Point estimation, confidence intervals, and P-values for optimal adaptive two-stage designs with normal endpoints.

Due to the dependency structure in the sampling process, adaptive trial designs create challenges in point and interval estimation and in the calculation of P-values. Optimal adaptive designs, which are designs where the parameters governing the adaptivity are chosen to maximize some performance criterion, suffer from the same problem. Various analysis methods which are able to handle this dependency structure have already been developed. In this work, we aim to give a comprehensive summary of these methods and show how they can be applied to the class of designs with planned adaptivity, of which optimal adaptive designs are an important member. The defining feature of these kinds of designs is that the adaptive elements are completely prespecified. This allows for explicit descriptions of the calculations involved, which makes it possible to evaluate different methods in a fast and accurate manner. We will explain how to do so, and present an extensive comparison of the performance characteristics of various estimators between an optimal adaptive design and its group-sequential counterpart.

Extension of a conditional performance score for sample size recalculation rules to the setting of binary endpoints.

BACKGROUND: Sample size calculation is a central aspect in planning of clinical trials. The sample size is calculated based on parameter assumptions, like the treatment effect and the endpoint's variance. A fundamental problem of this approach is that the true distribution parameters are not known before the trial. Hence, sample size calculation always contains a certain degree of uncertainty, leading to the risk of underpowering or oversizing a trial. One way to cope with this uncertainty are adaptive designs. Adaptive designs allow to adjust the sample size during an interim analysis. There is a large number of such recalculation rules to choose from. To guide the choice of a suitable adaptive design with sample size recalculation, previous literature suggests a conditional performance score for studies with a normally distributed endpoint. However, binary endpoints are also frequently applied in clinical trials and the application of the conditional performance score to binary endpoints is not yet investigated. METHODS: We extend the theory of the conditional performance score to binary endpoints by suggesting a related one-dimensional score parametrization. We moreover perform a simulation study to evaluate the operational characteristics and to illustrate application. RESULTS: We find that the score definition can be extended without modification to the case of binary endpoints. We represent the score results by a single distribution parameter, and therefore derive a single effect measure, which contains the difference in proportions [Formula: see text] between the intervention and the control group, as well as the endpoint proportion [Formula: see text] in the control group. CONCLUSIONS: This research extends the theory of the conditional performance score to binary endpoints and demonstrates its application in practice.

Optimization of the two-stage group sequential three-arm gold-standard design for non-inferiority trials.

If design parameters are chosen appropriately, group sequential trial designs are known to be able to reduce the expected sample size under the alternative hypothesis compared to single-stage designs. The same holds true for the so-called 'gold-standard' design for non-inferiority trials, a design involving an experimental group, an active control group, and a placebo group. However, choosing design parameters that maximize the advantages of a two-stage approach for the three-arm gold-standard design for non-inferiority trials is not a straightforward task. In particular, optimal choices of futility boundaries for this design have not been thoroughly discussed in existing literature. We present a variation of the hierarchical testing procedure, which allows for the incorporation of binding futility boundaries at interim analyses. We show that this procedure maintains strong control of the family-wise type I error rate. Within this framework, we consider the futility and efficacy boundaries as well as the sample size allocation ratios as optimization parameters. This allows the investigation of the efficiency gain from including the option to stop for futility in addition to the ability to stop for efficacy. To analyze the extended designs, optimality criteria that include the design's performance under the alternative as well as the null hypothesis are introduced. On top of this, we discuss methods to limit the allocation of placebo patients in the trial while maintaining relatively good operating characteristics. The results of our numerical optimization procedure are discussed and a comparison of different approaches to designing a three-arm gold-standard non-inferiority trial is provided.

Clinical pre-test probability for obstructive coronary artery disease: insights from the European DISCHARGE pilot study.

OBJECTIVES: To test the accuracy of clinical pre-test probability (PTP) for prediction of obstructive coronary artery disease (CAD) in a pan-European setting. METHODS: Patients with suspected CAD and stable chest pain who were clinically referred for invasive coronary angiography (ICA) or computed tomography (CT) were included by clinical sites participating in the pilot study of the European multi-centre DISCHARGE trial. PTP of CAD was determined using the Diamond-Forrester (D+F) prediction model initially introduced in 1979 and the updated D+F model from 2011. Obstructive coronary artery disease (CAD) was defined by one at least 50% diameter coronary stenosis by both CT and ICA. RESULTS: In total, 1440 patients (654 female, 786 male) were included at 25 clinical sites from May 2014 until July 2017. Of these patients, 725 underwent CT, while 715 underwent ICA. Both prediction models overestimated the prevalence of obstructive CAD (31.7%, 456 of 1440 patients, PTP: initial D+F 58.9% (28.1-90.6%), updated D+F 47.3% (34.2-59.9%), both p < 0.001), but overestimation of disease prevalence was higher for the initial D+F (p < 0.001). The discriminative ability was higher for the updated D+F 2011 (AUC of 0.73 95% confidence interval [CI] 0.70-0.76 versus AUC of 0.70 CI 0.67-0.73 for the initial D+F; p < 0.001; odds ratio (or) 1.55 CI 1.29-1.86, net reclassification index 0.11 CI 0.05-0.16, p < 0.001). CONCLUSIONS: Clinical PTP calculation using the initial and updated D+F prediction models relevantly overestimates the actual prevalence of obstructive CAD in patients with stable chest pain clinically referred for ICA and CT suggesting that further refinements to improve clinical decision-making are needed. TRIAL REGISTRATION: https://www.clinicaltrials.gov/ct2/show/NCT02400229 KEY POINTS: • Clinical pre-test probability calculation using the initial and updated D+F model overestimates the prevalence of obstructive CAD identified by ICA and CT. • Overestimation of disease prevalence is higher for the initial D+F compared with the updated D+F. • Diagnostic accuracy of PTP assessment varies strongly between different clinical sites throughout Europe.

Effect of iterative reconstruction and temporal averaging on contour sharpness in dynamic myocardial CT perfusion: Sub-analysis of the prospective 4D CT perfusion pilot study.

PURPOSE: Myocardial computed tomography perfusion (CTP) allows the assessment of the functional relevance of coronary artery stenosis. This study investigates to what extent the contour sharpness of sequences acquired by dynamic myocardial CTP is influenced by the following noise reduction methods: temporal averaging and adaptive iterative dose reduction 3D (AIDR 3D). MATERIALS AND METHODS: Dynamic myocardial CT perfusion was conducted in 29 patients at a dose level of 9.5±2.0 mSv and was reconstructed with both filtered back projection (FBP) and strong levels of AIDR 3D. Temporal averaging to reduce noise was performed as a post-processing step by combining two, three, four, six and eight original consecutive 3D datasets. We evaluated the contour sharpness at four distinct edges of the left-ventricular myocardium based on two different approaches: the distance between 25% and 75% of the maximal grey value (d) and the slope in the contour (m). RESULTS: Iterative reconstruction reduced contour sharpness: both measures of contour sharpness performed better for FBP than for AIDR 3D (d = 1.7±0.4 mm versus 2.0±0.5 mm, p>0.059 at all edges; m = 255.9±123.9 HU/mm versus 160.6±123.5 HU/mm; p<0.023 for all edges). Increasing levels of temporal averaging degraded contour sharpness. When FBP reconstruction was applied, contour sharpness was best without temporal averaging (d = 1.7±0.4 mm, m = 255.9±123.9 HU/mm) and poorest for the strongest levels of temporal averaging (d = 2.1±0.3 mm, m = 142.2±104.9 HU/mm; comparison between lowest and highest temporal averaging level: for d p>0.052 at all edges and for m p<0.001 at all edges). CONCLUSION: The use of both temporal averaging and iterative reconstruction degrades objective contour sharpness parameters of dynamic myocardial CTP. Thus, further advances in image processing are needed to optimise contour sharpness of 4D myocardial CTP.