Dissertations / Theses on the topic 'Trial Sequential Analysis'

Le revisioni sistematiche e meta-analisi sono i migliori disegni di studio per assistere il processo decisionale in ambito clinico. Pazienti, clinici, ricercatori e decisori politici dovrebbero privilegiare la revisione sistematica rispetto a altri tipi di disegno sperimentale in quanto permette di aggregare diversi studi primari contemporaneamente, incrementando la validità interna ed esterna dei risultati. Inoltre, la meta-analisi aumenta la validità statistica raggiungendo più alti livelli di potenza statistica e precisione rispetto ai singoli studi. Ciò significa comprendere meglio la grandezza dell’effetto del trattamento e identificare il miglior intervento tra due, nel caso di meta-analisi standard, o tra molteplici, nel caso di network meta-analisi. Nonostante i numerosi vantaggi sopra riportati, la meta-analisi ha delle limitazioni importanti. La potenza e la precisione, infatti, dipendono da diversi elementi quali la dimensione campionaria degli studi, il numero di eventi, la variabilità del campione e la relativa eterogeneità. Questi elementi sono condizioni imprescindibili e non modificabili dai revisori. Se uno o più di questi elementi sono problematici, la meta-analisi replicherà lo stesso problema, sebbene l’entità del problema stesso può diminuire grazie alla presenza di altri studi. Per esempio, le meta-analisi che includono studi con una numerosità campionaria molto bassa saranno più propensi a diversi bias, quali lo small study effect e il publication bias. Inoltre molti studi clinici si propongo di ricercare modesti effetti di un intervento, che, generalmente, sono difficili da identificare: infatti, anche limitate variazioni nei dati di uno studio, o tra uno studio e un altro, possono risultare in bias che nascondono o alterano gli effetti di un intervento. Alla luce di questo background, la prima domanda di ricerca si è focalizzata sull’adeguatezza del reporting degli studi randomizzati e controllati inclusi in revisioni sistematiche e meta-analisi, in particolare in termini di potenza statistica, rilevanza clinica e conclusività dei risultati. In secondo luogo, sono stati analizzati metodi innovativi per identificare e valutare meta-analisi sotto-potenziate, inconclusive e imprecise: tali metodi includono la Trial Sequential Analysis e il metodo GRADE (Grading of Recommendations Assessment, Development and Evaluation). Infine, è stata studiata e applicata una nuova tecnica meta-analitica per confrontare allo stesso tempo interventi multipli tra loro considerando sia l’evidenza diretta sia l’evidenza indiretta. Al fine di rispondere alle domande di ricerca proposte, si è passati dall’unità di analisi delle revisioni sistematiche, lo studio randomizzato e controllato, ai metodi di sintesi cumulativa delle evidenze. Il reporting inadeguato, le meta-analisi sotto-potenziate e le possibili discordanze tra evidenze dirette e indirette a partire dall’analisi dei confronti testa-a-testa, possono alterare le decisioni cliniche portando a errori se viene meno un’attenta disamina e ci si basa su un’analisi statistica grossolana. La presente tesi è organizzata in tre sezioni principali: - Sezione 1: è stata valutato come il calcolo della dimensione campionaria sia stato riportato negli studi randomizzati e controllati che valutano l’efficacia di interventi riabilitativi nella lombalgia e, tra gli studi adeguatamente riportati, come i risultati siano interpretati in termini di significatività statistica e rilevanza clinica. - Sezione 2: sono state descritte, analizzate e applicate in diverse aree cliniche la Trial Sequential Analysis (TSA) e il metodo GRADE. Inoltre, è stata comparata la loro concordanza nel valutare e giudicare la qualità dell’evidenza. - Sezione 3: infine, considerato un campione di studi randomizzati e controllati che valutano l’efficacia di interventi riabilitativi e farmacologici per la lombalgia, sono state combinate le evidenze dirette e indirette esistenti tramite la tecnica di network meta-analisi. Il focus di questa tesi è orientato ai metodi avanzati ed innovativi di sintesi delle evidenze. L’introduzione e applicazione di questi metodi rappresentano la necessità di rispondere al recente bisogno di risultati precisi e affidabili, a metodi trasparenti e di riferimento per valutare la qualità e affidabilità dell’evidenza, e alla necessità di confrontare numerosi interventi senza limitarsi a confronti binari.<br>Systematic reviews and meta-analysis are the most appropriate and preferred study designs to inform clinical decision-making. Patients, clinicians, researchers and policymakers rely on this type of design since it aggregates several studies at the same time, increasing internal and external validity. Furthermore meta-analyses can increase statistical validity, reaching higher levels of power and precision as compared to single studies. This often means better understanding the magnitude of the treatment effect and its uncertainty therefore identifying the best intervention in pairwise or network (multiple) comparisons. Despite these advantages, meta-analyses also have limitations and shortcomings. Power and precision depend on several elements, such as size of studies, number of events, sample variability and underlying heterogeneity. These elements are precondition and cannot be modified by reviewers. If one or more of these elements are problematic, the meta-analysis will replicate the same problem, despite the severity of the problem might be diminished by the co-presence of multiple studies. For instance meta-analyses including several small studies will be prone to several biases, eg, small study effect and publication biases. Moreover actual studies often explore modest intervention effects, which are difficult to be identified: even limited perturbations of study data can result in biases that can hide or inflate intervention effects, sabotaging the decision-making process of health professionals. Against this background, we first wondered if RCTs included in systematic reviews and meta-analysis are adequately reported in terms of power and relevance/conclusiveness of findings. Secondly we explored how detect and assess underpowered, inconclusive and imprecise meta-analyses, using and comparing two modern approaches - Trial Sequential Analysis and the GRADE (Grading of Recommendations Assessment, Development and Evaluation). Third, we explored new meta-analytic techniques to contrast multiple interventions through direct and indirect evidence, in an extreme attempt to solve major limitations of a priori literature and lack of head to head trials. In order to answer these research questions, we moved from the unit of analysis of systematic reviews, the randomized controlled trial, to the methods to accumulate evidence. Inadequate reporting, underpowered meta-analyses and conflicting direct and indirect evidence beyond head-to-head comparisons can alter the clinical decision-making process unless proper assessment, analysis and critical interpretation are put in place. My dissertation is organized in three main Sections: - Section 1: We focused on how sample size calculations were reported in RCTs exploring the efficacy of low back pain rehabilitation interventions and, among those adequately reported, how findings were interpreted in terms of statistical significance and clinical relevance. - Section 2: We explored Trial Sequential Analysis and the GRADE approach in several medical areas. We compared the agreement of the approaches in evaluating the overall quality of evidence. - Section 3. We finally combined direct and indirect evidence on a sample of RCTs assessing rehabilitation interventions for low back pain through a network meta-analysis. This dissertation focuses on innovative advanced methods used in evidence synthesis science. These methods are partial answers to the need for precise and reliable results, standard and transparent methods to assess the body of evidence, and comparisons of multiple interventions in addition to pairwise comparisons.

A recently developed group-sequential response-adaptive design to compare two treatments with immediate normally distributed responses and known variances is considered. The power function of the test is the same as that under non-adaptive sampling, and significant decreases in the inferior treatment number can be achieved with only minor increases in the average sample number. Reasonably accurate corrected confidence intervals for both the treatment mean difference and the individual means are obtained by constructing approximately pivotal quantities. An approximation to the bias of the maximum likelihood estimator of the treatment mean difference is also studied. When the variances of the response variables are unknown, inaccurate estimates of these can affect the Type II error rate considerably. A new modified version of an existing sample size re-estimation method is developed for group-sequential response-adaptive designs for normal data with unknown variances. The principal modifications involve updating the required sample size at each interim analysis and calculating the test statistic based on current estimates of the variances. Simulation is used to compare the performance of this test with modified versions of two other tests from the recent literature. The power is shown to be more accurately maintained in the new test. An analogous group-sequential response-adaptive design to compare two treatments with immediate dichotomous responses is then developed. Since the variances of the response variables are unknown in binary response trials, due to their dependence on the unknown success probabilities, the new sample size re-estimation method is incorporated into the design. Two parameters of interest are considered, the log odds ratio and the simple difference between the probabilities of success. Three adaptive urn models are studied and their properties are compared to a sequential maximum likelihood estimation rule that minimises the expected number of treatment failures. Simulation results favour the drop-the-loser rule

Thesis (Ph.D. in Biostatistics) -- University of Colorado Denver, 2007.<br>Typescript. Includes bibliographical references (leaves 104-106). Free to UCD affiliates. Online version available via ProQuest Digital Dissertations;

Adaptive clinical trials for new drugs or treatment options promise substantial benefits to both the pharmaceutical industry and the patients, but complicate resource allocation decisions. In this dissertation, we focus on sequential adaptive clinical trials with binary response, which allow for early termination of drug testing for benefit or futility at interim analysis points. The option to stop the trial early enables the trial sponsor to mitigate investment risks on ineffective drugs, and to shorten the development time line of effective drugs, hence reducing expenditures and expediting patient access to these new therapies. In this setting, decision makers need to determine a testing schedule, or the number of patients to recruit at each interim analysis point, and stopping criteria that inform their decision to continue or stop the trial, considering performance measures that include drug misclassification risk, time-to-market, and expected profit. In the first manuscript, we model current practices of sequential adaptive trials, so as to quantify the magnitude of drug misclassification risk. Towards this end, we build a simulation model to realistically represent the current decision-making process, including the utilization of the triangular test, a widely implemented sequential methodology. We find that current practices lead to a high risk of incorrectly terminating the development of an effective drug, thus, to unrecoverable expenses for the sponsor, and unfulfilled patient needs. In the second manuscript, we study the sequential resource allocation decision, in terms of a testing schedule and stopping criteria, so as to quantify the impact of interim analyses on the aforementioned performance measures. Towards this end, we build a stochastic dynamic programming model, integrated with a Bayesian learning framework for updating the drug’s estimated efficacy. The resource allocation decision is characterized by endogenous uncertainty, and a trade-off between the incentive to establish that the drug is effective early on (exploitation), due to a time-decreasing market revenue, and the benefit from collecting some information on the drug’s efficacy prior to committing a large budget (exploration). We derive important structural properties of an optimal resource allocation strategy and perform a numerical study based on realistic data, and show that sequential adaptive trials with interim analyses substantially outperform traditional trials. Finally, the third manuscript integrates the first two models, and studies the benefits of an optimal resource allocation decision over current practices. Our findings indicate that our optimal testing schedules outperform different types of fixed testing schedules under both perfect and imperfect information.<br>Ph. D.

This thesis deals with sequential and adaptive methods for clinical trials, and how such methods can be used to achieve efficient clinical trial designs. The efficiency gains that can be achieved through non-adaptive group sequential methods are well established, while the newer adaptive methods seek to combine the best of the classical group sequential framework with an approach that gives increased flexibility. Our results show that the adaptive methods can provide some additional efficiency, as well as increased possibilities to respond to new internal and external information. Care is however needed when applying adaptive methods. While sub-optimal rules for adaptation can lead to inefficiencies, the logistical challenges can also be considerable. Efficient non-adaptive group sequential designs are often easier to implement in practice, and have for the cases we have considered been quite competitive in terms of efficiency. The four problems that are presented in this thesis are very relevant to how clinical trials are run in practice. The solutions that we present are either new approaches to problems that have not previously been solved, or methods that are more efficient than the ones currently available in the literature. Several challenging optimisation problems are solved through numerical computations. The optimal designs that are achieved can be used to benchmark new methods proposed in this thesis as well as methods available in the statistical literature. The problem that is solved in Chapter 5 can be viewed as a natural extension to the other problems. It brings together methods that we have used to the design of individual trials, to solve the more complex problem of designing a sequence of trials that are the core part of a clinical development program. The expected utility that is maximised is motivated by how the development of new medicines works in practice.

Background Many randomised controlled trials (RCTs) are available to support using different pharmacotherapy agents in the management of various neuropathic pain conditions. However, choosing these pharmacotherapy agents for neuropathic pain is challenging, due to the limited evidence-based knowledge to support the use of different pharmacotherapy agents in different neuropathic pain conditions. Aims The aim of this PhD is to evaluate the efficacy and safety of oral and topical pharmacotherapies for managing neuropathic pain by deriving placebo and active comparative efficacy and safety evidence from RCTs. Methods This research used three approaches to summarise and synthesise evidence from randomised controlled studies including: a systematic review of placebo and active control RCTs to summarise and criticise the current evidence in neuropathic pain; a meta-analysis and sequential analysis of eligible studies to provide a more precise estimate of the overall treatment effects; and a network meta-analysis to estimate the relative effectiveness of the most commonly used interventions in neuropathic pain. Results Systematic review Two hundred placebo and active-controlled trials met the inclusion criteria. A wide range of different treatments were studied in these trials, including anticonvulsants, antidepressants, opioids and topical capsaicin and lidocaine. Most of the included studies were parallel placebo-controlled trials and commonly lasted for 3 to 12 weeks. In addition, the vast majority of the included RCTs were conducted in participants with painful diabetic neuropathy and post-herpetic neuralgia, while only a few trials were conducted in participants with central neuropathic pain conditions. Pairwise meta-analysis Sixty seven trials were eligible for the pairwise meta analysis of efficacy outcomes. Of the anticonvulsants group pregabalin and gabapentin compared with placebo demonstrated efficacy for 50% and 30% pain reduction and global improvement in patients with neuropathic pain. The efficacy of anticonvulsants varied in different types of neuropathic pain. Gabapentin when compared against a placebo was better than a placebo in PHN and PDN, while pregabalin was significantly effective in patients with post-herpetic neuralgia (PHN) and painful diabetic neuropathy (PDN) but not in patients with HIV associated neuropathic pain. Others anticonvulsant agents, such as lamotrigine, valproic acid, topiramate, levetiracetam and oxcarbazepine, were tested in a small number of trials. These did not provide useful benefits compared with a placebo for a 50% and 30% pain reduction. Of the antidepressant group, duloxetine when compared to a placebo demonstrated efficacy for 50% and 30% pain reduction in diabetic neuropathic pain. A few active comparison trials failed to demonstrate superior efficacy of one drug over another for a 50% and 30% reduction in neuropathic pain. Trial sequential analysis To examine the reliability and conclusiveness of the available evidence, trialsequential analysis has been applied in this study. The results show convincing evidence of the efficacy of some interventions (e.g. pregabalin, gabapentin and duloxetine) to reduce pain by 50% in some neuropathic pain conditions (e.g. diabetic neuropathic pain and post-herpetic neuralgia). The continuation of RCTs of pregabalin and duloxetine in diabetic neuropathy and gabapentin in post-herpetic neuralgia is not necessary as there appears to be sufficient evidence of the efficacy of these treatments in the management diabetic neuropathic pain and post herpetic neuralgia. Further RCTs of duloxetine, pregabalin and gabapentin are however required for central neuropathic pain. In contrast, the analysis failed to provide evidence that opioids and high concentration capsaicin demonstrate a 50% pain reduction. Network meta-analysis Twenty-eight trials were eligible for the network meta-analysis. The results incorporating both direct-comparison and indirect-comparison evidence suggested that there is no superiority of duloxetine over amitriptyline, pregabalin and gabapentin in achieving at least a 30% and 50% pain reduction with a treatment duration of 7 to 12 weeks in patients with neuropathic pain conditions, such as diabetic neuropathic pain, postherpetic neuralgia and spinal cord injury. Conclusions In summary, this research has found that some good quality trials provide good evidence regarding the efficacy of duloxetine, pregabalin and gabapentin in a minority of patients with neuropathic pain. Until advancements in developing mechanism-based approaches and improved clinical trial design become available, the routine use of these medications is unlikely to be changed. This may support the hypothesis that traditional RCTs might not be a suitable method of choice to address provisional health questions in routine clinical practice.

The research of my dissertation studies the methods of designing and analyzing sequential multiple assignment randomized trial (SMART) for comparing multiple adaptive interventions. As a SMART typically consists of numerous adaptive interventions, inferential procedures based on pairwise comparisons of all interventions may suffer substantial loss in power after accounting for multiplicity. I address this problem using two approaches. First, I propose a likelihood-based Wald test, study the asymptotic distribution of its test statistics, and apply it as a gate-keeping test for making an adaptive intervention selection. Second, I consider a multiple comparison with the best approach by constructing simultaneous confidence intervals that compare the interventions of interest with the truly best intervention, which is assumed to be unknown in inference; an adaptive intervention with the proposed interval excluding zero will be declared as inferior to the truly best with a pre-specified confidence level. Simulation studies show that both methods outperform the corresponding multiple comparison procedures based on Bonferroni's correction in terms of the power of test and the average width of confidence intervals for estimation. Simulations also suggest desirable properties of the proposed methods. I apply these methods to analyze two real data sets. As part of the dissertation, I also develop a user-friendly R software package that covers many statistical work related to SMART, including study design, data analysis and visualization. Both proposed methods can be implemented by using this R package. In the end of the dissertation, I show an application of designing a SMART to compare multiple patient care strategies for depression management based on one of the proposed methods.

碩士<br>國防醫學院<br>公共衛生學研究所<br>107<br>Introduction The proportion of elderly people aged over 65 in Taiwan has reached 13.9% in the end of 2017. The aging of the population will lead to the increase of chronic diseases such as osteoporosis, which is an important public health issue. Therefore, this study first search the meta- analysis of osteoporosis and gene polymorphism, and used trial sequential analysis to explore whether these SNPs provided by the above article can be affirmed with evidence. Among them, one of SNPs called VDR ApaI still has insufficient evidence, therefore, we first use the samples of our laboratory for case-control study, and then accumulate samples through meta- analysis and trial sequential analysis to explore whether the accumulated sample number can be affirmed and we don’t need more studies to explore in the future. Aim In this study, we screen the candidate genes of osteoporosis through trial sequential analysis and aim to assess the relationship between VDR ApaI gene polymorphisms and osteoporosis through case control study, meta- analysis and trial sequential analysis. Materials and methods First part, we search the meta- analysis literatures about relationship between gene polymorphisms and osteoporosis on PubMed, Embase and Cochrane (since the dates of article reception to September 2018). Then explore and screen the relationship between these SNPs in these literatures and osteoporosis through trial sequential analysis. Second part, we use case control study to explore the relationship between VDR ApaI gene polymorphisms which screen from the first part and osteoporosis. Third part, use meta- analysis and trial sequential analysis to explore the relationship between VDR ApaI gene polymorphisms and osteoporosis. The criteria of studies were included in this study were: (1) cross-sectional survey or case control study, (2) the definition of patients with osteoporosis had to be T score＜-2.5, (3) age of study population > 18 years, (4) article should report the detailed data of VDR ApaI genotypes distribution and (5) race of study population were Asian or White people. Results In the case control study, we find out that the gene polymorphism of VDR ApaI is not associated with osteoporosis. However, after adjust sex, age, BMI, bone mass and the intake of vitamin D, the gene polymorphism of VDR ApaI is associated with osteoporosis (OR = 2.16, 95% CI = 1.18- 3.94). In meta- analysis and trial sequential analysis, including data from 24 studies (with our case control study) were analyzed using a random effects model, we find out that the gene polymorphism of VDR ApaI is still not associated with osteoporosis in Asian population, and don’t need more studies to discuss their association in the future. Conclusion 1、The strategy of screening candidate genes of osteoporosis through trial sequential analysis find out there are 21 SNPs may be associated with osteoporosis in the race of Asian people. 2、In the case control study, meta- analysis and trial sequential analysis, we find out that the gene polymorphism of VDR ApaI is not associated with osteoporosis, and in the future we don’t need more studies to discuss their association.

碩士<br>國防醫學院<br>公共衛生學研究所<br>107<br>Introduction: Genes play an important role in osteoarthritis (OA). Although there are 80 gene-related papers to investigate the study of polymorphism and osteoarthritis, there are many different opinions in the same gene. However, the results of the study are inconsistent, and no study indicates whether the cumulative number of samples is enough. Therefore, the Trial Sequential Analysis (TSA) of this study verifies whether the cumulative number of samples at the locus has reached a certain conclusion, no need to invest more resources. Materials and methods: This study develops a new gene candidate strategy to find all meta-analysis articles in Pubmed, embase, cochrane databases related to osteoarthritis. Through these 80 articles, provided for 29 loci. Conduct TSA to determine "whether there is sufficient evidence to show whether the gene is not related to the disease, and no need to invest a lot of resources for research." It is found that only the locus of rs9340799 on ESR1 is insufficient due to the number of samples. It is expected that after adding this study, affirmative conclusions can be made, so TSA is conducted for this locus. Results: Through TSA candidates for all OA-related 29 loci, only 6 loci such as GDF5 are associated with OA. 9 articles (including this study), 7 Asians, 2 whites, showing ESR1 XbaI gene was no statistically significant difference in polymorphism (Allele model: OR=0.84 (95% CI=0.67-1.04)). Before adding this study , the number of Asian TSA cumulative samples reached 2,223, which is still not enough to conclude. A key sample of this study was included: 987 patients, and it is determined that the ESR1 XbaI polymorphism was not associated with osteoarthritis. Conclusion: 1.All locus related to OA only BtgI, SMAD3, DVWA, GDF, ApaI, MMP-1 are indeed associated with the disease, and 23 locus including CYP19A1 have sufficient sample number to confirm that there is no association with disease. 2. The ESR1 XbaI polymorphism and osteoarthritis in Asian populations were not associated after adding this study. Subsequent research scholars do not need to invest resources in this locus to explore the polymorphism of genes.

Background: There are a plethora of potential statistical designs which can be used to evaluate efficacy of a novel cancer treatment in the phase II clinical trial setting. Unfortunately, there is no consensus as to which design one should prefer, nor even which definition of efficacy should be used and the primary endpoint conclusion can vary depending on which design is chosen. It would be useful if an all-encompassing methodology was possible which could evaluate all the different designs simultaneously and allow investigators an understanding of the trial results under the varying scenarios. Methods: Finite Markov chain imbedding is a method which can be used in the setting of phase II oncology clinical trials but never previously evaluated in this scenario. Simple variations to the transition matrix or end-state probability definitions can be performed which allow for evaluation of multiple designs and endpoints for a single trial. A computer program is written in R which allows for computation of p-values and conditional power, two common statistical measures used for evaluation of trial results. A simulation study is performed on data arising from an actual phase II clinical trial performed recently in which the study conclusion regarding the efficacy of the potential treatment was debatable. Results: Finite Markov chain imbedding is shown to be useful for evaluating phase II oncology clinical trial results. The R code written for evaluating the simulation study is demonstrated to be fast and useful for investigating different trial designs. Further detail regarding the clinical trial results are presented, including the potential prolongation of stable disease of the treatment, which is a potentially useful marker of efficacy for this cytostatic agent. Conclusions: This novel methodology may prove to be an useful investigative technique for the evaluation of phase II oncology clinical trial data. Future studies which have disputable conclusions might become less controversial with the aid of finite Markov chain imbedding and the possible multiple evaluations which is now viable. Better understanding of activity for a given treatment might expedite the drug development process or help distinguish active from inactive treatments

Placebo response, an apparent improvement in the clinical condition of patients randomly assigned to the placebo treatment, is a major issue in clinical trials on psychiatric and pain disorders. Properly addressing the placebo response is critical to an accurate assessment of the efficacy of a therapeutic agent. The Sequential Parallel Comparison Design (SPCD) is one approach for addressing the placebo response. A SPCD trial runs in two stages, re-randomizing placebo patients in the second stage. Analysis pools the data from both stages. In this thesis, we propose a Bayesian approach for analyzing SPCD data. Our primary proposed model overcomes some of the limitations of existing methods and offers greater flexibility in performing the analysis. We find that our model is either on par or, under certain conditions, better, in preserving the type I error and minimizing mean square error than existing methods. We further develop our model in two ways. First, through prior specification we provide three approaches to model the relationship between the treatment effects from the two stages, as opposed to arbitrarily specifying the relationship as was done in previous studies. Under proper specification these approaches have greater statistical power than the initial analysis and give accurate estimates of this relationship. Second, we revise the model to treat the placebo response as a continuous rather than a binary characteristic. The binary classification, which groups patients into “placebo-responders” or “placebo non-responders”, can lead to misclassification, which can adversely impact the estimate of the treatment effect. As an alternative, we propose to view the placebo response in each patient as an unknown continuous characteristic. This characteristic is estimated and then used to measure the contribution (or the weight) of each patient to the treatment effect. Building upon this idea, we propose two different models which weight the contribution of placebo patients to the estimated second stage treatment effect. We show that this method is more robust against the potential misclassification of responders than previous methods. We demonstrate our methodology using data from the ADAPT-A SPCD trial.