BEGIN:VCALENDAR VERSION:2.0 METHOD:PUBLISH PRODID:-//Telerik Inc.//Sitefinity CMS 13.3//EN BEGIN:VEVENT DESCRIPTION:Venue: \; Novartis Campus\, Basel\, Switzerland\n\n\nPlease click here for a flyer on the event\n\nThis exciting one-day workshop wil l cover a wide range of statistical aspects relating to event-driven trial s. We have assembled a group of very knowledgeable speakers\, who will sha re their thoughts\, ideas and experiences\, including case studies\, on a range of particular issues relating to planning\, conduct and analysis of survival and recurrent event trials. The first half of the day will be ded icated to time-to-event endpoints and adverse events with the afternoon fo cusing on recurrent event endpoints that are associated with a terminal ev ent. \nSpeaker slides from the event have been added to the agenda below.\ n\n\nAgenda\n\n \n \n Time\n Agenda \; \n \n \n 08:30 - 09:00\n \n \n Registration\, Welcome and introduction\n \n \n \n 09:00 - 12:00\n \n Analy sis of time-to-event data and safety events\n \n Val entine Jehl (Novartis) \;on quantitative assessment of adverse events in clinical trials - comparison of methods at an interim and the final ana lysis.\n SPEAKER SLIDES \;Quantitative assessment of advers e events in clinical trials: comparison of methods at interim and final an alysis. V Jehl\, in collaboration with N Hollaender\, J Gonzalez-Maffe Bas el October 29\, 2019\n \n Qing Wang (Roche) \;&n bsp\;on comparison of time-to-first event and recurrent event methods in m ultiple sclerosis trials.\n SPEAKER SLIDES \;Comparison of Time-To-First-Event &\; Recurrent Event Methods in Multiple Sclerosis T rials \;\n \n Filip De Ridder (Janssen) \;on a Time to event model for early efficacy signal dose finding in epilepsy clinical trials.\n SPEAKER SLIDES \;A Time‐To‐Event Model f or Early Efficacy Signal Dose‐Finding in Epilepsy Clinical Trials PSI One Day Meeting ‐ 29 Oct 2019 Filip De Ridder\, Roy Twyman\, Marc Ceusters and Giacomo Salvadore\n \n Andrew Thomson (EMA) on  \;Estimators and Estimands for safety events in time-to-event studies: a r egulatory perspective.\n SPEAKER SLIDES \;Estimators and Es timands for safety events in time-to-event studies: a regulatory perspecti ve. Andrew Thomson\n \n \n \n \n 12:00 - 13:00\n \n \n Lunch break\n \n \n \n \n 13:00 - 16:30\ n \n Recurrent events with associated terminal event s\n \n Patrick Schlö\;mer \;&\; Arno Frit sch (Bayer) \;on the topic of estimands and estimators for recurrent e vents with an associated terminal event.\n SPEAKER SLIDES \ ;Estimands for recurrent events in the presence of a terminal event &ndash \; Considerations and simulations for chronic heart failure trials. Arno F ritsch Patrick Schlö\;mer\n \n \n Joh n Gregson (London School of Hygiene &\; Tropical Medicine) \;on the topic of practical experience of modelling repeat events in the REDUCE-IT and COAPT trials.\n \n Tobias Bluhmki (University o f Ulm) \;on the topic of simulating recurrent events with associated t erminal events.\n SPEAKER SLIDES \;Resampling complex time- to-event data without individual patient data\, with a view toward recurre nt events. Tobias Bluhmki\n \n Rob Hemmings (Consili um): Rejoinder\n \n \n \n\n\n\n\nAbstracts\n\n\n\n\n \n \n  \; \;\n \n Valen tine Jehl\n (Novartis)\n \n \n Quantitative assessment of adverse events in clinical trials &ndash\; com parison of methods at an interim and the final analysis.\n \n Abstract\n In clinical study reports\, adverse events (AEs) are commonly summarized using the incidence proportion despite cumu lative incidence function been advocated as the most appropriate method to account for different exposure time and competing events.\n In this presentation\, we compare different methods to estimate the probabil ity of one selected AE. Besides considering the final analysis at the time of the Clinical Study Report\, we especially investigate the capability o f the proposed methods to provide a reasonable estimate of the AE probabil ity at an early interim analysis. Robustness of the methods in the presenc e of a competing event is evaluated using data from a breast cancer study. The potential bias of each method is quantified in a simulation study.\n \n Biography\n Valentine Jehl is a senior quantitative safety scientist at Novartis. She received her Master&rsquo\ ;s degree in applied mathematics at the Louis Pasteur University in Strasb ourg.\n She started her carrier as statistician with a CRO in B russel. She then joined Novartis in Basel\, where she supported major subm issions and development programs for the oncology franchise. After 9 years in this role\, Valentine joined the quantitative safety group in April 20 16\, where she now promotes the use of quantitative methods for safety\, w ith a particular focus on Adverse Drug Reactions.\n \n \n \n \n \n \n \n Qing Wang\n (Roche)\n \n \n \n \n Comparison of time-to-first event and recur rent event methods in multiple sclerosis trials.\n \n Abstract\n Randomized clinical trials in multiple sclerosis ( MS) frequently use the time to the first confirmed disability progression (CDP) on the Expanded Disability Status Scale (EDSS) as an endpoint. Howev er\, especially in progressive forms of MS where CDP is typically the prim ary endpoint\, a substantial proportion of subjects may experience repeate d disability events. Recurrent event analyses could therefore increase stu dy power and improve clinical interpretation of results.\n We p resent results from two simulation studies which compare analyses of the t ime to the first event with recurrent event analyses (including negative b inomial\, Andersen-Gill\, and Lin\, Wei\, Ying\, and Yang models). The fir st simulation study is generic and recurrent events data is simulated acco rding to a mixed non-homogeneous Poisson process.  \;The second simula tion study is MS-specific: we first simulate longitudinal measurements of the ordinal EDSS scale using a multi-state model and then derive recurrent event data based on this. \; Simulation parameters are chosen to mimi c typical MS trial populations in relapsing-remitting or primary progressi ve MS\, respectively\, and include scenarios with heterogeneity (frailties ). Based on the results from the simulation studies\, the presentation wil l conclude with recommendations for the choice of the endpoint\, and analy sis method of MS trials with disability progression endpoints.\n \n Biography\n \n Qing is a statisticia n working at Roche Basel. She is currently the project lead statistician f or the Ocrevus (ocrelizumab) program\, and had been supporting the program from initial study readouts\, filing preparations\, US and EU approvals\, to market access and scientific communication over the past years. Before joining Roche in 2014 she has worked in HIV research at the Institute for Clinical Epidemiology and Biostatistics at University Hospital Basel. She received her Master in Mathematics and PhD in Biostatistics at the Univer sity of Cambridge (UK).\n \n \n \n \n \n Filip De Ridder\n (Janssen)\n \n A time to event model for early efficacy signal dose find ing in epilepsy clinical trials.\n \n Abstract\n Time to-event endpoints have been proposed as alternatives to esta blish the effect of anti-epileptic drugs in clinical trials. These endpoin ts may reduce exposure to placebo or ineffective treatments\, thereby faci litating trial recruitment and improving safety. Time to baseline seizure count is defined as the number of days until a subject experienced a numbe r of seizures equal to the baseline seizure count. \; A post hoc analy sis of completed Phase III trails with perampanel showed that an analysis of the time to baseline count endpoint is consistent with the classical en dpoints (median % seizure rate reduction\, percentage of patients achievin g a 50% or greater reduction in seizure frequency)1. \n We inve stigated the performance of the time to baseline seizure endpoint by (1) a post hoc analysis of topiramate and carisbamate clinical trial data and ( 2) clinical trial simulation using a longitudinal model for daily seizures counts. This model included key features of daily seizure count data\, su ch as a large between subject variability in baseline seizure rate and dru g response\, a large variability of the number of seizures per day and clu stering of seizures over time.\n The re-analysis of topiramate and carisbamate clinical trial data confirmed the relationship between the median time to baseline seizure count and the classical endpoint of media n % seizure rate reduction that was observed with perampanel. In addition\ , the observed relationship agreed with the one that was predicted by the simulation model.\n Clinical trial simulations were used to inv estigate the performance of a proof-of-concept study design using the time to baseline seizure count endpoint. The study consisted of a 4-week prosp ective baseline\, followed by a 4-week double blind treatment period\, aft er which subjects would exit the study if they had reached or exceeded the ir baseline seizure count\, or would continue for another 8-weeks. These s imulations showed that (1) with relatively small sample sizes (~ 20/arm) t he design is able to identify clinical relevant treatment effects (30% - 5 0% seizure rate reduction)\; (2) a 4-week baseline period provides enough information on the baseline seizure count and (3) the length of exposure o f subjects to placebo or an inactive treatment is strongly reduced as comp ared to a classical design. \;\n \n Biography\n \n Filip De Ridder is a Senior Scientific Director i n the Statistical Modeling &\; Methodology group of Janssen R&\;D. T wenty years ago\, he was one of founders of the Modeling &\; Simulation group at Janssen bringing together statisticians and pharmacometricians t o apply modeling &\; simulation techniques in clinical drug development . \; Since then he has worked on M&\;S projects in the context of P K/PD modeling\, dose response modeling and clinical trial design\, mainly in neuroscience and infectious diseases.\n \n \n \n  \; \;\n \n \n Andr ew Thomson\n (EMA)\n \n  \;\n Abstract\n The treatment of recurrent safety events and ter minal events\n requires careful consideration underlying the es timands in question\, and the\n assumptions in the methods used to estimate them. In this talk I shall give a\n regulatory per spective on these issues\, focussing on how and why the EU system\n summarises data as it does\, where the gaps are in the methodology\, and how we\n can progress to ensure that data are summarised ap propriately. I will consider\n whether we need to move beyond t he methods currently used\, and what questions\n we truly need to be answering (and how). In particular I shall argue that we\n need to be sure that when no true raised risk exists\, the method we use to\n summarise said risk should provide an unbiased average ef fect of 0\, but in\n time-to-event studies this is not always a s quite straightforward as it seems.\n  \;\n Bio graphy.\n Andrew Thomson is a statistician at the EMA Office of \n Biostatistics and Methodology Support\, joining in 2014. He supports the\n methodological aspects of the assessments of Mar keting Authorisation\n Applications\, as well as Scientific Adv ice\, and methodological aspects of\n Paediatric Investigationa l Plans. He has worked extensively on the\n methodological aspe cts of the EMA Reflection Paper on the use of extrapolation\n o f efficacy in paediatric studies. \n  \;\n Prior to the EMA\, he worked at the UK regulator\, the\n Medicines a nd Healthcare product Regulatory Agency. Here he worked initially as\n a statistical assessor in the Licensing Division\, assessing Marke ting\n Application Authorisations and providing Scientific Advi ce to companies. After\n rising to Senior Statistical Assessor\ , he moved to the Vigilance and Risk\n Management of Medicines Division\, to be Head of Epidemiology. Here he managed a\n team of statisticians\, epidemiologists and data analysts providing support to \n the assessment of post-licensing observational studies and m eta-analyses. He\n also managed the team&rsquo\;s design\, cond uct and analysis of epidemiology studies\,\n using the UK Clini cal Practice Research \n \n \n \n \n  \;\n \n Arno Fritsch &\; Patrick S chlö\;mer (Bayer)\n \n Estimands for recurrent e vents in the presence of a terminal event &ndash\; Considerations and simu lations for chronic heart failure trials.\n \n Abstr act\n \n \n In this presentation\, we wil l discuss potential estimands according to the ICH E9 addendum framework t hat can be addressed for recurrent events when there is a non-negligible r isk for a terminal event\, typically death. \n As an applicatio n\, we consider trials in chronic heart failure (HF). Here in the past\, t he standard (composite) primary endpoint was the time to either hospitaliz ation for HF or cardiovascular (CV) death. Since many patients experience recurrent HF hospitalizations\, there is interest to include these events into the primary endpoint. We consider two estimands\, one that focuses on ly on the total number of recurrent HF hospitalizations and another one th at includes CV death as an additional composite event. \n We pr esent results of an extensive simulation study that investigated which sta ndard methods for analyzing recurrent event data estimate the above-mentio ned estimands. In addition\, we compared the efficiency of recurrent event estimands and time-to-first event estimands. \n Biography\n Arno Fritsch received his PhD in Statistics from the University o f Dortmund\, Germany\, in 2010. Since then he has been working at Bayer as a clinical statistician\, mainly on the design\, analysis and submission of cardiovascular trials. Since 2017 he has the position as Group Leader E urope in the cardiovascular statistics department. His methodological inte rests include handling of missing data\, analysis of subgroups and recurre nt events. He is one of the co-authors of the application for an EMA quali fication opinion on use of recurrent events.\n Patrick Schl&oum l\;mer received his PhD in Statistics from the University of Bremen\, Germ any\, in 2014 for his work on group sequential and adaptive designs for th ree-arm non-inferiority trials. Since then he has been working at Bayer as a clinical statistician in the cardio-renal area with increasing responsi bilities\, now holding the position Lead Statistician. His methodological interests include group sequential and adaptive designs\, multiple compari son procedures and recurrent events. He is one of the co-authors of the ap plication for an EMA qualification opinion on use of recurrent events.\n \n \n \n \n \n Jo hn Gregson\n (London School of Hygiene &\; Tropical Medicine )\n \n \n \n \n The value of including recurrent events in the analysis of cardiovascular out comes trials.\n \n Abstract\n \n Including recurrent events in analyses of clinical trials can increase power and lead to a more complete assessment of treatment benefit. There a re several strategies to analysing repeat events\, but little practical gu idance as to which are best in any given scenario. Several methods for ana lyses of repeat events in trials will be compared\, including Andersen-Gil l\, Wei-Lin-Weissfeld\, negative binomial regression\, and joint frailty m odels. The assumptions underlying each of these methods\, and their variou s advantages and disadvantages will be outlined using data from recent lar ge cardiovascular trials. \n Biography\n John Gregso n is an Assistant Professor in Medical Statistics at the London School of Hygeine and Trpoical Medicine. He has a range of experience in the analysi s of cardiovascular clinical trials\, many of which have been published in high impact journals (e.g. NEJM\, Lancet\, JACC).  \;As well as an in terest in the applied analysis of randomised clinical trials and epidemiol ogical studies\, a major research interest of his is in methodological res earch into statistical issues which commonly arise in such studies. He hol ds a PhD in Epidemiology from Cambridge University and a Masters in Medica l Statistics from Southampton University.\n \n \n \n  \;\n Tobias Bluhmki (University of Ulm)\n \n \n Resampling complex time-to-event dat a without individual patient data\, with a view toward recurrent events.\n \n Abstract\n \n In this talk we consider non- and semi-parametric resampling of multistate event histo ries by simulating individual trajectories from an empirical multivariate hazard measure. \; \;\n \n One advantage is that it does not necessarily require individual patient data\, but may be based on published information. This is also attractive for both study pla nning and simulating realistic real‐world event history data in general.&n bsp\; A special focus is on simulating recurrent events data with associat ed terminal events. We demonstrate that our proposal gives a more natural interpretation of how such data evolve over the course of timethan many of the competing approaches. The multistate perspective avoids any latent fa ilure time structure and sampling spaces impossible in real life\, whereas its parsimony follows the principle of Occam's razor. We also suggest emp irical simulation as a novel bootstrap procedure to assess estimation unce rtainty in the absence of individual patient data. This is not possible fo r established procedures such as Efron's bootstrap. \n Biograph y\n \n \n Tobias Bluhmki studied Mathemat ical Biometry at Ulm University from 2009 to 2014 and was honored with the "Bernd-Streitberg Award" by the International Biometric Society - German Region for his Master's Thesis. Since then\, he has been research assistan t at the Institute of Statistics\, Ulm University\, Germany. He has recent ly defended his PhD thesis supervised by Jan Beyersmann at the Faculty of Mathematics and Economics and is now postdoctoral researcher. His research focuses on statistical methodology in clinical trials and epidemiological studies based on survival and event history techniques.\n \n He has published several articles in biostatistical\, epidemiolo gical and medical journals and is the current co-lead of the "Team of Youn g Statisticians" of the International Biometric Society - German Region.\n \n \n \n \n Rob Hemmings ( Consilium)\n \n Biography\n \n I am a partner at Consilium.  \;Consilium is my consultancy partnersh ip with Tomas Salmonson\, a long-standing member of the EMA&rsquo\;s CHMP and formerly the chair of that committee. \; Tomas and I support compa nies in the development\, authorisation and life-cycle management of medic ines.\n Previously I worked at AstraZeneca and for 19 years at the Medicines and Healthcare products Regulatory Agency\, heading the grou p of medical statisticians and pharmacokineticists. \; I am a statisti cian by background and whilst working at MHRA I was co-opted as a member o f EMA&rsquo\;s CHMP for expertise in medical statistics and epidemiology.& nbsp\; At CHMP I was Rapporteur for multiple products and was widely engag ed across both scientific and policy aspects of the committee&rsquo\;s wor k.  \;I was fortunate to chair the CHMP&rsquo\;s Scientific Advice Wor king Party for 8 years and have also chaired their expert groups on Biosta tistics\, Modelling and Simulation and Extrapolation. \; I wrote or co -wrote multiple regulatory guidance documents\, including those related to estimands\, subgroups\, use of conditional marketing authorisation\, deve lopment of fixed-dose combinations\, extrapolation and adaptive designs. & nbsp\;I have a particular interest in when and how to use data generated i n clinical practice to support drug development.\n \n \n \n\n\n\n\n\n\n\n\n \n \n Registration\n \n \n  \;PSI Member\n  \;£\;40 + V AT\n \n \n  \;Non-Member\n  \; £\;135 \;+ VAT \;(This includes PSI membership for 2019)\n \n \n\n\n\n\n\n \; DTEND;VALUE=DATE:20191030 DTSTAMP:20240328T182441Z DTSTART;VALUE=DATE:20191029 LOCATION: SEQUENCE:0 SUMMARY:PSI One Day Meeting: Time-to-event and Recurrent Event Endpoints in Clinical Trials UID:RFCALITEM638472470815103522 X-ALT-DESC;FMTTYPE=text/html:
Please c
lick here for a flyer on the event
\n
This exciting one-day works hop will cover a wide range of statistical aspects relating to event-drive n trials. We have assembled a group of very knowledgeable speakers\, who w ill share their thoughts\, ideas and experiences\, including case studies\ , on a range of particular issues relating to planning\, conduct and analy sis of survival and recurrent event trials. The first half of the day will be dedicated to time-to-event endpoints and adverse events with the after noon focusing on recurrent event endpoints that are associated with a term inal event.
\nSpeaker slides from the event have been added to the agenda below.
strong>
\n
\n
\nAgenda
Time | \nAgenda \; | \n \n
08:30 - 09:
00 \n | \n \n Registration\, Welcome and introduction \n | \n
09:00 - 12:00 | \n\n <
strong>Analysis of time-to-event data and safety events | \n
12:00 - 13:00 \n | \n
\n Lunch break | \n
13:00 - 16:30 | \n\n
Recurrent events
with associated terminal events | \n
Valentine Jehl
\n (Novartis)
Quantitative assessment of advers
e events in clinical trials &ndash\; comparison of methods at an interim a
nd the final analysis.
\n
\n Abstract
In cli nical study reports\, adverse events (AEs) are commonly summarized using t he incidence proportion despite cumulative incidence function been advocat ed as the most appropriate method to account for different exposure time a nd competing events.
\nIn this presentation\, we compare different methods to estimate t
he probability of one selected AE. Besides considering the final analysis
at the time of the Clinical Study Report\, we especially investigate the c
apability of the proposed methods to provide a reasonable estimate of the
AE probability at an early interim analysis. Robustness of the methods in
the presence of a competing event is evaluated using data from a breast ca
ncer study. The potential bias of each method is quantified in a simulatio
n study.
\n
\n Biography
Valentine J ehl is a senior quantitative safety scientist at Novartis. She received he r Master&rsquo\;s degree in applied mathematics at the Louis Pasteur Unive rsity in Strasbourg.
\nShe started her carrier as statistician with a CRO in Brussel. Sh e then joined Novartis in Basel\, where she supported major submissions an d development programs for the oncology franchise. After 9 years in this r ole\, Valentine joined the quantitative safety group in April 2016\, where she now promotes the use of quantitative methods for safety\, with a part icular focus on Adverse Drug Reactions.
\n \n \n
\n
Qing Wang
\n (Roche)
Comparison of time-to-first event and recurrent
event methods in multiple sclerosis trials.
\n
\n
Abstract
Randomized clinical trials in multiple sclerosis (MS) freq uently use the time to the first confirmed disability progression (CDP) on the Expanded Disability Status Scale (EDSS) as an endpoint. However\, esp ecially in progressive forms of MS where CDP is typically the primary endp oint\, a substantial proportion of subjects may experience repeated disabi lity events. Recurrent event analyses could therefore increase study power and improve clinical interpretation of results.
\n < p>We present results from two simulatio n studies which compare analyses of the time to the first event with recur rent event analyses (including negative binomial\, Andersen-Gill\, and Lin \, Wei\, Ying\, and Yang models). The first simulation study is generic an d recurrent events data is simulated according to a mixed non-homogeneous Poisson process.  \;The second simulation study is MS-specific: we fir st simulate longitudinal measurements of the ordinal EDSS scale using a mu lti-state model and then derive recurrent event data based on this. \; Simulation parameters are chosen to mimic typical MS trial populations in relapsing-remitting or primary progressive MS\, respectively\, and includ e scenarios with heterogeneity (frailties). Based on the results from the simulation studies\, the presentation will conclude with recommendations f or the choice of the endpoint\, and analysis method of MS trials with disa bility progression endpoints.A time to event model for early efficacy
signal dose finding in epilepsy clinical trials.
\n
\n Abstract
Time to-event endpoints have been proposed as alternat ives to establish the effect of anti-epileptic drugs in clinical trials. T hese endpoints may reduce exposure to placebo or ineffective treatments\, thereby facilitating trial recruitment and improving safety. Time to basel ine seizure count is defined as the number of days until a subject experie nced a number of seizures equal to the baseline seizure count. \; A po st hoc analysis of completed Phase III trails with perampanel showed that an analysis of the time to baseline count endpoint is consistent with the classical endpoints (median % seizure rate reduction\, percentage of patie nts achieving a 50% or greater reduction in seizure frequency)1 .
\nWe invest igated the performance of the time to baseline seizure endpoint by (1) a p ost hoc analysis of topiramate and carisbamate clinical trial data and (2) clinical trial simulation using a longitudinal model for daily seizures c ounts. This model included key features of daily seizure count data\, such as a large between subject variability in baseline seizure rate and drug response\, a large variability of the number of seizures per day and clust ering of seizures over time.
\nThe re-analysis of topiramate and carisbamate clinical tr ial data confirmed the relationship between the median time to baseline se izure count and the classical endpoint of median % seizure rate reduction that was observed with perampanel. In addition\, the observed relationship agreed with the one that was predicted by the simulation model.
\nClinical trial simula
tions were used to investigate the performance of a proof-of-concept study
design using the time to baseline seizure count endpoint. The study consi
sted of a 4-week prospective baseline\, followed by a 4-week double blind
treatment period\, after which subjects would exit the study if they had r
eached or exceeded their baseline seizure count\, or would continue for an
other 8-weeks. These simulations showed that (1) with relatively small sam
ple sizes (~ 20/arm) the design is able to identify clinical relevant trea
tment effects (30% - 50% seizure rate reduction)\; (2) a 4-week baseline p
eriod provides enough information on the baseline seizure count and (3) th
e length of exposure of subjects to placebo or an inactive treatment is st
rongly reduced as compared to a classical design. \;
\n
\n Biography
\n
Filip De Ridder i s a Senior Scientific Director in the Statistical Modeling &\; Methodol ogy group of Janssen R&\;D. Twenty years ago\, he was one of founders o f the Modeling &\; Simulation group at Janssen bringing together statis ticians and pharmacometricians to apply modeling &\; simulation techniq ues in clinical drug development. \; Since then he has worked on M& \;S projects in the context of PK/PD modeling\, dose response modeling and clinical trial design\, mainly in neuroscience and infectious diseases. span>
\n
\n
\n Andrew Thomson
\n (EMA)
Abstract p>\n
The treatment of recurrent safety events and terminal events\n requires careful consideration underlying the estimands in question\, an d the\n assumptions in the methods used to estimate them. In th is talk I shall give a\n regulatory perspective on these issues \, focussing on how and why the EU system\n summarises data as it does\, where the gaps are in the methodology\, and how we\n can progress to ensure that data are summarised appropriately. I will cons ider\n whether we need to move beyond the methods currently use d\, and what questions\n we truly need to be answering (and how ). In particular I shall argue that we\n need to be sure that w hen no true raised risk exists\, the method we use to\n summari se said risk should provide an unbiased average effect of 0\, but in\n time-to-event studies this is not always as quite straightforward as it seems.
\n \;
\nBiography.
\nAndrew Thomson is a statisticia n at the EMA Office of\n Biostatistics and Methodology Support\ , joining in 2014. He supports the\n methodological aspects of the assessments of Marketing Authorisation\n Applications\, as well as Scientific Advice\, and methodological aspects of\n Pae diatric Investigational Plans. He has worked extensively on the\n methodological aspects of the EMA Reflection Paper on the use of extrap olation\n of efficacy in paediatric studies.
\n \;
\nPrior to the EMA\, he worked at the UK regulator\, the\n Medicines and Healthcare product Regulatory Agency. Here he worked in itially as\n a statistical assessor in the Licensing Division\, assessing Marketing\n Application Authorisations and providing Scientific Advice to companies. After\n rising to Senior Stati stical Assessor\, he moved to the Vigilance and Risk\n Manageme nt of Medicines Division\, to be Head of Epidemiology. Here he managed a\n team of statisticians\, epidemiologists and data analysts prov iding support to\n the assessment of post-licensing observation al studies and meta-analyses. He\n also managed the team&rsquo\ ;s design\, conduct and analysis of epidemiology studies\,\n us ing the UK Clinical Practice Research
\n \;
\n
\n Arno Fritsch &\; Patrick Schl&o
uml\;mer (Bayer)
In this presentation\, we will discuss potential estimands according to the I CH E9 addendum framework that can be addressed for recurrent events when t here is a non-negligible risk for a terminal event\, typically death.
\nAs an applicati on\, we consider trials in chronic heart failure (HF). Here in the past\, the standard (composite) primary endpoint was the time to either hospitali zation for HF or cardiovascular (CV) death. Since many patients experience recurrent HF hospitalizations\, there is interest to include these events into the primary endpoint. We consider two estimands\, one that focuses o nly on the total number of recurrent HF hospitalizations and another one t hat includes CV death as an additional composite event.
\nWe present results of an exte nsive simulation study that investigated which standard methods for analyz ing recurrent event data estimate the above-mentioned estimands. In additi on\, we compared the efficiency of recurrent event estimands and time-to-f irst event estimands.
\nBiography
\nArno Fritsch received his PhD in Statistics from the University of Dortmund\, Germany\, in 2010. Since then he has been wo rking at Bayer as a clinical statistician\, mainly on the design\, analysi s and submission of cardiovascular trials. Since 2017 he has the position as Group Leader Europe in the cardiovascular statistics department. His me thodological interests include handling of missing data\, analysis of subg roups and recurrent events. He is one of the co-authors of the application for an EMA qualification opinion on use of recurrent events.
\nPatrick Schlö\;mer r eceived his PhD in Statistics from the University of Bremen\, Germany\, in 2014 for his work on group sequential and adaptive designs for three-arm non-inferiority trials. Since then he has been working at Bayer as a clini cal statistician in the cardio-renal area with increasing responsibilities \, now holding the position Lead Statistician. His methodological interest s include group sequential and adaptive designs\, multiple comparison proc edures and recurrent events. He is one of the co-authors of the applicatio n for an EMA qualification opinion on use of recurrent events.
\ n
\n John Gregson
\n (London School
of Hygiene &\; Tropical Medicine)
The value of including recurrent events in the analysis
of cardiovascular outcomes trials.
\n
\n
Abstract
\n
Including recurrent events in analyses of clinical trials can increase power and lead to a more complete assessment of treatment benefit. There are several strategies to analysin g repeat events\, but little practical guidance as to which are best in an y given scenario. Several methods for analyses of repeat events in trials will be compared\, including Andersen-Gill\, Wei-Lin-Weissfeld\, negative binomial regression\, and joint frailty models. The assumptions underlying each of these methods\, and their various advantages and disadvantages wi ll be outlined using data from recent large cardiovascular trials.
\nBiography< /strong>
\nJoh n Gregson is an Assistant Professor in Medical Statistics at the London Sc hool of Hygeine and Trpoical Medicine. He has a range of experience in the analysis of cardiovascular clinical trials\, many of which have been publ ished in high impact journals (e.g. NEJM\, Lancet\, JACC).  \;As well as an interest in the applied analysis of randomised clinical trials and e pidemiological studies\, a major research interest of his is in methodolog ical research into statistical issues which commonly arise in such studies . He holds a PhD in Epidemiology from Cambridge University and a Masters i n Medical Statistics from Southampton University.
\nResampling complex time-to-event data without individual patient dat
a\, with a view toward recurrent events.
\n
\n
Abstract
\n
In this talk we consider
non- and semi-parametric resampling of multistate event histories by simul
ating individual trajectories from an empirical multivariate hazard measur
e. \; \;
\n
\n One advantage is t
hat it does not necessarily require individual patient data\, but may be b
ased on published information. This is also attractive for both study plan
ning and simulating realistic real‐world event history data in general.&nb
sp\; A special focus is on simulating recurrent events data with associate
d terminal events. We demonstrate that our proposal gives a more natural i
nterpretation of how such data evolve over the course of timethan many of
the competing approaches. The multistate perspective avoids any latent fai
lure time structure and sampling spaces impossible in real life\, whereas
its parsimony follows the principle of Occam's razor. We also suggest empi
rical simulation as a novel bootstrap procedure to assess estimation uncer
tainty in the absence of individual patient data. This is not possible for
established procedures such as Efron's bootstrap.
Biography< /p>\n
\n \n
Tobias Bluhmki studied Mathematical Biometry at Ulm University fr
om 2009 to 2014 and was honored with the "Bernd-Streitberg Award" by the I
nternational Biometric Society - German Region for his Master's Thesis. Si
nce then\, he has been research assistant at the Institute of Statistics\,
Ulm University\, Germany. He has recently defended his PhD thesis supervi
sed by Jan Beyersmann at the Faculty of Mathematics and Economics and is n
ow postdoctoral researcher. His research focuses on statistical methodolog
y in clinical trials and epidemiological studies based on survival and eve
nt history techniques.
\n
\n He has publi
shed several articles in biostatistical\, epidemiological and medical jour
nals and is the current co-lead of the "Team of Young Statisticians" of th
e International Biometric Society - German Region.
I am a par tner at Consilium.  \;Consilium is my consultancy partnership with Tom as Salmonson\, a long-standing member of the EMA&rsquo\;s CHMP and formerl y the chair of that committee. \; Tomas and I support companies in the development\, authorisation and life-cycle management of medicines.
\nPreviously I work ed at AstraZeneca and for 19 years at the Medicines and Healthcare product s Regulatory Agency\, heading the group of medical statisticians and pharm acokineticists. \; I am a statistician by background and whilst workin g at MHRA I was co-opted as a member of EMA&rsquo\;s CHMP for expertise in medical statistics and epidemiology. \; At CHMP I was Rapporteur for multiple products and was widely engaged across both scientific and policy aspects of the committee&rsquo\;s work.  \;I was fortunate to chair t he CHMP&rsquo\;s Scientific Advice Working Party for 8 years and have also chaired their expert groups on Biostatistics\, Modelling and Simulation a nd Extrapolation. \; I wrote or co-wrote multiple regulatory guidance documents\, including those related to estimands\, subgroups\, use of cond itional marketing authorisation\, development of fixed-dose combinations\, extrapolation and adaptive designs.  \;I have a particular interest i n when and how to use data generated in clinical practice to support drug development.
\nRegistration | \n|
 \;PSI Member | \n \;£\;40 + VAT | \n \;Non-Membe r | \n \;£\;135 \;+ VAT \;(This in cludes PSI membership for 2019) | \n \n \n