Novel sequential methods in dose-finding designs : extensions of the continual reassessment method

Dose-finding trials are a form of clinical data collection process in which the primary objective is to estimate an optimum dose of an investigational new drug when given to a patient. This thesis develops and explores three novel dose-finding design methodologies. All design methodologies presented in this thesis are pragmatic. They use statistical models, incorporate clinicians' prior knowledge efficiently, and prematurely stop a trial for safety or futility reasons. Designing actual dose-finding trials using these methodologies will minimize practical difficulties, improve efficiency of dose estimation, be flexible to stop early and reduce possible patient discomfort or harm.

Sequential methods for pharmacogenetic studies

A study or experiment can be described as sequential if its design includes one or more interim analyses at which it is possible to stop the study, having reached a definitive conclusion concerning the primary question of interest. The potential of the sequential study to terminate earlier than the equivalent fixed sample size study means that, typically, there are ethical and economic advantages to be gained from using a sequential design. These advantages have secured a place for the methodology in the conduct of many clinical trials of novel therapies. Recently, there has been increasing interest in pharmacogenetics: the study of how DNA variation in the human genome affects the safety and efficacy of drugs. The potential for using sequential methodology in pharmacogenetic studies is considered and the conduct of candidate gene association studies, family-based designs and genome-wide association studies within the sequential setting is explored. The objective is to provide a unified framework for the conduct of these types of studies as sequential designs and hence allow experimenters to consider using sequential methodology in their future pharmacogenetic studies.

Comparative study of different formulations and solution methods for two phase flow in saturated porous media

Six models (Simulators) are formulated and developed with all possible combinations of pressure and saturation of the phases as primary variables. A comparative study between six simulators with two numerical methods, conventional simultaneous and modified sequential methods are carried out. The results of the numerical models are compared with the laboratory experimental results to study the accuracy of the model especially in heterogeneous porous media. From the study it is observed that the simulator using pressure and saturation of the wetting fluid (PW, SW formulation) is the best among the models tested. Many simulators with nonwetting phase as one of the primary variables did not converge when used along with simultaneous method. Based on simulator 1 (PW, SW formulation), a comparison of different solution methods such as simultaneous method, modified sequential and adaptive solution modified sequential method are carried out on 4 test problems including heterogeneous and randomly heterogeneous problems. It is found that the modified sequential and adaptive solution modified sequential methods could save the memory by half and as also the CPU time required by these methods is very less when compared with that using simultaneous method. It is also found that the simulator with PNW and PW as the primary variable which had problem of convergence using the simultaneous method, converged using both the modified sequential method and also using adaptive solution modified sequential method. The present study indicates that pressure and saturation formulation along with adaptive solution modified sequential method is the best among the different simulators and methods tested.

A practical comparison of group-sequential and adaptive designs

Sequential methods provide a formal framework by which clinical trial data can be monitored as they accumulate. The results from interim analyses can be used either to modify the design of the remainder of the trial or to stop the trial as soon as sufficient evidence of either the presence or absence of a treatment effect is available. The circumstances under which the trial will be stopped with a claim of superiority for the experimental treatment, must, however, be determined in advance so as to control the overall type I error rate. One approach to calculating the stopping rule is the group-sequential method. A relatively recent alternative to group-sequential approaches is the adaptive design method. This latter approach provides considerable flexibility in changes to the design of a clinical trial at an interim point. However, a criticism is that the method by which evidence from different parts of the trial is combined means that a final comparison of treatments is not based on a sufficient statistic for the treatment difference, suggesting that the method may lack power. The aim of this paper is to compare two adaptive design approaches with the group-sequential approach. We first compare the form of the stopping boundaries obtained using the different methods. We then focus on a comparison of the power of the different trials when they are designed so as to be as similar as possible. We conclude that all methods acceptably control type I error rate and power when the sample size is modified based on a variance estimate, provided no interim analysis is so small that the asymptotic properties of the test statistic no longer hold. In the latter case, the group-sequential approach is to be preferred. Provided that asymptotic assumptions hold, the adaptive design approaches control the type I error rate even if the sample size is adjusted on the basis of an estimate of the treatment effect, showing that the adaptive designs allow more modifications than the group-sequential method.

Interim analyses and sequential designs in phase III studies

Recruitment of patients to a clinical trial usually occurs over a period of time, resulting in the steady accumulation of data throughout the trial's duration. Yet, according to traditional statistical methods, the sample size of the trial should be determined in advance, and data collected on all subjects before analysis proceeds. For ethical and economic reasons, the technique of sequential testing has been developed to enable the examination of data at a series of interim analyses. The aim is to stop recruitment to the study as soon as there is sufficient evidence to reach a firm conclusion. In this paper we present the advantages and disadvantages of conducting interim analyses in phase III clinical trials, together with the key steps to enable the successful implementation of sequential methods in this setting. Examples are given of completed trials, which have been carried out sequentially, and references to relevant literature and software are provided.

Apply response adaptive randomization to group sequential with early stopping

Group sequential methods and response adaptive randomization (RAR) procedures have been applied in clinical trials due to economical and ethical considerations. Group sequential methods are able to reduce the average sample size by inducing early stopping, but patients are equally allocated with half of chance to inferior arm. RAR procedures incline to allocate more patients to better arm; however it requires more sample size to obtain a certain power. This study intended to combine these two procedures. We applied the Bayesian decision theory approach to define our group sequential stopping rules and evaluated the operating characteristics under RAR setting. The results showed that Bayesian decision theory method was able to preserve the type I error rate as well as achieve a favorable power; further by comparing with the error spending function method, we concluded that Bayesian decision theory approach was more effective on reducing average sample size.^

Monte Carlo-Based Bayesian group object tracking and causal reasoning

We present algorithms for tracking and reasoning of local traits in the subsystem level based on the observed emergent behavior of multiple coordinated groups in potentially cluttered environments. Our proposed Bayesian inference schemes, which are primarily based on (Markov chain) Monte Carlo sequential methods, include: 1) an evolving network-based multiple object tracking algorithm that is capable of categorizing objects into groups, 2) a multiple cluster tracking algorithm for dealing with prohibitively large number of objects, and 3) a causality inference framework for identifying dominant agents based exclusively on their observed trajectories.We use these as building blocks for developing a unified tracking and behavioral reasoning paradigm. Both synthetic and realistic examples are provided for demonstrating the derived concepts. © 2013 Springer-Verlag Berlin Heidelberg.

Sequentially testing for a gene-drug interaction in a genomewide analysis

Assaying a large number of genetic markers from patients in clinical trials is now possible in order to tailor drugs with respect to efficacy. The statistical methodology for analysing such massive data sets is challenging. The most popular type of statistical analysis is to use a univariate test for each genetic marker, once all the data from a clinical study have been collected. This paper presents a sequential method for conducting an omnibus test for detecting gene-drug interactions across the genome, thus allowing informed decisions at the earliest opportunity and overcoming the multiple testing problems from conducting many univariate tests. We first propose an omnibus test for a fixed sample size. This test is based on combining F-statistics that test for an interaction between treatment and the individual single nucleotide polymorphism (SNP). As SNPs tend to be correlated, we use permutations to calculate a global p-value. We extend our omnibus test to the sequential case. In order to control the type I error rate, we propose a sequential method that uses permutations to obtain the stopping boundaries. The results of a simulation study show that the sequential permutation method is more powerful than alternative sequential methods that control the type I error rate, such as the inverse-normal method. The proposed method is flexible as we do not need to assume a mode of inheritance and can also adjust for confounding factors. An application to real clinical data illustrates that the method is computationally feasible for a large number of SNPs. Copyright (c) 2007 John Wiley & Sons, Ltd.

Flexible selection of a single treatment incorporating short-term endpoint information in a phase II/III clinical trial

Seamless phase II/III clinical trials in which an experimental treatment is selected at an interim analysis have been the focus of much recent research interest. Many of the methods proposed are based on the group sequential approach. This paper considers designs of this type in which the treatment selection can be based on short-term endpoint information for more patients than have primary endpoint data available. We show that in such a case, the familywise type I error rate may be inflated if previously proposed group sequential methods are used and the treatment selection rule is not specified in advance. A method is proposed to avoid this inflation by considering the treatment selection that maximises the conditional error given the data available at the interim analysis. A simulation study is reported that illustrates the type I error rate inflation and compares the power of the new approach with two other methods: a combination testing approach and a group sequential method that does not use the short-term endpoint data, both of which also strongly control the type I error rate. The new method is also illustrated through application to a study in Alzheimer's disease. © 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

A saúde do homem: doenças crônicas não transmissíveis e vulnerabilidade social

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

A simulation study using the hybrid statistic and the meta-analysis t-test for data with matched and unmatched subjects in the interim analysis of clinical trials

An interim analysis is usually applied in later phase II or phase III trials to find convincing evidence of a significant treatment difference that may lead to trial termination at an earlier point than planned at the beginning. This can result in the saving of patient resources and shortening of drug development and approval time. In addition, ethics and economics are also the reasons to stop a trial earlier. In clinical trials of eyes, ears, knees, arms, kidneys, lungs, and other clustered treatments, data may include distribution-free random variables with matched and unmatched subjects in one study. It is important to properly include both subjects in the interim and the final analyses so that the maximum efficiency of statistical and clinical inferences can be obtained at different stages of the trials. So far, no publication has applied a statistical method for distribution-free data with matched and unmatched subjects in the interim analysis of clinical trials. In this simulation study, the hybrid statistic was used to estimate the empirical powers and the empirical type I errors among the simulated datasets with different sample sizes, different effect sizes, different correlation coefficients for matched pairs, and different data distributions, respectively, in the interim and final analysis with 4 different group sequential methods. Empirical powers and empirical type I errors were also compared to those estimated by using the meta-analysis t-test among the same simulated datasets. Results from this simulation study show that, compared to the meta-analysis t-test commonly used for data with normally distributed observations, the hybrid statistic has a greater power for data observed from normally, log-normally, and multinomially distributed random variables with matched and unmatched subjects and with outliers. Powers rose with the increase in sample size, effect size, and correlation coefficient for the matched pairs. In addition, lower type I errors were observed estimated by using the hybrid statistic, which indicates that this test is also conservative for data with outliers in the interim analysis of clinical trials.^

Residuos orgánicos como fuentes de fósforo

En este trabajo, se han llevado a cabo distintos experimentos en laboratorio, con el objetivo de estudiar el efecto de la aplicación de residuos orgánicos como fuentes de P en las pérdidas de este elemento, que se producen en suelo, tanto por escorrentía superficial como por lixiviación. El interés por evaluar las pérdidas de P se debe a la necesidad de conocer mejor los factores que influyen en los procesos de pérdidas de este elemento y así, poder reducir los problemas de eutrofización de aguas, tanto superficiales como subterráneas, provocadas por un exceso de este nutriente, junto con otros como el nitrógeno. Los trabajos experimentales que se han llevado a cabo se detallan a continuación: Se ha realizado el estudio de las formas de P contenidas en una serie de 14 residuos orgánicos, de distinto origen y tratamiento (compost, lodos, purines y digestato), comparando la información aportada por tres protocolos de fraccionamientos de P diferentes, seleccionados entre los principales métodos utilizados: protocolo de fraccionamiento de Ruttemberg (1992), protocolo de Normas, medidas y ensayos (Ruban et al., 2001a) y protocolo de Huang et al. (2008). Todos los métodos de fraccionamiento empleados aportaron información útil para conocer las formas de P de los residuos, a pesar de que alguno de ellos fue descrito para sedimentos o suelos. Sin embargo, resulta difícil comparar los resultados entre unos y otros, ya que cada uno emplea extractantes y tiempos de ensayos diferentes. Las cantidades de P total determinadas por cada método mantienen una relación lineal, aunque el método SMT, por ser más directo, obtiene las cantidades más elevadas para todos los residuos. Los métodos secuenciales (métodos de Huang y Ruttemberg), a pesar de ser más largos y tediosos, aportan información más detallada de la disponibilidad de las formas de P, y con ello, permiten obtener una mejor estimación de las pérdidas potenciales de este elemento tras su aplicación en suelo. Se han encontrado relaciones positivas entre las fracciones determinadas por estos dos métodos. Así mismo, se encuentra una correlación entre las formas solubles de P y la concentración de los iones [Ca + Fe + Al] de los residuos, útiles como indicadores de la disponibilidad de este elemento. Sin embargo, el protocolo SMT, no obtiene información de la solubilidad o disponibilidad de las formas de P contenidas, ni las formas determinadas mantienen relaciones directas con las de los otros métodos, con excepción del P total. Para el estudio del comportamiento de los residuos aplicados en suelos, se pusieron a punto sistemas de lluvia simulada, con el objetivo de caracterizar las pérdidas de P en la escorrentía superficial generada. Por otra parte, se emplearon columnas de suelos enmendados con residuos orgánicos, para el estudio de las pérdidas de P por lixiviación. Los ensayos de simulación de lluvia se llevaron a cabo de acuerdo al “National Phosphorus Research proyect“ (2001), que consigue simular eventos sucesivos de lluvia en unas condiciones semejantes a la realidad, empleando cajas llenas de suelo del horizonte superficial, con residuos aplicados tanto superficialmente como mediante mezcla con el propio suelo. Los ensayos se realizaron con seis residuos de diferente naturaleza y sometidos a distintos tratamientos. Se encontraron diferencias significativas en las pérdidas de las formas de P analizadas, tanto disueltas como particuladas, en las aguas de escorrentía generadas. En general, las pérdidas en el primer evento de lluvia tras la aplicación de los residuos fueron mayores a las generadas en el segundo evento, predominando las formas de P particuladas respecto a las disueltas en ambos. Se encontró una relación positiva entre las pérdidas de P en las aguas de escorrentía generadas en cada ensayo, con los contenidos de P soluble en agua y fácilmente disponible de los residuos empleados, determinados por los protocolos de fraccionamientos secuenciales. Además, se emplearon los modelos matemáticos desarrollados por Vadas et al. (2005, 2007), de evaluación de las pérdidas de P por escorrentía para fertilizantes y estiércoles. La predicción de estos modelos no se cumple en el caso de todos los residuos. Las distintas propiedades físicas de los residuos pueden afectar a las diferencias entre las pérdidas experimentales y las esperadas. Los ensayos de simulación del proceso de lixiviación se llevaron a cabo en columnas de percolación, con suelos enmendados con residuos orgánicos, de acuerdo a la norma “CEN/TS 14405–2004: Caracterización de los residuos – Test de comportamiento de lixiviación – Test de flujo ascendente”. Las pérdidas de P por procesos de lixiviación de agua, han sido despreciadas durante mucho tiempo respecto a las pérdidas por escorrentía. Sin embargo, se ha demostrado que deben tenerse en consideración, principalmente en algunos tipos de suelos o zonas cercanas a acuíferos. Se utilizaron tres suelos de distinta procedencia para los ensayos, de manera que se pudo estudiar la influencia del tipo de suelo en las pérdidas de P para cada tipo de residuo (purín, compost, digestato y lodo de EDAR). Los índices de adsorción de P determinados para cada suelo permiten evaluar aquellos que presentarán más riesgo de producir pérdidas de este elemento al aplicarse fuentes externas de P, encontrando una relación positiva entre ambos. Las pérdidas de P en los lixiviados varían en función tanto del residuo como del suelo empleado. Para el compost, el purín y el lodo, se encontró una relación entre las pérdidas generadas en el agua lixiviada de las columnas y las formas de P soluble contenidas en los residuos. Sin embargo, en el caso del digestato, no existía esta correlación. Las pérdidas para este residuo fueron en todos los casos menores a las estimadas, considerando las formas de P contenido. El estudio de la mojabilidad, propiedad física del residuo que evalúa la capacidad de interacción residuo-agua, permitió explicar el comportamiento anómalo de este residuo, con una mayor resistencia a que el agua entrara en su estructura y por tanto, una mayor dificultad de solubilizar el P contenido en el propio residuo, que en el caso de otros residuos. En general, podemos considerar que el estudio de las formas de P más disponibles o solubles en agua, aporta información útil de las pérdidas potenciales de P. Sin embargo, es necesario estudiar las propiedades físicas de los residuos orgánicos aplicados y la capacidad de adsorción de P de los suelos, para estimar las pérdidas de P y con ello, colaborar a controlar los procesos de eutrofización en aguas. ABSTRACT This dissertation explores the effect of organic wastes application as sources of P in losses of this element that occur by both surface runoff and leaching in soil. To do so, diverse laboratory experiments are conducted and presented here. Evaluating P losses is necessary to better understand the factors that influence the processes behind the loss of this element. Reducing P losses reduces eutrophication problems of both surface water and groundwater caused by an excess of this nutrient, along with other as nitrogen. Details of the experiments are presented below: The first experiment studies the forms of P contained in a series of 14 organic wastes of different origin and treatment (compost, sludge, slurry and digestate), comparing the information provided by three methods of P fractionation. The methods selected were: Ruttemberg protocol (1992); Standards, Measurements and Testing protocol (Ruban et al., 2001a); and Huang protocol (Huang et al., 2008). All fractionation methods employed successfully contribute to our knowledge of P forms in wastes, even though one of them was originally described for sediments or soils information. However, it is difficult to compare results among each other, as each protocol employs different extractants and time in the trials. Total amounts of P obtained by each method show a linear relationship, although the SMT method, which is more direct, obtains the highest amounts for all residues. Sequential methods (Huang and Ruttemberg’s protocols), despite being longer and more tedious, provide more detailed information on the availability of the forms of P. Therefore, allow the estimation of the potential losses of P after application in soil. Furthermore, positive relationships have been found among fractions obtained by these methods. Positive relationship has been found also among soluble forms of P and the concentration of ions Fe + Ca + Al, which is useful as an indicator of the availability of this element. However, the SMT protocol does not collect information about solubility or availability of forms of P contained; neither do certain forms maintain direct relations with the forms from other methods, with the exception of total P methods. To study the behavior of wastes applied to soils two experiments were conducted. Simulated rain systems were prepared to characterize P losses in the surface runoff generated. In addition, columns of soils amended with organic waste were developed for the study of P leaching losses. Simulated rain systems were carried out according to the ’National Phosphorus Research Project’ (2001), which manages to simulate successive rainfall events in conditions resembling reality. The experiment uses boxes filled with soil from the surface horizon amended with residues, both superficially and by mixing with the soil. Tests were conducted with six residues of different type and subjected to diverse treatments. Findings show significant differences in losses of the P forms analyzed in the generated runoff water, in both solution and particulate forms. In general, losses in the first rainfall event after application of waste were higher than the losses generated in the second event, predominating particulate forms of P over dissolved forms in both events. In all trials, a positive relationship was found between various P forms determined by sequential fractionation protocols (water soluble P and readily available P forms) and P losses in runoff. Furthermore, results from Vadas´s mathematical models (Vadas et al., 2005; 2007) to assess P losses by runoff fertilizers and manures indicate that the prediction of this model is not fulfilled in the case of all residues. The diverse physical properties of wastes may affect the differences between experimental and expected losses. Finally, leaching simulation processes were carried out in percolation columns, filled with soils amended with organic wastes, following the ‘CEN/TS 14405-2004 standard: Characterization of waste - Leaching behavior test - Test Flow ascending ’. P losses by leaching have been neglected for a long time with respect to runoff losses. However, findings corroborate previous studies showing that these P losses have to be taken into account, especially in certain types of soils and in zones near aquifers. To study the influence of soil type on P losses, experiments were carried out with three different soils and for each type of waste (manure, compost, digestate and sludge WWTP). Each soil’s P adsorption rates allow assessing which soils imply a higher risk of P losses when external sources of P are applied. P losses in leachate vary according to the type of soil employed and according to the specific residue. In the case of compost, manure and sludge, there is a relationship between leaching losses and residues’ soluble forms of P. The exception being the digestate, where there was no such correlation. Digestate P losses by leaching were lower than expected in all cases considering the forms of P contained. Moreover, examining digestate wettability -- that is, the physical property of the residue that assesses the capacity of waste-water interaction -- allowed explaining the anomalous behavior of this residue. Digestate has a high resistance to water entering its structure and thus higher difficulty to solubilize the P contained. Overall, studying the more available or soluble P forms provides useful information about the potential loss of P. However, this dissertation shows that it is necessary to examine the physical properties of organic residues applied as well as the P adsorption capacity of soils to estimate P losses, and thus to control eutrophication in water.

Adaptive designs in the time to event setting: The potential for benefit and risk

Thesis (Ph.D.)--University of Washington, 2016-06

Suboptimal Multistage Nonparametric Hypotheses Test

2000 Mathematics Subject Classification: 62L10, 62L15.