Simultaneous confidence intervals for an extended Koch-Röhmel design in three-arm non-inferiority trials

Three-arm 'gold-standard' non-inferiority trials are recommended for indications where only unstable reference treatments are available and the use of a placebo group can be justified ethically. For such trials, several study designs have been suggested that use the placebo group for testing 'assay sensitivity', that is, the ability of the trial to replicate efficacy. Should the reference fail in the given trial, then non-inferiority could also be shown with an ineffective experimental treatment and hence becomes useless. In this article, we extend the so-called Koch-Röhmel design where a proof of efficacy for the experimental treatment is required in order to qualify for the non-inferiority test. While the efficacy of the experimental treatment is an indication of assay sensitivity, it does not guarantee that the reference is sufficiently efficient to let the non-inferiority claim be meaningful. It has, therefore, been suggested to adaptively test the non-inferiority only if the reference demonstrates superiority to placebo and otherwise to test δ -superiority of the experimental treatment over placebo, where δ is chosen in such a way that it provides proof of non-inferiority with regard to the reference's historical effect. In this article, we extend the previous work by complementing its adaptive test with compatible simultaneous confidence intervals. Confidence intervals are commonly used and suggested by regulatory guidelines for non-inferiority trials. We show how to adopt different approaches to simultaneous confidence intervals from the literature to the setting of three-arm non-inferiority trials and compare these methods in a simulation study. Finally, we apply these methods to a real clinical trial example.

Change point detection for clustered expression data

BACKGROUND

To detect changes in biological processes, samples are often studied at several time points. We examined expression data measured at different developmental stages, or more broadly, historical data. Hence, the main assumption of our proposed methodology was the independence between the examined samples over time. In addition, however, the examinations were clustered at each time point by measuring littermates from relatively few mother mice at each developmental stage. As each examination was lethal, we had an independent data structure over the entire history, but a dependent data structure at a particular time point. Over the course of these historical data, we wanted to identify abrupt changes in the parameter of interest - change points.

RESULTS

In this study, we demonstrated the application of generalized hypothesis testing using a linear mixed effects model as a possible method to detect change points. The coefficients from the linear mixed model were used in multiple contrast tests and the effect estimates were visualized with their respective simultaneous confidence intervals. The latter were used to determine the change point(s). In small simulation studies, we modelled different courses with abrupt changes and compared the influence of different contrast matrices. We found two contrasts, both capable of answering different research questions in change point detection: The Sequen contrast to detect individual change points and the McDermott contrast to find change points due to overall progression. We provide the R code for direct use with provided examples. The applicability of those tests for real experimental data was shown with in-vivo data from a preclinical study.

CONCLUSION

Simultaneous confidence intervals estimated by multiple contrast tests using the model fit from a linear mixed model were capable to determine change points in clustered expression data. The confidence intervals directly delivered interpretable effect estimates representing the strength of the potential change point. Hence, scientists can define biologically relevant threshold of effect strength depending on their research question. We found two rarely used contrasts best fitted for detection of a possible change point: the Sequen and McDermott contrasts.

Recommendations for designing and analysing multi-arm non-inferiority trials: a review of methodology and current practice

Background and purpose

Multi-arm non-inferiority (MANI) trials, here defined as non-inferiority trials with multiple experimental treatment arms, can be useful in situations where several viable treatments exist for a disease area or for testing different dose schedules. To maintain the statistical integrity of such trials, issues regarding both design and analysis must be considered, from both the multi-arm and the non-inferiority perspectives. Little guidance currently exists on exactly how these aspects should be addressed and it is the aim of this paper to provide recommendations to aid the design of future MANI trials.

Methods

A comprehensive literature review covering four databases was conducted to identify publications associated with MANI trials. Literature was split into methodological and trial publications in order to investigate the required design and analysis considerations for MANI trials and whether they were being addressed in practice.

Results

A number of issues were identified that if not properly addressed, could lead to issues with the FWER, power or bias. These ranged from the structuring of trial hypotheses at the design stage to the consideration of potential heterogeneous treatment variances at the analysis stage. One key issue of interest was adjustment for multiple testing at the analysis stage. There was little consensus concerning whether more powerful p value adjustment methods were preferred to approximate adjusted CIs when presenting and interpreting the results of MANI trials.

We found 65 examples of previous MANI trials, of which 31 adjusted for multiple testing out of the 39 that were adjudged to require it. Trials generally preferred to utilise simple, well-known methods for study design and analysis and while some awareness was shown concerning FWER inflation and choice of power, many trials seemed not to consider the issues and did not provide sufficient definition of their chosen design and analysis approaches.

Conclusions

While MANI trials to date have shown some awareness of the issues raised within this paper, very few have satisfied the criteria of the outlined recommendations. Going forward, trials should consider the recommendations in this paper and ensure they clearly define and reason their choices of trial design and analysis techniques.

Inference for survival prediction under the regularized Cox model

When a moderate number of potential predictors are available and a survival model is fit with regularization to achieve variable selection, providing accurate inference on the predicted survival can be challenging. We investigate inference on the predicted survival estimated after fitting a Cox model under regularization guaranteeing the oracle property. We demonstrate that existing asymptotic formulas for the standard errors of the coefficients tend to underestimate the variability for some coefficients, while typical resampling such as the bootstrap tends to overestimate it; these approaches can both lead to inaccurate variance estimation for predicted survival functions. We propose a two-stage adaptation of a resampling approach that brings the estimated error in line with the truth. In stage 1, we estimate the coefficients in the observed data set and in [Formula: see text] resampled data sets, and allow the resampled coefficient estimates to vote on whether each coefficient should be 0. For those coefficients voted as zero, we set both the point and interval estimates to [Formula: see text] In stage 2, to make inference about coefficients not voted as zero in stage 1, we refit the penalized model in the observed data and in the [Formula: see text] resampled data sets with only variables corresponding to those coefficients. We demonstrate that ensemble voting-based point and interval estimators of the coefficients perform well in finite samples, and prove that the point estimator maintains the oracle property. We extend this approach to derive inference procedures for survival functions and demonstrate that our proposed interval estimation procedures substantially outperform estimators based on asymptotic inference or standard bootstrap. We further illustrate our proposed procedures to predict breast cancer survival in a gene expression study.

Multiple comparisons for survival data with propensity score adjustment

This article considers the practical problem in clinical and observational studies where multiple treatment or prognostic groups are compared and the observed survival data are subject to right censoring. Two possible formulations of multiple comparisons are suggested. Multiple Comparisons with a Control (MCC) compare every other group to a control group with respect to survival outcomes, for determining which groups are associated with lower risk than the control. Multiple Comparisons with the Best (MCB) compare each group to the truly minimum risk group and identify the groups that are either with the minimum risk or the practically minimum risk. To make a causal statement, potential confounding effects need to be adjusted in the comparisons. Propensity score based adjustment is popular in causal inference and can effectively reduce the confounding bias. Based on a propensity-score-stratified Cox proportional hazards model, the approaches of MCC test and MCB simultaneous confidence intervals for general linear models with normal error outcome are extended to survival outcome. This paper specifies the assumptions for causal inference on survival outcomes within a potential outcome framework, develops testing procedures for multiple comparisons and provides simultaneous confidence intervals. The proposed methods are applied to two real data sets from cancer studies for illustration, and a simulation study is also presented.

Recursive confidence band construction for an unknown distribution function

Given a sample X1,...,Xn of independent observations from an unknown continuous distribution function F, the problem of constructing a confidence band for F is considered, which is a fundamental problem in statistical inference. This confidence band provides simultaneous inferences on all quantiles and also on all of the cumulative probabilities of the distribution, and so they are among the most important inference procedures that address the issue of multiplicity. A fully nonparametric approach is taken where no assumptions are made about the distribution function F. Historical approaches to this problem, such as Kolmogorov's famous () procedure, represent some of the earliest inference methodologies that address the issue of multiplicity. This is because a confidence band at a given confidence level 1-α allows inferences on all of the quantiles of the distribution, and also on all of the cumulative probabilities, at that specified confidence level. In this paper it is shown how recursive methodologies can be employed to construct both one-sided and two-sided confidence bands of various types. The first approach operates by putting bounds on the cumulative probabilities at the data points, and a recursive integration approach is described. The second approach operates by providing bounds on certain specified quantiles of the distribution, and its implementation using recursive summations of multinomial probabilities is described. These recursive methodologies are illustrated with examples, and R code is available for their implementation.

Simultaneous confidence intervals for a success probability and intraclass correlation, with an application to screening mammography

This paper provides asymptotic simultaneous confidence intervals for a success probability and intraclass correlation of the beta-binomial model, based on the maximum likelihood estimator approach. The coverage probabilities of those intervals are evaluated. An application to screening mammography is presented as an example. The individual and simultaneous confidence intervals for sensitivity and specificity and the corresponding intraclass correlations are investigated. Two additional examples using influenza data and sex ratio data among sibships are also considered, where the individual and simultaneous confidence intervals are provided.