Improved group sequential Holm procedures for testing multiple correlated hypotheses over time.

Clinical trials can typically feature two different types of multiple inference: testing of more than one null hypothesis and testing at multiple time points. These modes of multiplicity are closely related mathematically but distinct statistically and philosophically. Regulatory agencies require strong control of the family-wise error rate (FWER), the risk of falsely rejecting any null hypothesis at any analysis. The correlations between test statistics at interim analyses and the final analysis are therefore routinely used in group sequential designs to achieve less conservative critical values. However, the same type of correlations between different comparisons, endpoints or sub-populations are less commonly used. As a result, FWER is in practice often controlled conservatively for commonly applied procedures.Repeated testing of the same null hypothesis may give changing results, when the hypothesis is rejected at an interim but accepted at the final analysis. The mathematically correct overall rejection is at odds with an inference theoretic approach and with common sense. We discuss these two issues, of incorporating correlations and how to interpret time-changing conclusions, and provide case studies where power can be increased while adhering to sound statistical principles.

Dapagliflozin in heart failure with preserved and mildly reduced ejection fraction: rationale and design of the DELIVER trial.

AIMS: Sodium-glucose co-transporter 2 (SGLT2) inhibitors, originally developed as glucose-lowering agents, have been shown to reduce heart failure hospitalizations in patients with type 2 diabetes without established heart failure, and in patients with heart failure with and without diabetes. Their role in patients with heart failure with preserved and mildly reduced ejection fraction remains unknown. METHODS: Dapagliflozin Evaluation to Improve the LIVEs of Patients With PReserved Ejection Fraction Heart Failure (DELIVER) is an international, multicentre, parallel group, event-driven, randomized, double-blind trial in patients with chronic heart failure and left ventricular ejection fraction (LVEF) >40%, comparing the effect of dapagliflozin 10 mg once daily, vs. placebo, in addition to standard of care. Patients with or without diabetes, with signs and symptoms of heart failure, a LVEF >40%, elevation in natriuretic peptides and evidence of structural heart disease are eligible. The primary endpoint is time-to-first cardiovascular death or worsening heart failure event (heart failure hospitalization or urgent heart failure visit), and will be assessed in dual primary analyses - the full population and in those with LVEF <60%. The study is event-driven and will target 1117 primary events. A total of 6263 patients have been randomized. CONCLUSIONS: DELIVER will determine the efficacy and safety of the SGLT2 inhibitor dapagliflozin, added to conventional therapy, in patients with heart failure and preserved and mildly reduced ejection fraction.

Statistical and practical considerations in designing of immuno-oncology trials.

The novel mechanism of action of immunotherapy agents, in treatment of various types of cancer, poses unique challenges during the designing of clinical trials. It is important to account for possibility of a delayed treatment effect and adjust sample size accordingly. This paper provides an analytical approach for computing sample size in the presence of a delayed effect using a piece-wise proportional hazards model. Failing to account for an anticipated treatment delay may result in considerable loss in power. The overall hazard ratio (HR), which now represents the average HR across the entire treatment period, can remain a meaningful measure of average benefit to patients in the trial. We show that, special consideration needs to be given for the designing of interim analyses related to futility, so as not to increase the probability of incorrectly stopping an effective agent. It is shown that the weighted log-rank test, using the Fleming-Harrington class of weights, can be used as supportive analysis to better reflect the impact of a delayed effect and possible long-term benefit in a subset of the overall population.

Prevalence of Viremic hepatitis C, hepatitis B, and HIV infection, and vaccination status among prisoners in Stockholm County.

BACKGROUND: Identification and knowledge of settings with high prevalence of hepatitis C virus (HCV) infection is important when aiming for elimination of HCV. The primary aim of this study was to estimate the prevalence of viremic HCV infection among Swedish prisoners. Secondary aims were to estimate the prevalence of hepatitis B surface antigen (HBsAg), human immunodeficiency virus (HIV), and the proportion who have received hepatitis B virus (HBV) vaccination. METHODS: A cross-sectional study of all incarcerated persons (n = 667) at all prisons (n = 9) in Stockholm County was conducted. All prisoners are routinely offered opt-in screening for HCV antibodies (anti-HCV), HCV RNA, HBsAg, anti-HBs, anti-HBc and HIV Ag/Ab at prison in Sweden. Data on the results of these tests and the number of received HBV vaccine doses were collected from the prison medical records. The parameters of HCV RNA, anti-HCV, and occurrence of testing for HCV were analysed in multiple logistic regression models in relation to age, sex and prison security class. RESULTS: The median age was 35 (IQR 26-44) years, and 93.4% were men. Seventy-one percent (n = 471) had been tested for anti-HCV, 70% (n = 465) for HBsAg and 71% (n = 471) for HIV. The prevalence of anti-HCV, HCV RNA, HBsAg and HIV Ag/Ab was 17.0, 11.5, 1.9, and 0.2%, respectively among tested persons. The proportion of prisoners who had received full HBV vaccination was 40.6% (n = 271) among all study subjects. CONCLUSIONS: The prevalence of viremic HCV infection among Swedish prisoners in Stockholm County was 11.5%, which is high in comparison to the general population. Therefore, when aiming for the WHO goal of HCV elimination, prisons could suit as a platform for identification and treatment of HCV infection. There is a need to increase testing for blood-borne viruses and to improve vaccination coverage against HBV in Swedish prisons.

Nonproportional Hazards for Time-to-Event Outcomes in Clinical Trials: JACC Review Topic of the Week.

Most major clinical trials in cardiology report time-to-event outcomes using the Cox proportional hazards model so that a treatment effect is estimated as the hazard ratio between groups, accompanied by its 95% confidence interval and a log-rank p value. But nonproportionality of hazards (non-PH) over time occurs quite often, making alternative analysis strategies appropriate. This review presents real examples of cardiology trials with different types of non-PH: an early treatment effect, a late treatment effect, and a diminishing treatment effect. In such scenarios, the relative merits of a Cox model, an accelerated failure time model, a milestone analysis, and restricted mean survival time are examined. Some post hoc analyses for exploring any specific pattern of non-PH are also presented. Recommendations are made, particularly regarding how to handle non-PH in pre-defined Statistical Analysis Plans, trial publications, and regulatory submissions.

Exposure vs. response of blood pressure in patients with rheumatoid arthritis following treatment with fostamatinib.

Fostamatinib is an oral spleen tyrosine kinase (SYK) inhibitor which has been evaluated as a potential treatment for rheumatoid arthritis (RA). Treatment with fostamatinib has been associated with an increase in blood pressure (BP). In this work, we present a pooled analysis of the pharmacokinetic-pharmacodynamic (PKPD) relationship for BP, based on 3 Phase III studies, aiming to increase the knowledge about fostamatinib's effect on BP in the RA population. Fostamatinib is rapidly and extensively converted to R406 after oral administration of fostamatinib, and the PK of R406 could be described by a two-compartment population PK model with first order absorption, with an estimated CL/F of 18.7 L/h. Average steady-state concentrations, predicted based on the individual CL/F estimates, were subsequently used in the PKPD analysis. The population PKPD analysis revealed a concentration dependent increase of BP with increasing R406 concentrations, where a power model and an Emax model best described the increase in SBP and DBP, respectively. The predicted increases were +5.2 mmHg for SBP and +4.2 mmHg for DBP, for a 100 mg bid dose. The impact of covariates on the PKPD relationship was investigated but covariates did only explain a minor part of the overall high variability in BP.

Optimal group-sequential designs for simultaneous testing of superiority and non-inferiority.

Confirmatory clinical trials comparing the efficacy of a new treatment with an active control typically aim at demonstrating either superiority or non-inferiority. In the latter case, the objective is to show that the experimental treatment is not worse than the active control by more than a pre-specified non-inferiority margin. We consider two classes of group-sequential designs that combine the superiority and non-inferiority objectives: non-adaptive designs with fixed group sizes and adaptive designs where future group sizes may be based on the observed treatment effect. For both classes, we derive group-sequential designs meeting error probability constraints that have the lowest possible expected sample size averaged over a set of values of the treatment effect. These optimized designs provide an efficient means of reducing expected sample size under a range of treatment effects, even when the separate objectives of proving superiority and non-inferiority would require quite different fixed sample sizes. We also present error spending versions of group-sequential designs that are easily implementable and can handle unpredictable group sizes or information levels. We find the adaptive choice of group sizes to yield some modest efficiency gains; alternatively, expected sample size may be reduced by adding another interim analysis to a non-adaptive group-sequential design.