A5044163787- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Lung Cancer Treatments and Mutations
- Advanced Causal Inference Techniques
- Hormonal and reproductive studies
- Statistical Methods in Clinical Trials
- Menopause: Health Impacts and Treatments
- Advanced Statistical Methods and Models
- Cancer Mechanisms and Therapy
- Cytokine Signaling Pathways and Interactions
- Ovarian function and disorders
- Estrogen and related hormone effects
- Pharmacology and Obesity Treatment
- Phytoestrogen effects and research
- Colorectal Cancer Treatments and Studies
- Health Systems, Economic Evaluations, Quality of Life
- Cancer Immunotherapy and Biomarkers
- Advanced Database Systems and Queries
- Ferroptosis and cancer prognosis
- Simulation Techniques and Applications
- Spatial and Panel Data Analysis
- PI3K/AKT/mTOR signaling in cancer
- Statistical Distribution Estimation and Applications
- Bayesian Methods and Mixture Models
- Lung Cancer Diagnosis and Treatment
Gilead Sciences (United States)
2016-2021
Purdue University West Lafayette
2014
University of Michigan
2012-2013
IntroductionWe report the final overall survival (OS) analyses of atezolizumab-carboplatin-paclitaxel (ACP [experimental arm]) and OS data with approximately 39.8 months median follow-up atezolizumab-bevacizumab-carboplatin-paclitaxel (ABCP) versus bevacizumab-carboplatin-paclitaxel (BCP) in chemotherapy-naive patients metastatic nonsquamous NSCLC phase 3 IMpower150 study (NCT02366143).MethodsIn this randomized, open-label (N = 1202), coprimary end points included investigator-assessed...
We consider finite sample properties of the regularized high-dimensional Cox regression via lasso. Existing literature focuses on linear models or generalized with Lipschitz loss functions, where empirical risk functions are summations independent and identically distributed (iid) losses. The summands in negative log partial likelihood function for censored survival data, however, neither iid nor Lipschitz.We first approximate by a sum non-Lipschitz terms, then derive non-asymptotic oracle...
Sex hormones may differ by race/ethnicity in postmenopausal women. Whether racial/ethnic differences also exist among those who are overweight and glucose intolerant is not clear. The objective of the study was to compare sex [non-Hispanic white (NHW), Hispanic, African-American (AA)] overweight, glucose-intolerant, This a secondary analysis randomized controlled trial. Participants included glucose-intolerant women participating Diabetes Prevention Program. Interventions intensive lifestyle...
We consider generalized linear regression with a covariate left-censored at lower detection limit. Complete-case analysis, where observations values below the limit are eliminated, yields valid estimates for coefficients but loses efficiency, ad hoc substitution methods biased, and parametric maximum likelihood estimation relies on models unobservable tail probability distribution may suffer from model misspecification. To obtain robust more efficient results, we propose semiparametric...
Objectives To examine the impact of a weight loss intervention upon follicle stimulating hormone (FSH) levels in postmenopause. Methods Participants were postmenopausal, overweight, glucose‐intolerant women not using exogenous estrogen ( n = 382) Diabetes Prevention Program. Women randomized to intensive lifestyle change (ILS) with goals reduction at least 7% initial and 150 min per week moderate‐intensity exercise, metformin 850 mg twice day, or placebo administered day. Results...
We consider a random effects model for longitudinal data with the occurrence of an informative terminal event that is subject to right censoring. Existing methods analyzing such include joint modeling approach using latent frailty and marginal estimating equation inverse probability weighting; in both cases effect on response variable not explicit thus easily interpreted. In contrast, we treat time as covariate conditional data, which provides straight-forward interpretation while keeping...
IntroductionSpecific treatment options are lacking for Kirsten rat sarcoma viral oncogene homolog (KRAS)-mutated non–small-cell lung cancer (NSCLC) despite advances in other mutation-driven subgroups.Patients and MethodsIn this study we evaluated the multitargeted Janus kinase/TANK-binding kinase 1 (TBK1) inhibitor momelotinib combined with mitogen/extracellular signal-related (MEK)1/MEK2 trametinib patients platinum-treated, refractory, metastatic, KRAS-mutated NSCLC. Dose escalations (3 +...
We consider the finite sample properties of regularized high-dimensional Cox regression via lasso. Existing literature focuses on linear models or generalized with Lipschitz loss functions, where empirical risk functions are summations independent and identically distributed (iid) losses. The summands in negative log partial likelihood function for censored survival data, however, neither iid nor Lipschitz. first approximate by a sum non-Lipschitz terms, then derive non-asymptotic oracle...
Preclinical evidence suggests the feedforward cytokine loop of interleukin-6/Janus kinases (JAK)/STAT3 plays a role in epidermal growth factor receptor tyrosine kinase inhibitor (EGFR TKI) resistance EGFR-mutated non-small cell lung cancer (NSCLC).In this phase 1b study, JAK1/2 and TANK-binding 1 (TBK1) momelotinib was evaluated combination with erlotinib patients EGFR TKI-naive, NSCLC. After lead-in (50, 75, 100, or 150 mg oral daily [QD]), combined dose escalated 3 + study design. The...
This research was motivated by studying anti-drug antibody (ADA) formation and its potential impact on long-term benefit of a biologic treatment in randomized controlled trial, which ADA status not only unobserved the control arm but also subset patients from experimental arm. Recent literature considers principal stratum estimand strategy to estimate effect groups defined an intercurrent status, that is, post-randomization variable observed one potentially associated with outcome. However,...
This research is motivated by studying the progression of age‐related macular degeneration where both a covariate and response variable are subject to censoring. We develop general framework handle regression with censored can be different types censoring random or (constant) detection limits. Multiple imputation popular technique missing data that requires compatibility between model substantive obtain valid estimates. With covariate, we propose novel multiple imputation‐based approach,...
Abstract We consider high‐dimensional inference for potentially misspecified Cox proportional hazard models based on low‐dimensional results by Lin and Wei (1989). A desparsified Lasso estimator is proposed the log partial likelihood function shown to converge a pseudo‐true parameter vector. Interestingly, sparsity of true can be inferred from that above limiting parameter. Moreover, each component (nonsparse) asymptotically normal with variance consistently estimated even under model...
This study aims to examine whether blood pressure reductions differ by estrogen use among overweight glucose-intolerant women.We conducted a secondary analysis of Diabetes Prevention Program postmenopausal participants who used oral with or without progestogen at baseline and 1-year follow-up (n = 324) versus those did not either time point 382). Changes in systolic (SBP) diastolic (DBP) were examined randomization arm (intensive lifestyle change [ILS], metformin 850 mg twice daily,...
This research was motivated by studying anti-drug antibody (ADA) formation and its potential impact on long-term benefit of a biologic treatment in randomized controlled trial, which ADA status not only unobserved the control arm but also subset patients from experimental arm. Recent literature considers principal stratum estimand strategy to estimate effect groups defined an intercurrent status, i.e. post-randomization variable observed one potentially associated with outcome. However,...
We consider high-dimensional inference for potentially misspecified Cox proportional hazard models based on low dimensional results by Lin and Wei [1989]. A de-sparsified Lasso estimator is proposed the log partial likelihood function shown to converge a pseudo-true parameter vector. Interestingly, sparsity of true can be inferred from that above limiting parameter. Moreover, each component (non-sparse) asymptotically normal with variance consistently estimated even under model...
To examine whether estrogen use potentiates weight loss interventions via sex steroid levels and endogenous predict response to among women not using estrogen.The Diabetes Prevention Program randomized overweight or obese dysglycemic participants lifestyle change with the goals of reduction >7% initial 150 minutes per week exercise, metformin, placebo. In this secondary analysis, we examined reductions in waist circumference (WC) postmenopausal (n = 324) 382) oral estrogen.Estrogen users...
We consider generalized linear regression analysis with left-censored covariate due to the lower limit of detection. Complete case by eliminating observations values below detection yields valid estimates for coefficients, but loses efficiency; substitution methods are biased; maximum likelihood method relies on parametric models unobservable tail probability distribution such covariate, thus may suffer from model misspecification. To obtain robust and more efficient results, we propose a...
e13564 Background: Traditional rPh2 trials have limitations that may yield suboptimal Ph3-GO. Compared to a of equivalent sample size, SAT+rwEC allows more patients receive experimental therapies while preserving the ability compare and control groups. Bias arising from measurement error confounding in rwEC, however, poses challenges statistical inference. Preliminary studies suggest higher response rates are observed rwEC than randomized controls. We compared Ph3-GO decisions between rPh2....