TheStaphylococcus aureusNetwork Adaptive Platform Trial Protocol: New Tools for an Old Foe

Clinical Infectious Diseases, Journal Year: 2022, Volume and Issue: 75(11), P. 2027 - 2034

Published: June 19, 2022

Staphylococcus aureus bloodstream (SAB) infection is a common and severe infectious disease, with 90-day mortality of 15%-30%. Despite this, <3000 people have been randomized into clinical trials treatments for SAB infection. The limited evidence base partly results from infections being difficult to complete at scale using traditional trial methods. Here we provide the rationale framework an adaptive platform applied infections. We detail design features Network Adaptive Platform (SNAP) that will enable multiple questions be answered as efficiently possible. SNAP commenced enrolling patients across countries in 2022 estimated target sample size 7000 participants. This approach may serve exemplar increase efficiency other disease syndromes.

Language: Английский

---

Why and how should we simulate platform trials? Learnings from EU-PEARL

BMC Medical Research Methodology, Journal Year: 2025, Volume and Issue: 25(1)

Published: Jan. 17, 2025

Platform trials are innovative clinical governed by a master protocol that allows for the evaluation of multiple investigational treatments enter and leave trial over time. Interest in platform has been steadily increasing last decade. Due to their highly adaptive nature, provide sufficient flexibility customize important design aspects requirements both specific disease under investigation different stakeholders. The trials, however, comes with complexities when designing such trials. In past, we reviewed existing software simulating found none them were suitable as they do not accommodate features inherent staggered entry We argued simulation studies crucial efficient developed proposed an iterative, simulation-guided "vanilla sprinkles" framework, i.e. from basic more complex design, addressed functionality limitations well unavailability coding therein developing suite open-source use based on R programming language. To give some examples, newly supports throughout trial, choosing options control data sharing, specifying stopping rules platform-level operating characteristics. is available through licensing enable users access modify code. separate two these packages implement same independent teams obtained results. tools necessary capture complexity

Language: Английский

---

Transforming the evidence landscape in mental health with platform trials

Nature Mental Health, Journal Year: 2025, Volume and Issue: unknown

Published: March 10, 2025

Language: Английский

---

Current state-of-the-art and gaps in platform trials: 10 things you should know, insights from EU-PEARL

EClinicalMedicine, Journal Year: 2023, Volume and Issue: 67, P. 102384 - 102384

Published: Dec. 26, 2023

Platform trials bring the promise of making clinical research more efficient and patient centric. While their use has become widespread, including prominent role during COVID-19 pandemic response, broader adoption platform been limited by lack experience tools to navigate critical upfront planning required launch such collaborative studies. The European Union-Patient-cEntric clinicAl tRial pLatform (EU-PEARL) initiative produced new methodologies expand with an overarching infrastructure services embedded into Integrated Research Platforms (IRPs), in collaboration representatives through consultation U.S. Food Drug Administration Medicines Agency stakeholders. In this narrative review, we discuss outlook for Europe, challenges related infrastructure, design, adaptations, data sharing regulation. Documents derived from EU-PEARL project, alongside a literature search PubMed relevant grey (e.g., guidance regulatory agencies health technology agencies) were used as sources multi-stage process which 10 important points based on lessons drawn project developed summarised setup trials. We conclude that early involvement stakeholder or patients are steps implementation later acceptance Addressing these gaps will be attaining full potential patients.

Language: Английский

---

Developing generic templates to shape the future for conducting integrated research platform trials

Trials, Journal Year: 2024, Volume and Issue: 25(1)

Published: March 21, 2024

Interventional clinical studies conducted in the regulated drug research environment are designed using International Council for Harmonisation (ICH) regulatory guidance documents: ICH E6 (R2) Good Clinical Practice-scientific guideline, first published 2002 and last updated 2016. This document provides an international ethical scientific quality standard designing conducting trials that involve participation of human subjects. Recently, there has been heightened awareness importance integrated platform (IRPs) to evaluate multiple therapies simultaneously. The use a single master protocol as key source fulfill trial conduct obligations resulted re-examination templates used dynamic modern development challenges.

Language: Английский

---

On the use of non-concurrent controls in platform trials: a scoping review

Trials, Journal Year: 2023, Volume and Issue: 24(1)

Published: June 15, 2023

Abstract Background Platform trials gained popularity during the last few years as they increase flexibility compared to multi-arm by allowing new experimental arms entering when trial already started. Using a shared control group in platform increases efficiency separate trials. Because of later entry some treatment arms, includes concurrent and non-concurrent data. For given arm, controls refer patients allocated arm before enters trial, while that are randomised concurrently arm. can result bias estimate case time trends if appropriate methodology is not used assumptions met. Methods We conducted two reviews on use trials: one statistical regulatory guidance. broadened our searches external historical review 43 articles identified through systematic search PubMed performed guidance 37 guidelines published EMA FDA websites. Results Only 7/43 methodological 4/37 focused With respect methodology, 28/43 articles, Bayesian approach was incorporate external/non-concurrent frequentist 8/43 considered both. The majority method downweights favour data (34/43), using for instance meta-analytic or propensity score approaches, 11/43 modelling-based approach, regression models In guidelines, critical but deemed acceptable rare diseases 12/37 accepted specific indications (12/37). Non-comparability (30/37) (16/37) were raised most often general concerns with controls. Indication found be instructive. Conclusions Statistical methods incorporating available literature, either means originally proposed incorporation mainly differ how combined temporary changes handled. Regulatory currently still limited.

Language: Английский

---

Statistical considerations for the platform trial in COVID-19 vaccine priming and boosting

Trials, Journal Year: 2024, Volume and Issue: 25(1)

Published: July 26, 2024

The Platform trial In COVID-19 priming and BOOsting (PICOBOO) is a multi-site, adaptive platform designed to generate evidence of the immunogenicity, reactogenicity, cross-protection different booster vaccination strategies against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) its variants, specific for Australian context. PICOBOO randomises participants receive one three vaccine brands (Pfizer, Moderna, Novavax) available use in Australia, where brand subtypes vary over time according national roll out strategy, employs Bayesian hierarchical modelling approach efficiently borrow information across consecutive doses, age groups subtypes. Here, we briefly describe structure report statistical considerations estimands, models decision making adaptations. This paper should be read conjunction with Core Protocol Sub-Study 1: Booster Vaccination. was registered on 10 February 2022 New Zealand Clinical Trials Registry ACTRN12622000238774.

Language: Английский

---

Covariate-adjusted response-adaptive designs for semiparametric survival models

Statistical Methods in Medical Research, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 25, 2024

Covariate-adjusted response adaptive (CARA) designs are effective in increasing the expected number of patients receiving superior treatment an ongoing clinical trial, given a patient's covariate profile. There has recently been extensive research on CARA with parametric distributional assumptions patient responses. However, range applications for such becomes limited real trials. Sverdlov et al. have pointed out that irrespective specific form survival outcomes, their proposed based exponential model provide valid statistical inference, provided final analysis is performed using appropriate accelerated failure time (AFT) model. In trials, however, planned primary rarely conducted AFT The developed obviating any about responses, relying only proportional hazards assumption between two arms. To meet multiple experimental objectives optimal allocation approach. covariate-adjusted doubly biased coin design and efficient-randomized used to randomize achieve derived targets expectation. These functions Cox regression coefficients estimated sequentially arrival every new into trial. merits assessed simulation studies operating characteristics then implemented re-design real-life confirmatory

Language: Английский

---

Examining the bias-efficiency tradeoff from incorporation of nonconcurrent controls in platform trials: A simulation study example from the adaptive COVID-19 treatment trial

Clinical Trials, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 8, 2025

Background: Platform trials typically feature a shared control arm and multiple experimental treatment arms. Staggered entry exit of arms splits the group into two cohorts: those randomized during same period in which was open (concurrent controls) outside that (nonconcurrent controls). Combining these groups may offer increased statistical power but can lead to bias if analyses do not account for time trends response variable. Proposed methods adjustment increase type I error rates when impact unequally or large, sudden changes rate occur. However, there has been limited exploration degree inflation one plausibly expect real-world scenarios. Methods: We use data from Adaptive COVID-19 Treatment Trial (ACTT) mimic realistic platform trial with remdesivir arm. compare four strategies estimating effect interferon beta-1a (the ACTT-3 arm) relative (data ACTT-1, ACTT-2, ACTT-3) on recovery death by day 29: utilizing concurrent controls only prespecified analysis), pooling all without “unadjusted-pooled” adjusting as categorical variable, Bayesian hierarchical model implementation adjusts using smoothing techniques “Bayesian machine”). efficiency each method simulation settings based observed ACTT data. Results: The unadjusted-pooled approach provided substantially different estimates compared concurrent-only model-based approaches, indicating over were ignorable across stages ACTT. approaches rely an assumption constant effects control; more than doubled where this satisfied. Relative analysis moderate. Conclusions: In key assumptions met, potential gains incorporation nonconcurrent outweighed risk substantial inflation. This leads us advise against primary confirmatory clinical trials, aligning current FDA guidance advising comparisons settings. be useful other settings, we recommend performing reference assessing drive results.

Language: Английский

---

Identification and Estimation of Causal Effects Using Non‐Concurrent Controls in Platform Trials

Statistics in Medicine, Journal Year: 2025, Volume and Issue: 44(6)

Published: March 15, 2025

ABSTRACT Platform trials are multi‐arm designs that simultaneously evaluate multiple treatments for a single disease within the same overall trial structure. Unlike traditional randomized controlled trials, they allow treatment arms to enter and exit at distinct times while maintaining control arm throughout. This comprises both concurrent controls, where participants concurrently either or arm, non‐concurrent who when under study is unavailable. While flexible, platform introduce challenge of using raising questions about estimating effects. Specifically, which estimands should be targeted? Under what assumptions can these identified estimated? Are there any efficiency gains? In this article, we discuss issues related identification estimation common choices estimand. We conclude most robust strategy increase without imposing unwarranted target average effect (cATE), ATE among only units, covariate‐adjusted doubly estimator. Our studies suggest that, purpose obtaining gains, collecting important prognostic variables more than relying on controls. also perils targeting due an untestable extrapolation assumption will often invalid. provide simulations illustrating our points application ACTT trial, resulting in 20% improvement precision compared naive estimator ignores controls variables.

Language: Английский

---