision process, where we jointly identify particles that initiate adverse biological outcomes and explain the probability of this event in terms of the particle physicochemical descriptors. The proposed inferential framework results in summaries that are easily interpretable as simple probability statements. We present the application of the proposed method to a data set on 24 metal oxides nanoparticles, characterized in relation to their electrical, crystal and dissolution properties.

Background: Differences in interlaboratory research protocols contribute to the conflicting data in the literature regarding engineered nanomaterial (ENM) bioactivity. Objectives: Grantees of a National Institute of Health Sciences (NIEHS)-funded consortium program performed two phases of in vitro testing with selected ENMs in an effort to identify and minimize sources of variability. Methods: Consortium program participants (CPPs) conducted ENM bioactivity evaluations on zinc oxide (ZnO), three forms of titanium dioxide (TiO2), and three forms of multiwalled carbon nanotubes (MWCNTs). In addition, CPPs performed bioassays using three mammalian cell lines (BEAS-2B, RLE-6TN, and THP-1) selected in order to cover two different species (rat and human), two different lung epithelial cells (alveolar type II and bronchial epithelial cells), and two different cell types (epithelial cells and macrophages). CPPs also measured cytotoxicity in all cell types while measuring inflammasome activation [interleukin-1β (IL-1β) release] using only THP-1 cells. Results: The overall in vitro toxicity profiles of ENM were as follows: ZnO was cytotoxic to all cell types at ≥ 50 μg/mL, but did not induce IL-1β. TiO2 was not cytotoxic except for the nanobelt form, which was cytotoxic and induced significant IL-1β production in THP-1 cells. MWCNTs did not produce cytotoxicity, but stimulated lower levels of IL-1β production in THP-1 cells, with the original MWCNT producing the most IL-1β. Conclusions: The results provide justification for the inclusion of mechanism-linked bioactivity assays along with traditional cytotoxicity assays for in vitro screening. In addition, the results suggest that conducting studies with multiple relevant cell types to avoid false-negative outcomes is critical for accurate evaluation of ENM bioactivity.

Frozen tissue, a gold standard biospecimen, can yield well preserved nucleic acids and proteins after over a decade but is vulnerable to thawing and has substantial fiscal, spatial, and environmental costs. A long-term room temperature biospecimen storage alternative that preserves broad analytical utility can potentially empower tissue-based research. As there is scant data on the analytical utility of lyophilized brain tumor biospecimens, we evaluated lyophilized (freeze-dried) samples stored for 1 year at room temperature. Lyophilized tumor tissue processed into paraffin sections produced good histology. Yields of extracted DNA, RNA, and protein approximated those of frozen tissue. After 1 year, lyophilized samples yielded high molecular weight DNA that permitted copy number variation analysis, IDH 1 mutation detection, and MGMT promoter methylation PCR. A 27 % decrease in RIN scores over the 1 year suggests that RNA degradation was inhibited though incompletely. Nevertheless, RT-PCR studies on lyophilized tissue performed similarly to frozen tissue. In contrast to FFPE tissues where protein bands were absent or shifted to a lower molecular weight, lyophilized samples showed similar protein bands as frozen tissue on SDS-PAGE analysis. Lyophilized tissue performed similarly to frozen tissue for Western blots and enzyme activity assays. Immunohistochemistry of lyophilized tissue that were processed into FFPE blocks often required longer incubation times for staining than standard FFPE samples but generally provided robust antigen detection. This preliminary study suggests that lyophilization has promise for long-term room temperature storage while permitting varied tests; however, further work is required to better stabilize nucleic acids particularly RNA.

The development of high throughput screening (HTS) assays in the field of nanotoxicology provide new opportunities for the hazard assessment and ranking of engineered nanomaterials (ENMs). It is often necessary to rank lists of materials based on multiple risk assessment parameters, often aggregated across several measures of toxicity and possibly spanning an array of experimental platforms. Bayesian models coupled with the optimization of loss functions have been shown to provide an effective framework for conducting inference on ranks. In this article we present various loss-function-based ranking approaches for comparing ENM within experiments and toxicity parameters. Additionally, we propose a framework for the aggregation of ranks across different sources of evidence while allowing for differential weighting of this evidence based on its reliability and importance in risk ranking. We apply these methods to high throughput toxicity data on two human cell-lines, exposed to eight different nanomaterials, and measured in relation to four cytotoxicity outcomes. This article has supplementary material online.

High-throughput functional proteomic technologies provide a way to quantify the expression of proteins of interest. Statistical inference centers on identifying the activation state of proteins and their patterns of molecular interaction formalized as dependence structure. Inference on dependence structure is particularly important when proteins are selected because they are part of a common molecular pathway. In that case, inference on dependence structure reveals properties of the underlying pathway. We propose a probability model that represents molecular interactions at the level of hidden binary latent variables that can be interpreted as indicators for active versus inactive states of the proteins. The proposed approach exploits available expert knowledge about the target pathway to define an informative prior on the hidden conditional dependence structure. An important feature of this prior is that it provides an instrument to explicitly anchor the model space to a set of interactions of interest, favoring a local search approach to model determination. We apply our model to reverse-phase protein array data from a study on acute myeloid leukemia. Our inference identifies relevant subpathways in relation to the unfolding of the biological process under study.