Handling Partially Observed Trial Data After Treatment Withdrawal: Introducing Retrieved Dropout Reference-Base Centred Multiple Imputation.

The ICH E9(R1) Addendum (International Council for Harmonization 2019) suggests treatment-policy as one of several strategies for addressing intercurrent events such as treatment withdrawal when defining an estimand. This strategy requires the monitoring of patients and collection of primary outcome data following termination of randomised treatment. However, when patients withdraw from a study early before completion this creates true missing data complicating the analysis. One possible way forward uses multiple imputation to replace the missing data based on a model for outcome on- and off-treatment prior to study withdrawal, often referred to as retrieved dropout multiple imputation. This article introduces a novel approach to parameterising this imputation model so that those parameters which may be difficult to estimate have mildly informative Bayesian priors applied during the imputation stage. A core reference-based model is combined with a retrieved dropout compliance model, using both on- and off-treatment data, to form an extended model for the purposes of imputation. This alleviates the problem of specifying a complex set of analysis rules to accommodate situations where parameters which influence the estimated value are not estimable, or are poorly estimated leading to unrealistically large standard errors in the resulting analysis. We refer to this new approach as retrieved dropout reference-base centred multiple imputation.

Comments on 'standard and reference-based conditional mean imputation': Regulators and trial statisticians be aware!

Accurate frequentist performance of a method is desirable in confirmatory clinical trials, but is not sufficient on its own to justify the use of a missing data method. Reference-based conditional mean imputation, with variance estimation justified solely by its frequentist performance, has the surprising and undesirable property that the estimated variance becomes smaller the greater the number of missing observations; as explained under jump-to-reference it effectively forces the true treatment effect to be exactly zero for patients with missing data.

Treatment policy estimands for recurrent event data using data collected after cessation of randomised treatment.

In the past, many clinical trials have withdrawn subjects from the study when they prematurely stopped their randomised treatment and have therefore only collected 'on-treatment' data. Thus, analyses addressing a treatment policy estimand have been restricted to imputing missing data under assumptions drawn from these data only. Many confirmatory trials are now continuing to collect data from subjects in a study even after they have prematurely discontinued study treatment as this event is irrelevant for the purposes of a treatment policy estimand. However, despite efforts to keep subjects in a trial, some will still choose to withdraw. Recent publications for sensitivity analyses of recurrent event data have focused on the reference-based imputation methods commonly applied to continuous outcomes, where imputation for the missing data for one treatment arm is based on the observed outcomes in another arm. However, the existence of data from subjects who have prematurely discontinued treatment but remained in the study has now raised the opportunity to use this 'off-treatment' data to impute the missing data for subjects who withdraw, potentially allowing more plausible assumptions for the missing post-study-withdrawal data than reference-based approaches. In this paper, we introduce a new imputation method for recurrent event data in which the missing post-study-withdrawal event rate for a particular subject is assumed to reflect that observed from subjects during the off-treatment period. The method is illustrated in a trial in chronic obstructive pulmonary disease (COPD) where the primary endpoint was the rate of exacerbations, analysed using a negative binomial model.

Missing data sensitivity analysis for recurrent event data using controlled imputation.

Statistical analyses of recurrent event data have typically been based on the missing at random assumption. One implication of this is that, if data are collected only when patients are on their randomized treatment, the resulting de jure estimator of treatment effect corresponds to the situation in which the patients adhere to this regime throughout the study. For confirmatory analysis of clinical trials, sensitivity analyses are required to investigate alternative de facto estimands that depart from this assumption. Recent publications have described the use of multiple imputation methods based on pattern mixture models for continuous outcomes, where imputation for the missing data for one treatment arm (e.g. the active arm) is based on the statistical behaviour of outcomes in another arm (e.g. the placebo arm). This has been referred to as controlled imputation or reference-based imputation. In this paper, we use the negative multinomial distribution to apply this approach to analyses of recurrent events and other similar outcomes. The methods are illustrated by a trial in severe asthma where the primary endpoint was rate of exacerbations and the primary analysis was based on the negative binomial model.

Platelet aggregation unchanged by lipoprotein-associated phospholipase A2 inhibition: results from an in vitro study and two randomized phase I trials.

BACKGROUND: We explored the theorized upregulation of platelet-activating factor (PAF)- mediated biologic responses following lipoprotein-associated phospholipase A2 (Lp-PLA2) inhibition using human platelet aggregation studies in an in vitro experiment and in 2 clinical trials. METHODS AND RESULTS: Full platelet aggregation concentration response curves were generated in vitro to several platelet agonists in human plasma samples pretreated with rilapladib (selective Lp-PLA2 inhibitor) or vehicle. This was followed by a randomized, double-blind crossover study in healthy adult men (n = 26) employing a single-agonist dose assay of platelet aggregation, after treatment of subjects with 250 mg oral rilapladib or placebo once daily for 14 days. This study was followed by a second randomized, double-blind parallel-group trial in healthy adult men (n = 58) also treated with 250 mg oral rilapladib or placebo once daily for 14 days using a full range of 10 collagen concentrations (0-10 µg/ml) for characterizing EC50 values for platelet aggregation for each subject. Both clinical studies were conducted at the GlaxoSmithKline Medicines Research Unit in the Prince of Wales Hospital, Sydney, Australia. EC50 values derived from multiple agonist concentrations were compared and no pro-aggregant signals were observed during exposure to rilapladib in any of these platelet studies, despite Lp-PLA2 inhibition exceeding 90%. An increase in collagen-mediated aggregation was observed 3 weeks post drug termination in the crossover study (15.4% vs baseline; 95% confidence interval [CI], 3.9-27.0), which was not observed during the treatment phase and was not observed in the parallel-group study employing a more robust EC50 examination. CONCLUSIONS: Lp-PLA2 inhibition does not enhance platelet aggregation. TRIAL REGISTRATION: 1) Study 1: ClinicalTrials.gov NCT01745458 2) Study 2: ClinicalTrials.gov NCT00387257.

Analysis of longitudinal trials with protocol deviation: a framework for relevant, accessible assumptions, and inference via multiple imputation.

Protocol deviations, for example, due to early withdrawal and noncompliance, are unavoidable in clinical trials. Such deviations often result in missing data. Additional assumptions are then needed for the analysis, and these cannot be definitively verified from the data at hand. Thus, as recognized by recent regulatory guidelines and reports, clarity about these assumptions and their implications is vital for both the primary analysis and framing relevant sensitivity analysis. This article focuses on clinical trials with longitudinal quantitative outcome data. For the target population, we define two estimands, the de jure estimand, "does the treatment work under the best case scenario," and the de facto estimand, "what would be the effect seen in practice." We then carefully define the concept of a deviation from the protocol relevant to the estimand, or for short a deviation. Each patient's postrandomization data can then be divided into predeviation data and postdeviation data. We set out an accessible framework for contextually appropriate assumptions relevant to de facto and de jure estimands, that is, assumptions about the joint distribution of pre- and postdeviation data relevant to the clinical question at hand. We then show how, under these assumptions, multiple imputation provides a practical approach to estimation and inference. We illustrate with data from a longitudinal clinical trial in patients with chronic asthma.

The use of baseline covariates in crossover studies.

It is our experience that in many settings, crossover trials that have within-period baseline measurements are analyzed wrongly. A "conventional" analysis of covariance in this setting uses each baseline as a covariate for the following outcome variable in the same period but not for any other outcome. If used with random subject effects such an analysis leads to biased treatment comparisons; this is an example of cross-level bias. Using a postulated covariance structure that reflects the symmetry of the crossover setting, we quantify such bias and, at the same time, investigate potential gains and losses in efficiency through the use of the baselines. We then describe alternative methods of analysis that avoid the cross-level bias. The development is illustrated throughout with 2 example trials, one balanced and orthogonal and one highly unbalanced and nonorthogonal.

Practical usage of O'Brien's OLS and GLS statistics in clinical trials.

Multivariate techniques of O'Brien's OLS and GLS statistics are discussed in the context of their application in clinical trials. We introduce the concept of an operational effect size and illustrate its use to evaluate power. An extension describing how to handle covariates and missing data is developed in the context of Mixed models. This extension allowing adjustment for covariates is easily programmed in any statistical package including SAS. Monte Carlo simulation is used for a number of different sample sizes to compare the actual size and power of the tests based on O'Brien's OLS and GLS statistics.