\; \;\n
\n Valentine
Jehl \n (Novartis) \n  \n Quantitative assessment of adverse events in clinical trial
s &ndash\; comparison of methods at an interim and the final analysis. \n \n Abstract \n
In clinical study reports\, adve
rse events (AEs) are commonly summarized using the incidence proportion de
spite cumulative incidence function been advocated as the most appropriate
method to account for different exposure time and competing events. \n In this presentat
ion\, we compare different methods to estimate the probability of one sele
cted AE. Besides considering the final analysis at the time of the Clinica
l Study Report\, we especially investigate the capability 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 competin
g 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 <
p>Valentine Jehl is a senior quantitati
ve safety scientist at Novartis. She received her Master&rsquo\;s degree i
n applied mathematics at the Louis Pasteur University in Strasbourg.\n She started her c
arrier as statistician with a CRO in Brussel. She then joined Novartis in
Basel\, where she supported major submissions and development programs for
the oncology franchise. After 9 years in this role\, Valentine joined the
quantitative safety group in April 2016\, where she now promotes the use
of quantitative methods for safety\, with a particular focus on Adverse Dr
ug Reactions. \n \
n \n 
\n \n Qing Wang \n
(Roche) \n \n \n  \n Comparison of timetofirst event and recurrent 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 S
tatus Scale (EDSS) as an endpoint. However\, especially in progressive for
ms of MS where CDP is typically the primary endpoint\, a substantial propo
rtion of subjects may experience repeated disability events. Recurrent eve
nt analyses could therefore increase study power and improve clinical inte
rpretation of results. \n We present results from two simulation studies which compare an
alyses of the time to the first event with recurrent event analyses (inclu
ding negative binomial\, AndersenGill\, and Lin\, Wei\, Ying\, and Yang m
odels). The first simulation study is generic and recurrent events data is
simulated according to a mixed nonhomogeneous Poisson process. \;Th
e second simulation study is MSspecific: we first simulate longitudinal m
easurements of the ordinal EDSS scale using a multistate model and then d
erive recurrent event data based on this. \; Simulation parameters are
chosen to mimic typical MS trial populations in relapsingremitting or pr
imary progressive MS\, respectively\, and include scenarios with heterogen
eity (frailties). Based on the results from the simulation studies\, the p
resentation will conclude with recommendations for the choice of the endpo
int\, and analysis method of MS trials with disability progression endpoin
ts. \n \n Biography \n \n Qing is a statistician working at Roche
Basel. She is currently the project lead statistician for the Ocrevus (oc
relizumab) program\, and had been supporting the program from initial stud
y 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 Epidemio
logy and Biostatistics at University Hospital Basel. She received her Mast
er in Mathematics and PhD in Biostatistics at the University of Cambridge
(UK). \n \n 
\n
Filip De Ridder \n (Janssen)  \n A time to event model for early efficacy signal dose finding in
epilepsy clinical trials. \n \n Abstract
\n Ti
me toevent endpoints have been proposed as alternatives to establish the
effect of antiepileptic drugs in clinical trials. These endpoints may red
uce exposure to placebo or ineffective treatments\, thereby facilitating t
rial recruitment and improving safety. Time to baseline seizure count is d
efined as the number of days until a subject experienced a number of seizu
res equal to the baseline seizure count. \; A post hoc analysis of com
pleted Phase III trails with perampanel showed that an analysis of the tim
e to baseline count endpoint is consistent with the classical endpoints (m
edian % seizure rate reduction\, percentage of patients achieving a 50% or
greater reduction in seizure frequency)^{1}. \n
We investigated the performance
of the time to baseline seizure endpoint by (1) a post hoc analysis of to
piramate and carisbamate clinical trial data and (2) clinical trial simula
tion using a longitudinal model for daily seizures counts. This model incl
uded key features of daily seizure count data\, such as a large between su
bject variability in baseline seizure rate and drug response\, a large var
iability of the number of seizures per day and clustering of seizures over
time. \n The r
eanalysis of topiramate and carisbamate clinical trial data confirmed the
relationship between the median time to baseline seizure count and the cl
assical endpoint of median % seizure rate reduction that was observed with
perampanel. In addition\, the observed relationship agreed with the one t
hat was predicted by the simulation model. \n Clinical trial simulations were used to inv
estigate the performance of a proofofconcept study design using the time
to baseline seizure count endpoint. The study consisted of a 4week prosp
ective baseline\, followed by a 4week 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 8weeks. 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 4week 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 in the Statistical Modeling &\; Methodology group of Janssen R
&\;D. Twenty years ago\, he was one of founders of the Modeling &\;
Simulation group at Janssen bringing together statisticians and pharmacome
tricians to apply modeling &\; simulation techniques in clinical drug d
evelopment. \; Since then he has worked on M&\;S projects in the co
ntext of PK/PD modeling\, dose response modeling and clinical trial design
\, mainly in neuroscience and infectious diseases. \n

\; \;\n \n \n Andrew Thomson \n (EMA) \n
 \;\n Abstract \n The treatment of
recurrent safety events and terminal events\n requires careful
consideration underlying the estimands in question\, and the\n
assumptions in the methods used to estimate them. In this talk I shall gi
ve a\n regulatory perspective on these issues\, focussing on ho
w and why the EU system\n summarises data as it does\, where th
e gaps are in the methodology\, and how we\n can progress to en
sure that data are summarised appropriately. I will consider\n
whether we need to move beyond the methods currently used\, and what quest
ions\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 shoul
d provide an unbiased average effect of 0\, but in\n timetoev
ent studies this is not always as quite straightforward as it seems. \n \; \n Biogr
aphy. \n Andrew Thomson is a statistician at the EMA Offic
e 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 Paediatric Investigat
ional Plans. He has worked extensively on the\n methodological
aspects of the EMA Reflection Paper on the use of extrapolation\n
of efficacy in paediatric studies. \n \; \n
Prior t
o the EMA\, he worked at the UK regulator\, the\n Medicines and
Healthcare product Regulatory Agency. Here he worked initially as\n
a statistical assessor in the Licensing Division\, assessing Marketi
ng\n Application Authorisations and providing Scientific Advice
to companies. After\n rising to Senior Statistical Assessor\,
he moved to the Vigilance and Risk\n Management of Medicines Di
vision\, to be Head of Epidemiology. Here he managed a\n team o
f statisticians\, epidemiologists and data analysts providing support to\n
the assessment of postlicensing observational studies and met
aanalyses. He\n also managed the team&rsquo\;s design\, conduc
t and analysis of epidemiology studies\,\n using the UK Clinica
l Practice Research \n 
\n
\; \n \n Arno Fritsch &\; Patrick Schlö\;mer (Bayer) \n  Estimands for recurrent events in the
presence of a terminal event &ndash\; Considerations and simulations for c
hronic heart failure trials. \n \n
Abstract \n \n \n In this presentation\
, we will discuss potential estimands according to the ICH E9 addendum fra
mework that can be addressed for recurrent events when there is a nonnegl
igible risk for a terminal event\, typically death. \n
As an application\, we consider t
rials in chronic heart failure (HF). Here in the past\, the standard (comp
osite) primary endpoint was the time to either hospitalization for HF or c
ardiovascular (CV) death. Since many patients experience recurrent HF hosp
italizations\, there is interest to include these events into the primary
endpoint. We consider two estimands\, one that focuses only on the total n
umber of recurrent HF hospitalizations and another one that includes CV de
ath as an additional composite event. \n We present results of an extensive simulation s
tudy that investigated which standard methods for analyzing recurrent even
t data estimate the abovementioned estimands. In addition\, we compared t
he efficiency of recurrent event estimands and timetofirst event estiman
ds. \n Biography \n Arno Fritsch received his PhD in Statistics from the University of
Dortmund\, Germany\, in 2010. Since then he has been working at Bayer as
a clinical statistician\, mainly on the design\, analysis and submission o
f cardiovascular trials. Since 2017 he has the position as Group Leader Eu
rope in the cardiovascular statistics department. His methodological inter
ests include handling of missing data\, analysis of subgroups and recurren
t events. He is one of the coauthors of the application for an EMA qualif
ication opinion on use of recurrent events. \n Patrick Schlö\;mer received his PhD in
Statistics from the University of Bremen\, Germany\, in 2014 for his work
on group sequential and adaptive designs for threearm noninferiority tr
ials. Since then he has been working at Bayer as a clinical statistician i
n the cardiorenal area with increasing responsibilities\, now holding the
position Lead Statistician. His methodological interests include group se
quential and adaptive designs\, multiple comparison procedures and recurre
nt events. He is one of the coauthors of the application for an EMA quali
fication opinion on use of recurrent events. \n 
\n
\
n John Gregson \n (London School of Hygiene &am
p\; Tropical Medicine) \n \n \n  \n The value of including recurrent events in the analysis of cardiovasc
ular outcomes trials. \n \n Abstract \n \n Including recurrent events in analyses of cl
inical trials can increase power and lead to a more complete assessment of
treatment benefit. There are several strategies to analysing repeat event
s\, but little practical guidance as to which are best in any given scenar
io. Several methods for analyses of repeat events in trials will be compar
ed\, including AndersenGill\, WeiLinWeissfeld\, negative binomial regre
ssion\, and joint frailty models. The assumptions underlying each of these
methods\, and their various advantages and disadvantages will be outlined
using data from recent large cardiovascular trials. \n
Biography \n John Gregson is a
n Assistant Professor in Medical Statistics at the London School of Hygein
e and Trpoical Medicine. He has a range of experience in the analysis of c
ardiovascular clinical trials\, many of which have been published in high
impact journals (e.g. NEJM\, Lancet\, JACC). \;As well as an interest
in the applied analysis of randomised clinical trials and epidemiological
studies\, a major research interest of his is in methodological research
into statistical issues which commonly arise in such studies. He holds a P
hD in Epidemiology from Cambridge University and a Masters in Medical Stat
istics from Southampton University. \n 
\; \n Tobias Bluhmki (University of Ulm) \n  \n Resampli
ng complex timetoevent data without individual patient data\, with a vie
w toward recurrent events. \n \n Abstract \n \n
In this talk we consider non and semi
parametric resampling of multistate event histories by simulating individu
al trajectories from an empirical multivariate hazard measure. \;
\; \n \n One advantage is that it does no
t necessarily require individual patient data\, but may be based on publis
hed information. This is also attractive for both study planning and simul
ating realistic real‐world event history data in general. \; A special
focus is on simulating recurrent events data with associated terminal eve
nts. 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 failure time stru
cture and sampling spaces impossible in real life\, whereas its parsimony
follows the principle of Occam's razor. We also suggest empirical simulati
on as a novel bootstrap procedure to assess estimation uncertainty in the
absence of individual patient data. This is not possible for established p
rocedures such as Efron's bootstrap. \n Biography \n
\n \n Tobia
s Bluhmki studied Mathematical Biometry at Ulm University from 2009 to 201
4 and was honored with the "BerndStreitberg Award" by the International B
iometric Society  German Region for his Master's Thesis. Since then\, he
has been research assistant at the Institute of Statistics\, Ulm Universit
y\, Germany. He has recently defended his PhD thesis supervised by Jan Bey
ersmann at the Faculty of Mathematics and Economics and is now postdoctora
l researcher. His research focuses on statistical methodology in clinical
trials and epidemiological studies based on survival and event history tec
hniques. \n \n He has published several a
rticles in biostatistical\, epidemiological and medical journals and is th
e current colead of the "Team of Young Statisticians" of the Internationa
l Biometric Society  German Region. \n 
\n Rob Hemmings (Consiliu
m) \n  Biography \n
\n I am a partner at Consilium. &n
bsp\;Consilium is my consultancy partnership with Tomas Salmonson\, a long
standing member of the EMA&rsquo\;s CHMP and formerly the chair of that c
ommittee. \; Tomas and I support companies in the development\, author
isation and lifecycle management of medicines. \n Previously I worked at AstraZeneca and
for 19 years at the Medicines and Healthcare products Regulatory Agency\,
heading the group of medical statisticians and pharmacokineticists. \
; I am a statistician by background and whilst working at MHRA I was coop
ted as a member of EMA&rsquo\;s CHMP for expertise in medical statistics a
nd epidemiology. \; At CHMP I was Rapporteur for multiple products and
was widely engaged across both scientific and policy aspects of the commi
ttee&rsquo\;s work. \;I was fortunate to chair the CHMP&rsquo\;s Scie
ntific Advice Working Party for 8 years and have also chaired their expert
groups on Biostatistics\, Modelling and Simulation and Extrapolation.&nbs
p\; I wrote or cowrote multiple regulatory guidance documents\, including
those related to estimands\, subgroups\, use of conditional marketing aut
horisation\, development of fixeddose combinations\, extrapolation and ad
aptive designs. \;I have a particular interest in when and how to use
data generated in clinical practice to support drug development.\n 