Timings: 09:30  17:00 UK
Time
\nLocation: Royal Statistical Society
\nAddress: 12 Errol Street\, London
\nBayesian method s are being increasingly used in the design and analysis of trials through all stages of clinical drug development. This meeting will include talks on two areas of application commonly used in early phase trials &ndash\; d ose selection and biomarkers. The meeting will include a range of case stu dies presented from different therapeutic areas. More details to follow.&n bsp\;
\nClick here \;to register.
\n15.10  15.30
\n \ nTime  \n Agenda \n 
09.30  09.55  Regis tration 
\n 09.55  10.00 \n  \n Welcome and Int roduction \n 
\n  \n
FirstinHuman Studies: Practice or Principle 
\n 10.40  11.20  \n
Model based dose escalation: in practice 
\n 11.20  11.40 \n \n  \n Break \n 
\n 11.40  12.20 \n  \n A Baye
sian information theoretic design for phase I dose finding trials without
monotonicity assumption 
\n
12.20  13.00 \n  \n

\n
13.00  13.50 \n  \n Lunch \n 
\n 13.50  14.30 \n  \n The use of form
ally elicited priors to aid the design of a phase 2B dose ranging study 
\n 14.30 15.10 \n  \n Bayesian m
ethods for design and analysis of clinical trials assessing multiple treat
ments and biomarkers 
\n Break \n  
\n 15.30  16.10< /p>\n  \n Error rat
es control via shrinkage priors in multivariate errorinvariable doseres
ponse models 
\n
16.10  16.50 \n  \n Prior elicitation and t
ranslation of biomarkers to aid clinical development plans and compound se
lection 
\n 16.50  17.00< /p>\n  \n Close \n 
Bayesian methods are being increasingly used in
the design and analysis of trials through all stages of clinical drug deve
lopment. This meeting will include talks on two areas of application commo
nly used in early phase trials &ndash\; dose selection and biomarkers. The
meeting will include a range of case studies presented from different the
rapeutic areas. \; \n  
FirstinHuman Studies: Practice or Principle \n \n Abstract: \; \n \n Bird SM\, Bailey RA\, Grieve AP\, Senn S. Statistical iss
ues in first‐in‐human studies on BIA 10‐2474: Neglected comparison of prot
ocol against practice. Pharmaceutical Statistics 2017\;16:1006. Expert Scientific Group on Phase One Clinical Trials (Final report of a group convened by U K's Medicines and Healthcare Regulatory Authority and chaired by Professor Sir Gordon Duff)\, Final report online 6 December 2006\, see http://webarchive.nationalarchives.gov.uk/< /a>+ /dh.gov.uk/ en/ publicationsandstatistics/ publications/publicationsp olicyandguidance/ dh_063117 (acc. Jan 2018). \nSenn S\, Amin D\, Bailey RA\, Bird SM \, Bogacka B\, Colman P\, Garrett A\, Grieve A\, Lachmann P. Statistical i ssues in first‐in‐man studies. Journal of the Royal Statistical Society: S eries A (Statistics in Society) 2007\; 170:51779. \n  
Phil Overend\, AstraZeneca  \n
Model base d dose escalation: in practice \nAbstract: A t AstraZeneca\, the implementation of a model based Bayesian dose escalati on design in Oncology phase 1 studies has been facilitated by putting prac tical considerations in place that address the operational challenges that arise due to increased statistical complexity that accompanies the method ology. We will discuss a practical case study\, describing the first time that this approach was implemented at AstraZeneca\, including the options explored for specifying the prior\, the advantages of the approach togethe r with practical and methodological learnings for future studies. \; T he extension to 2drug combination dose escalation studies will be introdu ced. \n 
A Bayesian informatio
n theoretic design for phase I dose finding trials without monotonicity as
sumption \n \n \n Abstra
ct: Methods for finding the highest dose that has an acceptable r
isk of toxicity in Phase I doseescalation clinical trials assuming a mono
tonic doseresponse relationship have been studied extensively in recent d
ecades. The assumption of monotonicity is fundamental in these methods. As
a result\, such designs fail to identify the correct dose when the doser
esponse relationship is nonmonotonic\, for example\, in doseschedules an
d dosecombinations trials. We propose a doseescalation method that does
not require monotonicity or any prespecified relationship between dose le
vels. The design is based on the Bayesian model incorporating an informati
ontheoretic criterion for dose selections. We will discuss the practical
aspects of the design such as prior and safety stopping. For small sample
size typical for doseescalation studies we will show in simulations that
the proposed method is comparable to wellstudied and used methods under t
he assumption of monotonicity and outperforms them when this assumption is
violated. We conclude with an extension of the design which allows to inc
lude an efficacy endpoint. Biography: Pavel did his undergraduate studies in the National Research Univer sity Higher School of Economics\, Moscow\, Russia. He worked on informatio ntheoretic aspects in statistics in the International Laboratory of Stoch astic Analysis. Currently\, Pavel is a PhD student in Lancaster University where he works on the project &ldquo\;Dose finding for combination trials with many treatments&rdquo\; under supervision of Professor Thomas Jaki. Pavel is a MarieCurie Fellow in the IDEAS network. The aim of his project is to develop novel dose finding designs which do not require parametric assumptions and can be applied to various clinical trials. \;\n  
Bayesian hierarchical mod
el for dose escalation \n \n \n
Abstract: Clinical trials with multiple strata (e.g. indi
cation\, dose regimen\, mutation status) are increasingly used in drug dev
elopment. In dose escalation trials\, where data are often sparse and a mu
ltiple strata analysis may be the only option to study a new treatment\, g
ood statistical inference and decisionmaking can be challenging. Inferenc
es from simple pooling or stratification are known to be inferior to hiera
rchical modeling methods\, which build on exchangeable strata parameters t
o allow borrowing information across strata. However\, the standard exchan
geability (EX) assumption bears the risk of too much shrinkage and excessi
ve borrowing for extreme strata. This talk will present the exchangeabilit
ynonexchangeability (EXNEX) approach in dose finding studies\, as a robus
t extension of the standard EX approach. It allows each stratumspecific p
arameter to be exchangeable with other similar strata parameters or nonexc
hangeable with any of them. A case study from a Phase I trial will also be
shown. Biography: After the de gree in Statistics for biomedicine\, environment and technology from the u niversity of Rome &ldquo\;La Sapienza&rdquo\; (2008)\, Alessandro joined t he pharmaceutical industry (Novartis). He is currently working as Associat e Director of Biostatistics in the area of Phase I/II Oncology clinical tr ials in Early Development Biostatistics (EDB) group. Prior to join the EDB Oncology group Alessandro has worked as trainee in the CISMethodology Gr oup\, supporting the methodological development and implementation of onco logy Bayesian single agent/combination dosefinding study design.\n  
\n \n Doug Thompson\, GSK  The use of formally elicited priors to aid the desig
n of a phase 2B dose ranging study \n \n \n \n Abstract: \;The use of a
ssurance to quantify the probability of success of a proposed parallel gro
up clinical trial is well understood. This framework extends naturally to
studies with other designs\; including when evaluating dose response is th
e primary focus. Prior distributions formally elicited about the mean resp
onse at certain dose levels can be used to set certain constraints such th
at a spread of plausible three parameter Emax curves may be generated. In
this presentation\, we will discuss our experiences in the development of
a dose response trial simulator. We illustrate how such an approach may b
e used to help characterise interim futility rules\, dose adaptation and a
ssurance. Biography: Doug Thomp son has been at GlaxoSmithKline for over three years working on the design \, analysis and reporting of clinical trials in rare diseases. He studied statistics at the University of Glasgow as an undergraduate before going o n to undertake a PhD at the University of Edinburgh on the development of a risk stratified approach for the treatment of acute stroke patients\, fu nded through the MRC Hubs for Trials Methodology Research. \n 
Bayesian methods for design and analysis of clinic
al trials assessing multiple treatments and biomarkers \n \n \n Abstract: Response to tre atments is often highly heterogeneous\, especially ones that are targeted at specific biological pathways. The increasing availability of biomarkers and targeted treatments has led to the need for trial designs that effici ently test new treatments in biomarkerstratified patient subgroups. Often new treatments are targeted at a specific biomarker subgroup\, but may in fact work in a narrower or broader set of patients. \nI will discuss
Bayesian adaptive methodology for trials that have multiple treatments an
d biomarkers. The proposed design incorporates biological hypotheses about
the links between treatments and biomarker subgroups\, but allows alterna
tive links to be formed during the trial. The statistical properties of th
e method compare well to alternative designs available. \; This design
has been developed for trials in ovarian cancer and breast cancer and som
e methodology issues specific to each application will be discussed. These
include the use of continuous biomarker information to allocate patients
and adding in new treatments and biomarkers during the trial. Biography: James Wason is a Professor of Bi ostatistics at Newcastle University and a Programme Leader at the MRC Bios tatistics Unit. He leads a research group focused on development of statis tical methods for more efficient trials. His research interests are adapti ve designs\, using biomarkers prospectively during trials and analysis of composite endpoints. As well as working on methodology\, James is very int erested in applying novel methods in real clinical trials and is currently involved in several ongoing trials using novel designs. \n  
Fabio Rigat\, GSK  Error rates control via shrink
age priors in multivariate errorinvariable doseresponse models
\n \n Abstract: When pl anning &\; analysing limited numbers of doseresponse data\, the risk o f false positive results may be considerable. In this case\, priors can be used to penalise inferences for doseresponse parameters so as to reduce this risk. We demonstrate a simple example of such priors for analysing th e multivariate relation between the dose of an investigational asset admin istered to a cohort of cancer patients and the associated average change i n the expression of several target genes. This analysis is further robusti fied by identifying error terms capturing the technical variability of the se biomarker data from those representing sampling variation. \; \n \n 
Prior elicitation and translation of biomarkers to aid clinical
development plans and compound selection \n \n Abstract: The talk will show how prior el icitation of a pharmacodynamic (PD) biomarker can be used to estimate a de sign prior to be used in designing clinical trials. \; In addition\, i f the translation of the biomarker to efficacy or safety is elicited\, the n this can help evaluate the benefits of the biomarker and how decisions m ay be made based on the biomarker. \; An example will be given where e fficacy and safety biomarkers are used to compare several compounds being considered for the same indication. \; The elicitation of the translat ion allows us to weigh up which biomarker is most relevant in making the d ecision over which compound to progress first. \; When this decision c an be made\, preclinical\, Phase 1 or Phase 2\, can be compared. \n \n \n Biography: Trevor has b een the Group Lead for Early Phase Neuroscience Statistics at Eli Lilly si nce September 2011 covering neurodegeneration\, psychiatric illnesses and pain. \; This covers the clinical planning after Candidate Identificat ion through to the decision to proceed to Phase 3. \; Prior to this he was at Pfizer in Kent from 1998\, where he worked on early clinical devel opment as well. \; Much of the work was in pain and many biomarker and methodology studies including: fMRI\, other imaging biomarkers\, healthy volunteer challenge models and patient biomarkers. Trevor has also worked for BioSS in Scotland and at Rothamsted. \n 
\n \; \n  \n
Registration Fees \n 
\n PSI Member \n  \n £\;40 + VAT \n 
\n Non Member \n  \n £\;135 + VAT 
\n
\nPl
ease register before Wednesday 21st February 2018. \;
\n
\nPlease contact the PSI secretariat on psi@mcigroup.com if you have any dietary requirements or queries ab
out the event.