BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:MRC Biostatistics Unit Seminars
SUMMARY:Predictive analytic modelling in clinical trials
(patient recruitment\, randomization and drug supp
ly) - Vladimir Anisimov\, Qunitiles
DTSTART;TZID=Europe/London:20121009T143000
DTEND;TZID=Europe/London:20121009T153000
UID:TALK39605AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/39605
DESCRIPTION:The advanced analytic methodology for modelling an
d predicting stochastic processes describing the b
ehaviour in time of different stages in Phase III
multicentre clinical trials is discussed. \nA stat
istical methodology for predictive patient's recru
itment modelling is developed in [1\,2\,4] where t
he patient's flows are modelled by using Poisson p
rocesses with random delays and gamma distributed
rates (Poisson-gamma model). ML and Bayesian techn
iques for estimating parameters and asymptotic app
roximations for the mean and predictive bounds in
time for the number of patients recruited in the r
egions and for time to complete trial are develope
d. The optimal number of clinical centres and tria
l performance can be also evaluated. \nAnalytic te
chnique for predicting in time the number of diffe
rent events in trials with waiting time to respons
e is developed [5]. The closed-form expressions fo
r predictive distributions are derived and used in
oncology trials. \nAnalytic technique for predict
ing in time randomisation processes in centres/reg
ions is developed and the impact of stratification
on sample size is investigated [3\,4]. A risk-bas
ed supply modelling tool based on these results is
developed [2]. \nSoftware tools in R supporting t
hese techniques are created. These tools are imple
mented on R&D GSK level and already led to signifi
cant benefits and cost savings.\n\nReferences\n\n1
. Anisimov\, V.V.\, Fedorov\, V.V.\, Modeling\, pr
ediction and adaptive adjustment of recruitment in
multicentre trials. Statistics in Medicine\, v. 2
6\, No. 27\, 2007\, pp. 4958-4975.\n\n2. Anisimov\
, V.V.\, Predictive modelling of recruitment and d
rug supply in multicenter clinical trials. Proc. o
f the Joint Statistical Meeting\, Washington\, USA
\, August\, 2009\, pp. 1248-1259.\n\n3. Anisimov\,
V.V.\, Effects of unstratified and centre-stratif
ied randomization in multicentre clinical trials.
Pharmaceutical Statistics\, v. 10\, iss. 1\, 2011\
, pp. 50-59.\n\n4. Anisimov\, V.V.\, Statistical m
odeling of clinical trials (recruitment and random
ization)\, Communications in Statistics - Theory a
nd Methods\, 40: 19-20\, 2011\, pp. 3684-3699. \n\
n5. Anisimov\, V.V.\, Predictive event modelling i
n multicentre clinical trials with waiting time to
response\, Pharmaceutical Statistics\, v. 10\, is
s. 6\, 2011\, pp. 517-522.\n
LOCATION:Large Seminar Room\, 1st Floor\, Institute of Pub
lic Health\, University Forvie Site\, Robinson Way
\, Cambridge
CONTACT:Dr Jack Bowden
END:VEVENT
END:VCALENDAR