\n

\n

\n

\n

\n

\n

\n

Registration fees for this event are as follows:

\n- Members o
f PSI = Free of charge

\n- Non-Members of PSI = £\;20+VAT

To register for this event\, please **click here**. \
;

The webinar aims to provide information on th e purpose and methods for covariate adjustment in randomized clinical tria ls (RCTs). The speakers will review and compare different approaches for a djusting for covariates in terms of the properties of the statistics\, est imands\, convergence and handling of missing data. The FDA's revised guida nce on adjusting for covariates in RCTs and relevant commentary will be di scussed with the objective to provide practical examples for analysis plan ning in studies. The speakers will offer insights into issues such as coll apsibility\, marginal and conditional effects and provide examples on how to use covariates to describe effects.

\n**Biography**

**Abstract**

\n Jonathan Bartlett

\n
(*London School of Hygiene &\; Tropical Medicine*)

Jonathan Bartlett is a Professor in Medical Statistics at the London School of Hygiene &\; Tropical Medicine. His research interests are fo cused around missing data and causal inference methods\, and more recently \, how these can be applied to target different estimands in clinical tria ls. He has held previous positions at AstraZeneca and the University of Ba th\, and maintains a blog https://thes tatsgeek.com/

\n \n \n**An introduction to covariate
adjustment in trials.\n **In this introductory tal
k I will begin by describing the motivations for covariate adjustment in r
andomised trials. I will then describe the usual way a covariate adjusted
analysis is performed\, namely via fitting a suitable outcome regression m
odel\, highlighting the type of estimand such an approach targets and the
statistical assumptions it relies on. Next\, I will describe the standardi
sation or G-computation approach to covariate adjustment\, as described in
the FDA&rsquo\;s covariate adjustment guidance. I will contrast it with t
he usual approach in terms of efficiency\, robustness and in terms of the
estimand it targets.

*\n Florian Voss ( Boehringer Ingelheim*)

Florian Voss is Expert The rapeutic Area (TA) Statistician at Boehringer Ingelheim. He works in all p hases of clinical development and provides statistical consultation and ad vice for the statisticians working for TA Inflammation at BI. His is inter ested in applications of statistical methods in clinical development\, e.g . paediatric extrapolation\, covariate adjustment\, interim analyses and i nclusion of historical data.

\n \n \n**Summary of regulat
ory guidances (FDA\, EMA) on covariate adjustment and comments from EFSPI/
PSI Regulatory SIG.\n **In many clinical trials th
ere are known baseline factors that are prognostic or predictive for outco
mes in the trial. It is important to suitably account for such covariates
in the planning and statistical analysis of a clinical trial to improve se
nsitivity for treatment effects\, and to minimise the impact of any chance
baseline imbalances between arms.

Regulatory agencies generally support inclusion of covariates in the primary analysis model. The ICH E9 guidance recommends to identify covariates that have an importa nt influence on the primary endpoint and to account for them in the analys is. More detailed guidance is provided in the EMA &ldquo\;Guideline on adj ustment for baseline covariates in clinical trials&rdquo\; from 2015 and t he more recent FDA Guidance for the Industry &ldquo\;Adjusting for Covaria tes in Randomized Clinical Trials for Drugs and Biological Products&rdquo\ ; from 2021.

\nThis presentation will summarize the avai lable regulatory guidance and the comments provided by the EFSPI/PSI regul atory SIG. We will review the difference between conditional and unconditi onal treatment effects for linear and some non-linear models and discuss t hem in the context of ICH E9 (R1) &ldquo\;Estimands and Sensitivity Analys is in Clinical Trials&rdquo\;.

\n \n \n \n \n

\n Rhian Daniel

\n (*Cardiff University*)

Rhia n Daniel is Professor of Statistics at the Division of Population Medicine \, Cardiff University. Her research interests include several areas of cau sal inference methods and their application.

\n \n \n**Ma
rginal versus conditional effects and collapsibility.\n <
/strong>This presentation will start with the difference between marginal
and conditional estimands\, and settings in which one or the other may be
more pertinent to the clinical question of interest. The distinction betwe
en marginal and conditional estimands will be contrasted with that between
unadjusted and adjusted analyses. The phenomenon of non-collapsibility wi
ll then be discussed in detail\, including an intuitive explanation of wha
t it is and situations in which we should be mindful of it.**

\n Tim Morris

\n
(*MRC Clinical Trials Unit at UCL*)

Tim Morris i s a principal research fellow based at the MRC Clinical Trials Unit at UCL . He works on the development\, evaluation and understanding of statistica l methods. His interests include simulation studies\, handling missing dat a\, sensitivity analysis\, covariate adjustment\, estimands\, and IPD meta -analysis.

\n \n \n**Planning a covariate adjustment meth
od for your SAP.\n **It has long been advised to a
djust for covariates in the analysis of phase III clinical trials. The key
justifications are: 1) adjustment increases power\; 2) when randomisation
was stratified\, ignoring covariates in the analysis leads to incorrect e
stimates of uncertainty (such as confidence intervals that are too wide).<
/p>\n

In practice we need to define how we will adjust for c ovariates in a statistical analysis plan. The usual approach is direct adj ustment\, where an outcome regression model includes the covariates and th e treatment effect is estimated as a parameter of this model. Two alternat ive approaches that have received less attention from the trials community are standardisation and inverse-probability-of-treatment weighting (IPTW) .

\nThis presentation will compare the three broad appro aches in terms of planning which to write into a SAP. Beginning with the s ummary measure attribute of the estimand\, I will discuss points to consid er such as efficiency\, variance estimation\, convergence issues\, and han dling of missing data. No single approach is best for all situations but s tandardisation and IPTW deserve far more consideration than they currently receive.

\n \n \n \n \n

\n Seth Seegobin (*AstraZeneca*)

Seth Seegobin is the Head of Biostatistics\, Vaccine a nd Immune Therapies at AstraZeneca. His statistical interests include prog ramming efficient simulations\, interim analyses and the use of sampling e rror during trial design.

\n \n \n**Immunogenicity and p
rognostic effects.\n **Immunobridging trials aim t
o infer vaccine effectiveness (in the absence of efficacy data) by compari
ng the immune response of a vaccine candidate with an approved vaccine.

Within an RCT\, typically we rely upon randomisation to safeguard against treatment arm imbalance for important response prognosti c variables\, however the prioritisation of demographic groups within nati onal vaccine programmes can create several challenges for immunobridging t rials that follow.

\nThis presentation will summarize th e challenges and possible solutions to accurately describe and compare imm unogenicity between booster and primary series treatment arms\, imbalanced with respect to prognostic baseline characteristics\, using covariate adj ustment.

\n \n \n \n\n\;

END:VEVENT END:VCALENDAR