Browsing by Subject "Clinical Trials"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Applications of ROC Type Curves in Clinical Trials
    (2023-01) Castro-Pearson, Sandra
    Randomized controlled trials are the most reliable form of evidence to drive medical practice and health policy. Yet, challenges remain in terms of understanding possible errors during protocol implementation and interpreting results of complex outcomes such as censored time-to-event outcomes with and without competing risks. In this dissertation, we study receiver operating characteristic (ROC) curves and use ROC concepts to create graphical methods that address these challenges.
  • Thumbnail Image
    Item
    Bayesian hierarchical modeling for adaptive incorporation of historical information In clinical trials.
    (2010-08) Hobbs, Brian Paul
    Bayesian clinical trial designs offer the possibility of a substantially reduced sample size, increased statistical power, and reductions in cost and ethical hazard. However when prior and current information conflict, Bayesian methods can lead to higher than expected Type I error, as well as the possibility of a costlier and lengthier trial. We develop several models that allow for the commensurability of the information in the historical and current data to determine how much historical information is used. First, we propose methods for univariate Gaussian data and provide an example analysis of data from two successive colon cancer trials that illustrates a linear models extension of our adaptive borrowing approach. Next, we extend the general method to linear and linear mixed models as well as generalized linear and generalized linear mixed models. We also provide two more sample analyses using the colon cancer data. Finally, we consider the effective historical sample size of our adaptive method for the case when historical data is available only for the concurrent control arm, and propose "optimal" use of new patients in the current trial using an adaptive randomization scheme that is balanced with respect to the amount of incorporated historical information. The approach is then demonstrated using data from a trial comparing antiretroviral strategies in HIV-1-infected persons. Throughout the thesis we present simulation studies that compare frequentist operating characteristics and highlight the advantages of our adaptive borrowing methods.
  • Thumbnail Image
    Item
    Bayesian Hierarchical Models for Data Extrapolation and Analysis in Rare and Pediatric Disease Clinical Trials
    (2017-08) Basu, Cynthia
    A rare disease is defined by the Rare Diseases Act of 2002 as a disease that currently affects fewer than 200,000 patients in the USA. A pediatric population is one where the subjects are of age 18 or less. These two crucial yet underserved types of populations come with their own limitations in clinical trials. The paucity of potential trial enrollees and sensitivity of these patients, combined with a lack of sufficient natural history and experience, presents several economical, logistical and ethical challenges when designing clinical trials. An increasingly well accepted approach to address these challenges has been data extrapolation; that is, the leveraging of available data from adults or older age groups to draw conclusions for the pediatric population. Bayesian hierarchical modeling facilitates the combining (or ``borrowing") of information across disparate sources, such as adult and pediatric data. In this thesis we begin by developing, illustrating, and providing suggestions for Bayesian statistical methods that could be used to design improved clinical trials for pediatric and rare disease populations that efficiently use all available information. A variety of relevant Bayesian approaches are described, several of which are illustrated through two case studies: extrapolating adult binary efficacy data to expand the labeling for the drug Remicade to include pediatric ulcerative colitis, and a simulated continuous longitudinal dataset patterned after an evaluation of the drug cinacalcet in treating pediatric secondary hyperparathyroidism (HPT). The thesis then turns to methods useful in the study of X-linked adrenoleukodystrophy (X-ALD), a rare neurodegenerative disease for which Lorenzo’s Oil (LO) is one of the few available treatments. We offer a hierarchical Bayesian statistical approach to understanding the pharmacokinetics (PK) and pharmacodynamics (PD) of LO, linking its %We experiment with individual- and observational-level errors and various choices of prior distributions and deal with the limitation of having just one observation per administration of the drug, as opposed to the more usual multiple observations per administration. dose to the plasma erucic acid concentrations by PK modeling, and then linking this concentration to a biomarker (C26, a very long-chain fatty acid) by PD modeling. Next, we design a Bayesian Phase IIa study to estimate precisely what improvements in the biomarker can arise from various LO doses while simultaneously modeling a binary toxicity endpoint. Our Bayesian adaptive algorithm emerges as reasonably robust and efficient while still retaining good classical (frequentist) operating characteristics. Future work in this area looks toward using the results of this trial to design a Phase III study linking LO dose to actual improvements in health status, as measured by the appearance of brain lesions observed via magnetic resonance imaging. Finally, the thesis shows how to utilize the rich PK/PD data to inform the borrowing of information from adults during pediatric drug development. Here we illustrate our approaches in the context of evaluating safety and efficacy of cinacalcet for treating HPT in pediatric and adult patients. We use population PK/PD modeling of the cinacalcet data to quantitatively assess the similarity between adults and children, and in turn use this information in various hierarchical Bayesian rules for borrowing from adults, statistical properties of which can then be evaluated. In particular, we simulate the bias and mean square error performance of our approaches in settings where borrowing is and is not warranted to inform guidelines for the future use of our methods.
  • Thumbnail Image
    Item
    Credible Subgroups: Identifying the Population that Benefits from Treatment
    (2017-05) Schnell, Patrick
    A single treatment may have a different effect on different patients. In particular, some patients may benefit from a given treatment while others do not. Often, some of the variation in effect among patients can often be explained by characteristics of those patients that are observable before treatment. Widespread acknowledgment of treatment effect variation due to observable patient characteristics has increased the health science community's interest in a broad field referred to as personalized or precision medicine. Among the aims of precision medicine are identifying the set of treatments that would benefit a given patient, and conversely, identifying the population of patients who would benefit from a given treatment. We treat the latter problem in the context of clinical trials run by treatment developers (e.g., pharmaceutical companies), with special attention paid to interactions between those developers and the relevant regulatory agencies (e.g., the US Food and Drug Administration). The primary difficulty in estimating the benefiting population in such settings is controlling the frequency with which at least one type of patient is incorrectly determined to benefit, and doing so in a way that does not render the approach excessively conservative. As a motivating application throughout this dissertation, we consider a battery of related clinical trials of treatments for Alzheimer's disease carried out by the pharmaceutical company AbbVie. These trials contain a small number of continuous and binary baseline patient characteristics that may influence the treatment effect. We apply standard and more novel regression models to the supplied data and develop methods of inference to accommodate the varied features of the datasets, such as nonlinear effects, multiple important endpoints, more than two treatments, and regions of the covariate space that are sparse in observations or lacking common support among treatment arms. We also discuss topics in practical implementation of these methods. Our approaches yield reliable and easily interpretable inferences regarding the population that benefits from treatment.
  • Thumbnail Image
    Item
    Statistical methods for time-varying adaptive interventions
    (2024-12) Sur , Aparajita
    The growing demand for personalized medicine has led to the popularity of sequential multiple assignment randomized trials (SMARTs) in developing and evaluating adaptive interventions (AIs). AIs are sequences of decision rules that guide whether, when, and how to adapt an intervention for a patient. A SMART design is longitudinal and sequential in nature, as participants may progress through multiple treatment stages: they are initially randomized to a treatment and may be subsequently re-randomized to adapted or additional treatments based on their response to a prior treatment or other measured characteristics. This dissertation addresses two key challenges when developing and evaluating AIs. The first challenge is handling unique missing data issues in SMARTs that arise from its complex sequential design, where earlier responses (and missing data) could directly impact subsequent treatment pathways. Although multiple imputation (MI) is a widely used missing data method that can facilitate unbiased estimation in longitudinal studies, it relies on the untestable assumption that data are missing at random (MAR). It is unclear how violations of MAR affect inference with MI about adaptive interventions in a SMART setting, eliciting the need for sensitivity analyses to evaluate the robustness of conclusions to deviations from MAR. We highlight challenges in conducting sensitivity analyses in SMARTs, evaluate the performance of MI under varying SMART-specific departures from MAR, and propose a tailored sensitivity analysis framework with metrics to synthesize the SMART sensitivity analysis for count and continuous outcomes. The second challenge lies in the resource-intensive process of identifying optimal AIs and then testing them against control or non-personalized conditions, which traditionally requires two separate studies. We propose two cost-effective and potentially less time-consuming alternatives to invesitgate personalized treatments. First, a group sequential method that combines the process of selecting the best AI and confirming it is superior to a control treatment within a single SMART, and also allows for early termination of the study if there is overwhelming evidence of the AI's superiority. Second, we apply a causal inference method (g-computation) to evaluate hypothetical interventions via simulation using existing observational data, offering a cost-effective alternative that can generate preliminary evidence about personalized treatments before initiating large-scale experimental studies. The methods in this dissertation were motivated by and are illustrated using data from two applications: a SMART aimed at reducing alcohol-related outcomes for college students and an observational cohort study examining mental health and financial outcomes in older adults from the Rush Alzheimer’s Center.