Current Practices in Choosing Estimands and Sensitivity Analyses in Clinical Trials: Results of the ICH E9 Survey.

BACKGROUND: An addendum to the International Conference on Harmonisation E9 (ICH E9) guidance document (Statistical Principles for Clinical Trials) is currently under development. The aim of the addendum is to promote harmonized standards on the choice of estimand (a well-defined measure of the treatment effect that is being estimated) in clinical trials and to describe a consensual framework for planning, conducting, and interpreting sensitivity analyses of clinical trial data. METHODS: In order to help understand current practices relating to the choice of estimands and sensitivity analyses for clinical trials, the ICH E9 working group developing the addendum conducted a survey with a primary focus on clinical trials involving drugs, vaccines, and biologics. The survey was distributed electronically between May 19, 2015, and June 11, 2015, to various stakeholder groups within ICH, including industry, regulatory, and academic communities. A total of 1305 respondents participated. RESULTS: Of the 1305 respondents 547 (42%), 344 (26%) and 283 (22%) were from Europe, USA and Japan respectively. Over half of the respondents work in pharmaceutical companies, and approximately a quarter of respondents noted oncology as the primary therapeutic area they work in. Over half of the respondents (595, 55%) noted the treatment effect being estimated was 'in the entire target population of patients regardless of whether they will take treatment as instructed'. The most common methods for handling missing data in primary analyses were mixed-models repeated measures (555, 56% respondents) and last observation carried forward (549, 55% respondents). The majority of respondents (816, 83%) noted they conducted sensitivity analyses to estimate treatment effects in different ways compared to the primary analysis by using alternative assumptions (627, 78%) and/or using alternative statistical methods (616, 76%). CONCLUSIONS: The survey results have provided useful information to the ICH E9 working group on current practices on the choice of primary estimands for measuring treatment effects in confirmatory clinical trials, and approaches used to select sensitivity analyses.

Rank-based principal stratum sensitivity analyses.

We describe rank-based approaches to assess principal stratification treatment effects in studies where the outcome of interest is only well-defined in a subgroup selected after randomization. Our methods are sensitivity analyses, in that estimands are identified by fixing a parameter and then we investigate the sensitivity of results by varying this parameter over a range of plausible values. We present three rank-based test statistics and compare their performance through simulations, and provide recommendations. We also study three different bootstrap approaches for determining levels of significance. Finally, we apply our methods to two studies: an HIV vaccine trial and a prostate cancer prevention trial.

Vaccine clinical trials--a statistical primer.

We discuss some of the key statistical issues that arise in all stages of vaccine development and, where necessary, contrast drug and vaccine clinical trials.

Statistical considerations for noninferiority/equivalence trials in vaccine development.

Noninferioritylequivalence designs are often used in vaccine clinical trials. The goal of these designs is to demonstrate that a new vaccine, or new formulation or regimen of an existing vaccine, is similar in terms of effectiveness to the existing vaccine, while offering such advantages as easier manufacturing, easier administration, lower cost, or improved safety profile. These noninferioritylequivalence designs are particularly useful in four common types of immunogenicity trials: vaccine bridging trials, combination vaccine trials, vaccine concomitant use trials, and vaccine consistency lot trials. In this paper, we give an overview of the key statistical issues and recent developments for noninferioritylequivalence vaccine trials. Specifically, we cover the following topics: (i) selection of study endpoints; (ii) formulation of the null and alternative hypotheses; (iii) determination of the noninferioritylequivalence margin; (iv) selection of efficient statistical methods for the statistical analysis of noninferioritylequivalence vaccine trials, with particular emphasis on adjustment for stratification factors and missing pre-or post-vaccination data; and (v) the calculation of sample size and power.

Relative risk estimation and inference using a generalized logrank statistic.

When comparing two survival distributions with proportional hazard functions, the logrank test is optimal for testing the null hypothesis that the constant hazard ratio (relative risk) is one. In this paper, we focus on (i) testing for departures from a relative risk other than one, and (ii) estimation of the relative risk. The standard tool to address both (i) and (ii) is the Cox proportional hazards model. However, the performance of the Cox model can be less than optimal with small samples. We show why this is the case, and propose a simple alternative method of estimation and inference based on a generalized logrank (GLR) statistic. While the GLR and Cox model approaches are asymptotically similar, empirical results reveal that the GLR approach is notably more efficient than the Cox model when the number of subjects is small (< 100 subjects per treatment group). An example based on survival times of cervical cancer patients is used to illustrate the proposed methodology.

Simultaneous testing strategy for comparing two treatments in a stratified binomial trial.

The Mantel-Haenszel (M-H) procedure is commonly used to compare two treatments in a stratified binomial trial. However, this procedure is asymptotically optimal only if the odds ratio is constant across strata. We propose an alternative analytic strategy based on the simultaneous use of two statistics, ZS and ZI, each involving a weighted averaging of within-stratum differences between proportions. The two treatments are declared significantly different at overall level alpha if either min(ZS, ZI) > Zalpha/2 or max(ZS, ZI) > Zalpha*/2, where alpha* is data dependent. Our strategy is shown to be more powerful than the M-H and other related procedures. Numerical examples are provided for illustration.

Minimum risk weights for comparing treatments in stratified binomial trials.

When comparing two treatments in a stratified trial with a binary endpoint, data are commonly analysed using a weighted averaging of the stratum-specific differences between proportions. Two popular sets of weights are the harmonic means of the stratum-specific sample sizes (SSIZE) and the reciprocals of the variances of the stratum-specific differences (INVAR). Either the SSIZE or INVAR weights are chosen and prespecified in the data analysis plan. We show that the 'wrong' choice between SSIZE and INVAR can result in a significantly inefficient analysis. To circumvent this potential problem, we propose a 'minimum risk' (MR) weighting strategy. The easy-to-compute MR weights are designed to yield more precise and less biased estimates of the overall treatment difference relative to the SSIZE and INVAR weights, respectively. We show, via a simulation study, that the proposed weights are an attractive compromise between the SSIZE and INVAR weights in terms of statistical power. Numerical examples are presented to illustrate the utility of the MR weights.

Effects of controlled-onset extended-release verapamil on nocturnal blood pressure (dippers versus nondippers). COER-Verapamil Study Group.

Approximately 1 in 4 patients with systemic hypertension have a 24-hour blood pressure (BP) profile characterized by a blunted or absent nocturnal decline in pressure. We evaluated the effects of a chronotherapeutic delivery system of controlled-onset extended-release (COER) verapamil hydrochloride and placebo in 257 hypertensive patients according to their circadian BP pattern in an 8-week prospective, multicenter, randomized, and double-blind clinical trial. Patients were stratified into 193 dippers (>10% decline in BP during the period of 10 P.M. to 5 A.M. compared with the hours of 5 A.M. to 10 P.M.) and 64 nondippers (<10% decline in BP during nighttime). During daytime, placebo-subtracted BP was similarly decreased in dippers and nondippers by COER verapamil. During nighttime, the placebo increased nocturnal BP in dippers (baseline nocturnal BP, 133/78 mm Hg) by 3/3 +/- 2/2 mm Hg and reduced BP by -5/-3 +/- 2/2 mm Hg in nondippers (baseline nocturnal BP, 152/94 mm Hg) (p = NS between groups). After controlling for age, gender, ethnicity, and the regression to the mean observed on placebo for all doses, COER verapamil reduced nocturnal BP more in nondippers than dippers -5.8/-2.4 mm Hg, p <0.0001 for systolic BP and p = 0.09 for diastolic BP). Additionally, a significant dose-related reduction in systolic and diastolic nocturnal BP (r = 0.56, p <0.0001 for systolic BP and r = 0.62, p <0.0001 for diastolic BP) was observed with COER verapamil after controlling for baseline covariates. These data demonstrate that nocturnal BP is decreased by a greater extent in nondipper hypertensives than in dipper hypertensives following treatment with COER verapamil HCL.

Non-iterative robust estimators of variance components in within-subject designs.

Classic estimators of variance components break down in the presence of outliers and perform less efficiently under non-normality. In this article I present simple non-iterative estimators of variance components that are resistant to outliers and robust to systematic departures from normality, such as heavy tailedness of the distribution of responses. The proposed estimators are based on a robust extension of Hocking's AVE approach and are thus called RAVE estimators. I present results from a Monte Carlo comparison of RAVE versus classic estimation methods including maximum likelihood (ML), restricted maximum likelihood (REML) and minimum variance quadratic unbiased estimation (MIVQUE). Under simulated deviations from normality, RAVE estimators are associated with smaller mean squared errors than all the comparators, and in the normal case they exhibit a minimal loss in relative efficiency. A numerical example illustrates the proposed methodology.

Robust elementwise estimation of a dispersion matrix.

Most of the existing robust methods for estimating a covariance or correlation matrix involve a multivariate approach in which matrix elements are estimated via simultaneous manipulation of all variables. These methods are generally based on complex iterative algorithms and hence are rather difficult to implement. The purpose of this article is to recommend an easy to implement noniterative robust method for estimating a dispersion matrix, based on an elementwise estimation approach. Simple expressions are provided for robust estimators of variances and covariances based, in part, on a modified A-estimator of scale discussed previously by Lax (1985, Journal of the American Statistical Association 80, 736-741). A Monte Carlo study is used to compare the performance of the proposed noniterative method with that of some iterative procedures studied in the literature. A numerical example involving robust estimation of variance components is presented as an application of the proposed methodology.

Polymeric delivery of the active isomer of misoprostol reduces systemic availability and uterotonic activity.

SC-30249 is the active isomer of misoprostol responsible for its mucosal protective effects against nonsteroidal anti-inflammatory drugs (NSAIDS). Linkage of SC-30249 to a polybutadiene polymer results in a delivery system (SC-55307) that releases the active component only under the acidic conditions of the stomach. This approach could be used to minimize side effects and systemic availability of synthetic prostaglandins. These studies were done to determine whether uterotonic activity could be recorded after treatment with SC-55307. Female beagles were implanted with uterine strain gauge force transducers, allowed 10 days for recovery and treated with estrogen to sensitize the uterus to the actions of prostaglandins. Base-line responses were determined with SC-30249, i.v., and then a randomized series of four treatments were given: SC-30249, IG, 10 micrograms/kg; SC-55307, IG, equivalent to 30 and 100 micrograms/kg of SC-30249; and a blank polymer control. HCI was given IG to provide an acid environment in the stomach, uterine responses were obtained for up to 4 h and plasma concentrations of SC-30249 free acid was determined. No uterotonic effect was seen after a low dose of SC-55307, whereas the high dose caused a brief but statistically significant increase equal to 8.8% and 17.8% of the responses to SC-30249, i.v. and IG, respectively. Peak plasma levels of SC-30249 free acid were 176.4 +/- 17.4 and 59.5 +/- 10.6 pg/ml after SC-30249, i.v. and IG, respectively, but were only 3.9 +/- 1.7 and 15.5 +/- 6.6 pg/ml after low and high doses of SC-55307, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Extended inhibition of platelet aggregation with the orally active platelet inhibitor SC-54684A.

BACKGROUND: Platelet aggregation is important in thrombotic events, and platelets play a major role in the etiology of several cardiovascular diseases. Platelet aggregation requires the binding of fibrinogen (fgn) to activated platelets. Synthetic peptides modeled after the RGD binding sequence on the fgn alpha-chain block the platelet glycoprotein (GP) IIb/IIIa receptor for fgn and effectively inhibit aggregation. SC-54684A (SCp, orally active prodrug of the active moiety SC-54701, SCa) is a mimetic of the RGD-containing peptide sequence that is recognized by the platelet GPIIb/IIIa receptor. SCa blocks the binding of fgn to the platelet and therefore prevents platelet aggregation in response to all agonists. METHODS AND RESULTS: SCp was administered orally at 1.25, 2.5, 5.0, and 7.5 mg/kg in a single-dose, dose-ranging study. Blood samples were taken periodically for 24 hours, and platelet-rich plasma was prepared and tested for inhibition of ex vivo collagen-induced platelet aggregation. The plasma concentration of active moiety was determined by bioassay. The time, inhibition, and concentration data were used to predict two doses that would result in minimum daily inhibition levels of 30% and 70% when administered twice daily (0.6 and 2.4 mg/kg, respectively). SCp was administered orally to conscious dogs twice daily for 14 days (after dose adjustment). Blood samples were obtained at daily peak and trough plasma levels (predicted from dose-ranging study). Inhibition of ex vivo collagen-induced platelet aggregation and concentration of active moiety in the plasma were determined. Average inhibition of platelet aggregation and plasma concentration of active moiety amounted to approximately 21% and 14 ng/mL at 1.5 mg/kg BID and 75% and 24 ng/mL at 2.4 ng/kg BID at daily minimum plasma levels (trough) in steady state. Platelet counts in the 2.4-mg/kg group declined from 3.2 x 10(5)/microL to 2.5 x 10(5)/microL in the first 9 days of dosing, with no further decline despite continued administration of compound. No changes were observed in the animals receiving 1.5 mg/kg. CONCLUSIONS: The results of the dose-ranging study show that oral administration of SCp results in dose-dependent inhibition of platelet aggregation. As shown in the 14-day administration, this dose-dependent inhibition can be maintained for an extended period while exhibiting no adverse effects. SCp is a leading candidate for development and is currently in clinical trials.

A new method for measurement of plasma concentration of orally active glycoprotein IIb/IIIa antagonists.

A bioassay for determining concentrations of antiplatelet compounds in plasma or aqueous solution has been developed. The method uses an aliquot of plasma from treated animals to inhibit collagen-induced platelet aggregation in pooled platelet-rich plasma (PRP) obtained from donor dogs. The concentration in plasma from treated animals was estimated using a standard curve of inhibition established using plasma from untreated animals which had been spiked with known amounts of compound. For independent validation, plasma concentrations of certain compounds were determined in identical dog plasma samples by both bioassay and HPLC. Results from the two methods were concordant. The bioassay provides an accurate and sensitive method for measuring antiplatelet activity without the need for extraction of plasma samples and may be used to measure activity in any solution which is compatible with PRP. This assay is routinely used to provide an estimate of absorption of prodrugs and systemic conversion to active compound after oral dosing. Some of the compounds of interest are ester-acid pairs with the inactive ester prodrug being cleaved to the active acid following administration. Compounds were administered orally (ester) or IV (acid) and blood samples were taken periodically for 24 hours. Plasma concentration of active moiety was determined for each time point and the area under the curve (AUC) of concentration vs. time was calculated. Comparing the AUCs for oral and IV routes of administration yielded the Oral Systemic Activity (OSA), a measure of active compound available after oral dosing.

Species variation in the effect of glycoprotein IIb/IIIa antagonists on inhibition of platelet aggregation.

Differences exist between platelets of different species in their reaction to pharmaceutical agents, such as inhibitors of platelet aggregation. Understanding these differences is critical in the interpretation of data from experimental animal models of thrombosis. Platelet aggregation, essential in the hemostatic process, requires that fibrinogen (fgn) bind to activated platelets. Analogs of Arginine-Glycine-Aspartic acid-Phenylalanine (RGDF), a peptide sequence of fgn, block fgn binding to its receptor known as glycoprotein (GP) IIb/IIIa on activated platelets and prevent aggregation. We studied the inhibition resulting from Arginine-Glycine-Aspartic acid-Serine (RGDS) and two analogs of RGDF, (SC-46749 and SC-47643) on aggregation of human, rat, guinea pig, dog, and rhesus monkey platelets in vitro using ADP as the agonist. The inhibitory potency of RGDS, SC-46749, ad SC-47643 was species dependent. The rank order of potency was rhesus monkey, dogs, and human followed by guinea pig and rat. In order to study the relative inactivity of the compounds in rat platelets compared to human, we diluted rat platelet-rich plasma (PRP) to yield platelet levels approximating that of humans. Platelet inhibition was not significantly changed in diluted rat PRP nor did changing concentration appear to affect activity in human PRP. Our data suggest that the platelet response of some species may better represent human response with regard to inhibition of GP IIb/IIIa by (RGDX) analogs.