Incomplete datasets due to missing data is an issue th at has been\, and will be\, around for a long time. At this meeting we wil l present the evolution of missing data approaches\, looking at how they h ave been handled in the past\, the current established missing data approa ches and the impact of the new ICH E9 R1 addendum on the handling of missi ng data\, focussing in particular on the treatment policy estimand.
\n< h4>Registration\nYou can now register for this event. Registration
fees are as follows:
\n Members of PSI = £\;20+VAT
\n No
nMembers of PSI = £\;115+VAT*
\n*Please note: NonMember ra
te includes membership for the rest of the 2021 calendar year.
\
nTo register for the \;session\, please \;click here.
\n
DAY 1 &ndash\; Tuesday 4^{th} May 2021 \n  
\n Speaker \n  \n Biography \n  \n Abstract \n 
\n
 \n Khadija Rantell (nee Rerhou) is a Senior Statist ical Assessor at the Medicines and Healthcare Products Regulatory Agency ( MHRA) where she has worked since 2013. \nKhadija is also a member of the EFSPI (European Federation for Statisticians in the Pharm aceutical Industry) Scientific Committee and a member of the EFPIA/EFSPI E stimands Implementation and Pharmaceutical Industry Biostatistics Working Groups. \n  \n History of missing data in regulatory s ettings. \nThe problem of handling missing data has evolved greatly over the last decade along with statistical methodolo gies for dealing with missing data. Several recent documents have been dev eloped that lay out a set of general principles and techniques for address ing the problems raised by missing data in clinical trials. However\, pros pective prevention of missing data occurrence\, through carefully designin g and implementing of a research study\, remains the single best approach. This session will focus on the evolution of regulatory framework for hand ling missing data in confirmatory trials and the impact of the ICH E9 (R1) addendum on the handling of missing data. \n 
\n
 \n Dr. Jiawei Wei joined Novart is in 2011\, where she is currently a Director Biostatistician in the Adva nced Methodology and Data Science group. She is interested in supporting t he methodological development in various areas of pharmaceutical statistic s\, including estimand\, recurrent event data\, multiple testing\, etc. Be fore joining Novartis\, Jiawei got her PhD in statistics from Texas A&\ ;M University in 2010\, and then one year assistant professor. Jiawei is a warded leading scientist at Novartis\, she is associate editor of Statisti cs in Biopharmaceutical Research\, and also a part time advisor at Fudan U niversity in China. \n  \n On the role of hypothetical estimand in clinical trials and its estimation. \nThe ICH E9(R1) Addendum on 'Estimands and Sensitivity Analysis in Clin ical Trials' introduced various strategies for addressing intercurrent eve nts when defining the clinical question of interest. For hypothetical esti mand strategies\, a scenario is envisaged in which the intercurrent event would not occur. The value of the variable that reflects the clinical ques tion of interest is the value which the variable would have taken in the h ypothetical scenario. If a hypothetical strategy is proposed\, it should b e made clear what hypothetical scenario is envisaged. The ICH E9(R1) Adden dum acknowledges that a wide variety of hypothetical scenarios can be envi saged\, but it also clarifies that some scenarios are likely to be of more clinical or regulatory interest than others. \nIn this talk we will not only discuss the role of hypothetical estimand in clinic al trials\, but also the estimation of hypothetical estimand. One basic co nsideration is that the estimation of hypothetical estimand requires the p rediction of hypothetical trajectories for all patient who have the interc urrent event. Importantly\, the assumptions for the predictions need to be aligned with the hypothetical strategy\, which begs the questions: From w here or from whom do we borrow information to predict the hypothetical mea surements of interest? Do we have sufficient data or information to borrow from? Moreover\, the prediction uncertainty needs to be adequately accoun ted for\, e.g. through multiple predictions. And finally\, sensitivity ana lyses need to be performed to investigate the robustness of our conclusion s. All the above considerations will be discussed and illustrated by a cas e study. \n 
\n

\n Bohdana is a Principal Statistician at Bayer. She has ove r 15 years of experience in biopharmaceutical industry including statistic al methodology research and consulting\, with numerous publications in mis sing data\, estimands\, subgroup identification\, and machine learning. Sh e is a coauthor of two books\, &ldquo\;Estimands\, Estimators and Sensiti vity Analysis in Clinical Trials&rdquo\; (2020) and &ldquo\;Clinical Trial s with Missing Data: a Guide for Practitioners&rdquo\; (2014). Bohdana is a member of the DIA SWG on Missing data and EFSPI Subgroup Special Interes t Group. \n  \n Statistical Issues and Recommendations for Clinical Trials Conducted During the COVID19 Pandemic. \ nThe COVID19 pandemic continues to impact planned and ongo ing clinical trials. Any study that is conducted in part or in whole durin g the pandemic must anticipate\, assess\, and mitigate these impacts to as sure safety of participants and address operational issues. From a statist ical perspective\, the study teams need to ensure the ability to conduct t he trial and collect data in alignment with the study objectives. Statisti cal analysis plans should incorporate strategies to deal with potential in creases or distinct patterns of treatment discontinuations\, interruptions \, protocol deviations\, and missing data. In this presentation\, we will discuss potential impacts of COVID19 pandemic disruptions on clinical tri als with a focus on statistical aspects and mitigation strategies. It is b eneficial to do so in a structured and systematic way through the estimand framework\, regardless of whether the estimand is formally defined in the trial protocol. We will also discuss issues of missing data handling and supplemental analyses that may be needed for the trial. \n < /td>\n 
\n Day 2 &ndash\; Wednesday 5^{th} May 2021 \n  
\n
 \n David became the Head of Statistical Innovation at Astra Zeneca in September 2016. David leads a team of expert statistical methodo logists who advise colleagues within AstraZeneca on novel trial design and analysis issues. Between 1999 and 2016 David worked for the Medicines and Healthcare products Regulatory Agency (MHRA) as a Statistical Assessor an d was Chair of the Biostatistics Working Party from 20112016. David was h eavily involved in the revision to the CHMP missing data guideline. He is currently involved in how the ICH E9 Addendum on Estimands impacts the des ign\, analysis and reporting of clinical trials within AstraZeneca and is also a member of the EFPIA/EFSPI Estimand Implementation Working group. \n 
\n Aligning how subjects with missing data due to st udy discontinuation are handled in the primary analysis with the primary e stimands. \nICH E9 R1 explains the distinction between treatment discontinuation and study withdrawal. The former is an i ntercurrent event whilst the latter gives rise to missing data to be addre ssed in the statistical analysis. However\, E9 R1 also states that &ldquo\ ;methods to address the problem presented by missing data can be selected to align with the estimand&rdquo\;. In this presentation I will stress tha t in fact the chosen methods should be selected to align with the estimand and failure to do this would lead to an incoherent analysis strategy. The treatment policy strategy will be considered and it will be highlighted t hat there are a number of methods of estimation available to address missi ng data. Which methods are fully aligned with the treatment policy estiman d will be discussed. \n 
\n
 \n James originally trained as a chemis t\, receiving his PhD from the University of Cambridge in 2009. Following a short spell in industry in computational drug design\, he completed his MSc in statistics at UCL in 2013. Starting at Boehringer Ingelheim\, he sp ent three years as trial statistician before joining its newlyformed stat istical methodology group. James now works as a consultant methodology sta tistician for the CRO Elderbrook Solutions GmbH\, providing services to a major pharmaceutical company. James is an active member of crossindustry WGs in his two main areas of research\; estimands and missing data handlin g. Other topics of particular interest include event prediction and the pr ocess of how clinical trials are designed. \n  \n The Practicalities of Treatment Policy Estimation. \n < p>Treatment policy estimands were introduced by ICH E9(R1) as an approach to include the effects of intercurrent events within the treatment effect of interest. They have been commonly thought of as both equivalent to ITT analysis and robust to estimate. However\, it has become increasingly clea r that these properties start to break down in the presence of missing dat a &ndash\; almost an inevitability in real clinical trials. Indeed\, due t o missing data typically being associated with discontinuation of randomis ed treatment\, treatment policy estimation will usually deviate from tradi tional ITT approaches and in doing so\, situations can easily arise where no reliable estimation is possible.\nThis talk will des cribe the practical difficulties of treatment policy estimation that arise due to missing data and discuss potential solutions. Topics covered will include trial conduct\, approaches to estimation\, assumptions\, how much data is needed and sample size calculations. \n 
\n
 \n Michael O&rsquo\;Kelly has worked as a s tatistician in the pharmaceutical industry for 27 years. He has been invol ved in many areas of biostatistics\, and has a special interest in statist ical modelling and simulation. He coauthored a Best Practice proposal for projects involving Modelling and Simulation\, which was adopted by the PS I board in 2017. With colleagues Michael O&rsquo\;Kelly has developed new methods for missing data that are now widely used in clinical trials. His book authored with Bohdana Ratitch\, &ldquo\;Clinical trials with missing data: a guide for practitioners&rdquo\;\, was published in 2014 by Wiley. He has given courses on missing data for PSI and at many conferences and f or many pharmaceutical companies. For his work in best practice and in mis sing data\, Michael received the RSS/PSI award for Excellence in Pharmaceu tical Statistics in 2017\; he is Senior Director with IQVIA&rsquo\;s Centr e for Statistics in Drug Development. \n  \n Even a &l dquo\;treatment policy&rdquo\; estimand may have missing data: how can we take account of this? \nUnder the &ldquo\;treat ment policy&rdquo\; estimand\, outcomes contribute to the estimate of effe ct irrespective of treatment actually taken after randomisation\, and all subjects are to be followed up for the full scheduled followup period. It is hoped that under this estimand missing data would be limited. However\ , in practice\, subjects may avail of their right to discontinue completel y from a trial. Even under a &ldquo\;treatment policy&rdquo\; estimand\, t his leaves the analyst with missing data. How is the analyst to take accou nt of this residual missing data under the &ldquo\;treatment policy&rdquo\ ; estimand? The new ICH E9 R1 Addendum suggests &ldquo\;for subjects who d iscontinue treatment without further data being collected\, a model may us e data from other subjects who discontinued treatment but for whom data co llection has continued&rdquo\;. However\, numbers of appropriate subjects with data available may be small\, and compromises or assumptions may be r equired to do the kind of modelling suggested by ICH E9 R1. This presentat ion evaluates some of the options that use statistical modelling to take a ccount of this residual missing data. \n 
\n
 \n Dan worked as a statistician at GS K since 2016. During this time he has worked on RCTs investigating the eff ect of various treatments in severe asthma\, COPD and nasal polyps\, and i s now working in trials assessing therapies for COVID19. Dan started his career as a medical statistician at the Medical Research Council Clinical Trials Unit (MRC CTU) working on RCTs in respiratory diseases and dermatol ogy. He then completed a PhD at the same unit through UCL investigating st atistical issues in the design of multiarm multistage clinical trials wh ich are increasingly being used in practice to accelerate the evaluation o f new therapies. Prior to starting at GSK\, Dan worked as a statistician i n the Department of Pulmonology at University Hospital Zurich for one year \, focusing mainly on network metaanalyses of RCTs investigating treatmen ts for sleep apnoea. \n  \n Treatment policy estimands for recurrent event data using data collected after cessation of randomis ed treatment. \nTrials now increasingly collect data from subjects following premature discontinuation of study treatment \, as this event is irrelevant for the purposes of a treatment policy esti mand. However\, despite efforts to keep subjects in a trial\, some will st ill choose to withdraw. Publications for sensitivity analyses of recurrent event data have often focused on reference‐based imputation methods\, mor e commonly applied to continuous outcomes\, where imputation for the missi ng data for one treatment arm is based on the observed outcomes in another arm. The existence of data following premature discontinuation of treatme nt now raises the opportunity to impute missing data for subjects who with draw from study using this observed &lsquo\;offtreatment&rsquo\; data\, p otentially allowing more plausible assumptions for the missing post‐study‐ withdrawal data than other reference‐based approaches. In this poster\, we describe a recent imputation method (1) for recurrent event data in which the missing post‐study‐withdrawal event rate for a subject is assumed to reflect the rate observed from subjects during the off‐treatment period. T he method is illustrated in a trial in chronic obstructive pulmonary disea se (COPD\; GSK funded NCT02105961) where the primary endpoint was the rate of exacerbations\, analysed using a negative binomial model. \n 