• Introduction To Industry Training Course 2017

    Dates: 01 Oct, 2017 – 31 Jul, 2018

    Are you a PSI member with approx. 1-3 years experience as a Statistician or a Statistical Programmer within the industry?



    AIM: To describe the drug development process from research right through to research, toxicology, data management & role of the CRO, clinical trials, product development & manufacture and marketing.

    Limited places available!

    Application forms must be received by 30th June 2017!

    Please discuss your application with your manager
    Final dates to be confirmed.


    For further information contact:

    Alex Godwood

    MedImmune Ltd, Milstein Building, Granta Park

    Great Abington, Cambridge, CB21 6GH

    Tel: 0203 7496241

    Email: godwooda@MedImmune.com

  • PSI Webinar: Patient Preferences – a webinar with Kevin Marsh presented by the Benefit-Risk SIG

    Dates: 24 – 24 Oct, 2017
    This webinar is free to attend. Please click here to register.

    For more information please click here.
  • EFPSI/EFPIA Webinar: New draft ICH E9 addendum on Estimands and Sensitivity Analysis

    Dates: 30 – 30 Oct, 2017

    14:00 - 15:30 UK Time

    In this webinar, the EU regulatory and Industry members of the ICH E9(R1) working group will present the new draft addendum for ICH E9 on estimands and sensitivity analysis.  The addendum introduces a new framework for designing and analysing clinical trials aligned to the trial objectives.  

    Click here to view the flyer.

    Rob Hemmings (MHRA) will present the motivation behind the new draft addendum, define estimands and sensitivity analysis, and explain different strategies that can be used in constructing an estimand.

    Frank Bretz (Novartis) will present case studies to illustrate how the new framework can be implemented in designing clinical trials and defining the appropriate analysis methods.

    A Q&A session will be chaired by Frank Pétavy (EMA) and Chrissie Fletcher (Amgen).

    This webinar is free to attend. 

    Click here to register.

  • PSI Webinar: Causal Inference

    Dates: 02 – 02 Nov, 2017

    Defining the estimand of interest in a clinical trial is crucial to align its planning, design, conduct, analysis, and interpretation. The need for more precise specifications of estimands is highlighted in the draft addendum ICH E9(R1) which was published for public consultation in August 2017. Although not explicitly mentioned in ICH E9(R1), the addendum  brings causal reasoning – besides randomization and ITT – into our world of pharmaceutical statistics. In this webinar, we will discuss the link between the ICH E9(R1) and causal inference. Furthermore, per protocol analyses will be discussed from a causal inference perspective and a case study where a principal strata estimand was investigated will be presented.

    For more information please click here.

    Baldur Magnusson

    Novartis Pharma AG

    Using principal stratification to address post-randomization events: A case study

    In a randomized controlled trial, occurrence of post-randomization events associated with treatment and the primary endpoint may complicate the interpretation of the overall treatment effect. In this presentation, we discuss how these events may be accounted for at the estimand and the estimator level in the context of a recent case study. We define a principal stratification estimand derived from the scientific question of interest. Consideration is given to identifying assumptions, model-based derivation of an estimator, handling of covariates and missing data. We also discuss the role of sensitivity analyses.

    Daniel Scharfstein
    Professor of Biostatistics, Johns Hopkins Bloomberg School of Public Health

    Estimands and Causal Inference

    Recently, the ICH proposed an addendum to the E9 Guidance: Statistical Principles for Clinical Trials. This addendum is focused on estimands and sensitivity analysis for randomized trials with intercurrent events. In this webinar, I will discuss the potential outcomes framework for causal inference and use it to formally define estimands that address different types of intercurrent events. I will then discuss the assumptions required to identify these estimands from the observable data and discuss the important role of sensitivity analysis.

    Wanjie Sun

    Estimating Causal Effects in Clinical Endpoint Bioequivalence Studies in the Presence of Treatment Noncompliance and Missing Data

    In clinical endpoint bioequivalence (BE) studies, the primary analysis for assessing equivalence between a generic and an innovator product is usually based on the observed per-protocol (PP) population (i.e., completers and compliers in general). The FDA Missing Data Working Group and the ICH E9 Revision 1 Working Group recommended using “causal estimands of primary interest.” The analysis based on the PP population, however, is not generally causal because PP is determined post-treatment, hence conditioning on it may introduce selection bias. To date, no causal inference has been proposed to assess to equivalence. In this paper, we propose a causal framework and co-primary causal estimands to test equivalence by applying Frangakis and Rubin (2002)’s principal stratification in causal inference. We identify three conditions when the current PP estimator is unbiased for one of the proposed co-primary causal estimands – the“Survivor Average Causal Effect” (SACE) estimand. Simulation was used to demonstrate the bias, type 1 error, and power associated with the PP estimator when these three conditions are not met. We also propose a tipping point sensitivity analysis to evaluate the robustness of the current PP estimator (primary analysis) in testing equivalence when the underlying sensitivity parameters vary across a clinically meaningful range. Data from a clinical endpoint BE study is used to illustrate the proposed co-primary causal estimands and sensitivity analysis method. Our work starts causal evaluation of equivalence assessment in clinical endpoint BE studies with non-compliance and missing data, and can be applied to clinical biosimilar and non-inferiority studies.

    *The views expressed in this article represent the opinions of the authors, and do not represent the views and/or policies of the U.S. Food and Drug Administration.

    This webinar is free to attend, please click here to register. 

  • Statistics Fundamentals for Clinical Trials for Non-Statisticians (or ‘How to speak stats in a day!’)

    Dates: 14 – 14 Nov, 2017

    Location: Premier Inn Reading Central

    Address: Premier Meetings Reading Central, Letcombe Street, Reading, Berkshire, RG1 2HN, United Kingdom 

    Presented by
    Gemma Hodgson
    (Qi Statistics Ltd)

    This basic but wide-ranging course covers techniques for investigating, visualising and performing basic statistical techniques on data sets typical to industry settings. There are many basic concepts that need to be understood before statistics can be used to its full potential to give useful and informative answers. This course ensures that these concepts are understood in a non-technical way and then demonstrated using data examples. 

    Mathematical details are kept to a necessary minimum and we focus on the interpretation of statistical output and illustrate applications with data from dummy clinical trials or published data. The objective of the course is not to teach you how to become a statistician, but to help you work with statisticians and get the maximum value from statistical output. 

    The course will consist of lectures, practical examples and discussions. There will not be any computer exercises. 

    Target Audience: 

    This is a 1-day course, aimed to introduce statistics to people who work on Clinical Trials, but who are not Statisticians. No previous knowledge of Statistics is assumed as we start right at the beginning with the basics. Many practical examples are given and the emphasis is on application and understanding rather than the equations and the technical background. 

    The basics of statistics are discussed to give background and a common base to start from and the applications and use of statistics in drug development is then discussed. The role of the statistician and their ability to help with decision making is also discussed. 

    It also serves as a useful refresher course to those who once studied statistics as part of a college course. 

    The following key topics will be addressed: 

    1. Types of Data 

    2. Measures of location and variability 

    3. Basic Inference 

    4. Power calculations and Sample Sizing 

    5. Design Issues 

    For more information on specific topics, please contact the presenter direct on gemma@qistatistics.co.uk

    Please click here to view the flyer.

    About the presenter: 

    Gemma Hodgson, Qi Statistics Ltd. http://www.qistatistics.co.uk 

    Gemma Hodgson has worked in the Pharmaceutical industry for 20 years. After receiving her first degree from Imperial College (Maths with Statistics) and then an MSc in Medical Statistics from London School of Tropical Hygiene and Medicine, Gemma began her career at Pfizer in Sandwich working in experienced global teams on major phase 3 projects. After 13 years at Pfizer and working in all phases of development, from phase 1 to phase 4, Gemma then moved to Takeda R &D in London where she worked on later phase projects, focussing on close liaison with other departments within the organisation. In 2012 Gemma left Takeda to work for a statistical training and consultancy firm, Qi Statistics Ltd, where training of non-statisticians and explaining statistical concepts to non-scientific audiences is key. Gemma has a broad interest in the application of statistics and is an experienced trainer to all types of audience, specialising in translating technical concepts into everyday English.


    Course runs from:            09:45 – 17:00 (registration from 9:15)


    Please register online at www.psiweb.org and click on Events; payment now available online.

    Registration costs (includes lunch and refreshments)

    Registration before 13th October 2017

    £425 plus vat

    Registration on or after 13th October 2017

    £495 plus vat

     Please click here to register

    PSI aims to be fully inclusive and endeavours to accommodate delegates with disabilities wherever possible.  Please help us to help you by letting us know if you require additional facilities or have any special requirements.  Please contact us on +44 (0)1730 715 235 or at PSI@mci-group.com for further information.

  • PSI One Day Meeting: Extrapolation

    Stevenage | Dates: 22 Nov, 2017
    Use of extrapolation techniques is playing an increasingly important role in the development of new medicines particularly with regard to special populations such as paediatrics and rare diseases. This meeting will include speakers from industry, academia and regulatory (including Rob Hemmings from MHRA).

    Speakers include:
    • Peter Milligan - Pfizer
    • Kristin Karlsson - Medical Products Agency, Sweden
    • Rob Hemmings - MHRA
    • Nicky Best - GSK
    • Dawn Edwards - GSK
    • Adrian Mander - MRC Biostatistics Unit, University of Cambridge
    • Lisa Hampson - AstraZeneca
    • Ian Wadsworth - Lancaster University
    Please click here to view the flyer.


    Rob Hemmings, MHRA

    Extrapolation; regulatory need, examples and emerging guidance.

    Abstract: Extrapolation is defined as ‘extending information and conclusions available from studies in one or more subgroups of the patient population (source population(s)), or in related conditions or with related medicinal products, to make inferences for another subgroup of the population (target population), or condition or product, thus reducing the amount of, or general need for, additional information (types of studies, design modifications, number of patients required) needed to reach conclusions for the target population, or condition or medicinal product’.  The talk will illustrate the potential need for, and benefits of, this concept in regulatory work with a primary focus on extrapolation from adults to children.  An overview of the EMA Reflection Paper on this topic will be presented and discussed, highlighting areas for further discussion and research.

    Ian Wadsworth, Lisa V. Hampson, Thomas Jaki and Graeme J. Sills

    Using historical data to inform extrapolation decisions in children

    When developing a new medicine for children, the potential to extrapolate from adult efficacy data is well recognised. However, significant assumptions about the similarity of adults and children are needed for extrapolations to be biologically plausible. One such assumption is that pharmacokinetic-pharmacodynamic (PK-PD) relationships are similar in these different groups. In this presentation, we consider how ‘source’ data available from historical trials completed in adults and adolescents treated with a test drug, can be used to quantify prior uncertainty about whether PK-PD relationships are similar in adults and younger children. A Bayesian multivariate meta-analytic model is used to synthesise the PK-PD data available from the historical trials which recruited adults and adolescents. The model adjusts for the biases that may arise since these existing data are not perfectly relevant to the comparison of interest, and we propose a strategy for eliciting expert prior opinion on the size of these external biases. From the fitted bias-adjusted meta-analytic model we derive prior distributions which quantify our uncertainty about the similarity of PK-PD relationships in adults and younger children. These prior distributions can then be used to calculate the probability of similar PK-PD relationships in adults and younger children which, in turn, may be used to inform decisions as to whether complete extrapolation of efficacy data from adults to children is currently justified, or whether additional data in children are needed to reduce uncertainty. Properties of the proposed methods are assessed using simulation, and their application to epilepsy drug development is considered.

    Clara Domínguez-Islas1, Adrian Mander1, Rebecca Turner2, Nicky Best3

    A Bayesian framework for extrapolation using mixture priors 

    1 MRC Biostatistics Unit, University of Cambridge, UK.

    2 MRC Clinical Trials Unit, University College London, UK.

    3 GlaxoSmithKline, UK.

    As defined by the European Medicines Agency (EMA), extrapolation refers to the extension of information and conclusions available from studies in a source population to make inferences in a target population, in order to reduce the amount of additional information needed to reach conclusions for the latter.  Bayesian inference seems to provide a natural framework to implement the extrapolation principle, as the information from the source population can be used as the prior beliefs for the target population. However, intrinsic to extrapolation principle, there is also the belief that the source and target populations, although similar enough to allow one of them to inform the other, are not exactly the same and important differences, not known a priori, might exist. Therefore, along with informative priors, we also need to incorporate a certain degree of scepticism. This could be achieved by the use of mixture priors.  Although mixture priors have been already proposed in different extrapolation contexts (bridging studies, historical controls, paediatric extrapolation), we identify some gaps in the research conducted and reported so far. In this presentation, we intend to further explore and better understand the potential of mixture priors to provide a quantitative framework for extrapolation. First we present the mixture prior model with special emphasis on the interpretation and type of inference that it allows, providing a connection with Bayesian model averaging. We then address some of the challenges that arise when constructing a mixture prior, including the choices to be made for each of the components of the model, as well as technical aspects of the estimation and computation. Finally, we discuss the frequentist operating characteristics of this approach and identify the trade-offs that come with the flexibility and robustness of the mixture priors. 

    Lisa V Hampson, Franz Koenig

    Use of frequentist and Bayesian approaches for extrapolating from adult efficacy data to design and interpret confirmatory trials in children

    New medicines for children should be subject to rigorous examination whilst taking steps to avoid unnecessary experimentation. Extrapolating from adult data can reduce uncertainty about a drug’s effects in younger patients meaning smaller trials may suffice.

    We consider how to design a confirmatory trial in children intended to compare the efficacy of a new drug, E, against control. Assuming that conduct of this trial is conditional on having demonstrated a significant beneficial effect in adults, we adopt a Bayesian approach to incorporate these adult data into the design and analysis of the paediatric trial. At each stage, inferences are made using all available data to update a Bayesian mixture model for prior opinion on the degree of similarities between adults and children. Using this framework, we propose designs for the paediatric trial which are specified by calibrating the sample size and final decision rule to: a) achieve a high frequentist power and high minimum (or average) Bayesian positive predictive value of a significant result in children; or b) ensure that a final decision to adopt (abandon) drug E in children is always associated with a minimum positive (negative) predictive value. Operating characteristics of our Bayesian designs are evaluated and compared with those of a recently proposed hybrid approach (Hlavin et al. Statistics in Medicine 2016; 35: 2117) where the sample size and significance level of a frequentist confirmatory trial in children are set to achieve a high frequentist power and high average positive predictive value of a significant result in children.

    Nicky Best, Dawn Edwards

    A case study using Bayesian methods to leverage existing clinical efficacy data in paediatric trials

    Recently there has been increased regulatory interest in partial extrapolation of adult efficacy information to paediatrics populations to reduce data collection requirements in children.  In this talk we will present a case study describing plans to use partial extrapolation of adult efficacy data from a phase III trial of an experimental drug in adolescents with a respiratory disease.  We will demonstrate how adult data on the treatment difference for the endpoint of interest can be included via an informative prior distribution to increase the probability of success of the study in adolescents and the precision of the estimated treatment difference. A method which incorporates dynamic borrowing will be used to define the level of extrapolation using a 2-step approach whereby information from the adult data is first incorporated into a prior distribution before being integrated with the data from the adolescent population. We propose a 3-component weighted robust mixture prior with the informative components based on (1) the adult efficacy data, (2) rescaled adult efficacy data to reflect the expected response for the adolescent population, and (3) a flat component to ensure that, in the event the adolescent and adult data are in clear conflict, the latter will have minimal influence on the posterior distribution of the treatment difference, thus also preventing excessive inflation of type 1 error. We will present results of a simulation study investigating operating characteristics for different choices of success criteria and prior weights.

    Peter Milligan

    Utilizing a Quantitative Framework to support Extrapolation



    Registration on or before 22nd October
    PSI Member  £120 (plus VAT)
    Non-Member  £160 (plus VAT)
    Academic   £60 (plus VAT) 
    Registration after 22nd October
    PSI Member  £160 (plus VAT)
    Non-Member  £220 (plus VAT)
    Academic  £90 (plus VAT)

    Please click here to register.
  • European Statistical Meeting: Latest Trends in Health Technology Assessments

    Two Pancras Square | Dates: 28 Nov, 2017

    This 1-day scientific meeting will provide an update on latest trends in HTA, including: the Real-World Evidence Navigator tool created by the IMI GetReal project; the EUnetHTA Joint Action 3 initiative and methodology being researched; introduction to value-based frameworks and estimands in HTA.  Patient perspectives in HTA will be discussed including how to involve patients in HTA and latest methods in patient reported outcomes.  HTA related methodological considerations will be highlighted including approaches to handle treatment switching in HTA.  Industry HTA case studies will also be presented. 

    Speakers include well known representatives from academia, European regulatory bodies and industry. The day will end with a panel discussion.

    To download the flyer please click here.

    Outline of the Agenda

     09:30 Welcome
    Chrissie Fletcher (Amgen, Chair HTA SIG), William Malbecq (MSD)
     09:50 Session: Trends in HTA(1)

    Introducing the RWE Navigator - Heather Stegenga (NICE)

    EUnetHTA Joint Action 3 activities - to be confirmed

     11:10 Coffee Break
     11:30 Session: HTA research and methods

    Adjusting for treatment switching in randomised controlled trials - Nick Latimer (University of Sheffield)

    Accumulated Industry experience in bridging regulatory and HTA research methodologies - William Malbecq, Kristel Vandormael (MSD)
     12:50  Lunch
     13:30 Session: Patient perspectives in HTA

    Involving the patient in HTA - Karen Facey (Evidence Based Health Policy Consultant)

    Benefit-Risk assessments in HTA - Shahrul Mt-Isa (MSD), Susan Talbot (Amgen)
     14:50 Coffee Break
     15:10 Session: Trends in HTA(2)

    Value-based frameworks - Jan McKendrick (PRMA Consulting)

    Estimands in HTA - Jason Wang (Celgene), Chrissie Fletcher (Amgen)

     16:30 Panel Discussion
     17:00 Summary and meeting close


    Fee includes lunch & refreshments.

    Registration on or before 14th October
    PSI Member  £100 (plus VAT)
    Non-Member  £140 (plus VAT)
    Academic  £70 (plus VAT)
    Registration after 14th October
    PSI Member  £120 (plus VAT)
    Non-Member  £160 (plus VAT)
    Academic  £90 (plus VAT)

    Please click here to register.
  • A PSI Training Course on Missing Data

    Heathrow | Dates: 06 – 07 Mar, 2018
    The aim of this course is to provide participants with an understanding of missing data, its link with what is to be estimated in a study (the “estimand”), and statistical modelling approaches. The 2 day course includes workshops: participants will undertake a number of practical exercises on missing data in SAS. The course will provide participants the opportunity to gain insight into some of the more useful new methodologies for missing data, with a view to being at the service of the real scientific question of interest. Multiple imputation (MI) will be emphasised – due to this method’s flexibility.

    Attendees will require a laptop with access to SAS.

    The following topics will be covered:
    - History of research into missing data
    - Prevention of missing data and impact on study power
    - Missing Data and its relation to the estimand
    - Estimands and their models
    - Multiple imputation I: models for missing data
    - Weighting I: weighting for missing data
    - Multiple imputation II: methods for non-continuous endpoints
    - Weighting II: augmenting weighed data with model estimates
    - Composite endpoints
    - Case studies

    Course runs from:
    10:00 - 17:00 (registration from 09:00) on Day 1
    09:00 - 16:00 on Day 2


    Registration costs include lunch and refreshments. PSI are holding a limited number of hotel rooms until the 31st January 2018 which will be allocated on a first come first served basis.

    Registration BEFORE 31st January 2018 
     PSI Member  £495 plus VAT
     Non-Member  £590 plus VAT
     Registration AFTER 31st January 2018
     PSI Member  £595 plus VAT
     Non-Member  £690 plus VAT

    Please click here to register.
    Please click here to view the flyer.