Active-control trials with binary data: a comparison of methods for testing superiority or non-inferiority using the odds ratio

This paper considers methods for testing for superiority or non-inferiority in active-control trials with binary data, when the relative treatment effect is expressed as an odds ratio. Three asymptotic tests for the log-odds ratio based on the unconditional binary likelihood are presented, namely the likelihood ratio, Wald and score tests. All three tests can be implemented straightforwardly in standard statistical software packages, as can the corresponding confidence intervals. Simulations indicate that the three alternatives are similar in terms of the Type I error, with values close to the nominal level. However, when the non-inferiority margin becomes large, the score test slightly exceeds the nominal level. In general, the highest power is obtained from the score test, although all three tests are similar and the observed differences in power are not of practical importance. Copyright (C) 2007 John Wiley & Sons, Ltd.

Estimation following selection of the largest of two normal means

This paper considers the problem of estimation when one of a number of populations, assumed normal with known common variance, is selected on the basis of it having the largest observed mean. Conditional on selection of the population, the observed mean is a biased estimate of the true mean. This problem arises in the analysis of clinical trials in which selection is made between a number of experimental treatments that are compared with each other either with or without an additional control treatment. Attempts to obtain approximately unbiased estimates in this setting have been proposed by Shen [2001. An improved method of evaluating drug effect in a multiple dose clinical trial. Statist. Medicine 20, 1913–1929] and Stallard and Todd [2005. Point estimates and confidence regions for sequential trials involving selection. J. Statist. Plann. Inference 135, 402–419]. This paper explores the problem in the simple setting in which two experimental treatments are compared in a single analysis. It is shown that in this case the estimate of Stallard and Todd is the maximum-likelihood estimate (m.l.e.), and this is compared with the estimate proposed by Shen. In particular, it is shown that the m.l.e. has infinite expectation whatever the true value of the mean being estimated. We show that there is no conditionally unbiased estimator, and propose a new family of approximately conditionally unbiased estimators, comparing these with the estimators suggested by Shen.

Fitting models for the joint action of two drugs using SAS(R)

Combinations of drugs are increasingly being used for a wide variety of diseases and conditions. A pre-clinical study may allow the investigation of the response at a large number of dose combinations. In determining the response to a drug combination, interest may lie in seeking evidence of synergism, in which the joint action is greater than the actions of the individual drugs, or of antagonism, in which it is less. Two well-known response surface models representing no interaction are Loewe additivity and Bliss independence, and Loewe or Bliss synergism or antagonism is defined relative to these. We illustrate an approach to fitting these models for the case in which the marginal single drug dose-response relationships are represented by four-parameter logistic curves with common upper and lower limits, and where the response variable is normally distributed with a common variance about the dose-response curve. When the dose-response curves are not parallel, the relative potency of the two drugs varies according to the magnitude of the desired effect and the models for Loewe additivity and synergism/antagonism cannot be explicitly expressed. We present an iterative approach to fitting these models without the assumption of parallel dose-response curves. A goodness-of-fit test based on residuals is also described. Implementation using the SAS NLIN procedure is illustrated using data from a pre-clinical study. Copyright © 2007 John Wiley & Sons, Ltd.

A 25-year review of sequential methodology in clinical studies

This paper explores the theoretical developments and subsequent uptake of sequential methodology in clinical studies in the 25 years since Statistics in Medicine was launched. The review examines the contributions which have been made to all four phases into which clinical trials are traditionally classified and highlights major statistical advancements, together with assessing application of the techniques. The vast majority of work has been in the setting of phase III clinical trials and so emphasis will be placed here. Finally, comments are given indicating how the subject area may develop in the future.

The genome of Rhizobium leguminosarum has recognizable core and accessory components

Background: Rhizobium leguminosarum is an alpha-proteobacterial N-2-fixing symbiont of legumes that has been the subject of more than a thousand publications. Genes for the symbiotic interaction with plants are well studied, but the adaptations that allow survival and growth in the soil environment are poorly understood. We have sequenced the genome of R. leguminosarum biovar viciae strain 3841. Results: The 7.75 Mb genome comprises a circular chromosome and six circular plasmids, with 61% G+C overall. All three rRNA operons and 52 tRNA genes are on the chromosome; essential protein-encoding genes are largely chromosomal, but most functional classes occur on plasmids as well. Of the 7,263 protein-encoding genes, 2,056 had orthologs in each of three related genomes ( Agrobacterium tumefaciens, Sinorhizobium meliloti, and Mesorhizobium loti), and these genes were overrepresented in the chromosome and had above average G+C. Most supported the rRNA-based phylogeny, confirming A. tumefaciens to be the closest among these relatives, but 347 genes were incompatible with this phylogeny; these were scattered throughout the genome but were over-represented on the plasmids. An unexpectedly large number of genes were shared by all three rhizobia but were missing from A. tumefaciens. Conclusion: Overall, the genome can be considered to have two main components: a 'core', which is higher in G+C, is mostly chromosomal, is shared with related organisms, and has a consistent phylogeny; and an 'accessory' component, which is sporadic in distribution, lower in G+C, and located on the plasmids and chromosomal islands. The accessory genome has a different nucleotide composition from the core despite a long history of coexistence.

Nitrogen management is essential to prevent tropical oil palm plantations from causing ground-level ozone pollution

More than half the world's rainforest has been lost to agriculture since the Industrial Revolution. Among the most widespread tropical crops is oil palm (Elaeis guineensis): global production now exceeds 35 million tonnes per year. In Malaysia, for example, 13% of land area is now oil palm plantation, compared with 1% in 1974. There are enormous pressures to increase palm oil production for food, domestic products, and, especially, biofuels. Greater use of palm oil for biofuel production is predicated on the assumption that palm oil is an "environmentally friendly'' fuel feedstock. Here we show, using measurements and models, that oil palm plantations in Malaysia directly emit more oxides of nitrogen and volatile organic compounds than rainforest. These compounds lead to the production of ground-level ozone (O-3), an air pollutant that damages human health, plants, and materials, reduces crop productivity, and has effects on the Earth's climate. Our measurements show that, at present, O-3 concentrations do not differ significantly over rainforest and adjacent oil palm plantation landscapes. However, our model calculations predict that if concentrations of oxides of nitrogen in Borneo are allowed to reach those currently seen over rural North America and Europe, ground-level O-3 concentrations will reach 100 parts per billion (10(9)) volume (ppbv) and exceed levels known to be harmful to human health. Our study provides an early warning of the urgent need to develop policies that manage nitrogen emissions if the detrimental effects of palm oil production on air quality and climate are to be avoided.

A combined score test for binary and ordinal endpoints from clinical trials

There is growing interest, especially for trials in stroke, in combining multiple endpoints in a single clinical evaluation of an experimental treatment. The endpoints might be repeated evaluations of the same characteristic or alternative measures of progress on different scales. Often they will be binary or ordinal, and those are the cases studied here. In this paper we take a direct approach to combining the univariate score statistics for comparing treatments with respect to each endpoint. The correlations between the score statistics are derived and used to allow a valid combined score test to be applied. A sample size formula is deduced and application in sequential designs is discussed. The method is compared with an alternative approach based on generalized estimating equations in an illustrative analysis and replicated simulations, and the advantages and disadvantages of the two approaches are discussed.

NOx and O3 above a tropical rainforest: an analysis with a global and box model

A cross-platform field campaign, OP3, was conducted in the state of Sabah in Malaysian Borneo between April and July of 2008. Among the suite of observations recorded, the campaign included measurements of NOx and O3 – crucial outputs of any model chemistry mechanism. We describe the measurements of these species made from both the ground site and aircraft. We then use the output from two resolutions of the chemistry transport model p-TOMCAT to illustrate the ability of a global model chemical mechanism to capture the chemistry at the rainforest site. The basic model performance is good for NOx and poor for ozone. A box model containing the same chemical mechanism is used to explore the results of the global model in more depth and make comparisons between the two. Without some parameterization of the nighttime boundary layer – free troposphere mixing (i.e. the use of a dilution parameter), the box model does not reproduce the observations, pointing to the importance of adequately representing physical processes for comparisons with surface measurements. We conclude with a discussion of box model budget calculations of chemical reaction fluxes, deposition and mixing, and compare these results to output from p-TOMCAT. These show the same chemical mechanism behaves similarly in both models, but that emissions and advection play particularly strong roles in influencing the comparison to surface measurements.

Induced-fit binding of π-electron donor substrates to macrocyclic aromatic ether-imide-sulfones: a versatile approach to molecular assembly

Novel macrocyclic receptors which bind electron-donor aromatic substrates via π-stacking donor- acceptor interactions are obtained by cyclo-imidization of an amine-functionalized arylether-sulfone with pyromellitic- and 1,4,5,8-naphthalene-tetracarboxylic dianhydrides. These macrocycles complex with a wide variety of π-donor substrates including tetrathiafulvalene, naphthalene, anthracene, pyrene, perylene, and functional derivatives of these polycyclic hydrocarbons. The resulting supramolecular assemblies range from simple 1:1 complexes, to [2]- and [3]-pseudorotaxanes, and even (as a result of crystallographic disorder) an apparent polyrotaxane. Direct, five-component self-assembly of a metal-centred [3]pseudorotaxane is also observed, on complexation of a macrocyclic ether-imide with 8-hydroxyquinoline in the presence of palladium(II) ions. Binding studies in solution were carried out by 1H NMR and UV-visible spectroscopy, and the stoichiometries of binding were confirmed by Job plots based on charge-transfer absorption bands. The highest association constants are found for strong π-donor guests with large surface-areas, notably perylene and 1-hydroxypyrene, for which Ka values of 1.4 x 103 and 2.3 x 103 M-1 respectively are found. Single crystal X-ray analyses of the receptors and their derived complexes reveal large, induced-fit distortions of the macrocyclic frameworks as a result of complexation. These structures provide compelling evidence for the existence of strong, attractive forces between the electronically-complementary aromatic π-systems of host and guest.

Refinement of the Clinical Scenario Evaluation Framework for Assessment of Competing Development Strategies with an Application to Multiple Sclerosis

When competing strategies for development programs, clinical trial designs, or data analysis methods exist, the alternatives need to be evaluated in a systematic way to facilitate informed decision making. Here we describe a refinement of the recently proposed clinical scenario evaluation framework for the assessment of competing strategies. The refinement is achieved by subdividing key elements previously proposed into new categories, distinguishing between quantities that can be estimated from preexisting data and those that cannot and between aspects under the control of the decision maker from those that are determined by external constraints. The refined framework is illustrated by an application to a design project for an adaptive seamless design for a clinical trial in progressive multiple sclerosis.

Design considerations in the sequential analysis of matched case–control data

A role for sequential test procedures is emerging in genetic and epidemiological studies using banked biological resources. This stems from the methodology's potential for improved use of information relative to comparable fixed sample designs. Studies in which cost, time and ethics feature prominently are particularly suited to a sequential approach. In this paper sequential procedures for matched case–control studies with binary data will be investigated and assessed. Design issues such as sample size evaluation and error rates are identified and addressed. The methodology is illustrated and evaluated using both real and simulated data sets.