Alleviating spatial confounding in frailty models.

The confounding between fixed effects and (spatial) random effects in a regression setup is termed spatial confounding. This topic continues to gain attention and has been studied extensively in recent years, given that failure to account for this may lead to a suboptimal inference. To mitigate this, a variety of projection-based approaches under the class of restricted spatial models are available in the context of generalized linear mixed models. However, these projection approaches cannot be directly extended to the spatial survival context via frailty models due to dimension incompatibility between the fixed and spatial random effects. In this work, we introduce a two-step approach to handle this, which involves (i) projecting the design matrix to the dimension of the spatial effect (via dimension reduction) and (ii) assuring that the random effect is orthogonal to this new design matrix (confounding alleviation). Under a fully Bayesian paradigm, we conduct fast estimation and inference using integrated nested Laplace approximation. Both simulation studies and application to a motivating data evaluating respiratory cancer survival in the US state of California reveal the advantages of our proposal in terms of model performance and confounding alleviation, compared to alternatives.

Luminal A Versus B After Choosing Wisely: Does Lymph Node Surgery Affect Oncologic Outcomes?

BACKGROUND: In 2016, the Choosing Wisely campaign recommended against routine sentinel lymph node biopsy (SLNB) in women ≥ 70 years old diagnosed with early-stage hormone receptor positive (HR+), HER2 negative (HER2-) breast cancer. No distinction is made between luminal A and luminal B phenotypes, despite luminal B being considered more aggressive. This study evaluates the effect of SLNB on oncologic outcomes in HER2- luminal B versus luminal A breast cancer. PATIENTS AND METHODS: We performed an IRB-approved, single institution, retrospective cohort study from 2010 to 2020 of women aged ≥ 70 years with clinically node negative, HR+ breast cancer undergoing definitive surgical treatment. Luminal status was defined by gene expression panel testing, Ki67%, and/or pathologic grading. Primary endpoints included locoregional recurrence (LRR), disease free survival (DFS), and overall survival (OS). RESULTS: SLNB did not correlate with significant differences in LRR in luminal A (p = 0.92) or luminal B (p = 0.96) disease. SLNB correlated with improved DFS (p < 0.01) and OS (p < 0.001) in luminal A disease, but not in luminal B disease (DFS p = 0.73; OS p = 0.36). On multivariate analysis, age (HR = 1.17; p < 0.01) and tumor size (HR = 1.03; p < 0.05) were associated with DFS, while SLNB was not (p = 0.71). Luminal status (HR = 0.52, p < 0.05), age (HR = 1.15, p < 0.01), and comorbidities (HR = 1.35, p < 0.05) were associated with OS, but not SLNB (p = 0.71). CONCLUSIONS: Our results suggest that SLNB may be safely omitted in patients aged ≥ 70 years with luminal B disease given similar LRR in luminal A disease. Our findings suggest that DFS and OS are driven by tumor biology, patient age, and comorbidities rather than receipt of SLNB.

A phase 1 study of regorafenib and sildenafil in adults with advanced solid tumors.

The purpose of this study is to establish the recommended phase 2 dose for regorafenib in combination with sildenafil for patients with advanced solid tumors. Secondary outcomes included identification of antitumor effects of regorafenib and sildenafil, toxicity of the combination, determination of PDE5 expression in tumor samples, and the impact of sildenafil on the pharmacokinetics of regorafenib. This study was a phase 1, open-label single-arm dose-escalation trial using a 3 + 3 design. Additional patients were enrolled at the maximum tolerated dose (MTD) until a total of 12 patients were treated at the MTD. A total of 29 patients were treated in this study. The median duration of treatment was 8 weeks. The recommended phase 2 doses determined in this study are regorafenib 160 mg daily with sildenafil 100 mg daily. The most common toxicities included palmar-plantar erythrodysesthesia syndrome (20 patients, 69%) and hypophosphatemia (18 patients, 62%). Two patients (7%) experienced grade 4 lipase increase. Objective responses were not observed; however, 14 patients (48%) had a period of stable disease during the study. Stable disease for up to 12 months was observed in patients with ovarian cancer as well as up to 20 months for a patient with cervical cancer. The combination of regorafenib and sildenafil at the recommended phase 2 dose is safe and generally well tolerated. Disease control in patients with gynecologic malignancies was especially encouraging. Further evaluation of the combination of regorafenib and sildenafil in gynecologic malignancies is warranted. Clinical Trial Registration Number: NCT02466802.

Using Oncotype DX breast recurrence score® assay to define the role of neoadjuvant endocrine therapy in early-stage hormone receptor-positive breast cancer.

PURPOSE: The role of neoadjuvant endocrine therapy in the treatment of patients with early-stage, hormone receptor-positive (HR +) breast cancer is not well defined. Tools to better determine which patients may benefit from neoadjuvant endocrine therapy versus chemotherapy or upfront surgery remain an unmet need. METHODS: We assessed the rate of clinical and pathologic complete response (cCR, pCR) among a pooled cohort of patients with early-stage HR + breast cancer who had been randomized to neoadjuvant endocrine therapy or neoadjuvant chemotherapy in two earlier studies to understand better how outcomes varied by Oncotype DX Breast Recurrence Score® assay. RESULTS: We observed that patients with intermediate RS results had no statistically significant differences in pathologic outcomes at the time of surgery based on whether they received neoadjuvant endocrine therapy or neoadjuvant chemotherapy, suggesting that a subgroup of women with a RS 0-25 may omit chemotherapy without compromising outcomes. CONCLUSION: These data suggest that Recurrence Score® (RS) results may serve as a useful tool in treatment decision-making in the neoadjuvant setting.

Racial/Ethnic Disparities in the Era of Minimally Invasive Surgery for Treatment of Colorectal Cancer.

BACKGROUND: Minimally invasive (laparoscopic and robotic) surgery (MIS) for colorectal cancer is associated with improved outcomes. We sought to characterize possible disparities in surgical approach and outcomes. PATIENTS AND METHODS: In this cross-sectional study, colorectal adenocarcinoma cases among non-Hispanic white (NHW), non-Hispanic Black (NHB), and Hispanic patients were identified using the National Cancer Database (2010-2017). Logistic and Poisson regressions, generalized logit models, and Cox proportional hazards were used to assess outcomes, with reclassification of surgery type if converted to open. RESULTS: NHB patients were less likely to undergo robotic surgery. After multivariable analysis, NHB patients were 6% less likely, while Hispanic patients were 12% more likely to undergo a MIS approach. Lymph node retrieval was higher (> 1.3% more, p < 0.0001) and length of stay was shorter (> 17% shorter, p < 0.0001) for MIS approaches. Unplanned readmission was lower for MIS colon cancer operations compared with open operations, but not for rectal cancer. Race/ethnicity-adjusted risk of death was lower with MIS approaches for colon as well as rectal cancer. After adjusting for surgery type, risk of death was 12% lower for NHB and 35% lower for Hispanic patients compared with NHW patients. Hispanic patients had 21% lower risk of death, while NHB patients had 12% higher risk of death than NHW patients with rectal cancer, after adjusting for surgery type. CONCLUSIONS: Racial/ethnic disparities exist in utilization of MIS for colorectal cancer treatment, disproportionately affecting NHB patients. Since MIS has the potential to improve outcomes, suboptimal access may contribute to harmful and thus unacceptable disparities in survivorship.

Outcomes of Cancer Patients with COVID-19: The Virginia Commonwealth University Experience.

PURPOSE: The cancer population is significantly impacted by coronavirus disease 2019 (COVID-19) due to inherent risks of infection imposed by malignancy and therapeutic agents. Evaluating risk factors in this group will lead to improved guidelines for the treatment of malignancy in the setting of a COVID-19 pandemic. PATIENTS AND METHODS: This retrospective study reviewed 295 inpatient cancer patients positive for COVID-19 between February 2020 and December 2021 to determine specific risk factors of mortality and associated complications. Various patient characteristics were collected to evaluate outcomes in patient death, oxygen requirement, ventilatory support, and increased length of stay. RESULTS: 31 (10.5%) of 295 patients died due to COVID-19. Of those that died, the majority had hematologic cancer (48.4%). There was no difference in the odds of death among the cancer groups. Those vaccinated had a reduced risk of death (OR 0.04, CI 0-0.23). Patients with lung cancer (OR 3.69, CI 1.13-12.31), obesity (OR 3.27, CI 1.18-9.27), CHF (OR 2.68, CI 1.07-6.89) were more likely to require ventilation. Those treated with hormonal therapy had higher odds of having a prolonged admission (OR 5.04, CI 1.17-2.53). Otherwise, cancer therapy had no significant difference in any outcome. CONCLUSION: The mortality rate of cancer patients was 10.5%, lower than in other studies. Vaccinations had mortality benefits, but no effect on hypoxia, ventilator use, or LOS. Delaying cancer therapy during peak infection is likely not necessary based on the results of this study. With improved knowledge in the risks of infection and the utility of personalized precautions, both providers and patients can better prepare for another potential wave of COVID-19.

Kelvin-Helmholtz Instability Augmented by von Kármán Vortex Shedding during an Oil Droplet Impact on a Water Pool.

The impact of an oil droplet on a water surface has been explored with the aid of computational fluid dynamics simulations. The study reveals the details of the spatiotemporal evolution of such a ternary system with a triplet of interfaces, e.g., air-water, oil-water, and oil-air, when the impact velocity of the oil droplet with the water surface is high. The oil droplet is found to flatten, spread, stretch, and eventually dewet on the water surface of the deep crater to show a host of interesting post-impact flow morphologies. Furthermore, at higher impact velocities, the formation of a biphasic oil-water crown is observed followed by the ejection of secondary water droplets from the crown tip due to capillary instability. The rapidly spreading oil film on the "crater" of the water surface is found to undergo Kelvin-Helmholtz instability before dewetting the same due to cohesion failure. Subsequently, the formation of an array of secondary oil droplets is observed during the process of dewetting. The dominant wavelength evaluated from the linear stability analysis of a representative flow system could faithfully predict the simulated spacing of dewetted oil droplets floating on the crater. Importantly, the variations in Laplace pressure around the curvatures of the undulatory interfaces along with sharp viscosity gradients across the three-phase contact line is found to engender interesting recirculation patterns, which eventually shed to form a coherent wake region in air near the crater. We also uncover the conditions under which the counter-rotating vortices shed along the oil-water interface resembling a von Kármán vortex street.

A semiparametric joint model for cluster size and subunit-specific interval-censored outcomes.

Clustered data frequently arise in biomedical studies, where observations, or subunits, measured within a cluster are associated. The cluster size is said to be informative, if the outcome variable is associated with the number of subunits in a cluster. In most existing work, the informative cluster size issue is handled by marginal approaches based on within-cluster resampling, or cluster-weighted generalized estimating equations. Although these approaches yield consistent estimation of the marginal models, they do not allow estimation of within-cluster associations and are generally inefficient. In this paper, we propose a semiparametric joint model for clustered interval-censored event time data with informative cluster size. We use a random effect to account for the association among event times of the same cluster as well as the association between event times and the cluster size. For estimation, we propose a sieve maximum likelihood approach and devise a computationally-efficient expectation-maximization algorithm for implementation. The estimators are shown to be strongly consistent, with the Euclidean components being asymptotically normal and achieving semiparametric efficiency. Extensive simulation studies are conducted to evaluate the finite-sample performance, efficiency and robustness of the proposed method. We also illustrate our method via application to a motivating periodontal disease dataset.

Bayesian regression analysis of skewed tensor responses.

Tensor regression analysis is finding vast emerging applications in a variety of clinical settings, including neuroimaging, genomics, and dental medicine. The motivation for this paper is a study of periodontal disease (PD) with an order-3 tensor response: multiple biomarkers measured at prespecified tooth-sites within each tooth, for each participant. A careful investigation would reveal considerable skewness in the responses, in addition to response missingness. To mitigate the shortcomings of existing analysis tools, we propose a new Bayesian tensor response regression method that facilitates interpretation of covariate effects on both marginal and joint distributions of highly skewed tensor responses, and accommodates missing-at-random responses under a closure property of our tensor model. Furthermore, we present a prudent evaluation of the overall covariate effects while identifying their possible variations on only a sparse subset of the tensor components. Our method promises Markov chain Monte Carlo (MCMC) tools that are readily implementable. We illustrate substantial advantages of our proposal over existing methods via simulation studies and application to a real data set derived from a clinical study of PD. The R package BSTN available in GitHub implements our model.

Bayesian additive regression trees for multivariate skewed responses.

This paper introduces a nonparametric regression approach for univariate and multivariate skewed responses using Bayesian additive regression trees (BART). Existing BART methods use ensembles of decision trees to model a mean function, and have become popular recently due to their high prediction accuracy and ease of use. The usual assumption of a univariate Gaussian error distribution, however, is restrictive in many biomedical applications. Motivated by an oral health study, we provide a useful extension of BART, the skewBART model, to address this problem. We then extend skewBART to allow for multivariate responses, with information shared across the decision trees associated with different responses within the same subject. The methodology accommodates within-subject association, and allows varying skewness parameters for the varying multivariate responses. We illustrate the benefits of our multivariate skewBART proposal over existing alternatives via simulation studies and application to the oral health dataset with bivariate highly skewed responses. Our methodology is implementable via the R package skewBART, available on GitHub.

Sieve estimation of a class of partially linear transformation models with interval-censored competing risks data.

In this paper, we consider a class of partially linear transformation models with interval-censored competing risks data. Under a semiparametric generalized odds rate specification for the cause-specific cumulative incidence function, we obtain optimal estimators of the large number of parametric and nonparametric model components via maximizing the likelihood function over a joint B-spline and Bernstein polynomial spanned sieve space. Our specification considers a relatively simpler finite-dimensional parameter space, approximating the infinite-dimensional parameter space as n → ∞, thereby allowing us to study the almost sure consistency, and rate of convergence for all parameters, and the asymptotic distributions and efficiency of the finite-dimensional components. We study the finite sample performance of our method through simulation studies under a variety of scenarios. Furthermore, we illustrate our methodology via application to a dataset on HIV-infected individuals from sub-Saharan Africa.

Electric Field-Induced "Tentillar" Bridging of a Droplet Twin.

Experimental investigations with high-speed imaging reveal that two unequal sized oppositely charged droplets suspended in an insulating oil can come in contact in an asymmetric manner under an electric field. The approaching poles of the droplets undergo asymmetric "cone-cone" to "cone-groove" deformations during noncoalescence under an electric field. Nonlinear three-dimensional simulations confirm the occurrence of a third "groove-groove" configuration at close proximity. A general linear stability analysis confirms the cone-cone to cone-groove transitions of the oil-water interfaces with decreasing thickness of the oil film. Experiments together with simulations confirm the bifurcation of the Taylor cone on the smaller droplet into a number of liquid "tentilla" bridges prior to contact with the bigger droplet. Simulations also predict that the length scales of the formation of such tentillar bridges from the initial Taylor cone match well with the predictions from the nonlinear bifurcation theory.

Geostatistical modeling of positive-definite matrices: An application to diffusion tensor imaging.

Geostatistical modeling for continuous point-referenced data has extensively been applied to neuroimaging because it produces efficient and valid statistical inference. However, diffusion tensor imaging (DTI), a neuroimaging technique characterizing the brain's anatomical structure, produces a positive-definite (p.d.) matrix for each voxel. Currently, only a few geostatistical models for p.d. matrices have been proposed because introducing spatial dependence among p.d. matrices properly is challenging. In this paper, we use the spatial Wishart process, a spatial stochastic process (random field), where each p.d. matrix-variate random variable marginally follows a Wishart distribution, and spatial dependence between random matrices is induced by latent Gaussian processes. This process is valid on an uncountable collection of spatial locations and is almost-surely continuous, leading to a reasonable way of modeling spatial dependence. Motivated by a DTI data set of cocaine users, we propose a spatial matrix-variate regression model based on the spatial Wishart process. A problematic issue is that the spatial Wishart process has no closed-form density function. Hence, we propose an approximation method to obtain a feasible Cholesky decomposition model, which we show to be asymptotically equivalent to the spatial Wishart process model. A local likelihood approximation method is also applied to achieve fast computation. The simulation studies and real data application demonstrate that the Cholesky decomposition process model produces reliable inference and improved performance, compared to other methods.

Weighted least-squares regression with competing risks data.

The semiparametric accelerated failure time (AFT) model linearly relates the logarithm of the failure time to a set of covariates, while leaving the error distribution unspecified. This model has been widely investigated in survival literature due to its simple interpretation and relationship with linear models. However, there has been much less focus on developing AFT-type linear regression methods for analyzing competing risks data, in which patients can potentially experience one of multiple failure causes. In this article, we propose a simple least-squares (LS) linear regression model for a cause-specific subdistribution function, where the conventional LS equation is modified to account for data incompleteness under competing risks. The proposed estimators are shown to be consistent and asymptotically normal with consistent estimation of the variance-covariance matrix. We further extend the proposed methodology to risk prediction and analysis under clustered competing risks scenario. Simulation studies suggest that the proposed method provides rapid and valid statistical inferences and predictions. Application of our method to two oncology datasets demonstrate its utility in routine clinical data analysis.

Influence of the pre-impact shape of an oil droplet on the post-impact flow dynamics at air-water interface.

We computationally explore the effects of pre-impact shape of an oil droplet on the spatiotemporal dynamics after the droplet impacts an air-water interface. Simulations reveal that the initial shape of the impacting oil-droplet alters the post-impact transient flow structures during the evolution. The spherical and oblate drop spreads over the crater to manifest interesting flow morphologies including the formation of oil-toroids and compound oil-droplets. However, the prolate drop impinges much deeper into the water pool after impact to create a few more exclusive flow features, such as, interface overturning, vortex shedding and formation of secondary droplets. The temporal variation of the crater depth shows distinct three stage dynamics, which can be explained by the generic energy analysis of the entire system. The combined theoretical and numerical energy analyses reveal the influences of the pre-impact drop shape and their effects on the subsequent energy conversion after the impact takes place. The analysis also reveals that the initial surface and kinetic energies are different for non-spherical droplets than for the spherical ones. The conversion of such excess surface energy due to the non-spherical curvature into kinetic energy dictates the impact and subsequently the crater dynamics of such systems. Such influences largely lead to the exclusive flow patterns demonstrated here. Concisely, this study presents a tri-phasic computational model, which is capable of analyzing the salient features of the impact and splash dynamics of the non-spherical droplets into a water continuum.

Doubly-robust methods for differences in restricted mean lifetimes using pseudo-observations.

In clinical studies or trials comparing survival times between two treatment groups, the restricted mean lifetime (RML), defined as the expectation of the survival from time 0 to a prespecified time-point, is often the quantity of interest that is readily interpretable to clinicians without any modeling restrictions. It is well known that if the treatments are not randomized (as in observational studies), covariate adjustment is necessary to account for treatment imbalances due to confounding factors. In this article, we propose a simple doubly-robust pseudo-value approach to effectively estimate the difference in the RML between two groups (akin to a metric for estimating average causal effects), while accounting for confounders. The proposed method combines two general approaches: (a) group-specific regression models for the time-to-event and covariate information, and (b) inverse probability of treatment assignment weights, where the RMLs are replaced by the corresponding pseudo-observations for survival outcomes, thereby mitigating the estimation complexities in presence of censoring. The proposed estimator is double-robust, in the sense that it is consistent if at least one of the two working models remains correct. In addition, we explore the potential of available machine learning algorithms in causal inference to reduce possible bias of the causal estimates in presence of a complex association between the survival outcome and covariates. We conduct extensive simulation studies to assess the finite-sample performance of the pseudo-value causal effect estimators. Furthermore, we illustrate our methodology via application to a dataset from a breast cancer cohort study. The proposed method is implementable using the R package drRML, available in GitHub.

Nonparametric tests for multistate processes with clustered data.

In this work, we propose nonparametric two-sample tests for population-averaged transition and state occupation probabilities for continuous-time and finite state space processes with clustered, right-censored, and/or left-truncated data. We consider settings where the two groups under comparison are independent or dependent, with or without complete cluster structure. The proposed tests do not impose assumptions regarding the structure of the within-cluster dependence and are applicable to settings with informative cluster size and/or non-Markov processes. The asymptotic properties of the tests are rigorously established using empirical process theory. Simulation studies show that the proposed tests work well even with a small number of clusters, and that they can be substantially more powerful compared to the only, to the best of our knowledge, previously proposed test for this problem. The tests are illustrated using data from a multicenter randomized controlled trial on metastatic squamous-cell carcinoma of the head and neck.

A graphical model for skewed matrix-variate non-randomly missing data.

Epidemiological studies on periodontal disease (PD) collect relevant bio-markers, such as the clinical attachment level (CAL) and the probed pocket depth (PPD), at pre-specified tooth sites clustered within a subject's mouth, along with various other demographic and biological risk factors. Routine cross-sectional evaluation are conducted under a linear mixed model (LMM) framework with underlying normality assumptions on the random terms. However, a careful investigation reveals considerable non-normality manifested in those random terms, in the form of skewness and tail behavior. In addition, PD progression is hypothesized to be spatially-referenced, i.e. disease status at proximal tooth-sites may be different from distally located sites, and tooth missingness is non-random (or informative), given that the number and location of missing teeth informs about the periodontal health in that region. To mitigate these complexities, we consider a matrix-variate skew-$t$ formulation of the LMM with a Markov graphical embedding to handle the site-level spatial associations of the bivariate (PPD and CAL) responses. Within the same framework, the non-randomly missing responses are imputed via a latent probit regression of the missingness indicator over the responses. Our hierarchical Bayesian framework powered by relevant Markov chain Monte Carlo steps addresses the aforementioned complexities within an unified paradigm, and estimates model parameters with seamless sharing of information across various stages of the hierarchy. Using both synthetic and real clinical data assessing PD status, we demonstrate a significantly improved fit of our proposition over various other alternative models.

Cardioprotective medication use in Black and white breast cancer survivors.

PURPOSE: Racial disparities in cardiovascular disease and cardiac dysfunction exist amongst breast cancer survivors. This study examined the prevalence of cardioprotective medication use in survivors and identified factors associated with use by race. METHODS: The analysis included women enrolled in the Women's Hormonal Initiation and Persistence study, a longitudinal observational trial of breast cancer survivors. The study outcome, angiotensin converting enzyme inhibitor (ACEi) or ß-Blocker (BB) use, were ascertained from pharmacy records. Demographic, psychosocial, healthcare, and quality of life factors were collected from surveys and clinical data were abstracted from medical records. Bivariate associations by race and ACEi/BB use were tested using chi square and t tests; logistic regression evaluated multivariable-adjusted associations. RESULTS: Of the 246 survivors in the sample, 33.3% were Black and most were < 65 years of age (58.4%). Most survivors were hypertensive (57.6%) and one-third received ACEi/BBs. In unadjusted analysis, White women (vs. Black) (OR 0.33, 95% 0.19-0.58) and women with higher ratings of functional wellbeing (OR 0.94, 95% 0.89-0.99) were less likely to use ACEi/BBs. Satisfaction with provider communication was only significant for White women. In multivariable-adjusted analysis, ACEi/BB use did not differ by race. Correlates of ACEi/BB use included hypertension among all women and older age for Black women only. CONCLUSIONS: After adjusting for age and comorbidities, no differences by race in ACEi/BB use were observed. Hypertension was a major contributor of ACEi/BB use in BC survivors.

A microfluidic viscometer: Translation of oscillatory motion of a water microdroplet in oil under electric field.

The electric field induced motion of a charged water droplet suspended in a low-dielectric oil medium is exploited to evaluate the rheological properties of the suspending medium. The time-periodic electrophoretic motion of the droplet between the electrodes decorated in a polymeric micro-well is translated into a proof-of-concept microfluidic prototype, which can measure viscosities of the unknown fluid samples. The variations in the instantaneous velocities of the migrating droplet have been measured inside silicone oil of known physical properties at different electric field intensities. Subsequently, a balance between the electric field to the viscous force has been employed to evaluate the experimental charge density on the droplet surface. Thereafter, a comprehensive scaling law has been devised to find a correlation between the charge on the droplet to the dielectric permittivity of the surrounding medium, size of the water droplet, and the applied electric field intensity. Following this, the scaling law and force balance have been employed together to evaluate the unknown viscosity of an array of suspending mediums by simply analyzing the electrophoretic motion of water droplet. The model proposed is also found to be consistent when a solid amberlite microparticle has been employed as a probe instead of the water droplet. In such a scenario, minor changes in the exponents of the scaling law are found to be necessary to reproduce the results obtained using the water droplet. The method paves the way for the making of an economical and portable microfluidic rheometer with further finetuning and translational developments.